Download - The increased use of mopeds and scooters · motorcycles in all Australian jurisdictions. While an extensive body of knowledge exists on motorcycle safety, some of which is relevant

The increased popularity of mopeds and motor scooters:

Exploring usage patterns and safety outcomes

Ross Alexander Blackman

Bachelor of Social Science (Hons)

A thesis submitted as fulfilment for the Degree of Doctor of Philosophy

Centre for Accident Research & Road Safety – Queensland

School of Psychology and Counselling

Queensland University of Technology

2012

The increased popularity of mopeds and motor scooters i

Keywords

Moped; scooter; motorcycle; powered two-wheeler; motorised two-

wheeler; road safety; rider licence; vulnerable road user; mode choice;

Australia.

The increased popularity of mopeds and motor scooters ii

The increased popularity of mopeds and motor scooters iii

Abstract

Increased use of powered two-wheelers (PTWs) often underlies increases in

the number of reported crashes, promoting research into PTW safety. PTW riders

are overrepresented in crash and injury statistics relative to exposure and, as such, are

considered vulnerable road users. PTW use has increased substantially over the last

decade in many developed countries. One such country is Australia, where moped

and scooter use has increased at a faster rate than motorcycle use in recent years.

Increased moped use is particularly evident in the State of Queensland which is one

of four Australian jurisdictions where moped riding is permitted for car licence

holders and a motorcycle licence is not required.

A moped is commonly a small motor scooter and is limited to a maximum

design speed of 50 km/h and a maximum engine cylinder capacity of 50 cubic

centimetres. Scooters exceeding either of these specifications are classed as

motorcycles in all Australian jurisdictions.

While an extensive body of knowledge exists on motorcycle safety, some of

which is relevant to moped and scooter safety, the latter PTW types have received

comparatively little focused research attention. Much of the research on moped

safety to date has been conducted in Europe where they have been popular since the

mid 20th

century, while some studies have also been conducted in the United States.

This research is of limited relevance to Australia due to socio-cultural, economic,

regulatory and environmental differences. Moreover, while some studies have

compared motorcycles to mopeds in terms of safety, no research to date has

specifically examined the differences and similarities between mopeds and larger

scooters, or between larger scooters and motorcycles.

To address the need for a better understanding of moped and scooter use and

safety, the current program of research involved three complementary studies

designed to achieve the following aims: (1) develop better knowledge and

understanding of moped and scooter usage trends and patterns; and (2) determine the

factors leading to differences in moped, scooter and motorcycle safety.

Study 1 involved six-monthly observations of PTW types in inner city

parking areas of Queensland’s capital city, Brisbane, to monitor and quantify the

types of PTW in use over a two year period. Study 2 involved an analysis of

Queensland PTW crash and registration data, primarily comparing the police-

The increased popularity of mopeds and motor scooters iv

reported crash involvement of mopeds, scooters and motorcycles over a five year

period (N = 7,347). Study 3 employed both qualitative and quantitative methods to

examine moped and scooter usage in two components: (a) four focus group

discussions with Brisbane-based Queensland moped and scooter riders (N = 23); and

(b) a state-wide survey of Queensland moped and scooter riders (N = 192).

Study 1 found that of the PTW types parked in inner city Brisbane over the

study period (N = 2,642), more than one third (36.1%) were mopeds or larger

scooters. The number of PTWs observed increased at each six-monthly phase, but

there were no significant changes in the proportions of PTW types observed across

study phases. There were no significant differences in the proportions or numbers of

PTW type observed by season.

Study 2 revealed some important differences between mopeds, scooters and

motorcycles in terms of safety and usage through analysis of crash and registration

data. All Queensland PTW registrations doubled between 2001 and 2009, but there

was an almost fifteen-fold increase in moped registrations. Mopeds subsequently

increased as a proportion of Queensland registered PTWs from 1.2 percent to 8.8

percent over this nine year period. Moped and scooter crashes increased at a faster

rate than motorcycle crashes over the five year study period from July 2003 to June

2008, reflecting their relatively greater increased usage. Crash rates per 10,000

registrations for the study period were only slightly higher for mopeds (133.4) than

for motorcycles and scooters combined (124.8), but estimated crash rates per million

vehicle kilometres travelled were higher for mopeds (6.3) than motorcycles and

scooters (1.7). While the number of crashes increased for each PTW type over the

study period, the rate of crashes per 10,000 registrations declined by 40 percent for

mopeds compared with 22 percent for motorcycles and scooters combined.

Moped and scooter crashes were generally less severe than motorcycle

crashes and this was related to the particular crash characteristics of the PTW types

rather than to the PTW types themselves. Compared to motorcycle and moped

crashes, scooter crashes were less likely to be single vehicle crashes, to involve a

speeding or impaired rider, to involve poor road conditions, or to be attributed to

rider error. Scooter and moped crashes were more likely than motorcycle crashes to

occur on weekdays, in lower speed zones and at intersections. Scooter riders were

older on average (39) than moped (32) and motorcycle (35) riders, while moped

riders were more likely to be female (36%) than scooter (22%) or motorcycle riders

The increased popularity of mopeds and motor scooters v

(7%). The licence characteristics of scooter and motorcycle riders were similar, with

moped riders more likely to be licensed outside of Queensland and less likely to hold

a full or open licence. The PTW type could not be identified in 15 percent of all

cases, indicating a need for more complete recording of vehicle details in the

registration data.

The focus groups in Study 3a and the survey in Study 3b suggested that

moped and scooter riders are a heterogeneous population in terms of demographic

characteristics, riding experience, and knowledge and attitudes regarding safety and

risk. The self-reported crash involvement of Study 3b respondents suggests that

most moped and scooter crashes result in no injury or minor injury and are not

reported to police. Study 3 provided some explanation for differences observed in

Study 2 between mopeds and scooters in terms of crash involvement. On the whole,

scooter riders were older, more experienced, more likely to have undertaken rider

training and to value rider training programs. Scooter riders were also more likely to

use protective clothing and to seek out safety-related information.

This research has some important practical implications regarding moped and

scooter use and safety. While mopeds and scooters are generally similar in terms of

usage, and their usage has increased, scooter riders appear to be safer than moped

riders due to some combination of superior skills and safer riding behaviour. It is

reasonable to expect that mopeds and scooters will remain popular in Queensland in

future and that their usage may further increase, along with that of motorcycles.

Future policy and planning should consider potential options for encouraging moped

riders to acquire better riding skills and greater safety awareness. While rider

training and licensing appears an obvious potential countermeasure, the effectiveness

of rider training has not been established and other options should also be strongly

considered. Such options might include rider education and safety promotion, while

interventions could also target other road users and urban infrastructure.

Future research is warranted in regard to moped and scooter safety,

particularly where the use of those PTWs has increased substantially from low levels.

Research could address areas such as rider training and licensing (including program

evaluations), the need for more detailed and reliable data (particularly crash and

exposure data), protective clothing use, risks associated with lane splitting and

filtering, and tourist use of mopeds. Some of this research would likely be relevant

to motorcycle use and safety, as well as that of mopeds and scooters.

The increased popularity of mopeds and motor scooters vi

The increased popularity of mopeds and motor scooters vii

Table of Contents

CHAPTER 1: INTRODUCTION ............................................................................. 1

1.1 Background .................................................................................................... 1

1.2 Powered two-wheeler (PTW) definitions ...................................................... 3

1.3 Rider licensing and training ........................................................................... 5

1.4 Demarcation of scope .................................................................................... 6

1.5 An outline of the thesis .................................................................................. 8

1.6 Chapter One summary ................................................................................. 10

CHAPTER 2: AN OVERVIEW OF MOPED AND SCOOTER USE AND

SAFETY .................................................................................................................... 13

2.1 Introduction .................................................................................................. 13

2.1.1 Literature search methods ................................................................. 13

2.1.2 Background ....................................................................................... 14

2.2 Moped, scooter and motorcycle usage ......................................................... 17

2.2.1 Trends in sales and registration ........................................................ 18

2.2.2 Development and expansion of PTW types ..................................... 22

2.2.3 Patterns of moped and scooter use ................................................... 23

2.2.4 Demographic characteristics of riders .............................................. 25

2.2.5 Motivations and reasons for PTW use .............................................. 30

2.2.6 Traffic congestion, fuel consumption and emissions ....................... 31

2.3 Motorcycle, moped and scooter (PTW) safety ............................................ 34

2.3.1 Crash rates and crash severity .......................................................... 34

2.3.2 Risk factors ....................................................................................... 39

2.3.3 Approaches to understanding PTW rider risk .................................. 52

2.3.4 Potential of licensing and training to improve rider safety .............. 55

2.4 Chapter Two summary ................................................................................. 59

CHAPTER 3: RESEARCH DESIGN .................................................................... 61

3.1 Introduction .................................................................................................. 61

3.2 Research aims .............................................................................................. 62

3.3 Research location ......................................................................................... 63

3.4 Research questions ....................................................................................... 63

3.5 Research studies ........................................................................................... 66

3.5.1 Study 1: Observation of powered two-wheeler types in an inner

city area ............................................................................................. 67

3.5.2 Study 2: Analysis of Queensland crash and registration data .......... 68

3.5.3 Study 3: Exploration of moped and scooter rider characteristics ..... 69

3.6 Chapter Three summary ............................................................................... 71

CHAPTER 4: STUDY 1 – AN OBSERVATION OF POWERED TWO-

WHEELER TYPES IN AN INNER CITY AREA ................................................ 73

4.1 Introduction and rationale ............................................................................ 73

The increased popularity of mopeds and motor scooters viii

4.2 Study design and methods ............................................................................ 74

4.3 Results .......................................................................................................... 78

4.3.1 PTW’s observed over the study period ............................................. 78

4.3.2 PTW’s observed by geographic location .......................................... 81

4.4 Discussion .................................................................................................... 84

4.5 Limitations ................................................................................................... 87

4.6 Chapter Four summary ................................................................................. 88

CHAPTER 5: STUDY 2 – ANALYSIS OF CRASH AND REGISTRATION

DATA ......................................................................................................................... 89

5.1 Introduction .................................................................................................. 89

5.1.1 Research questions ............................................................................ 90

5.2 Study design and methods ............................................................................ 91

5.2.1 Acquisition of registration data ........................................................ 91

5.2.2 Acquisition, cleaning and coding of crash data ................................ 92

5.2.3 Data analysis ..................................................................................... 95

5.2.4 Mapping crash data ........................................................................... 98

5.3 PTWs registered in Queensland ................................................................. 100

5.4 Results ........................................................................................................ 101

5.4.1 Trends in PTW crashes ................................................................... 101

5.4.2 Crash characteristics ....................................................................... 105

5.4.3 PTW controlling characteristics ..................................................... 123

5.4.4 Fault attribution and contributing circumstances ........................... 127

5.5 Discussion .................................................................................................. 139

5.5.1 Patterns of usage as indicated by crash data ................................... 139

5.5.2 Crash rates and related characteristics ............................................ 141

5.5.3 Main contributors to crash and injury risk ...................................... 145

5.5.4 Research questions .......................................................................... 154

5.6 Limitations of Study 2 ................................................................................ 157

5.7 Chapter Five summary ............................................................................... 159

CHAPTER 6: STUDY 3A – FOCUS GROUPS WITH MOPED AND

SCOOTER RIDERS .............................................................................................. 161

6.1 Introduction ................................................................................................ 161

6.2 Methods ...................................................................................................... 162

6.2.1 Setting ............................................................................................. 162

6.2.2 Selection criteria, recruitment and participation ............................. 163

6.2.3 Data collection ................................................................................ 164

6.2.4 Guiding questions ........................................................................... 164

6.2.5 Analysis .......................................................................................... 165

6.3 Results ........................................................................................................ 166

6.3.1 Participation .................................................................................... 166

6.3.2 Topics on moped and scooter usage ............................................... 167

6.3.3 Topics concerning safety ................................................................ 170

The increased popularity of mopeds and motor scooters ix

6.4 Discussion .................................................................................................. 178

6.4.1 PTW usage ...................................................................................... 179


6.4.3 Research questions ......................................................................... 187

6.4.4 Limitations of Study 3a .................................................................. 189

6.5 Chapter Six summary ................................................................................. 189

CHAPTER 7: STUDY 3B – QUEENSLAND SCOOTER AND MOPED RIDER

SURVEY ................................................................................................................. 191

7.1 Introduction ................................................................................................ 191

7.2 Study design and methods ......................................................................... 192

7.2.1 Survey content and delivery ........................................................... 192

7.2.2 Recruitment and participation ........................................................ 193

7.2.3 Data processing and analysis .......................................................... 195

7.3 Results ........................................................................................................ 196

7.3.1 Characteristics of respondents ........................................................ 197

7.3.2 Moped and scooter characteristics .................................................. 203

7.3.3 Travel patterns ................................................................................ 204

7.3.4 Trip purpose and motivations for riding ......................................... 206

7.3.5 Risk perception and risk management ............................................ 209

7.3.6 Crash involvement .......................................................................... 215

7.3.7 Comments on moped and scooter safety and transport planning ... 221

7.4 Discussion .................................................................................................. 222

7.4.1 PTW usage ...................................................................................... 223

7.4.2 Crash involvement .......................................................................... 224


7.4.4 Research questions ......................................................................... 233

7.4.5 Limitations ...................................................................................... 236

7.5 Chapter Seven summary ............................................................................ 237

CHAPTER 8: DISCUSSION ................................................................................ 239

8.1 Introduction ................................................................................................ 239

8.2 Review of findings ..................................................................................... 240

8.2.1 RQ1: Why has moped and scooter usage increased? ..................... 240

8.2.2 RQ2: How does the usage of mopeds, scooters and motorcycles

differ? .............................................................................................. 243

8.2.3 RQ3: How does the safety of mopeds, scooters and motorcycles

differ? .............................................................................................. 246

8.2.4 RQ4: Why does the safety of mopeds, scooters and motorcycles

differ? .............................................................................................. 249

8.3 Implications of the research ....................................................................... 253

8.3.1 Implications for policy and planning .............................................. 255

8.4 Strengths and limitations of the research ................................................... 257

8.5 Potential topics for further research ........................................................... 258

The increased popularity of mopeds and motor scooters x

8.6 Concluding remarks ................................................................................... 259

REFERENCES ....................................................................................................... 262

APPENDICES ........................................................................................................ 274

The increased popularity of mopeds and motor scooters xi

List of Figures

Fig. 1.1 Conceptual categorisation of motorcycles, mopeds and scooters .............. 4

Fig. 4.1 Aggregate PTW type distribution across Brisbane city parking areas ..... 83

Fig. 7.1 Flyer used for recruitment of survey participants ................................... 194

The increased popularity of mopeds and motor scooters xii

The increased popularity of mopeds and motor scooters xiii

List of Tables

Table 1.1 Licensing requirements for moped riding in Australia .......................... 6

Table 2.1 Mopeds and motorcycles per 1,000 inhabitants in European

countries ................................................................................................ 18

Table 3.1 Relevance of the studies to the research aims and questions ................ 67

Table 4.1 Parking areas included in observational study of PTW use .................. 76

Table 4.2 PTW type classification ........................................................................ 78

Table 4.3 Mopeds, scooters and motorcycles observed in Brisbane CBD by

data collection phase ............................................................................. 79

Table 4.4 PTW subcategory information for all PTW’s observed ....................... 80

Table 4.5 PTW’s observed in Brisbane CBD by type and location

(aggregate) ............................................................................................ 82

Table 5.1 Reclassification of PTW types using vehicle make and model

details .................................................................................................... 93

Table 5.2 Grouping of contributing circumstances into like categories ............... 98

Table 5.3 Queensland PTW registrations by type and year, June 2001-June

2009 .................................................................................................... 100

Table 5.4 Queensland PTW crashes by type and year, July 2003-June 2009 .... 102

Table 5.5 PTW crashes involving other PTWs ................................................... 103

Table 5.6 Crashes per 10,000 QLD registrations by financial year and

registration (ADR) category ............................................................... 104

Table 5.7 Crash rates per VKT for 5 years (using data from Harrison and

Christie, 2006) .................................................................................... 105

Table 5.8 Injured road user type by PTW type for reported crashes, July

2003-June 2008 (including fatally injured) ........................................ 105

Table 5.9 Crash severity by PTW type for report crashes, July 2003-June

2008 .................................................................................................... 106

Table 5.10 Crashes per 10,000 registration years by crash severity level and

registration category ........................................................................... 107

Table 5.11 Location characteristics of moped, scooter and motorcycle

crashes ................................................................................................. 109

Table 5.12 Moped crashes by LGA and place licensed, July 2003-June 2008 .... 111

Table 5.13 Day of week for moped, scooter and motorcycle crashes .................. 112

Table 5.14 Time of day for moped, scooter and motorcycle crashes ................... 113

The increased popularity of mopeds and motor scooters xiv

Table 5.15 Roadway characteristics of moped, scooter and motorcycle

crashes ................................................................................................. 115

Table 5.16 Number of units involved in moped, scooter and motorcycle

crashes ................................................................................................. 116

Table 5.17 Number of units involved in crashes by crash severity ...................... 117

Table 5.18 Road user types involved in multi-unit crashes with a PTW .............. 117

Table 5.19 Parameter coefficient estimates of ordered probit model of

severity by PTW type .......................................................................... 119

Table 5.20 Crash configuration of moped, scooter and motorcycle crashes ........ 120

Table 5.21 Crash group description of moped, scooter and motorcycle

crashes ................................................................................................. 121

Table 5.22 Breakdown of crash description – ‘Same direction’ ........................... 122

Table 5.23 Breakdown of crash description – ‘Off path on straight’.................... 122

Table 5.24 Breakdown of crash description – ‘Opposite approach’ ..................... 123

Table 5.25 Breakdown of crash description – ‘Manoeuvring’ ............................. 123

Table 5.26 Age and gender characteristics of PTW riders in crashes ................... 125

Table 5.27 Moped rider age by gender crosstabulation ........................................ 125

Table 5.28 Licence characteristics of PTW controllers in crashes ....................... 127

Table 5.29 Attribution of Unit 1 (most at fault) .................................................... 128

Table 5.30 Proportions of crash configuration with PTW designated Unit 1

(multi-unit crashes) ............................................................................. 129

Table 5.31 Proportions of crash group descriptions with PTW designated

Unit 1 (multi-unit crashes) .................................................................. 130

Table 5.32 Number of contributing circumstances attributed to all PTWs........... 131

Table 5.33 Contributing circumstances (CCs) attributed to a Unit 1 PTW .......... 131

Table 5.34 Contributing circumstances attributed to a PTW (all crashes) ........... 132

Table 5.35 Contributing circumstances attributed to a PTW (multi-unit

crashes, PTW Unit 1) .......................................................................... 133

Table 5.36 Contributing circumstances (single vehicle crashes) .......................... 133

Table 5.37 Number of contributing circumstances attributed to other road

users in multi-unit crashes (whether Unit 1 or not) ............................ 134

Table 5.38 Number of contributing circumstances attributed to other (Unit 1)

road users in multi-unit crashes (excluding animals, and other

PTWs) ................................................................................................. 134

The increased popularity of mopeds and motor scooters xv

Table 5.39 Contributing circumstances attributed to other (Unit 1) road users

in multi-unit crashes (excluding crashes with animals and other

PTWs) ................................................................................................. 135

Table 5.40 Main circumstances attributed to other (Unit 1) road users in

multi-unit crashes (excluding crashes with animals and other

PTWs) ................................................................................................. 135

Table 5.41 Entire logistic regression predicting odds of PTW being Unit 1 ........ 137

Table 5.42 Logistic regression analysis predicting odds of moped being Unit

1 .......................................................................................................... 138

Table 7.1 Age and gender of respondents ........................................................... 197

Table 7.2 Place of residence of respondents ....................................................... 198

Table 7.3 Weekly individual income and employment status of respondents .... 199

Table 7.4 General demographic characteristics of respondents .......................... 200

Table 7.5 Licence characteristics of respondents ............................................... 202

Table 7.6 Riding experience* and training undertaken ...................................... 203

Table 7.7 Riding frequency ................................................................................. 204

Table 7.8 Distance travelled by moped and scooter riders ................................. 205

Table 7.9 Proportion (mean %) of riding by speed zone and

weekday/weekend ............................................................................... 206

Table 7.10 Riding purpose as a mean proportion of usage ................................... 206

Table 7.11 Importance of factors influencing moped and scooter use

generally .............................................................................................. 207

Table 7.12 Importance of factors influencing moped and scooter use for a

particular journey ................................................................................ 208

Table 7.13 Self rated level of riding skill ............................................................. 209

Table 7.14 Perceived risk associated with riding scenarios .................................. 210

Table 7.15 Accessing safety-related information and resources .......................... 211

Table 7.16 Frequency of use of upper body clothing items while riding ............. 212

Table 7.17 Frequency of use of lower body clothing items while riding ............. 213

Table 7.18 Rating of factors influencing choice of clothing ................................ 214

Table 7.19 Response to introduction of a PTW licence for moped riders ............ 215

Table 7.20 Self-reported crash involvement and injury severity .......................... 216

Table 7.21 Vehicle involvement, vehicle damage and police attendance ............ 217

Table 7.22 Road characteristics in self-reported crashes ...................................... 217

The increased popularity of mopeds and motor scooters xvi

Table 7.23 Temporal characteristics in self-reported crashes ............................... 218

Table 7.24 Self-reported crash description (coded) .............................................. 218

Table 7.25 Licence characteristics of crash-involved riders ................................. 219

Table 7.26 Age, gender and training involvement of crash-involved riders ......... 220

Table 7.27 Self-reported crash rates per million vehicle kilometres travelled ..... 221

Table 7.28 Police-reported crash rates per million vehicle kilometres

travelled ............................................................................................... 221

The increased popularity of mopeds and motor scooters xvii

List of Abbreviations

Abbreviation/Symbol Definition

ABS Australian Bureau of Statistics

ADR Australian Design Rule

ACEM Association of European Motorcycle Manufacturers

BCC Brisbane City Council

CARRS-Q Centre for Accident Research and Road Safety – Queensland

ERSO European Road Safety Observatory

EU European Union

FCAI Federal Chamber of Automotive Industries

FEMA Federation of European Motorcyclists’ Associations

LGA Local Government Area

MAIDS Motorcycle Accident In-Depth Study

MSF Motorcycle Safety Foundation (US)

NHTSA National Highway Traffic Safety Administration (US)

PTW Powered two-wheeler

SLA Statistical Local Area

SMIDSY Sorry mate I didn’t see you

SWOV Dutch Institute for Road Safety

TMR Transport and Main Roads, Queensland Department of

US United States (of America)

VKT Vehicle kilometres travelled

VRU Vulnerable road user

The increased popularity of mopeds and motor scooters xviii

The increased popularity of mopeds and motor scooters xix

Statement of original authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To the

best of my knowledge and belief, the thesis contains no material previously

published or written by another person except where due reference is made.

Signature: ____________________

Date: 30th

March, 2012

The increased popularity of mopeds and motor scooters xx

The increased popularity of mopeds and motor scooters xxi

Acknowledgements

Firstly I would like to thank my Principal Supervisor, Professor Narelle

Haworth, for her unwavering guidance and support, for the wealth of knowledge she

has contributed, and for keeping me on track, more or less, throughout the lengthy

and often difficult research process. I am also grateful to Narelle for keeping me

employed on other research projects, which has enabled me to eat more than just

instant noodles over the gruelling nine or so months since my final seminar.

Thanks also to my Associate Supervisors, Emeritus Professor Mary Sheehan

and Associate Professor Jonathan Bunker, for providing input where required and for

general encouragement and positive remarks. Particular thanks go to Mary for

recognising me as potential PhD candidate even before I had seriously considered the

possibility myself. I am also grateful to the other members of my Final Seminar

panel for their suggestions and (warranted) criticisms: Professor Barry Watson,

Professor Simon Washington and Dr Nerida Leal.

The assistance of many departments and organisations is greatly appreciated.

In particular, the Queensland Department of Transport and Main Roads (TMR)

provided data that were fundamental to the overall research. Queensland University

of Technology (QUT), the Centre for Accident Research and Road Safety –

Queensland (CARRS-Q), the Institute of Health and Biomedical Innovation (IHBI)

and the Motor Accident Insurance Commission (MAIC) provided essential funding

in the form of scholarships, top-ups and other financial assistance, among other

things. Staff at the Federal Chamber of Automotive Industries (FCAI) were also

helpful in providing information in a timely manner when requested.

Sincere thanks go to focus group and survey participants, and the moderators

of Scooteroo and Scooter Community online forums for their assistance in promoting

the rider survey. Special thanks also to Joe D’Ercole and Ben Silver of Scooters

Scooters and Scootopia for the chance to experience first-hand what all the fuss is

about – I still can’t decide between the Fuoco and the Vespa GTS300.

Assistance with data collection, analysis and thesis formatting was greatly

appreciated and many thanks go to Angela Watson, Hollie Wilson, Pete Rowden,

Adjunct Professor Vic Siskind, Veronica Baldwin and Md. Mazharul (Shimul)

Haque. The opportunity to visit and present my research to road safety researchers at

the University of Pavia, Italy, and the University of California, Berkeley, provided an

The increased popularity of mopeds and motor scooters xxii

exciting and educational experience. Particular thanks go to Dr Anna Morandi,

Chiara Orsi (Italy) and Professor Simon Washington (formerly Berkeley).

To my family in Melbourne and Sydney, thanks for your encouragement and

support and for generally having faith that I could achieve this goal. Hopefully we

can see more of each other in future. To my friends at CARRS-Q and elsewhere who

are too numerous to name, including members of the famous Watson et al. ensemble,

thanks for reminding me that there is always more to life than writing a thesis.

Finally, to my wonderful partner Hollie, thank you for being Amy

Winehouse, and for not really being Amy Winehouse. More seriously though, you

have given me more than a little extra incentive to finally finish writing this thesis!

You have also endured some periods of me at my grumpy best, which must take

some effort. At the end of one journey begins another, and I can’t thank you enough

for embarking on that with me.

The increased popularity of mopeds and motor scooters

1

CHAPTER 1: INTRODUCTION

1.1 Background

Riders of powered two-wheelers (PTWs), including motorcycle, moped and

scooter riders, are at substantially greater risk of death and injury from road crashes

than car and other vehicle occupants. In the last three years, PTW riders comprised

13, 16 and 21 percent of traffic fatalities in the United States (US), Australia and the

United Kingdom (UK) respectively, where PTWs account for less than five percent

of registered vehicles (BITRE, 2010; Department for Transport, 2010; NHTSA,

2010). A similar situation exists in many European countries, although statistics and

PTW usage are highly variable across the region (World Health Organization, 2009).

Estimates of relative risk per distance travelled show that PTW riders are about 30

times more likely to die in a crash than car occupants in Australia, and 41 times more

likely to be seriously injured (Johnston, Brooks, & Savage, 2008). Similarly high

risk levels are reported in other developed countries. As such, PTW riders are often

referred to as vulnerable road users (VRUs), along with pedestrians and cyclists

(Constant & Lagarde, 2010; Naci, Chisholm, & Baker, 2009; SWOV, 2006a). The

vulnerability of PTW riders stems from a combination of relatively high crash risk

and a lack of protection from collision impacts.

Inherent PTW design characteristics arguably contribute to a higher crash risk

of PTWs compared to other motorised vehicles. Mopeds, scooters and motorcycles

share common characteristics of single-track vehicles which make them more

difficult to control than two-track vehicles such as cars. Compared with cars and

other four-wheeled vehicles, PTWs are inherently unstable, have minimal contact

with road surfaces and, subsequently, longer braking distances. Due to their

relatively high power to weight ratios, PTWs are often also capable of accelerating

more rapidly than other vehicles, while their small frontal area makes them difficult

for other road users to see. All PTW riders are largely unprotected from collision

impacts as they are not enclosed within the structure of a vehicle body, nor further

protected by passive safety devices such as airbags and seatbelts. PTW riders are

therefore more likely than other vehicle occupants to be seriously injured in the event

of a crash. Additionally, all PTW riders are potentially exposed to adverse weather


2

conditions such as rain, wind, heat and cold, the various effects of which may

increase crash risk.

The PTW characteristics outlined above are among the factors which

contribute to the greater vulnerability of PTW riders compared to other vehicle

occupants. However, examining PTW types collectively in terms of safety does not

identify or address the potential differences in safety of moped, scooter and

motorcycle riding. These differences may relate to particular vehicle design and

performance characteristics, but also to the characteristics and motivations of riders,

their preferences for particular PTW types, and the regulations that govern their use.

As the vulnerability and disproportionate involvement of PTW riders in

crashes has long been recognised, there is a considerable body of research literature

on PTW safety and usage in developed countries. However, there are several gaps in

knowledge which provide the rationale for the current program of research. First,

most research into PTW safety has focused primarily on motorcycles and has not

addressed moped and scooter use in depth. Moreover, despite some comparison of

mopeds to motorcycles, there has been no comprehensive comparison of mopeds to

larger scooters in regard to safety and usage. Second, most of the research that has

focused on moped and scooter safety originates from Europe where these vehicles

have been traditionally popular for many decades. While some of this research is

transferrable to the Australian context, socioeconomic, cultural, legislative and

environmental differences between Australia and elsewhere mean that there are

limits regarding the transferability of findings (Naci, Chisholm et al., 2009). Third,

there are also differences between Australian jurisdictions in terms of regulations and

environment which impact the use of mopeds and scooters. The situation in

Queensland is therefore likely to differ from that in other Australian jurisdictions

including Victoria and New South Wales (Australia’s two most populous States), as

well as from that in other countries, where moped and scooter safety research has

been conducted.

As mopeds and scooters have historically comprised a very small proportion

of road traffic in Australia, until recently they have not been a major concern in

safety research. However, recent increases in moped and scooter use in Australia,

and Queensland in particular, have generated greater interest in their safety compared

to motorcycles, and also compared to each other. Although mopeds and scooters still


3

comprise less than one percent of registered vehicles in Australia, there has been an

almost 15-fold increase in Queensland moped registrations over the last 10 years. As

might be expected in light of this increased usage, a notable increase in reported

moped crashes has also been observed over this period.

1.2 Powered two-wheeler (PTW) definitions

Some research conceptualises mopeds and scooters as a sub-category of

motorcycles and, subsequently, moped and scooter riders as a sub-population of

motorcyclists (Tunnicliff, 2006). By contrast, the approach taken in the current

research was to define mopeds, scooters and motorcycles each as a sub-category of

powered two-wheelers (PTWs).

In most cases scooters and mopeds share typical features such as a step-

through chassis and automatic transmission. Among users, mainstream media and

some industry sources, the term scooter is often used to refer to both mopeds and

larger capacity scooters. However, the relevant Australian Design Rules (ADR)

dictates that a scooter may be either a MOPED (LA) or MOTORCYCLE (LC),

depending on its engine cylinder capacity and/or maximum speed (see Figure 1.1).

The ADRs provide the following definitions of the PTW types that are in common

use on public roads and that are referred to throughout this thesis (Australian

Government, 2008):

MOPED - 2 Wheels (LA category)

A 2-wheeled motor vehicle, not being a power-assisted pedal cycle,

with an engine cylinder capacity not exceeding 50 ml (cc) and a

‘Maximum Motor Cycle Speed’ not exceeding 50 km/h; or a 2-

wheeled motor vehicle with a power source other than a piston engine

and a ‘Maximum Motor Cycle Speed’ not exceeding 50 km/h.

MOTOR CYCLE (LC category)

A 2-wheeled motor vehicle with an engine cylinder capacity exceeding

50 ml (cc) or a ‘Maximum Motor Cycle Speed’ exceeding 50 km/h.


4

Figure 1.1. Conceptual categorisation of motorcycles, mopeds and scooters1

These ADR definitions correspond closely with the European classifications

L1 and L3 for moped and motorcycle respectively, although a European L1 vehicle

may also be pedal-assisted and may also be defined a mofa which is limited to 25

km/h (ACEM, 2008a). While the proposed research focuses on two-wheeled

vehicles, additional ADR categories (LE and LB) which may be encountered during

data collection and therefore discussed cover three-wheeled motorcycles, scooters

and mopeds:

LE – MOTOR TRICYCLE

A motor vehicle with 3 wheels symmetrically arranged in relation to

the longitudinal median axis, with a ‘Gross Vehicle Mass’ not

exceeding 1.0 tonne and either an engine cylinder capacity exceeding

50 ml (cc) or a ‘Maximum Motor Cycle Speed’ exceeding 50 km/h..

LB – MOPED – 3 Wheels

A 3-wheeled motor vehicle, not being a power-assisted pedal cycle,

with an engine cylinder capacity not exceeding 50 ml (cc) and a

‘Maximum Motor Cycle Speed’ not exceeding 50 km/h; or a 3-

wheeled motor vehicle with a power source other than a piston engine

and a ‘Maximum Motor Cycle Speed’ not exceeding 50 km/h.

1 Photographs illustrate basic styles only. The scooter/moped illustrated represents a general style

available as LA Moped (up to 50cc engine) and LC Motorcycle (scooter, >50cc engine).


5

Several mopeds and scooters are currently available in Australia which are

powered by electric or hybrid electric/internal combustion engines. ADR

classification of these vehicles as either LA Moped or LC Motorcycle is generally

dependant on their maximum speed (up to 50 km/h, or above 50 km/h) rather than on

a combination of maximum speed and engine characteristics.

1.3 Rider licensing and training

Patterns of use and safety outcomes for mopeds, scooters and motorcycles are

strongly influenced by rider education, training and licensing requirements which

vary considerably across developed countries. Moped riding is permitted from 14

years of age in some European countries and many countries allow mopeds and light

motorcycles (up to 125cc engine cylinder capacity) to be ridden on a car licence.

Countries in the European Union are currently moving towards a uniform system for

moped rider licensing and training under the European Directive on driving licences.

Industry sources suggest that mopeds will be incorporated into the Directive in 2013,

though full adoption by EU member States depends on compliance of individual

jurisdictions (ACEM, 2010a). The historical situation in Europe is summarised in a

table sourced from the SWOV Institute for Road Safety Research (Schoon, 2004) in

Appendix A1.

In other jurisdictions, requirements for moped riding may include moped

rider permits or licences (sometimes obtainable at an earlier age than a car licence),

special endorsement of an existing car licence (with some training and testing

attached), or a standard motorcycle licence (usually subject to some form of

graduated licensing process). These and other regulatory differences make it

difficult to compare moped safety between jurisdictions, due to their likely influence

on usage patterns and motivations, and the age and experience of riders.

Rider licensing is a responsibility of State and Territory governments in

Australia and the licence requirements for moped and motorcycle riding vary across

these jurisdictions. In the States of New South Wales, Victoria, Tasmania and the

Australian Capital Territory, a motorcycle licence is required to ride a moped or a

larger scooter (provision for an automatic motorcycle licence exists in New South

Wales). In Queensland, South Australia, Western Australia and the Northern


6

Territory, a person may legally ride a moped if they hold a provisional or full car

driver’s licence, while scooter riders require a motorcycle licence (provision for an

automatic motorcycle licence exists in Queensland). Consequently, in these latter

jurisdictions mopeds are accessible to people with potentially no motorcycling

experience, skills, training or education. Recent proposals for improving PTW safety

in Queensland include introduction of mandatory PTW licensing for moped riders

(Queensland Transport, 2008). Such a change in legislation has the potential to alter

both the number of mopeds in use and the characteristics of moped riders. The

current situation in Australian jurisdictions is summarised below in Table 1.1.

Table 1.1 Licensing requirements for moped riding in Australia

State/

Territory

Min.

age*

Moped riding

permitted on

car licence

Automatic

permitted in

motorcycle test

Mandatory tests &

training for minimum

licence required

New South

Wales 16 years 9

months No

Yes, may ride auto

only

Theory/road rules test

Pre-learner rider training

Practical riding test

Australian

Capital

Territory

16 years 9

months No

Yes, may ride auto

only




Victoria 18 No Yes, may ride

manual motorcycle


M’cycle knowledge test


Queensland 17 Yes Yes, may ride auto

only Theory/road rules test

Practical driving test

South

Australia 17 Yes No Theory/road rules test


Western

Australia 16 (moped

learner)

Yes

(Moped licence

also available) No Theory/road rules test


Tasmania 16 years 6

months No No



Northern

Territory 16 Yes

Yes, may ride

manual motorcycle Theory/road rules test


*Including for learner motorcycle licence where applicable and moped licence (WA)

1.4 Demarcation of scope

This research focuses on gaining a better understanding of the use and safety

of mopeds, scooters and motorcycles which can be registered for use on public roads.

It excludes power-assisted bicycles, mobility scooters and other personal


7

transportation devices such as Segways. As moped and scooter use is predominantly

an on-road activity, off-road PTW use is not considered in this research.

The primary objectives of the research are to discover what differences exist

between mopeds, scooters and motorcycles in terms of safety and usage, and to

examine those differences with a view to identifying potential safety improvement.

As the research essentially seeks to better understand differences between PTW types

in regard to safety, there is no strong theoretical focus on predicting the intentions or

behaviour of riders or other road users.

The research employed both quantitative and qualitative methods to achieve

the overall objectives. Analysis of crash and registration data covered a period of

five years from July 2003 to June 2008 inclusive, this being the most recent five year

period for which complete data were available. The analysis explored trends and

differences between PTW types in usage, crash rates, crash severity and

characteristics, contributing factors and rider demographics. Identification of crash-

involved PTW types required a novel approach in which a new dataset was

developed to reveal moped, scooter and motorcycle make and model details.

Analyses included Chi Square tests for statistical significance, Cramer’s V

calculations for estimated effect size, and post-hoc analyses using an adjusted

standardised residual statistic.

An observational study conducted in the Brisbane inner city area sought to

determine the types of PTW in use, as well as changes observed over the two-year

study period commencing in August 2008. Focusing on the major hub of moped and

scooter activity in Queensland, the study was essentially a measure of the proportions

of mopeds and scooters in use relative to other PTW types in selected designated

parking areas. The study did not quantify all of the PTWs in use in that area, nor

provide a measure of exposure (distance travelled) for the different PTW types, but

provided useful and timely information on usage patterns.

Focus groups were used to explore the beliefs, attitudes and experiences of

riders relating to the use and safety of mopeds, scooters and motorcycles, with

participants including regular commuters, students and industry representatives.

This qualitative study informed development of a questionnaire survey instrument

which was used to profile Queensland moped and scooter riders’ demographic,

social, motivational and attitudinal characteristics. The survey targeted those who


8

had ridden a moped or scooter in Queensland at least monthly over a period of three

months prior to completing the survey. These two studies used self-report measures

which, while imperfect due to potential for self-selection and response bias, are

widely used and considered valid in road safety research (Tubre, Bell, Arthur,

Edwards, Tubre, & Day, 2005; Lajunen & Summala, 2003; Kaiser, Frick, & Stoll-

Kleemann, 2001).

The research is focused geographically within the State of Queensland,

Australia, for the purposes of crash and registration data analysis and for establishing

a profile of moped and scooter rider characteristics, beliefs and experiences. The

observational study of PTW use and a series of focus groups with moped and scooter

riders focused on the Brisbane inner city area, which represents the major hub of

moped and scooter activity in Queensland.

The review of the literature relevant to moped and scooter use and safety

generally focused on research from developed countries. This was due to key

differences between Australia and developing countries regarding crash rates and

characteristics, socioeconomic circumstances, legislation, levels of use and

motivations for use.

1.5 An outline of the thesis

The thesis begins with an examination of current knowledge regarding the

use and safety of PTWs, focusing in particular on moped and scooter use in

developed countries as well as on motorcycle use where appropriate. The published

research reviewed addresses differences and similarities between PTW types in terms

of crash characteristics, crash rates and severity, contributory factors, rider behaviour

and characteristics, and other crash and injury risk factors. The literature on PTW

usage, including trends in usage, patterns of use, exposure estimates, motivations for

use and rider characteristics, is also reviewed. Also discussed are safety-oriented

countermeasures including licensing and training, the role of PTWs in urban

transport systems, aspects of PTW design and performance, and theoretical

perspectives and methods guiding relevant research. The literature review provides

the foundation for the development of a series of research questions which are

addressed through three separate studies.


9

Chapter Three describes the research aims, research questions and research

design including the three studies in detail. In summary, the research aims and

questions arise from what is known and what remains to be known about moped and

scooter use and safety in the context of Queensland, Australia, according to the

literature. The research aims (RA) are as follows:

RA1. To develop better knowledge and understanding of moped and

scooter usage trends and patterns.

RA2. To determine the factors leading to differences in moped, scooter and

motorcycle safety.

Chapter Four describes Study 1, an observation of PTW types in inner city

Brisbane. Previous research indicates that this is an area of concentrated PTW use

relative to other Queensland locations. With limited exceptions, the review of the

literature in Chapter Two identifies a lack of information specific or clearly relevant

to this location on the patterns and frequency of use of various PTW types, including

mopeds and scooters. While not providing a measure of exposure by distance

travelled, the observational study provides baseline data relating to frequency of use

of different PTW types, as well as measuring trends over a two year period. The

study also examines the distribution of PTW types across different parking areas, the

locations of which may relate to motivations for moped and scooter use.

Chapter Five presents the rationale, design, methods and results of Study 2,

an analysis of Queensland PTW crash and registration data. As noted above, this

analysis covered a five-year period from July 2003 to June 2008, and necessitated a

novel approach to determine the crash-involvement of different PTW types. The

analysis primarily explored trends, differences and similarities between mopeds,

scooters and motorcycles in usage, crash rates, severity, characteristics, contributing

factors and rider demographics. Issues surrounding data quality are also explored in

this study. The introductory and discussion sections of this chapter link the results

and design of Study 2 with the findings and other relevant information contained in

the literature reviewed in Chapter Two.

Chapter Six describes Study 3a, a qualitative exploration through focus

groups of the beliefs, attitudes and experiences of Brisbane riders relating to the use


10

and safety of mopeds, scooters and motorcycles. As with Study 1, Study 3a

concentrated on Brisbane as a major hub of moped and scooter use, though some

participants also regularly rode mopeds or scooters outside of Brisbane. The focus

groups provided participants with an opportunity to discuss key issues relating to the

use and safety of mopeds, scooters and motorcycles in an open forum guided by

questions delivered within a semi-structured format. The open-ended questions

guiding the focus groups were developed by reference to a range of issues identified

in the research literature as relevant to moped and scooter safety.

The focus groups were used to inform development of a survey, titled The

Queensland scooter and moped rider survey 2010, referred to as Study 3b and

described in detail in Chapter Seven of the thesis. The survey was designed to

collect information on the demographic, social, motivational, attitudinal and other

characteristics, including crash involvement and licensing, of Queensland moped and

scooter riders. These data could then be compared with moped and scooter rider

profiles from other jurisdictions, as well as with motorcycle rider profiles from

Queensland, to assist in identifying specific risk factors.

Chapter Eight provides a discussion and summary of the overall findings and

limitations of the research in relation to the research aims and questions and the

literature reviewed. The main points discussed include trends in PTW usage and

crashes, crash characteristics and risk factors, and the rationale and potential for

safety-oriented interventions including moped rider licensing and training.

Additional discussion considers the conflicts between safety, personal mobility and

transport planning objectives in the context of urban transport systems.

1.6 Chapter One summary

Riders of powered two-wheelers (PTWs) have been described as vulnerable

road users, being at substantially greater risk of death and injury from road crashes

than car and other vehicle occupants. Moped and scooter riders are defined here as

sub-populations of PTW riders about which relatively little is known in regard to

their safety, particularly in jurisdictions where moped and scooter usage has

increased substantially from a low base. The primary aim of this research is to

discover and elucidate the similarities and differences between moped, scooter and

motorcycle riders in terms of safety in Queensland. This will help to identify ways


11

in which safety may be improved for Queensland moped and scooter riders.

This first chapter has outlined the rationale and scope of the current program

of research, based on the review of literature which identified gaps in knowledge

regarding moped and scooter safety. Powered two-wheelers (PTWs) as referred to

throughout this thesis have been defined, followed by a summary of the ways in

which regulations vary in regard to rider licensing across jurisdictions. An outline of

the research aims, research design, specific studies undertaken, and thesis structure in

terms of chapter content was provided. Chapter Two provides a review of the

literature relevant to the current research, followed by a detailed description of the

research design in Chapter Three. Chapters Four through Seven describe the four

studies undertaken to achieve the research aims. Chapter Eight presents a discussion

and summary of the overall research findings, implications and limitations.


12

The increased popularity of mopeds and motor scooters 13

CHAPTER 2: AN OVERVIEW OF MOPED AND SCOOTER USE

AND SAFETY

2.1 Introduction

Powered two-wheelers (PTWs) contribute to a substantial proportion of

global road fatalities and injuries and they are overrepresented in crashes relative to

their usage. As PTW riders, moped and scooter riders share much with motorcyclists

in terms of the issues which have led to their description as ‘vulnerable road users’

(Constant & Lagarde, 2010; Mihailovic, 2010; World Health Organization, 2009;

Haworth, 2006). The extensive literature on motorcycle safety and usage is therefore

highly relevant to mopeds and scooters, which have received comparatively little

research attention.

This chapter presents a review of research and related literature relevant to

moped and scooter safety, including that concerning motorcycle safety where

appropriate. Following some background in the current section, three main sections

cover the issues regarding moped and scooter usage and safety, as well as theoretical

approaches and perspectives that are relevant to the current program of research.

Safety issues are a consequence of the extent and nature of usage. Therefore, section

2.2 outlines PTW usage in Australia and other developed countries, including usage

trends, rider characteristics and motivations, and the role of PTWs in urban transport.

section 2.3 synthesises what is known about PTW safety, with a general focus on

mopeds and scooters. Section 2.4 concludes the chapter with a summary of the

review and the attendant implications for the current research.

2.1.1 Literature search methods

This literature review drew primarily on published academic literature and

reports from government and key international organisations. Scientific journal

articles, books, book sections and conference papers were identified using Google

Scholar, Science Direct and transport-related research databases. Reports and other

official material were identified through websites of relevant organisations and via

Google search engines. Grey literature from PTW industry sources was also


monitored and reviewed where relevant. Following initial searches, further material

was identified in the reference lists of literature already obtained.

The review was limited to material written in English and for which a full

text version was available. No date limits were set in the search for academic

literature and new literature was incorporated as identified up until June 2011.

2.1.2 Background

PTW riders generally comprise between 10 and 25 percent of road traffic

fatalities in developed countries, and more than 60 percent of all road deaths in some

developing countries (World Health Organization, 2009). In the last five years, PTW

riders comprised between 10 and 20 percent of traffic fatalities in the United

Kingdom (UK), United States (US) and Australia, where PTWs account for less than

five percent of registered vehicles (ATSB, 2007; Evans, 2004; Johnston, Brooks et

al, 2008; NHTSA, 2007, 2009; World Health Organization, 2009). PTW riders

represented between eight percent (Poland) and 28 percent (Greece) of fatalities in

Europe in 2009-2010 (IRTAD, 2010). While the fatality statistics provide a key

indicator of PTW safety for comparative purposes, the burden of non-fatal injury

from PTW use is also considerable and should not be understated (Ameratunga,

Hijar, & Norton, 2006; Constant & Lagarde, 2010; Johnston, Brooks et al., 2008).

In Australia, PTWs currently account for approximately one percent of all

vehicle kilometres travelled (VKT) and four percent of registered vehicles (including

commercial vehicles) (ABS, 2009). Yet PTW riders (including pillion passengers)

represented 16 percent of road user fatalities in Australia in the two years to July

2010 (BITRE, 2010). In 2009-2010, there were more rider fatalities in Queensland

(110) than any other Australian jurisdiction, with the proportion of PTW riders in all

road user deaths reflecting the previous five year average of 19 percent (TMR,

2011a).

The number of reported PTW crashes has increased substantially in some

developed countries in recent decades, while PTW crashes have stabilised in other

countries. In absolute numbers, rider fatalities in the UK have been relatively stable

over the last 15 years (Department for Transport, 2010). In the US there was a 144

percent increase in fatalities involving motorcycles over 10 years to 2007 (Morris,

2009). The steady increase in frequency of PTW crashes in Australia and elsewhere


has occurred against a background of consistent increases in PTW sales and usage

(Christie & Newland, 2001; Department for Transport, 2010; Johnston, Brooks et al.,

2008; Morris, 2009; Paulozzi, 2005; Tunnicliff, 2006).

In Australia and many European countries, reported crash rates per registered

PTW and per distance travelled have declined, suggesting that PTW usage has

become safer relative to exposure. In the last 15 years in the UK, the fatality rate per

billion miles travelled by PTWs has fallen by 25 percent (Department for Transport,

2010). In contrast, in the decade to 2008 rider fatalities increased in the US at a

faster rate than the number of registered PTWs and the distance travelled (Morris,

2009). More recently in the US, the last two years have seen an unprecedented

decline in PTW rider fatalities (and car occupant fatalities) (NHTSA, 2010; Hedlund,

2011). This reduction in rider fatalities does not appear to have resulted directly

from declining sales and usage, as the number of motorcycles in use and distance

travelled both increased by about five percent from 2008 to 2009 (MIC, 2011).

Hedlund (2011) suggests some reasons for the decline, including more participation

in rider training and higher priority for education, publicity and enforcement.

While fatality statistics provide an outline of trends for the most serious

crashes, non-fatal injury crashes do not necessarily follow those trends. For

example, the aforementioned 144 percent increase in fatal crashes observed in the US

was accompanied by a 94 percent increase in non-fatal injury crashes over the same

period (Morris, 2009). In Australia from 1999 to 2004, motorcycle rider fatalities

increased by 2.6 percent annually on average, while serious injuries increased by 4.4

percent annually over the same period (Johnston, Brooks et al., 2008).

As well as increased sales and usage, there have also been changes in the

demographic characteristics of riders, with increasing proportions of older riders seen

in crash and survey data in recent decades. This may be attributable in part to an

ageing population (Schulze & Koßmann, 2010), but has resulted more from an

increase in popularity of PTWs among older riders (Broughton & Walker, 2009;

Haworth, Mulvihill, & Rowden, 2006). The proportion of female riders has also

increased in some instances, though generally males continue to comprise a large

majority (~90%) of all PTW riders in developed countries.

Considering the situation briefly outlined above, it is understandable that

motorcycle safety has attracted considerable research attention in recent decades and

there is increasingly abundant literature on the topic. However, much of the


literature is less applicable to mopeds and scooters than to other PTW types, in that it

focuses primarily on larger and faster motorcycles (Watson, Tunnicliff, White,

Schonfeld, & Wishart, 2007). Within the general body of PTW safety literature,

studies with a specific focus on mopeds or scooters are not only relatively few, but

also originate from outside Australia with limited exceptions (Faberi, Martuzzi, &

Pirrami, 2004; Haworth & Nielson, 2008). The increased popularity of scooters and

mopeds in Australia has been comparatively recent and there is a subsequent paucity

of research on the topic from Australian jurisdictions. The need for such research is

acknowledged in other recent work (Haworth, Nielson, & Greig, 2008; Tunnicliff,

2006; Watson, Tunnicliff et al., 2007).

Many European countries have a long history of widespread moped and

scooter use (Noordzij, Forke, Brendicke, & Chinn, 2001), particularly those in

southern Europe where scooter production escalated in the late 1940s (Shattuck &

Peterson, 2005). Much of the early literature on moped and scooter safety therefore

originates from Europe, which continues to be the primary source of relevant

literature to the present day. Some early research was also conducted in the US

when moped and scooter use increased there in the 1970s and 80s, along with

concern over related safety issues (Evans, 1978; Matzsch & Karlsson, 1986;

McHugh & Stinson, 1984). While some research was also conducted in Australia

during this period, itself identifying a need for further study (Wigan & Carter, 1980),

there appears to have been little further research until relatively recently.

While much of the European and other literature appears relevant to moped

and scooter safety in Australia, certain contextual differences need to be considered

(Naci, Chisholm et al., 2009). Among these are differences in legislation, such as

licensing requirements (including minimum rider age), helmet use laws, drink-

driving legislation and vehicle performance restrictions (Haworth, Nielson et al.,

2008). Behavioural interactions between riders and other road users may also be

different in places which have a relatively long history of moped and scooter use

compared to Australia. Further, environmental differences may variously influence

usage and behaviour, both in terms of natural environment (climate, topography) and

built environment (infrastructure, services, population density). As well as

producing contrasts between Australia and other countries, environmental factors

may also influence usage and behaviour differentially within Australia given its

environmental diversity.


PTW use in developing and often densely populated countries in Asia, South

America, Africa and Middle East regions is not considered specifically in this

literature review. However, it is noted that PTW riders in many of these places (a

large proportion of whom ride mopeds and scooters) represent a much higher

proportion of road users than they do in developed countries (Ameratunga, Hijar et

al., 2006; World Health Organization, 2009). Consequently, due to high levels of

PTW use in large populations, PTW-related road trauma in developing countries is of

a scale unseen in Australia, Europe or North America (Naci, Chisholm et al., 2009).

While this clearly represents a major global health problem, the economic, political,

regulatory and socio-cultural differences between these countries and Australia

render their inclusion here beyond the scope of this research.

2.2 Moped, scooter and motorcycle usage

Many aspects of PTW use and safety have changed since the first

comprehensive studies of these issues were conducted in the 1970s and early 1980s.

For most developed countries, arguably the most significant changes relate to

increased usage, developments in PTW design and manufacture, and the changing

demographic characteristics of riders. Population growth and increasing traffic

congestion appears to have driven increased PTW use by commuters in cities, while

there has also been an increase in recreational PTW use in many countries.

Accurate estimates of PTW usage require reliable exposure data, usually

defined by distance travelled over a given timeframe, but potentially also defined by

time spent riding in a given period. Although exposure data are most useful for

measuring PTW usage, and arguably provide the most reliable basis for calculation

of crash rates, they are relatively scarce and difficult to obtain. Some PTW usage

trends can be identified from sales, registration and (to a lesser degree) licensing

data, which are obtained relatively easily. However, as well as revealing little about

the actual amount of usage, these data often lack detail regarding PTW types. For

example, scooters are often not distinguished from motorcycles where the two PTW

types share a vehicle category for registration purposes (see section 1.2).

Nonetheless, while registration, sales and exposure data do not provide consistent

estimates of usage, they collectively provide an overall view of usage trends

(Haworth, 2003).


2.2.1 Trends in sales and registration

2.2.1.1 European trends in sales and registration

There is a long history of moped and scooter use in many European countries,

particularly those in southern Europe where the climate is more conducive to PTW

riding (Faberi, Martuzzi et al., 2004; Noordzij, Forke et al., 2001). European

countries differ in terms of usage and also crash rates, and it is arguably misleading

to review them collectively in terms of PTW usage and safety (Wigan, 2000). In

terms of PTW ownership, the European Road Safety Observatory (ERSO) notes that

the number and ratio of mopeds to motorcycles per 1,000 inhabitants varies across

European countries (SafetyNet, 2009). Moped and motorcycle ownership data

sourced from the ERSO for selected European countries are presented in Table 2.1.

In 2005 when these data were collected, rates of moped ownership were clearly

highest in Greece, followed by Italy. Rates of motorcycle ownership were highest in

Greece, Italy and Switzerland. There were more mopeds than motorcycles in Greece

and Italy, as well as in the Czech Republic, Norway, Portugal, Slovenia and Spain.

Table 2.1 Mopeds and motorcycles per 1,000 inhabitants in European

countries

Country PTWs per 1,000 inhabitants 2005

Moped Motorcycle Austria 36 38

Czech Rep. 43 31

Denmark 12 18

Finland 25 27

France 19 22

Germany 22 46

Great Britain 2 19

Greece 150 101

Ireland (incl. moped) 8

Italy (2004) 90 79

Netherlands 34 33

Norway (2004) 32 21

Portugal 40 14

Slovenia 17 7

Spain 53 42

Sweden 18 26

Switzerland 24 80 Source: European Road Safety Observatory (ERSO) (SafetyNet, 2009)


Overall PTW sales and registrations increased in Europe from 1994 to 2008,

with the motorcycle sector growing at a faster rate than the scooter sector to

comprise 60 percent of all PTWs. More recently, the global economic downturn has

impacted the European PTW market. In 2009 there was a 25 percent decline in PTW

sales on the previous year, and a 30 percent decline in moped sales (ACEM, 2010a).

Other motor vehicle sales also declined in Europe over the same period, with

considerable variation between countries, though it appears that new passenger cars

sales generally declined less sharply than new PTW sales (ACEA, 2010).

The Department for Transport in Great Britain reported that the number of

licensed (registered) PTWs increased in all categories from 1999 to 2009, with the

rate of increase differing according to engine size. Over this 10 year period, PTWs

up to 50cc (mostly mopeds) increased by seven percent, 50-125cc (light

motorcycles) by 51 percent, 125-500cc PTWs by 2.5 percent, while larger PTWs

(over 500cc) increased by 77 percent (Department for Transport, 2009). Recent sales

data for the UK in particular indicate that overall PTW sales there fell by 20 percent

in 2009 over 2008 levels (excluding ATVs), again reflecting impacts of the global

economic downturn. While the recent decline in European sales has been

substantial, the PTW market there has been more resilient than the US market in the

most recent sales period (AMCN, 2010), as outlined in the following section.

2.2.1.2 North American trends in sales and registration

Over the 10 years to 2006 there was an increase of nearly 250 percent in sales

of new PTWs in the US (excluding unregistered and off-road vehicles). The increase

in PTW sales and use has generally been in large capacity motorcycles (over 600cc

engine capacity), but scooter sales for the first three quarters of 2008 increased by 50

percent over 2007 levels. Overall, the percentage share of new sales by PTW type

was fairly stable over the three years to 2007, with scooters representing

approximately six percent of new PTW sales in the US (Morris, 2009).

The recent global economic downturn has had considerable impact on PTW

sales in the US, with the Motorcycle Industry Council reporting an estimated decline

of about 40 percent in 2009 compared to 2008 sales (AMCN, 2010; Madson, 2010).

The scooter segment was the most negatively affected of the US PTW market during

this period, with scooter sales reportedly falling by about 60 percent over the


previous year. It should be noted that this decline in 2009 followed a record year in

2008, during which scooter sales increased by about 41 percent compared to 2007

(Madson, 2010). The recent fluctuation observed in the US scooter market reflects a

similar yet more extreme situation to that observed in Australia and Europe. In the

US, mopeds and light scooters up to 50cc are not usually required to be registered for

use on public roads (Morris, 2009). Subsequently, registration data do not provide

information on usage of these vehicles.

As in Europe, moped and scooter use in the US appears to vary considerably

by geographic region and also by location within regions. For example, the number

of reported moped crashes in Honolulu from 2002 to 2004 exceeded those involving

motorcycles (Kim, Pant, & Yamashita, 2010), suggesting that mopeds comprise a

large proportion of PTWs used in Hawaii. However, light PTWs (up to 125cc,

including mopeds) represented less than five percent of new PTW sales across the

US in 2003 (Morris, 2009). Different moped usage rates according to location may

relate to differences in parking availability, occupation and relative access to

alternative transport modes, as well as climate as previously noted. For example,

moped and scooter use appears popular with students attending some university

campuses in the US, particularly those which provide more parking for PTWs

relative to that available for cars (Kennedy, 2007).

There has been strong growth in PTW sales in Canada in recent years, driven

in part by a substantial increase in the popularity of scooters. While all PTW sales

(excluding ATVs) increased by 12 percent from 2004 to 2008, scooter sales

increased by nearly 70 percent over the same period. Scooters comprised around

eight percent of PTW sales in Canada in 2004, yet by 2008 they accounted for 12

percent of PTW sales. A decline in the popularity of mid-capacity (250-600cc) PTW

engines was also reported for the same period (Motorcycle and Moped Industry

Council, 2009). The growth in PTW sales in Canada represents a trend observed

over more than a decade, with the Canada Safety Council (2009) reporting a

threefold increase in motorcycle sales from 1996 to 2003.


2.2.1.3 Australian trends in sales and registration

There has been a consistent increase in combined PTW sales in Australia in

the last decade, with a 70 percent increase in new sales over the five years to 2008.

An unprecedented rise in the popularity of motor scooters and mopeds has

contributed to the overall increase in PTW sales in Australia (FCAI, 2008; Haworth

& Nielson, 2008). Scooter and moped sales have tripled since 2004, and the sales

growth of these vehicles has generally exceeded that of other PTW types over the

last decade. In recent years, approximately one third of all new scooter and moped

sales in Australia have occurred in Queensland. In the first half of 2005, mopeds

comprised 84 percent of new moped and scooter sales (Haworth & Nielson, 2008).

More recently, the share of mopeds dropped to 63 percent of new moped and scooter

sales in Queensland in the 12 months to September 2009 (FCAI, 2009b). This

suggests that scooter use is increasing at a faster rate than moped use in Queensland,

which is perhaps surprising given that moped riding does not require a motorcycle

licence.

Consistent with these sales increases, PTW registrations increased

substantially across Australia in the five year period from 2004 to 2009. Motorcycle

registrations across Australia increased by around 50 percent over this period,

compared with all other vehicles which increased by 14 percent to 23 percent

(depending on vehicle type). Approximately half of all new PTWs sold in Australia

in recent years have been off-road motorcycles, many of which cannot be registered

for use on public roads. This accounts for much of the difference between sales and

registration increases. Over the five years to March 2009, Queensland recorded a 70

percent increase in PTW registrations, compared with 53 percent in NSW and 44

percent in Victoria (ABS, 2009). As scooters are defined as LC Category

motorcycles for the purposes of Australian Design Rule (ADR) classification, and

recorded as such in registration databases, trends in above 50cc scooter registrations

can only be estimated from the available sales data at this point in time. Trends in

moped registrations are more accessible. In Queensland, an increase in moped

registrations of 150 percent occurred between 2004 and 2007 (TMR, 2009).

After a sustained period of growth over the last decade, overall sales of new

PTWs in Australia have recently declined. The Federal Chamber of Automotive

Industries (FCAI) reported a decline of 13.6 percent for 2009 over 2008 figures for


all new PTWs and all-terrain vehicles (ATVs), while a decline of 7.4 percent for all

other vehicle types was reported for the same period (FCAI, 2010). The FCAI

reports that this overall trend is as expected in light of the recent global economic

downturn (FCAI, 2009a). It is worth noting that, according to the FCAI, tax-related

business incentives produced a late surge in sales of other vehicle types in 2010,

prior to which the overall yearly sales decline was comparable to that for PTWs. The

data suggest that, while growth in the PTW sector has peaked in the last couple of

years, these vehicles remain popular relative to other vehicle types in the face of the

recent global economic pressures.

Scooter and moped sales have declined more dramatically than other PTWs

and cars, with a 32 percent drop in scooter and moped sales for 2009 compared to

2008 sales (FCAI, 2010). By comparison, new on-road motorcycle sales (excluding

scooters) fell by 14.2 percent over the same period, suggesting that scooter and

moped sales are more prone to market fluctuation in an uncertain economic climate.

For the first half of 2009, the scooter and moped segment represented 10.1 percent of

new sales in the PTW market, well behind the leading cruiser segment which held a

22.5 percent share (FCAI, 2009a). However, in early 2007 scooters and mopeds

comprised the largest segment of new sales for the on-road motorcycle market

(Haworth, Greig, & Wishart, 2007).

2.2.2 Development and expansion of PTW types

Research has identified relationships between crash risk and PTW type

(Harrison & Christie, 2003; Teoh & Campbell, 2010), and trends in PTW design are

therefore of general interest to this research. The number and range of PTW types in

the global market has expanded considerably over the last decade. Traditional

motorcycle and scooter designs remain the most popular among buyers of new

vehicles, but innovation and consumer demand has underpinned the emergence and

expansion of niche markets within the PTW sector. Large capacity cruising

motorcycles have become increasingly popular, possibly in association with the

increasing number of older riders observed in recent years. Several medium capacity

motorcycles with automatic transmission have also recently entered the Australian

market. Among these is the Honda DN01 (680cc), which appears a deliberate

attempt to bridge the traditional divide between ‘motorcycle’ and ‘scooter’ concepts


(Duke, 2009). Traditional or ‘retro’ styled scooters and mopeds remain popular,

though many riders evidently prefer contemporary ‘sport’ style scooters and mopeds,

with the (often) associated superior performance characteristics. There is now a

wider range of large capacity scooters suitable for touring than was previously the

case. New designs have also appeared in the form of three-wheeled scooters

(Piaggio MP3 for example) and three-wheeled motorcycles (Can Am Spyder for

example), with these vehicles having two front wheels and one rear wheel, reversing

the ‘trike’ configuration historically (and still) applied to traditional cruisers.

In the scooter and moped market, electric and hybrid electric/petrol motors

are now readily available and have become cost-competitive, with major

manufacturers increasingly involved in their production (Bowdler, 2011b). While

electric and hybrid technology for scooters and mopeds is still in the early stages of

development, such vehicles have been prominent at recent international trade shows

according to industry reports (Round, 2010). In terms of vehicle type classification,

existing electric and hybrid vehicles meet requirements for both LA moped and LC

motorcycle categorisation under Australian Design Rules. With increasing

international focus on reducing carbon emissions, it is possible that electric vehicles

will eventually displace many of those with traditional combustion engines, although

some industry observers are highly sceptical (Purvis, 2010).

In recent years many mopeds and scooters manufactured in China have

entered the Australian market and these may be particularly popular amongst riders

with limited budgets. Anecdotal reports cast doubts upon the quality of some of the

Chinese PTWs. There is no research or other reliable evidence available from

Australia to substantiate these claims, but industry reports from Europe indicate that

some of the Chinese PTWs (a large proportion of which are mopeds and scooters)

are potentially unsafe and polluting, failing to comply with EU standards (ACEM,

2010a).

2.2.3 Patterns of moped and scooter use

Sales and registration data are useful for identifying trends in PTW usage and

for estimating crash rates per registered vehicle. It is generally accepted that the

number of crashes per distance travelled, or crashes relative to exposure, is a more

useful measure but exposure data for moped and scooter use are generally scarce


(ERSO, 2006; Harrison & Christie, 2003). Research has estimated annual distance

travelled by mopeds in the UK at 3,600 kilometres (Huang & Preston, 2004) and in

the Netherlands at approximately 2,900 kilometres (Schoon, 2004). More recent

literature suggests that exposure is highly variable across European countries, that

the situation may have changed since prior to 2004 and/or that the data may be

somewhat unreliable. In recent years, moped use in the Netherlands accounted for

approximately one percent of all VKT (SWOV, 2006b), while in the UK all PTWs

combined accounted for roughly the same proportion (~1%) of all distance travelled

by vehicles in 2009 (Department for Transport, 2010). Despite this pattern, the

average kilometres travelled annually per moped in the UK was reported in 2006 to

be about double that in the Netherlands (ERSO, 2006). The Netherlands exposure

data also suggest that male moped riders aged 15-17 years travel more than twice as

far as female riders of the same age (SWOV, 2006b). Research in Australia has also

found that males ride further than females on average, though this finding primarily

concerns motorcycle use rather than moped and scooter use (Harrison & Christie,

2003).

The travel exposure data for moped and scooter use in Australia are limited in

terms of reliability and currency. The average distance travelled by motorcycles in

Australia in 2007 was estimated by the Australian Bureau of Statistics (ABS) at

3,745 kilometres, though this is likely an underestimate due to the inclusion in

calculations of registered motorcycles that did not travel (ABS, 2008). An earlier

estimate for 1998 showed an average annual distance of around 4,400 kilometres for

PTWs in Australia (Wigan, 2000). Research conducted 30 years ago in Victoria,

South Australia and Western Australia suggested mopeds in these States travelled

about 2,300 kilometres per year on average, though this finding is not assumed to

reflect current exposure (Wigan & Carter, 1980). Responses to a self-report survey

conducted in New South Wales in 2002 indicated that, compared with motorcycles,

VKT by scooters was relatively low at a median of about 1,800 kilometres per year,

roughly half to one third of the distance travelled by motorcycles (Harrison &

Christie, 2003). A motorcycle usage survey conducted in Queensland in 2005

(Harrison & Christie, 2006) reported average annual motorcycle travel distances

between 6,500 and 7,300 kilometres. Reported odometer readings taken 12 months

apart indicated an average annual distance for mopeds of about 2,600 kilometres,

while scooters travelled around 4,000 kilometres. The reported distances travelled


by mopeds and scooters spanned a wider range than those of motorcycles.

The survey conducted in New South Wales (Harrison & Christie, 2003)

indicated that, consistent with other research (ACEM, 2008a; Moskal, Martin, &

Laumon, 2010; Sexton, Baughan, Elliott, & Maycock, 2004), mopeds and scooters

were more often ridden on weekdays and in urban areas (mainly for commuting)

compared to motorcycles. This research did not separate scooter and moped riders,

but it was found that scooter riders were slightly older and slightly less experienced

than other PTW riders as a whole. However, the data must be treated with caution

due to self-reporting and low numbers of scooters in the research (n = 31). The

survey targeted a randomly selected sample of 6,000 riders, with only 794 valid

responses valid for final analysis (a 13% response rate).

Exposure data which rely on the distance travelled by PTW types do not

account for the average amount of time spent covering a given distance. Given that

mopeds generally travel at lower speeds than motorcycles, the time spent riding per

kilometre is likely to be greater for mopeds than for motorcycles. Therefore, in

terms of actual time spent riding, the ‘exposure’ of moped riders may be greater than

that of motorcycle riders over the same distance. This further complicates the

comparison of crash rates by PTW type relative to exposure, and is a potential

problem that is generally overlooked in the literature.

2.2.4 Demographic characteristics of riders

The demographic characteristics of riders affect their riding patterns and

crash risks. Information on demographic characteristics may be obtained from a

number of sources, including licensing data, survey responses, crash data and direct

observation. All of these sources are methodologically limited in various ways and

may not provide entirely representative samples. For example, Wigan (2000)

suggests that an over-reliance on crash data for PTW usage information has led to

‘unbalanced’ assessments through the exclusion of riders who had not crashed. In

any case, it is clearly useful if possible to know the number of active riders in a

population as a starting point for addressing safety issues. However, the total

number of active and potentially active PTW riders in Queensland and elsewhere is

very difficult to estimate, for reasons outlined below.


In any jurisdiction, data on motorcycle licences held will include people who

have a licence but are not active riders (Watson, Tunnicliff et al., 2007), often termed

‘dormant’ licence holders. In Queensland there are approximately five motorcycle

licence holders for every registered PTW (Queensland Transport, 2008). Therefore it

appears that a large proportion of current motorcycle licence holders do not ride

regularly and many may not ride at all. Research suggests that inactive or dormant

licensed riders are more likely than active riders to be older and (possibly) female

(Haworth, Mulvihill, & Symmons, 2002). Licensing data may therefore provide an

underestimate of young male riders as a proportion of (legally) active riders, while at

the same time overestimating the total number of (legally) active riders (Haworth,

Smith, Brumen, & Pronk, 1997).

Motorcycle licence data do not include unlicensed but active riders,

sometimes referred to as ‘unriders’ (Haworth, 2003). Research has found unlicensed

riding to be more common in younger riders and male riders, potentially further

confounding estimated gender and age distributions of active riders based on

licensing data. The potential distortion may be considerable where unlicensed riding

is common, such as in Queensland. Research suggests that more than 10 percent of

riders involved in reported Queensland crashes were effectively unlicensed (no

current licence, or inappropriate class of licence) between 1996 and 2004 (Watson &

Steinhardt, 2007).

The characteristics of motorcycle riders are somewhat better known than

those of moped riders. This may be partly due to the relatively high volume of

research on motorcycle use, but is also related to problems in identifying moped

riders in the data. In jurisdictions where moped riding is permitted for all car licence

holders, moped riders who do not hold a motorcycle licence cannot be identified in

licensing data as PTW riders. However, car licence holders contribute to the total

number of potential and actual PTW riders.

As scooter riders in Queensland are required to hold a motorcycle licence,

they cannot be reliably separated from motorcycle riders in the licensing data. It

appears likely that holders of a motorcycle licence with an ‘A’ (automatic only)

condition would be mostly scooter riders, as there are very few motorcycles with

automatic transmission. However, such data are not readily available and, moreover,

would not identify scooter riders who hold a motorcycle licence without an ‘A’

condition.


The two other main sources of information on the demographic

characteristics of riders are survey data and crash data. Survey data may be

compromised by volunteer bias, where participants who complete a survey are not

representative of the target population. There are also questions of accuracy in self-

reported survey data, including differences in underreporting of risky driving

behaviours by different social and cultural groups (although self-report data has been

shown in some studies to be largely accurate) (Tomaskovic-Devey, Pfaff Wright,

Czaja, & Miller, 2006; Tubre, Bell et al., 2005). Crash data may not be

representative of active riders as a whole as there is likely to be a bias toward those

more likely to crash (Wigan, 2000).

2.2.4.1 Age

In recent years there has been an apparent shift in the age profile toward older

riders in many developed countries including Australia, North America and some

European countries. In the US there was an increase in the median age of PTW

owners from 27 years in 1985 to 41 years in 2003 (Morris, 2009). Similarly, in

Canada the average age of PTW buyers increased from mid-20s in 1990 to mid-40s

in 2005 (Canada Safety Council, 2009). In recent years Australia has also seen an

apparent ageing of the PTW rider population as a whole (Haworth, Mulvihill et al.,

2006), at least as indicated by crash data. Surveys in Australia indicated average

rider ages in the early 40s between 2000 and 2002 (Harrison & Christie, 2005).

From 1998 to 2007 in Australia, the number of fatally injured riders aged 45 years or

older increased by 12 percent annually on average, while the number of riders aged

under 24 remained stable (Johnston, Brooks et al., 2008). By contrast, an ageing

rider population is not evident in Great Britain, where 2008 rider age profiles are

consistent with 1996-2003 averages (Department for Transport, 2009; 2004).

Rider age distributions appear to differ between PTW types and also between

jurisdictions (ERSO, 2006). In particular, the general shift toward older riders

outlined above is not necessarily reflected among moped riders. Mopeds have

traditionally attracted younger riders in many places, particularly in jurisdictions

where licensing regulations permit moped riding at a relatively young age (Antonio

& Matos, 2008; Kopjar, 1999; Noordzij, Forke et al., 2001). This has been the case

in many European countries, as previously discussed in the section on licensing


(section 1.3). Where the minimum age for moped riding is closer to that for larger

PTWs or for car driving, there may be a greater proportion of older moped riders

(Harrison & Christie, 2003; Wigan & Carter, 1980). A recent postal survey of newly

licensed riders in the UK showed that new riders tended to be younger, tended to be

commuters, and were moped or scooter riders in around 40 percent of cases (Jamson

& Chorlton, 2009). The implications of age-related factors for safety are discussed

in the later section on risk factors (section 2.3.2).

2.2.4.2 Gender

Males comprise the majority of moped and scooter riders in developed

countries, though compared with motorcycle riding, moped and scooter use is

relatively more popular among females. The European Motorcycle Accident In-

depth Study (MAIDS) report indicates that 77.6 percent of (L1) moped riders in a

control sample (N = 923) were male, compared with 93.5 percent of (L3) motorcycle

riders (ACEM, 2008a). The second phase (2005-2008) of a study in Barcelona

observed that 65 percent of injured moped riders were male according to police

reports (Perez, Mari-Dell'Olmo, Borrell, Villalbi, Santamarina, & Tobias, 2009).

The same study reported a similar gender distribution for light motorcycle injuries

(68% male), while injuries involving heavy motorcycles (over 125cc) were more

likely to involve males (84%). Research on scooter use at a university campus in the

US reported that 59 percent of riders were male (Kennedy, 2007). A slight increase

in female ownership of motorcycles has been observed in the US during the last

decade, rising from eight percent in 1998 to 10 percent in 2003, though these data are

not broken down by motorcycle type and do not reflect the gender distribution

among scooter and moped riders (Morris, 2009).

During the last decade in Australia, males represented around 62 percent of

crash-involved moped riders in Queensland (N = 306) (Haworth, Nielson et al.,

2008) and around 66 percent of scooter riders in Victorian crash data (N = 337)

(Christie, 2008). Both of these studies found that males represented around 93

percent of crash-involved motorcycle riders. Christie (2008) notes that that the

gender distributions likely reflect usage rates rather than a tendency among female

riders to crash scooters and mopeds more than motorcycles. It can therefore be

assumed that about one third of Queensland moped riders are female and that the


overall characteristics and motivations of moped riders will thus differ from those of

motorcyclists, around 90 percent of whom are male.

Research also suggests that male moped riders travel further on average than

female riders of the same age (Harrison & Christie, 2005; SWOV, 2006b). For all

PTW riders in NSW, Harrison and Christie (2005) found that males travelled an

average annual distance of 3,637 kilometres, compared to 2,760 kilometres for

females. If males and females have comparable crash rates per distance travelled,

then crash data may not provide a reliable estimate of usage frequency by gender

(i.e., the proportion of active female riders may be underestimated).

2.2.4.3 Socioeconomic status

Low cost of use is among the primary motivators for PTW use, as discussed

in the following section on motivations for riding. Although PTWs and mopeds in

particular can be a relatively cheap form of private motorised transport compared to

cars, it is not clear that they are used disproportionately by those of low

socioeconomic status in developed countries. Certainly many young moped riders

may have incomes below the average for their place of residence (Nja & Nesvag,

2007). However, in many places (such as Europe) it is likely that moped use in

younger age groups is largely driven by regulations which permit moped riding at an

earlier age than car driving, rather than by lower costs involved.

A recent survey of PTW riders in the UK found that, overall, their income

levels were above the national average (Broughton & Walker, 2009). Riders tended

to be employed in managerial and skilled employment, with only two percent of

respondents unemployed and two percent students. Lower income earners tended to

spend less on their PTW and associated expenses, though there was no specific

information on differences between motorcyclists and moped and scooter riders with

regard to income. An unpublished survey on moped and scooter use in Brisbane,

Queensland, found no significant differences between moped and scooter riders and

non-riders in terms of income (Haworth & Rodney, unpublished).


2.2.5 Motivations and reasons for PTW use

The numerous motivations for PTW riding identified in the research literature

broadly relate to demands for mobility and also recreation and leisure. More

specifically, riders are motivated by low vehicle purchase and running costs, time-

efficiency, parking costs and availability, convenience, ability to negotiate congested

traffic, enjoyment, excitement, sense of freedom and self-image (ACEM, 2010a;

Broughton & Walker, 2009; Jamson & Chorlton, 2009; Nja & Nesvag, 2007; Wigan,

2000). The motivations for riding may differ considerably depending on the type of

PTW ridden. In particular, research has shown that larger PTWs are more likely to

be used on weekends and for leisure and recreation compared with smaller PTWs,

including mopeds and scooters, which are more likely to be used on weekdays and

for commuting (Harrison & Christie, 2003; Jamson & Chorlton, 2009; Sexton,

Baughan et al., 2004). Moped and scooter riding evidently satisfies the mobility

needs of many riders who are dependent to some degree on private vehicles, partly

due to a lack of suitable or attractive alternatives (Ibrahim, Radin, Habshah, Kassim,

Stevenson, & Hariza, 2006).

Despite generally overlooking PTWs, studies on travel mode choice and

transport planning offer some explanations of PTW usage motivations and why

usage has increased in many places (Steg, 2005; Steg, Geurs, & Ras, 2001; Steg &

Tertoolen, 1999). Increased PTW use may be driven by what Steg (2005) describes

as ‘instrumental’ motives associated with commuting and general transport needs, as

well as by ‘symbolic’ and ‘affective’ motives more often associated with recreation

(Steg, 2005). It appears likely that instrumental motives are a stronger driver of

moped and scooter use than of motorcycle use in most places, given their

comparatively greater use for commuting than recreation (Jamson & Chorlton, 2009;

Harrison & Christie, 2003). However, noting that data regarding travel mode choice

are typically obtained from surveys, Steg (2005, p. 148) argues that:

...symbolic-affective motives are better expressed when the aim of the

research task is not too apparent. If respondents are asked to explicitly

evaluate the attractiveness of various aspects of car use, they especially

mention instrumental aspects. Apparently, they are not likely to admit that

symbolic and affective aspects make car use attractive. However, if the


research task is rather ambiguous, respondents indicate that especially

symbolic and affective aspects make car use attractive.

It seems possible that expressed motives for PTW use may be similarly

influenced by the stated aims of a research task. Specifically, instrumental motives

for moped or scooter use may be overstated, or symbolic-affective motives

understated, in research that explicitly seeks to examine usage motivations.

Travel mode choice has also been explained as the ‘combined result of

travellers’ economic concerns, psychological preferences and habitual behaviour’

(Chang & Wu, 2008, p. 370). These various concerns, preferences and behaviours

are related in some combination to instrumental, symbolic and affective motives

(where the latter two are closely related) (Steg, 2005).

Importantly, the various motives (instrumental, symbolic, affective)

influencing PTW use do not exist in isolation from one another. For example, while

riders may see a moped or scooter as the most practical and efficient means of

commuting to work, they may also see it as fun, an expression of self-identity or sub-

cultural belonging, a challenge to be met, and so on. In this sense, an

anthropological perspective may see these various motives as contained with a

culturally referenced ‘web of meaning’ (Geertz 1973, cited in Nja 2007). From this

perspective it is arguably not possible to fully disentangle the different motives that

drive PTW use for individuals. Moreover, in some cases even a specific motive may

be impossible to define singularly in Steg’s (2005) terms. A desire to reduce one’s

environmental footprint for example, by shifting from car to scooter use, may stem

simultaneously from symbolic, affective and instrumental motives.

2.2.6 Traffic congestion, fuel consumption and emissions

Traffic congestion is a major problem in large cities and an increasing

problem in cities which are experiencing growth, to the extent that current transport

systems are often deemed unsustainable (de Groot & Steg, 2006; Mackett & Ahern,

2000). Increased congestion results in increased travel times and increased fuel

consumption and, therefore, increased emissions from motorised vehicles. A range

of respiratory and other illnesses are linked to motor vehicle-related pollutants,

resulting in premature death rates that may be of a magnitude comparable to those


from road crashes in some places (although this is difficult to measure for numerous

reasons) (Delhomme, Chappé, Grenier, Pinto, & Martha; Jacobson, 2008; Litman,

2003). As well as directly impacting physical health, increasing congestion may also

negatively impact overall quality of life by increasing stress levels and vehicle

running costs, and reducing the time people have for other activities (ACEM, 2010a;

Clarke & Hawkins, 2006; de Groot & Steg, 2006).

In the literature that focuses on sustainable urban transport systems and

planning, PTWs receive little attention compared to other transport modes which

present as alternatives to car use, such as public transport, cycling and walking

(Musso, Vuchic, Bruun, & Corazza, 2010; Wigan, 2000). For example, a report

titled ‘Potential for mode transfer of short trips’, published by the Centre for

Transport Studies, London, makes no mention of PTWs as a potential alternative to

car use (Mackett & Ahern, 2000). References to PTWs are similarly absent in other

research focused on reducing or modifying car use (Myers & Ridout, 2010; Ogilvie,

Egan, Hamilton, & Petticrew, 2004). This is despite the strong and enduring trend

towards increased PTW use observed in Australia and other developed countries, as

outlined earlier in this chapter (section 2.2). With limited exceptions, organisations

that advocate a key role for PTWs in urban transport are likely to be PTW industry-

based groups, though they may also be active in safety promotion (ACEM, 2008b;

FEMA, 2007; Wigan, 2000).

The use of PTWs as an alternative to cars can help to relieve traffic

congestion as they occupy less road space than other motorised passenger vehicles

(Wigan, 2000). On public roads the estimated space occupied by a PTW is 20m²,

compared with 70m² for a passenger car according to ACEM reports. The average

space occupied by a PTW in parking is 1.4m² to 2.6m² (with mopeds and scooters at

the lower end of this range), compared with 7.4m² to 12.5m² for one passenger car

(ACEM, 2008b, 2010b). However, if this claimed benefit in regard to congestion

relief is to be accurately quantified, vehicle occupancy must be taken into account.

While they are usually capable of carrying two people, in most cases PTWs carry

only one person (in developed countries). This is especially true of mopeds, which

are more limited than larger PTWs in their ability to accommodate pillions, as the

weight of a second occupant may severely restrict acceleration, suspension and

braking power. Private cars in urban areas will also often carry only one person at a

time, and although cars are usually capable of carrying of four or more people,


attempts to encourage car pooling have not been very successful (Fellows & Pitfield,

2000; Ferguson, 1997; PTUA, 2010).

PTWs generally consume much less fuel than passenger cars, with claimed

consumption of 50 kilometres per litre not uncommon for mopeds (ACEM, 2008b;

Wigan, 2000). A study of fuel consumption of light PTWs (under 150cc) in

Malaysia found average fuel consumption of 45km per litre for four-stroke models

and 35km per litre for two-stroke models. Fuel consumption was also found to

depend on travel speed, gear selection (where applicable), vehicle mass, frontal area,

and tyre pressures (Lee, Chong, & Gitano, 2010). Some larger PTWs may compare

in fuel consumption with some small cars, but mopeds and even larger scooters are

clearly more efficient by comparison (Wigan, 2000).

Despite their typically lower fuel consumption, PTWs have not historically

represented a necessarily ‘clean’ alternative to cars in terms of carbon emissions and

other pollutants. In particular, the traditional two-stroke engines used in mopeds

have been heavily criticised for their contribution to air (and noise) pollution, as well

as their higher fuel consumption compared to four-stroke engines (Lee, Chong et al.,

2010; Musso, Vuchic et al., 2010; Wigan, 2000). PTW manufacturers have

reportedly lagged behind the passenger car industry in this regard, having been

largely exempt from increasingly stringent emissions controls in Europe until

recently (ACEM, 2010a).

It appears that the PTW industry is now responding somewhat to pressure

regarding emissions reduction, producing more efficient and less polluting engines

and replacing carburettors with electronic fuel injection, among other measures. A

recent edition of Scooter Magazine in Australia listed 40 mopeds currently available

on the new PTW market, 35 percent of which used four-stroke engines (mopeds

comprised 33 percent of 122 scooters listed) (Anonymous, 2010). Attempting to

address simultaneously the problems of declining sales and environmental

sustainability, Italy provided a ‘scrappage incentive’ to encourage the removal of old

polluting PTWs from service, a program which was at least partly successful.

Similar measures were not taken in other European jurisdictions (ACEM, 2010a).


2.3 Motorcycle, moped and scooter (PTW) safety

As mentioned previously, increased PTW usage has been associated with

increases in the number of reported crashes. Trends in overall reported PTW crashes

in developed countries have been outlined in the introduction to this chapter (section

2.1). However, the trends for all PTW crashes do not necessarily reflect trends for

moped and scooter crashes specifically. Analyses of motivations and usage patterns

have been used in research to identify different behavioural characteristics among

riders, and to subsequently develop risk models on the basis of group segmentation

or categorisation (Christmas, Young, Cookson, & Cuerden, 2009; Harrison &

Christie, 2005; Sexton, Hamilton, Baughan, Stradling, & Broughton, 2006). Some of

these models are discussed below in the current section on PTW safety, but at this

point it is worth noting that motivations for riding, riding patterns and choice of PTW

type have been found to influence crash risk (Morris, 2009; Sexton, Baughan et al.,

2004).

2.3.1 Crash rates and crash severity

Much of the research on crash and injury rates compares PTWs collectively

with cars. This research suggests that fatality rates per distance travelled in

developed countries are 20 to 40 times higher for PTWs than for cars (Department

for Transport, 2010; Huang & Preston, 2004; Jamson & Chorlton, 2009; Johnston,

Brooks et al., 2008; NHTSA, 2007; Lin & Kraus, 2009). Studies of the relative crash

and injury risks of mopeds and motorcycles are comparatively scarce and the results

are somewhat inconsistent across developed countries. Research on moped crash

risk compared to that of larger scooters has not been identified in this literature

review, as larger scooters (over 50cc) are either subsumed within motorcycle data, or

grouped with mopeds for comparison against motorcycles.

2.3.1.1 Crash rates

Research from Europe suggests that mopeds have a lower crash risk and

lower injury severity risk than motorcycles in some countries (Koornstra, Lynam,

Nilsson, Noordzij, Petterson, Wegman, & Wouters, 2002; Noordzij, Forke et al.,


2001; Yannis, Golias, & Papadimitriou, 2005), but the reverse pattern is found in

others (Aare & Holst, 2003; Koornstra, Lynam et al., 2002; Sexton, Baughan et al.,

2004). The fatality rate for moped riders per distance travelled in the Netherlands

was approximately double that of the UK, while the opposite was true of motorcycle

fatality rates (Koornstra, Lynam et al., 2002). A study of hospital and police-

reported fatalities in Sweden found a relatively high crash risk per distance travelled

for mopeds, with moped riders being twice as likely as motorcyclists and 20 times as

likely as car occupants to be killed (Aare & Holst, 2003). The fatality rate for

mopeds was also reported to be twice that of motorcycles in Denmark, while the

fatality and injury rate per registered vehicle was reported as 2.6 times higher for

mopeds compared to motorcycles (Christie, 2008). The MAIDS report on PTW

crashes in five European countries (France, Germany, Italy, the Netherlands and

Spain) showed that mopeds and motorcycles were similarly crash-involved relative

to exposure (ACEM, 2008a).

Research in the UK on motorcycle crash risk (Sexton, Baughan et al., 2004)

found that motorcycles with engine size above 125cc had a 15 percent lower crash

risk than those with smaller engines. This suggests that mopeds and small scooters

may be more likely to be involved in a crash than larger PTWs in the UK. However,

separation of the PTW types is required to confirm this and the MAIDS report

indicated no relationship between engine size and crash risk for neighbouring

European countries (ACEM, 2008a). Other research suggests that crash risk is

associated more with motorcycle type rather than engine size, with greater risk

observed among sport motorcycles (Morris, 2009; Teoh & Campbell, 2010).

The inconsistent findings on moped and motorcycle crash rates likely relate

to differences in usage and regulations between research locations. The reliability of

data and the research methods used may also contribute to the different research

findings. In Australia, the reliability of estimated relative crash risks of mopeds,

scooters and motorcycles is compromised by insufficient denominator data (Christie,

2008; Haworth, Nielson et al., 2008). The different licensing requirements for

moped riding across jurisdictions also limit the transferability of findings to

Queensland from other locations.

In an exposure study conducted in New South Wales in 2002-2003, the self-

reported crash rate of moped and scooter riders combined (n = 31) was the lowest of

all the PTW types. The reported crash rate per million VKT was 4.0 for mopeds and


scooters, compared with 10.1 for sport motorcycles and 23.5 for trail/dual use

motorcycles. Crash rates of around 6.0 per million VKT were reported for other

motorcycle types, and 9.6 for the entire sample (Harrison & Christie, 2005). Moped

and scooter riders in New South Wales require a motorcycle licence, which may be

obtained using a PTW with automatic transmission during testing (which will almost

always be a scooter or moped).

Reviewing other jurisdictions in a report on scooter crashes in Victoria,

Christie (2008) cited data from Western Australia where the reported moped and

motorcycle crash rates were 137 and 289 respectively per 10,000 registered vehicles.

These figures represent the 10 year average to 2004, but again the reliability is

compromised by low numbers of moped crashes (a range of 2-9 crashes per year).

Given that motorcycles travel much further on average than mopeds, the moped

crash rate per distance travelled may be still be higher than that for motorcycles,

despite these reported crash rates per registered vehicle.

Another issue concerning estimated comparative crash rates is that moped

exposure in terms of time over a given distance is likely to be greater than that of

motorcycles and larger scooters. This difference can result from the performance

restrictions applied to mopeds, as well as from their use in predominantly lower

speed zones. In the relevant literature identified, estimated crash rates per distance

travelled do not account for the greater amount of time spent by moped riders to

cover a given distance compared to motorcycle riders.

As noted above, exposure estimates using numbers of registered vehicles are

limited in their usefulness for comparing crash rates of the PTW types. Arguably

more reliable are crash rates based on distance travelled (despite some problems as

mentioned above), but these data are scarce for moped and scooter use, particularly

in Australia. In addition to these measures of exposure and related crash risk, the use

of a quasi-induced exposure procedure is also considered a potential means to

strengthen research methodologies with regard to crash risk. The method essentially

selects from multi-vehicle crashes those drivers (or riders) who were considered not

to be at fault by reporting authorities. These data are then used to estimate the

exposure of particular groups in terms of their presence in the traffic system,

providing a reference against which their involvement in all crashes can be

compared.


In particular, the quasi-induced exposure method has been used to compare

the crash risk of licensed and unlicensed drivers (DeYoung, Peck, & Helander, 1997;

Watson, 2004). In this context, the method suggests that if five percent of not at fault

drivers in multi-vehicle crashes are unlicensed, then approximately five percent of all

active drivers are unlicensed. Applying the method to PTW use, if a quarter of all

multi-vehicle PTW crashes involving a not at fault rider involve mopeds, then

mopeds are assumed to represent a quarter of all active PTWs. For the particular

groups of interest, overall crash involvement is then measured against their estimated

presence in the traffic system. The estimated presence is assumed to be roughly

indicative of time spent in the traffic system rather than of distance travelled, which

partly overcomes the problem regarding distance over time as noted above.

Although potentially useful, the quasi-induced exposure method has some

inherent limitations which impact its usefulness in the current research. As with

exposure measures using registration or licensing data, the quasi-induced exposure

method does not measure the extent of actual usage. Using multi-vehicle crashes

only, the method relies on an assumption that of all crashes for each of the groups to

be compared (in this case mopeds, scooters and motorcycles), similar proportions are

multi-vehicle crashes (DeYoung, Peck et al., 1997). Some research has found

statistically significant differences between mopeds and motorcycles in multi-vehicle

crash involvement (ACEM, 2008a), while other research has not (Haworth &

Nielson, 2008). Neither of these studies separated larger scooters (over 50cc) from

motorcycles. DeYoung et al. (1997, p. 20) note also that ‘innocent’ groups may

differ in terms of ‘characteristics which make them more likely to be involved in

traffic crashes, in other words, if they are crash prone’. Therefore, if moped riders

(for example) are more ‘crash prone’ than motorcycle or scooter riders regardless of

fault (due to poorer hazard perception, for example), their presence in the traffic

system will be overestimated by the quasi-induced exposure method.

2.3.1.2 Crash severity

Numerous studies have found that injuries to scooter, moped and light

motorcycle riders are less severe than those to other PTW riders, due in part to their

lower travel speeds and subsequently lower speeds of impact with obstacles (ACEM,

2008a; Albalate & Fernandez-Villadangos, 2009; Bostrom, Wladis, & Nilsson, 2002;


Christie, 2008; de Lapparent, 2006; Langley, Mullin, Jackson, & Norton, 2000;

Moskal, Martin, Lenguerrand, & Laumon, 2007; Noordzij, Forke et al., 2001; Otte,

Willeke, Chinn, Doyle, & Schuller, 1998). In particular, moped riders generally

appear less likely to be killed if involved in a crash. In the European MAIDS data,

around six percent of L1 (moped and mofa) crashes were fatal, compared with 15

percent of L3 (motorcycle and larger scooter) crashes (ACEM, 2008a). The relative

risk of non-fatal injury may also be somewhat lower in moped crashes, though this is

less clear and the difference appears less than that observed in fatal crashes.

Research using Swedish hospital data found slightly lower injury severity

among moped riders compared with motorcycle riders. Hospitalised moped riders

spent 11.3 days in treatment on average, compared with 12.3 days for motorcycle

riders (Aare & Holst, 2003). Research in the UK also found that smaller engine size

was associated with lower injury severity and fewer fatalities (Sexton, Baughan et

al., 2004), though this research does not separate mopeds, scooters and light

motorcycles in the data. Examining motorcycle crash severity in Singapore, Quddus,

Noland and Chin (2002) also found a large (202%) increase in the probability of a

fatality for large (1500cc) motorcycles relative to small motorcycles (170cc), but this

study does not separate mopeds or scooters either. The general finding of lower

severity in moped crashes compared to motorcycle crashes is not supported in all

research (Haworth, Nielson et al., 2008; Kopjar, 1999) and there are contextual

differences (i.e. rural/urban, licensing age, helmet laws) which influence research

findings.

Previous research in Queensland found that distributions of crash severity

were similar for reported moped and motorcycle crashes from 2001 to 2005

(Haworth, Nielson et al., 2008). A fatality resulted from 2.8 percent of police-

reported motorcycle crashes and 1.3 percent of moped crashes. Hospitalisation

resulted from 46 percent of motorcycle crashes and 42 percent of moped crashes. A

further 32 percent of motorcycle crashes and 38 percent of moped crashes resulted in

medical treatment. The differences in crash severity were not statistically significant.

The authors note that the number of moped crashes was relatively small and the data

did not identify injury types or duration of hospitalisation, which may have differed

between moped and motorcycle riders (Haworth, Nielson et al., 2008).

Crash data have been examined using logistic regression, including ordered

probit and stepwise models, among others, to determine how various factors


including rider and roadway characteristics interact to influence injury and vehicle

damage severity (Quddus et al., 2002; Zambon & Hasselberg, 2006). Quddus et al.

(2002) opted to use an ordered probit model because it accounts for the ordinal

nature of the categorical dependent variable, in this case severity. Traffic

environment, roadway type and geometry, collision type, time of day, day of week,

motorcycle engine capacity, and rider age and gender were included in the model.

With these variables controlled for, results included that motorcycle crashes were

more severe on bends, during early morning hours (12am-8am), on higher speed

roadways, involving at-fault riders, involving riders over 60 years of age, for crashes

into stationary objects (single vehicle) and for larger motorcycle crashes (1500cc).

As noted above, the study did not attempt to classify or separate PTW types.

Using stepwise logistic regression, Zambon and Hasselberg (2006) included a

similar range of variables to Quddus et al. (2002) to model factors influencing

motorcycle crash severity. After controlling for these variables in the model, the

adjusted odds of severe or fatal injury were higher in crashes at night time, on

Saturdays, in rural areas and higher speed zones (>50 km/h), and where alcohol

involvement was suspected. Lower odds (OR 0.8) of a severe or fatal injury were

reported for larger engine sizes (over 125cc) compared to smaller engines (up to

125cc). The result is suggestive of a protective effect of larger motorcycles, though

the study did not consider any other vehicle factors in the regression analysis.

Zambon and Hasselberg (2006) reported no significant effects of rider age or gender

on crash severity.

An inherent problem with comparing crash severity by PTW type is that it

generally relies on analyses of reported crashes only. Many (or indeed most) crashes

which do not result in injury or significant property damage are unlikely to be

reported. As some level of injury is often the factor which determines whether or not

a crash is reported, analysing only reported crashes does not reveal the severity of all

crashes relative to PTW type. The severity of reported crashes may therefore differ

more or less than the severity of all crashes by PTW type.

2.3.2 Risk factors

Australia has been successful at substantially improving safety for most of its

road users through a focus on known risk factors, but it has been argued that this


success has been less for motorcyclists (de Rome, 2006b). While there are many

factors that increase crash and injury risk for PTW riders, some are more prevalent

than others, and multiple risk factors often combine in crash events. Six main

categories of risk factors have been outlined, including: inexperience or lack of

recent experience; risk taking; driver failure to see motorcyclists; instability and

braking difficulties; road surface and environmental hazards; and vulnerability to

injury (Greig, Haworth, & Wishart, 2007). These groupings of risk factors have been

used as framework (with minor modifications) for the following discussion of the

literature and related material regarding risk and PTW use.

2.3.2.1 Vulnerability to injury

Along with cyclists and pedestrians, PTW riders are often referred to as

vulnerable road users (VRUs) in the literature, due to their lacking the protection

which is usually afforded to car and other vehicle occupants (Constant & Lagarde,

2010; World Health Organization, 2009; Haworth, 2006). For PTW riders, both the

likelihood and severity of injury are greater than for road users who are enclosed

within a vehicle body and thereby relatively more protected. Cars and other two-

track vehicles are now also usually fitted with safety systems such as seatbelts and

airbags which are known to significantly reduce injury. As mentioned previously,

PTW riders in Australia are up to 30 times more likely to die in a crash than car

occupants per distance travelled (Johnston, Brooks et al., 2008).

The most common cause of death for injured PTW riders is an injury to the

head or thorax (as well as multiple injuries), regardless of whether a helmet was

worn or not (Johnston, Brooks et al., 2008; Lin & Kraus, 2009; Morris, 2009).

According to coroners’ reports, head injuries were present in 30 percent of

motorcycle rider fatalities in Australia from 2001-2003, while multiple injuries and

thorax injuries were reported in 21 percent and 12 percent of rider fatalities

respectively (Johnston, Brooks et al., 2008). While helmets do not always prevent

head injury, their use reduces both the frequency and severity of head injuries among

crash-involved PTW riders. Helmets have been shown to reduce the likelihood of

death in a PTW crash by up to 40 percent (Liu, Ivers, Norton, Boufous, Blows, & Lo,

2008; Morris, 2009). Hospital data from the US, where mandatory helmet use laws

were introduced and later repealed in some States, also demonstrate the effectiveness


of helmets in reducing head injuries in PTW riders (Brown, Hejl, Bui, Tips, &

Coopwood, 2011; Lin & Kraus, 2009).

There is almost full compliance with mandatory helmet use laws which

operate throughout Australia for PTW riding on public roads (de Rome, 2006a),

although usage rates are lower off-road and on private property where helmet use is

not mandatory (Blackman, Cheffins, Veitch, & O'Connor, 2009). However, even

where usage rates are high, crash and injury data demonstrate a failure among some

riders to securely fasten helmets, leading to their dislodgement during or prior to

impact (ACEM, 2008a; de Lapparent, 2006; de Rome, 2006a; Johnston, Brooks et

al., 2008). The small proportion (~1%) of riders found not wearing helmets in

Australia are often also found engaged in other forms of risk taking such as riding

unlicensed or riding while over prescribed BAC limits (Haworth, Greig, & Nielson,

2009; Haworth, Smith et al., 1997).

Injuries to the lower extremities are the most common for PTW riders, yet

this area of the body is often poorly protected (de Rome, 2006a; Lin & Kraus, 2009;

Lateef, 2002). In a review of research on motorcycle rider injuries, Lin and Kraus

(2009) reported that lower extremity injuries affected 30 to 70 percent of injured

riders. Lateef (2002) reported that 58 percent of injured riders presenting to a city

hospital emergency department over 12 months had sustained lower limb injuries

(18% sustained head injury). Although jackets, gloves, enclosed footwear and other

items are known to prevent injury and reduce its severity (Haworth, de Rome,

Varnsverry, & Rowden, 2007), their use is not mandatory anywhere in Australia. An

Australian study found moped and scooter riders wore motorcycle pants and boots at

rates of 38 percent and 51 percent respectively, compared with other PTW riders

(motorcyclists) at 61 percent and 78 percent (de Rome, 2006a). Other research

supports the finding that moped and scooter riders use less protective clothing than

motorcyclists in Australia (Christie, 2008). There is currently only limited and

incomplete data with which to analyse the relative injury risk of moped and

motorcycle riding in Queensland (Haworth, Nielson et al., 2008), but it appears that

non-use of protective clothing is a key contributor to vulnerability to injury.


2.3.2.2 Inexperience or lack of recent experience

Inexperience in general and inexperience with a particular vehicle or vehicle

type is a risk factor for PTW crashes (ACEM, 2008a; Haworth, Smith et al., 1997;

Mullin, Jackson, Langley, & Norton, 2000). In terms of general inexperience, this is

evidenced by the overrepresentation of young riders in crashes seen consistently

across developed countries. As young riders generally lack experience and are more

prone to risk-taking behaviour than older riders, the age at which moped riding is

permitted appears to be a critical factor with regard to safety. It has been suggested

that age is a more critical factor than experience, where a propensity among younger

riders for risk-taking places them at greater risk than older riders who are similarly

inexperienced (Rutter & Quine, 1996). Moped crashes in particular tend to involve

high proportions of young riders in many countries. A recent report from the UK

indicated that 57 percent of injured moped (under 50cc) riders were aged below 18

years, while 56 percent of injured riders on larger PTWs (over 500cc) were aged

between 30 and 50 years (Department for Transport, 2010).

In some European countries, moped riding is permitted as early as 14 years of

age and some research suggests that riders are at too great a risk at that age to be

permitted to ride mopeds (SWOV, 2009). However, MAIDS data indicate that PTW

riders below the age of 18 were not overrepresented in crashes relative to riders in a

control group, while riders between 18 and 25 were significantly overrepresented

(ACEM, 2008a). In the MAIDS data, the vast majority of riders younger than 22

years of age were L1 moped riders, with moped riding permitted from 14 years of

age in three of the five countries included in the study (Italy, France and Spain).

Over the last decade, increases in PTW rider fatalities across Australia have

mostly been among riders over 30 years and, particularly, over 40 years of age

(Johnston, Brooks et al., 2008). Similar trends have been reported in other developed

countries including the US (Morris, 2009) and UK (Jamson & Chorlton, 2009),

reflecting changes in rider demographics. There has been an overall decline in

fatalities among riders under 25 years of age in Australia, though they are still

overrepresented relative to their estimated exposure (Johnston, Brooks et al., 2008).

Moreover, younger riders have remained at significantly greater risk per kilometre

travelled despite the increasing popularity (and crash rates) of motorcycling among

older people (ATSB, 2002). It is thought that a return to riding of those who have


had an extended break from motorcycling may partly explain rising fatality rates

among older riders, but so far the evidence for this is inconclusive (Queensland

Transport, 2008). Even if this is the case, some older riders are new riders who may

suffer from a lack of experience in general, rather than a lack of recent experience or

experience with a particular vehicle (Haworth & Rowden, 2010).

Lack of experience is mostly addressed formally through a range of licensing,

training and education programs. These programs generally target new riders

regardless of age, though tend to capture mostly younger riders who comprise the

majority of those seeking a licence. For riders who already hold a licence for PTW

riding, including moped riders requiring only a car licence, such programs are

undertaken voluntarily. Participation in voluntary rider training appears to depend

on a range of factors, though it seems that inexperience is often not the key motivator

for self selection (Haworth, Mulvihill et al., 2006). Some research has shown a

preference among PTW riders for informal learning processes, supported by a belief

that skills and knowledge (and by extension, safety) are accumulated though

experience over time (Natalier, 2001). Rider licensing, training and education are

addressed in greater detail later in this chapter.

2.3.2.3 Risk-taking

In terms of both actual and self-perceived behaviour, the majority of PTW

riders do not generally choose to adopt an unsafe riding style or to deliberately

engage in risky riding (Bellaby & Lawrenson, 2001; Noordzij, Forke et al., 2001).

However, this area of research remains contentious and it is clear that some riders

regularly engage in risky riding by choice, while others do so only occasionally

(Broughton & Walker, 2009). Importantly, research has shown that safe and risky

riding intentions may coexist and are not necessarily mutually exclusive (Watson,

Tunnicliff et al., 2007). Analyses of risk and PTW riding often focus on sensation-

seeking behaviours and emotive factors, but deliberate risk-taking may also be driven

by motives which are rather more utilitarian or instrumental. This section examines

the literature on common forms of deliberate risk-taking among PTW riders, though

the depth and complexity of the topic unfortunately precludes an exhaustive review.


Speeding

Speeding is the most common form of deliberate risk-taking on PTWs, either

by exceeding specified limits or exceeding appropriate speeds for given

circumstances and conditions. The small engine size and low power output of

mopeds relative to larger PTWs suggests that speeding may be less likely for moped

riders than motorcycle riders, and there is evidence in the literature to confirm this

(ACEM, 2008a; Haworth, Greig et al., 2009; Lardelli-Claret, Jimenez-Moleon, de

Dios Luna-del-Castillo, Garcia-Martin, Bueno-Cavanillas, & Galvez-Vargas, 2005).

It is also possible that moped riders are less inclined to ride at speeds excessive for

given circumstances due to psychosocial factors which differentiate them from many

motorcyclists in terms of a propensity for sensation-seeking (Watson, Tunnicliff et

al., 2007). However, despite the performance restrictions applied to mopeds in

standard form, speeding still appears to be a risk factor for moped crashes. The

modification of mopeds to increase engine performance, power output and/or

driveline function is common in Europe (and probably elsewhere), suggesting an

intention among some riders to achieve faster acceleration and/or higher maximum

speeds (ACEM, 2008a; Schoon, 2004). Performance modifications may relate to

attempts to conform with surrounding traffic, to maintaining speeds uphill or while

carrying pillion passengers, therefore not necessarily reflecting intentions to speed.

MAIDS data indicate that modified mopeds were overrepresented in crashes

(ACEM, 2008a). Recent research in the Netherlands identified speeding violations

as the ‘most common aberrant behaviour among moped riders’ (Steg & van Brussel,

2009), although these were all young riders (age 16 – 25).

Alcohol and drug impairment

Riding under the influence of alcohol and/or drugs is another prominent risk

factor for PTW crashes. According to MAIDS data, riders under the influence of

alcohol were overrepresented in crashes by a factor of 2.7 relative to exposure data

(ACEM, 2008a). As well as clearly reducing a rider’s ability to control a PTW and

to perceive and respond to hazards, research has also found an association between

alcohol use and other risk factors such as speeding and non-use of helmets (Haworth,

Greig et al., 2009; Haworth, Smith et al., 1997; Lardelli-Claret, Jimenez-Moleon et

al., 2005).


The prevalence of alcohol and drug use among riders varies internationally

and also within Australia. In Australia, some research has found lower rates of drink

driving among riders compared to car drivers, while there is also some evidence that

riders are more likely than car drivers to be under the influence of other drugs,

particularly cannabis (Haworth, Smith et al., 1997; Sheehan, Siskind, Turner, Veitch,

O’Connor, Steinhardt et al., 2008; Watson, Tunnicliff et al., 2007). Recent data from

the UK also shows lower rates of alcohol impairment among fatally injured riders

(9%) compared to other vehicle drivers (20%) (Department for Transport, 2010).

The situation is different in the US, where higher rates of alcohol impairment have

been observed among riders. Motorcyclist fatalities in the US in 2006 involved an

impaired rider (BAC >.08) in 27 percent of cases, compared with 23 percent of

drivers in fatal car crashes (NHTSA, 2007).

There has been only limited research that compares the prevalence of alcohol

or drug impairment between moped and motorcycle riders. MAIDS data indicate

that 6.6 percent of crashed moped riders were impaired by alcohol or drugs (mostly

alcohol), compared with 3.3 percent of motorcyclists, though overall rates of

impairment were low (4.4%) compared to other studies (ACEM, 2008a). Analysis of

police-reported data in Queensland suggests similarly low rates of alcohol

impairment among PTW riders (~5%), with no significant difference between moped

and motorcycle riders (Haworth, Greig et al., 2009).

Unlicensed riding

Unlicensed riders have been shown to be significantly overrepresented in

crashes in Australia and elsewhere and unlicensed riding is therefore considered a

risk factor for PTW crashes (Haworth, Smith et al., 1997). While unlicensed riding

itself is somewhat intangible as a risk factor, associations of unlicensed riding with

other known risk factors such as alcohol and drug impairment, speeding and non-use

of helmets has been demonstrated (Haworth, Smith et al., 1997; Watson &

Steinhardt, 2006). Although the contribution of unlicensed riding to crashes when

isolated from other risk factors is unclear, it is possible if not likely that some

unlicensed riders will lack experience and/or skills which enable them to effectively

control a PTW in hazardous situations (Haworth, Greig et al., 2009).

The prevalence of unlicensed PTW riding in Australia varies according to

jurisdiction and PTW type. As noted previously, moped riders in some Australian


jurisdictions including Queensland require a car licence only, not a PTW licence. In

Queensland crash data from 2001 to 2005, 5.2 percent of moped riders and 4.5

percent of motorcycle riders were found to be unlicensed. Scooters (LC) were

included in the data set as motorcycles in this research, but the low number of

scooters relative to motorcycles is unlikely to have affected findings significantly

(Haworth, Greig et al., 2009). Other research in Queensland over the last decade

found that more than 10 percent of PTW riders involved in police-reported crashes

were effectively unlicensed, although these studies did not separate mopeds and

scooters from other PTWs (Blackman, Veitch, & Steinhardt, 2008; Watson &

Steinhardt, 2007). Research in Victoria found approximately six percent of all

injured riders to be unlicensed in 1996 (Haworth, Smith et al., 1997). A more recent

analysis of police-reported scooter crashes in Victoria between 2001 and 2006 was

roughly consistent with this, showing five percent of motorcycle riders to be

unlicensed (Christie, 2008). Interestingly, this later study separated PTW types and

found that, in contrast to motorcycle riders, less than one percent of scooter riders

(including moped riders) were known to be unlicensed (licence type was unknown

for 11% of scooter riders).

2.3.2.4 Conspicuity issues and other road users

The failure of other drivers to see PTWs is a prominent contributor to PTW

crashes with other vehicles, often involving ‘right of way’ violations. Right of way

violations involving another vehicle turning in front of a PTW proceeding straight

ahead have been cited as the most frequent multi-vehicle PTW crash configuration

(Huang & Preston, 2004). This was identified in early research (Hurt, Ouellet, &

Thom, 1981) and continues to be seen as a primary factor in multi-vehicle PTW

crashes (ACEM, 2008a; Comelli, Morandi, Magazzu, Bottazzi, & Marinoni, 2008;

de Rome, 2006a; Johnston, Brooks et al., 2008; Wells, Mullin, Norton, Langley,

Connor, Lay-Yee, & Jackson, 2004). The failure of other drivers to see or respond

appropriately is related to the conspicuity of the PTW. ‘Conspicuity’ has been

defined as ‘the susceptibility of an entity to be detected by a road user’ (Comelli,

Morandi et al., 2008, p. 71).

Crashes in which another vehicle driver has failed to see or failed to respond

to the presence of an approaching PTW have been referred to as ‘looked but failed to


see’ crashes in research literature (Broughton & Walker, 2009; Brown, 2005). The

phrase ‘sorry mate I didn’t see you’ (and its variants) as a response from other

vehicle drivers who have collided with PTWs has been noted so frequently that these

crashes are also sometimes referred to as ‘SMIDSY’ crashes in research and other

literature (Broughton & Walker, 2009; MAG UK, 2006; South Gloucestershire

Council, 2008).

The difficulty that other road users have in detecting and appropriately

responding to PTWs evidently stems from two main factors: the small frontal area of

the vehicle and rider combined, and the tendency to misjudge the approaching PTW

speed. These factors may be exacerbated where there is a lack of contrast of PTW

and rider with the background and surrounding environment. The use of bright,

reflective or fluorescent colours for rider apparel and vehicle bodies may help to

reduce conspicuity-related PTW crashes (Comelli, Morandi et al., 2008). However,

a more effective solution seems rather more complex according to some research

(Gershon & Shinar, 2010) and would take into account changing contrasts as the

PTW and rider move through different environments.

Some research has suggested that mopeds and scooters may be even more

difficult than motorcycles for other road users to detect due to overall smaller

dimensions and smaller frontal area (de Rome, 2006b). This may be generally true,

but some mopeds and scooters have a larger frontal area than many motorcycles.

Moreover, analysis of MAIDS data showed that conspicuity-related crashes

increased with PTW engine size and that this was not related to a difference in speed

(though may have been related to differences in acceleration). This research

suggests that mopeds were more conspicuous in traffic than motorcycles due to the

riders’ choice of clothing and equipment (Comelli, Morandi et al., 2008). While

differences in conspicuity may warrant further attention, it is nonetheless clear that

motorcycles, scooters and mopeds alike are comparatively small in relation to other

vehicles and are therefore difficult for other road users to detect.

Failure to see PTWs explains very broadly of one of the main causes of PTW

crashes with other vehicles, but although other vehicle drivers are found to be at fault

in the majority of such crashes (ACEM, 2008a; Kim & Boski, 2001), PTW riders are

at fault approximately one third of the time. The relationship between fault

attribution and crash characteristics in multi-vehicle PTW crashes has been explored

using multivariate probit, binary logit and other logistic regression models to explore


potential interactions and correlations among various factors (Kim & Boski, 2001;

Haque, Chin & Huang, 2009; Scneider, Savolainen, Van Boxel & Beverley, 2012).

Schneider, Savolainen et al. (2012) found after controlling for other factors in a

multivariate probit model that for both riders and other vehicle drivers alike, multiple

risk factors were correlated and increased the odds of being at fault. These included

being a younger driver or rider, riding/driving under the influence of alcohol, being

uninsured, and being unrestrained/unhelmeted. Other vehicle drivers had higher

odds of being at fault at intersections and driveways, and in conditions that decreased

PTW conspicuity (such as darkness), while the odds of at-fault PTW riders were

higher in rear-end crashes. The study found that riders of newer PTWs were at

higher odds to be at fault, but PTWs were not separated by type or engine capacity to

examine other vehicle characteristics.

Haque, Chin and Huang (2009) used a logistic regression with at-fault/not-at-

fault as the outcome variable to examine the influence of various PTW crash

characteristics and factors on the odds of either outcome. The conclusions include a

negative relationship between at-fault status and increasing rider age (up to 60 years).

Night time crashes were also associated with lower odds of the PTW rider being at-

fault and this was attributed at least partly to poor conspicuity in dark conditions.

Higher speed roads (>70 km/h) were associated with higher odds of an at-fault rider,

while lower odds were reported for lower speed roads (<70 km/h). PTW riders were

more likely to be at fault in crashes with pedestrians, which was also attributed at

least partly to poor conspicuity. Higher PTW engine capacity was associated with

higher odds of an at-fault rider on expressways but not on other roads.

Examining fault in motorcycle crashes in Hawaii using logistic regression

similarly to Haque, Chin and Huang (2009), Kim and Boski (2001) reported that for

PTW riders and other vehicle drivers alike, inattention and misjudgement were the

main factors contributing to fault, and younger and older operators were more likely

to be at fault for both drivers and riders. Differences between riders and drivers

emerged, including that risky behaviours such as speeding, tailgating and improper

overtaking were more associated with at-fault PTW riders, while inattention and

failure to give way were more characteristic of at-fault other vehicle drivers.


2.3.2.5 Vehicle instability and braking performance

PTWs are inherently unstable and exhibit poor braking performance in

comparison with cars (Elliott, Baughan, Broughton, Chinn, Grayson, Knowles,

Smith, & Simpson, 2003). Mopeds and most scooters have smaller wheels and

shorter wheelbases in comparison to motorcycles, making them arguably less stable

than their larger counterparts. In relation to the gyroscopic effect, whereby a

spinning wheel maintains a measure of stability due to dynamic forces produced

(Fajans, 2000), the relative stability is less for a smaller wheel. Thus a small-

wheeled moped or scooter is inherently less stable at a given speed than a motorcycle

with larger wheels, though tyre properties and other factors must also be considered

in any thorough assessment (Evangelou, 2003). While stability might really only be

significantly improved by adding an extra wheel, braking, suspension and other

technology for PTWs has developed considerably over time (Noordzij, Forke et al.,

2001).

Research suggests that many PTW crashes are attributable in part to poor

braking performance and application (ACEM, 2008a, 2010a). Half of all crashes in

MAIDS data involved a PTW braking in collision avoidance manoeuvres, where loss

of control was mainly related to braking (ACEM, 2008a). Measures to address poor

braking performance (including poor brake application by riders) include the use of

advanced braking systems. While all PTWs are fitted with both front and rear

brakes, the front brake provides most of the stopping power and needs to be used

effectively, particularly where emergency braking is required (Broughton & Walker,

2009; Corno, Savaresi, Tanelli, & Fabbri, 2008). A typical moped now has at least

one if not two hydraulic disc brakes, and front and rear brakes are sometimes linked

to be activated together in combined braking systems (CBS). This can help to

overcome deficiencies in riding skill with regard to appropriate brake application, but

is not currently a feature of many mopeds and scooters (ACEM, 2010a, 2008b).

Other technological improvements include anti-lock braking systems (ABS), which

can also prevent loss of control crashes under heavy braking. ABS frequently

appears on larger and costlier motorcycles, but as yet does not commonly appear on

either scooters or mopeds (Greig, Haworth et al., 2007).

Both ABS and CBS are endorsed and promoted by sections of the PTW

industry and their effectiveness has been demonstrated in research. With ABS and


CBS as the cornerstones of current advanced braking system technology, the

European Association of Motorcycle Manufacturers (ACEM) advocates greater use

of these technologies on all PTWs. For mopeds and smaller scooters, from which

advanced braking systems have been largely absent to date due largely to the cost of

components, ACEM suggests that CBS may be more appropriate than ABS (ACEM,

2010a).

2.3.2.6 Environment and road conditions

Environmental hazards for PTWs include poor or contaminated road surfaces,

obstructed field of vision, temporary obstacles, roadside furniture, adverse weather

conditions and poor road alignment and delineation (Andrea, 2006; Berg, Rucker,

Gartner, Konig, Grzebieta, & Zou, 2005; ERSO, 2006; Greig, Haworth et al., 2007).

Some of these hazards may be more likely to affect smaller PTWs and those with

smaller wheels, such as mopeds and scooters. In particular, the smaller wheels and

tyres of scooters and mopeds relative to motorcycles may be more susceptible to

road-based hazards such as potholes, loose surfaces and reverse crossfall (off-

camber) horizontal curves, among others (Cossalter, Doria, Lot, Ruffo, & Salvador,

2003). Poor road and weather conditions are the primary contributing factor in very

few serious crashes overall, but they are seen as secondary factors somewhat more

often (Haworth & Mulvihill, 2006; Haworth, Mulvihill, Wallace, Symmons, &

Regan, 2005; Johnston, Brooks et al., 2008).

Research conducted in various jurisdictions over the last 15 years shows that

between 40 and 60 percent of all PTW crashes analysed occurred at intersections

(ACEM, 2008a; Johnston, Brooks et al., 2008; Kim, Takeyama, & Nitz, 1995). Most

relevant to the current research, Haworth et al. (2007) found approximately half of all

Queensland moped and motorcycle crashes alike occurred at intersections.

Consistent with this, the majority also occurred in low to moderate speed zones

where intersections are more likely to be present (ACEM, 2008a; de Rome, 2006a).

These road environments are of particular interest for moped safety research as they

evidently represent high risk sites in which mopeds frequently operate.


2.3.2.7 Other risk factors

Some of the literature suggests that limiting the speed of mopeds and other

PTWs to below that at which the surrounding traffic travels produces a hazardous

situation, potentially placing riders at risk (Noordzij, Forke et al., 2001; Schoon,

2004; Wegman & Aarts, 2006). The ideal solution appears to be separation of

vulnerable road user groups from other traffic (and each other), but this is extremely

difficult to achieve in practice (Wegman & Aarts, 2006). The Dutch Institute for

Road Safety (SWOV) has recommended that speed limits for mopeds should equal

those imposed on other traffic in most situations (Schoon, 2004). This general

argument is based on a principle of homogeneity, whereby vehicles sharing a

roadway should conform to the speed and direction of the most vulnerable road user.

Whether that means raising the speed of mopeds or lowering the speed of other

vehicles depends on the particular traffic context. Wegman (2006) also notes that a

safe speed for PTWs is yet to be defined on the basis of scientific evidence.

The ability of PTWs to move relatively quickly through congested traffic is

seen as advantageous by riders and is one of the motivating factors for PTW use.

Although the issue is somewhat contentious, ‘lane-splitting’ (riding between lanes of

moving traffic) and ‘filtering’ (riding between lanes of stationary traffic) by PTW

riders is seen as another form of deliberate risk-taking by some observers (Haworth,

Mulvihill & Clark, 2006). In contrast, other research has suggested that the practice

of filtering by PTW riders is generally safe and this particular behaviour warrants

further research attention (Wigan, 2000). It is difficult to obtain and analyse data on

lane splitting and filtering as these behaviours are generally not specifically reported

by police. Further, although these practices often entail one or more traffic violations

in most jurisdictions, the relevant regulations are complex and poorly understood.

Fatigue is an issue that has received relatively little attention in PTW safety

research, despite being a recognised risk factor in other vehicle crashes (Haworth &

Rowden, 2006). The factors contributing to rider fatigue differ somewhat to those

which impact car drivers. However, it is likely that fatigue is related to travel

patterns and trip purpose for both riders and drivers alike. For example, fatigue may

be more prevalent among riders who travel long distances, ride at particular times of

day or night, choose to adopt a physically demanding riding style, or ride particular

PTW types. In a survey of Queensland motorcycle riders, unpublished research


(Tunnicliff, 2006) found that while some participants thought riding while tired to be

dangerous, most did not consider fatigue to be a serious safety issue. No research

has been identified that specifically explores the issue of fatigue in moped or scooter

crashes.

2.3.3 Approaches to understanding PTW rider risk

This literature review has considered theoretical perspectives and approaches

from the fields of psychology, sociology, engineering and urban planning (among

others). In order to adequately understand moped and scooter use in Queensland and

the attendant safety implications of increased usage, it is necessary to examine a

broad range of factors concerning behaviour and environment. While there is

arguably no single theoretical approach or perspective which accommodates the

overall objectives of the current research, the following approaches and models have

been useful.

Social Cognitive Theory, Health Belief Models, the Theory of Reasoned

Action and its variant the Theory of Planned Behaviour, among others, are all well

established approaches to explaining and predicting behaviour from a psychological

perspective (Tunnicliff, 2006; Weinstein, 1993). A general assumption among the

approaches is that motivation to adopt protective behaviour will be provided by a

desire for positive health outcomes. Individual, social and environmental influences

have been considered and accounted for to varying degrees, while over time

perceived deficiencies in a given theory or model have led to various attempts at

refinement (Tunnicliff, 2006). In practical terms the approaches typically involve

recording responses to a range of structured questions or tests, the results of which

can be measured using various statistical analyses. These methods are useful for

predicting risky behaviour with reasonable accuracy through rigorous statistical

analyses, but have also been criticised for revealing little about social and cultural

contexts within which individuals make decisions (Nja & Nesvag, 2007; Tunnicliff,

2006).

Social-cultural and qualitative perspectives on PTW safety have been offered

in relatively few road safety research papers (Coxon, 2002; Nja & Nesvag, 2007;

Natalier, 2001). Much of the motorcycle safety literature identifies social

characteristics that influence risk exposure, ‘but fail(s) to relate those characteristics


to social action’ (Natalier 2001, p. 66). Natalier stresses that the ‘lived experience’ is

important in terms of how motorcyclists perceive and interpret risk, and suggests it

cannot be adequately explained by statistical analyses. A tension between risk as

perceived by riders and ‘the system oriented perspective’ has also been observed

using an anthropological approach (Nja & Nesvag, 2007). Further research on social

and cultural influences on road safety, including motorcycles and scooters, has been

advocated in a number of other works (Factor, Mahalel, & Yair, 2007; Haworth,

Nielson et al., 2008; Tunnicliff, 2006; Vick, 2006). The current research seeks to

improve the understanding in this area

Research has used surveys to explore PTW rider characteristics, motivations,

behaviours and self-reported crash involvement and to develop models in which

groups of riders are separated on the basis of risk and related characteristics. For

example, through focus groups and rider surveys, qualitative research for the UK

Department for Transport (Christmas, Young et al., 2009, p. 1) identified seven

groups of riders, to each of which was assigned a risk management approach the

riders were most likely to employ:

Performance disciples: precautionary fatalism; Performance hobbyists:

cautious attraction; Riding disciples: active management of risks; Riding

hobbyists: personal responsibility for avoiding risk; Car rejecters: high

awareness and high unhappiness; Car aspirants: low awareness but high

educability; Look-at-me enthusiasts: blasé confidence.

The authors note that this segmentation of groups of riders based on shared

characteristics is an ‘imperfect simplification’. However, they argue that it reflects

the diversity of riders in the population and that it is preferable to the stereotypes

derived from anecdotes of riders (Christmas, Young et al., 2009). Moped and

scooter riders are not identified in this summary report, but on the basis of other

research it seems likely they would be classified as Car rejecters or Car aspirants

based on the following descriptions:

• Car rejecters. These are escapees (a higher proportion of women than in

any other segment) from traffic jams, parking tickets, fuel costs and other

problems of car use – who don’t care for motorcycles, but do care for low-

cost mobility. Risk management approach: very sensitive to the risks of


riding, and see this as a strong argument against riding.

• Car aspirants. These are young people looking forward to getting their first

car when age/finances allow – but for the time being just happy to have got

their own wheels. Risk management approach: tend not to think about the

risks of riding and as a result may not take steps to manage them; but signs

that they will take steps when the risks are pointed out to them.

In another approach, data on self-reported crash involvement in New South

Wales were used to perform a cluster analysis to identify groups of PTW riders who

were at greater risk of crashing per distance travelled than other groups (Harrison &

Christie, 2003). Nine groups were identified based on similar demographic

characteristics and riding patterns, three of which (groups 1, 2 and 7) had higher

crash rates than the remaining six. These three clusters are as follows (Harrison &

Christie, 2003, p. 30):

• Cluster 1 consists of 14% of the sample of participants. They ride less

distance each year than average, tend to ride on two-way roads in urban

areas, tend to ride on weekends for pleasure, are more likely to ride a

traditional-style motorcycle, and are more likely to live in Sydney.

• Cluster 2 consists of 12% of the sample. Riders in this cluster tend to ride

more often on urban and rural freeways and multilane highways. They ride

more often on weekends and for pleasure than the sample as a whole, and are

more likely to ride sports-style motorcycles and to live in the Sydney area.

• Cluster 7 consists of 6% of the sample. Riders in this cluster rode less than

other riders, rode on weekends for pleasure on trail bikes, in off-road

contexts.

The authors of this report suggest that the cluster analysis is useful for

identifying high risk groups for the purpose of developing targeted interventions.

However, the clusters as defined may not be useful for targeting scooter riders who

represented only a small proportion of the overall sample. The largest proportion of

scooter riders were identified as belonging in Cluster 1, seemingly due to their

relatively low exposure (VKT) and their tendency to ride mainly on urban two-way


roads. However, the tendency of riders in Cluster 1 to ride on weekends and for

leisure is not consistent with what is known about scooter riders from other research.

2.3.4 Potential of licensing and training to improve rider safety

There are numerous countermeasures and initiatives aimed at reducing PTW

crashes and related trauma. These include rider training and licensing programs,

awareness campaigns for riders and other road users, infrastructure and

environmental treatments, promoting use and knowledge of protective clothing,

educational resources and modifications to legislation (Queensland Transport, 2008;

Greig, Haworth et al., 2007). Of particular interest to the current research are

licensing and training issues since they are potentially the major factor influencing

safety as well as the extent of PTW usage. The current section discusses the research

on rider licensing and training, most of which has focused on motorcycle rather than

moped or scooter use.

Licensing requirements and conditions for moped and scooter use in

Australia, Europe and North America have been summarised in the previous chapter

(Introduction). As noted, licensing systems vary considerably among and within

countries. In Australia, licensing is currently regulated at State or Territory level and

the regulatory differences among jurisdictions appear to have promoted differences

in the usage of PTW types. Compared to Queensland (and possibly other

jurisdictions) where mopeds can be ridden on a car licence, riders in Victoria and

NSW tend to ride larger capacity scooters rather than mopeds as all riders require a

motorcycle licence. Training providers in Victoria have estimated that up to one

third of PTW riders attending weekend training courses are scooter riders (Haworth,

Greig et al., 2008).

PTW rider licensing systems often incorporate compulsory practical rider

training components as well as theoretical and educational components. Other

licensing systems only require demonstration of competency through practical

testing, as well as knowledge tests, while rider training is undertaken voluntarily.

Hazard perception tests are also incorporated into licence tests in some jurisdictions.

Notwithstanding these variations, PTW rider licensing systems usually incorporate

some combination of rider training, skills testing and education elements into

licensing processes. It has been suggested in previous research that licensing and


training systems have greater potential to reduce PTW rider crashes and injuries by

reducing the amount of riding rather than by reducing crash risk per distance

travelled (Haworth & Mulvihill, 2005). This is despite the explicit objective of rider

licensing and training to ultimately produce safer riders.

Historically, moped riders in many jurisdictions have been exempt from some

or all of the testing and training requirements which apply to riders of larger PTWs

seeking a licence. Such exemptions continue to apply in many places, including

Queensland. The rationale for such exemptions essentially relies on the low power

and limited speed of mopeds in comparison to other PTWs, as well as a presumption

that knowledge of basic road rules has been attained through completion of basic

road safety education and/or acquisition of a car licence or learners permit. Riders of

scooters with larger engines and higher power outputs are typically required to obtain

a standard motorcycle licence, though in some jurisdictions an ‘automatic only’

condition may exempt them from testing on geared PTWs. While it seems intuitive

to expect that moped riders without a specific PTW licence would be at greater risk

of crashing than those who hold a PTW licence, this has not been clearly

demonstrated in the literature.

Graduated licensing systems in their various forms are essentially designed to

reduce the exposure of inexperienced riders and drivers to high risk situations, and

are generally seen as representing best practice in novice and young driver licensing

(de Rome, Ivers, Haworth, Heritier, Fitzharris, & Du, 2010; Reeder, Alsop, Langley,

& Wagenaar, 1999; Vanlaar, Mayhew, Marcoux, Wets, Brijs, & Shope, 2009).

Reductions in injuries and fatalities in target groups (including young and

inexperienced PTW riders) have been observed in association with the

implementation of graduated licensing systems (Reeder, Alsop et al., 1999).

However, as with training programs, the actual effects of graduated licensing systems

are difficult to separate from the effects of other factors including various

interventions which may be implemented simultaneously. Additionally, graduated

licensing systems for PTW riders differ in some ways from those which regulate

novice car drivers and there is a need for these differences to be more clearly

understood and articulated (de Rome, Ivers et al., 2010; Haworth & Rowden, 2010).

Where moped riding is permitted for car licence holders, access to mopeds is granted

in the early stages of the graduated licensing process.


Regardless of whether or not it is required for licensing, rider training and

education has historically been seen as important for improving rider safety, and

continues to be widely promoted by researchers and the PTW industry (ACEM,

2010a; Bowdler, 2011a; Buche, Williams, & Ochs, 2010; Hurt, Ouellet et al., 1981;

Haworth, Mulvihill et al., 2006). However, the actual effectiveness of particular

programs has not been clearly demonstrated in many cases and some training

programs have been associated with elevated crash risk (Haworth & Rowden, 2010;

Haworth & Schulze, 1996; Savolainen & Mannering, 2007). The lack of positive

training program evaluations is not generally seen to reflect the failure of training per

se, but the need for more effective program design (Rowden, Watson, & Haworth,

2007). Training program evaluations have also been typically compromised by

methodological problems, which may help to explain the limited number of

evaluations published in the literature to date. Where evaluations of individual

programs have shown a positive or negative effect, the relevance to other

jurisdictions depends on a degree of similarity in regulatory, economic and cultural

environments.

In Australia, rider training and education mostly targets motorcyclists

generally and does not often specifically address moped or scooter use (Greig,

Haworth et al., 2007). Additionally, rider training has historically focused on vehicle

control skills, with little attention to attitudinal and behavioural issues. Such an

approach may fail to recognise different training needs of riders of different PTW

types. A review of PTW crash countermeasures potentially relevant for Queensland

noted that moped and scooter riders may have specific training needs due to different

performance and design characteristics compared with motorcycles (Haworth &

Rowden, 2010). Moped and scooter rider training is available in some jurisdictions

including Queensland, generally consisting of a modified (shorter) version of basic

motorcycle rider training courses. However, such training is voluntary and

discussions with training providers suggest that uptake of these courses by new and

existing moped riders is low (Haworth, Greig, & Wishart, 2008).

There are few formal evaluations of moped rider training, education and

skills testing. Two evaluations in Europe failed to show clear safety improvements

from moped rider training programs. While these evaluations are arguably of limited

relevance to Australia as they involved young riders (14-16 years of age) who were

not car licence holders, they are nonetheless summarised below in the absence of


more relevant research.

A study in the Netherlands compared a group of trained riders with a group of

untrained riders in 2000 – 2001 (Goldenbeld, Twisk, & de Craen, 2004). Both

groups of riders had successfully completed theory-based testing. Trained riders

tested two weeks after training demonstrated better riding performance than

untrained riders, but these benefits were not always retained in follow-up testing 11

months later. Riders without formal training generally improved their skills over one

year, while trained riders showed either no further improvement or deterioration in

skills on follow-up testing (Goldenbeld, Twisk et al., 2004). Moped riders in the

Netherlands have been required to gain a moped certificate through completion of a

theory test since 1996, but this policy measure does not appear to have reduced

moped crash involvement (Steg & van Brussel, 2009). Further, reports published in

2001-2002 suggest that 29 percent of Netherlands moped riders were effectively

unlicensed well after the introduction of the moped certificate (SWOV, 2006b).

An evaluation of a moped rider training program in Portugal showed that in

the four years after commencement, 52 percent of the experimental group (trained

riders, N = 190) reported crashes, compared with 31 percent in the control group

(untrained riders, N = 84) (Antonio & Matos, 2008). Although this is a relatively

small sample of the riders trained since program commencement (over 2,000 riders

submitted in the first year), the result was contrary to the expectation that training

would reduce moped rider crashes. The experimental group showed higher levels of

rule-following and self awareness of internal risk factors (psychological and

physical), and were less likely to have traffic offences compared to the control group,

but these differences reduced over time. Program participants also showed higher

levels of vehicle knowledge and awareness of rider equipment and safety accessories

(in line with program goals). Different rates of exposure were suggested as a

possible explanation for the unexpected results on crash involvement (Antonio &

Matos, 2008). The lack of exposure data combined with the small sample size

further highlights the lack of rigour often encountered in evaluation studies.

The MAIDS report indicates some differences between trained and untrained

riders in crash data, although conclusions regarding the benefits of training are

compromised by confounding factors. Of crash-involved riders, 75 percent of L1

(moped) riders had undertaken no training, compared with 14 percent of L3

(motorcycle and larger scooter) riders. For both vehicle categories combined, a


collision avoidance manoeuvre (braking and/or swerving) was attempted by 67

percent and 53 percent of trained and untrained riders respectively. While this

suggests that trained riders may demonstrate better hazard perception and response,

in many cases there was insufficient time for any such manoeuvre and the results are

therefore difficult to interpret (ACEM, 2008a).

Despite the lack of positive training program evaluations, the US-based

Motorcycle Safety Foundation (MSF) currently advocates ‘lifelong training’ through

its rider education and training system (Buche, Williams et al., 2010). In some cases

research has found elevated crash and fatality risks among riders who had undertaken

MSF rider training (Savolainen & Mannering, 2007). However, noting that many

evaluations, both positive and negative, have been methodologically flawed, the

MSF claims that rider training is effective in improving rider safety if properly

designed and delivered. The MSF is critical of isolated ‘one off’ training

experiences which it claims are of limited value (Buche, Williams et al., 2010). As

the MSF is a commercial training provider this is perhaps to be expected, although

their point appears to be soundly argued.

The MSF provides a scooter-specific rider training course aimed at entry-

level riders to provide them with basic skills and knowledge. ‘Scooter School’

comprises four hours of education and practical training in a controlled setting on

scooters of up to 200cc engine capacity. This course differs from the two-day Basic

Rider Course (BRC) offered by MSF in that it is less comprehensive and does not

qualify graduates for a licence test waiver. The Scooter School course is currently

offered in 12 locations across eight US States (Motorcycle Safety Foundation, 2010),

though the safety effects of the program have not been evaluated.

2.4 Chapter Two summary

The vulnerability of PTW riders is well documented, with the risk of fatal

injury commonly estimated to be 20 to 40 times higher than that of car occupants per

kilometre travelled in developed countries. PTW use has increased over the last

decade or more in many developed countries, including Australia, the US, Canada,

the UK and other European jurisdictions. Coincident with increased usage, some

countries have seen substantial increases in PTW rider death and injury, although


usage and crashes have not necessarily risen at similar rates. In some places,

including Australia and North America, moped and scooter use has increased sharply

relative to that of other PTW types. This has stimulated greater interest in the safety

of moped and scooter riders compared to motorcyclists and also each other.

Most PTW safety research to date has focused on motorcycles and there is

therefore relatively little knowledge about moped and scooter safety. Research that

has focused on moped safety is mostly of European origin, so of limited relevance to

jurisdictions with vastly different cultural, socioeconomic and natural environments.

Some research has considered the relative safety of motorcycles and mopeds, but no

published literature has comprehensively compared mopeds with scooters in regard

to safety. Two key issues which impede a better understanding of moped and scooter

safety are the lack of separation of PTW types in crash data, and the difficulty of

obtaining reliable exposure data with which to estimate crash rates.

Given these gaps in knowledge, greater attention to the use and safety of

mopeds and scooters is warranted, particularly in places where their use has

increased from a low base. Moreover, in places where licensing requirements differ

for moped and scooter riding, it is important to understand the relative safety of those

PTW types. The call for greater research into this topic is partly met by the current

program of research, the design of which is described and discussed in the following

chapter (Chapter Three).

.


CHAPTER 3: RESEARCH DESIGN

3.1 Introduction

This thesis offers a comprehensive examination of the use of mopeds and

motor scooters and the factors which influence rider safety. The previous chapter

showed that while there is a substantial body of motorcycle safety literature from

both international and Australian research settings, there is less known about moped

and scooter safety and the sources of risk to riders. Much of the PTW safety research

does not separate (or include) all of the PTW types and therefore does not allow the

researcher or policy analyst to be able to articulate differences between them with

regard to safety. With few exceptions, those studies that have compared mopeds

with motorcycles have included scooters in one or the other of these groups based on

either legal or conceptual definitions. Further, the transferability of research

findings across countries is limited by differences in regulatory, socio-cultural,

economic and physical environments. Differences in minimum rider age, licensing

and training requirements, usage rates, motivations and environmental conditions

present particular problems for international data comparisons.

In places where moped and scooter use has increased rapidly from

traditionally low levels, safe accommodation of these vehicles in the transport system

presents a challenge for all stakeholders. Understandably, it is generally in such

places that mopeds and scooters have received little research attention due to their

minimal historical contribution to road trauma. As seen in the literature, increased

usage has brought concomitant increases in moped and scooter crashes and related

injury, subsequently generating greater interest in the safety of these PTWs.

Drawing on previous research (Greig, Haworth et al., 2007), the literature

review in the preceding chapter identified and discussed six main contributors to

crash and injury risk for motorcycle riders, including: inexperience or lack of recent

experience; risk taking; driver failure to see motorcyclists; instability and braking

difficulties; road surface and environmental hazards; and vulnerability to injury.

The literature review also suggested that these risk factors do not necessarily apply

equally to moped, scooter and motorcycle use, due to differences among the PTW

types in usage patterns and rider and vehicle characteristics. The two risk factors that


do appear to apply similarly to all PTW riders are other driver failure to see

motorcyclists and inexperience or lack of recent experience of the rider. The

universal applicability of the other four risk factors remains questionable as the

relative safety of scooters, mopeds and motorcycles has not been comprehensively

examined in earlier research.

The limited research on moped and scooter safety and the constraints on

transferring that research across countries indicate a need for further study,

particularly in places where moped and scooter use has increased from traditionally

low levels. For moped and scooter use and related crashes, detailed quantitative

crash and injury data are difficult to obtain, while qualitative material on moped and

scooter rider motivations, experiences, attitudes and behaviour is generally scarce.

The current program of research therefore sought to remedy current gaps in

knowledge regarding moped and scooter use and safety. This chapter sets out the

aims of the research, the research questions and the studies undertaken to answer

them, as well as describing the relationships of these components to each other.

3.2 Research aims

There are two primary aims of this thesis:

1. To develop better knowledge and understanding of moped and scooter usage

trends and patterns.

2. To determine the factors leading to differences in moped, scooter and

motorcycle safety.

To achieve these aims the researcher undertook an analysis of crash and

registration data; an exploration of rider characteristics, motivations, beliefs, attitudes

and experiences; and an observational study in an area of concentrated moped and

scooter usage. It was anticipated that potential areas for further research would be

identified and potential areas for safety-oriented interventions be explored.

The research involved the collection of both quantitative and qualitative data.

A key aspect was concerned with riders’ perceptions of risk and the relationship

between these perceived risks and the objectively determined crash causes, as

discussed in the literature and analysed from the crash data. The cross-disciplinary


approach employed in this research enabled the necessary identification of

behavioural, vehicle and environmental factors in crash and injury causation.

3.3 Research location

The research was geographically focused on Australia’s north eastern State of

Queensland. Previous research has shown that moped and scooter use within

Queensland is concentrated primarily in the urban south east including the Brisbane

and adjacent Gold Coast areas (Haworth & Nielson, 2008). Brisbane is Australia’s

third largest city with a population of approximately two million people. Therefore,

while all of Queensland was included for purposes of data collection and analysis,

the research is mainly focused on the larger urban areas.

As noted previously in the introduction and literature review, licensing

requirements in Queensland differ from those in other Australian States and

Territories and this is thought to underlie differences in usage across jurisdictions.

While moped riding is permitted with a car licence in Queensland, in Australia’s two

most populous States of New South Wales and Victoria a motorcycle licence is

required. It was therefore considered outside the scope of the present study to extend

data collection beyond the Queensland border. The climate in Queensland is more

tropical than in southern Australian States, a factor which also has a likely influence

on PTW usage.

3.4 Research questions

This section presents the four specific research questions and their rationale

based on the review of literature. The introduction to the four research questions is

followed by an outline of the three research studies undertaken in order to answer

them. The rationale and methods employed for each of these studies are described in

full in each of the relevant chapters to follow (Chapters Four to Seven). The current

chapter includes a table summarising the relevance of each of the studies to the

research aims and questions (see Table 3.1 in the following section).


Research question 1: Why has moped and scooter usage increased?

According to the literature, overall PTW use has generally increased across

developed countries over the last decade (ACEM, 2010a; Broughton & Walker,

2009; Johnston, Brooks et al., 2008; Morris, 2009). This is at least partly associated

with urban growth and increasing traffic congestion. While the rate of increase in

PTW use has been rapid in some countries and relatively slight in others, the overall

trend has been consistent. However, within this overall trend, the proportional

increase by PTW type is less consistent. Relative to motorcycle use, moped use has

increased in some locations but decreased in others. Over the last decade, in some

countries (UK and Spain for example) (Department for Transport, 2009; Perez, Mari-

Dell'Olmo, et al., 2009) motorcycle use has increased at a faster rate than moped and

scooter use, while the reverse is true in other locations (Canada and Australia for

example) (Haworth & Nielson, 2008; Motorcycle and Moped Industry Council,

2009). Within Australia over this period, moped use increased at a faster rate in

Queensland than in other Australian States and Territories, and at a faster rate than

motorcycle use within Queensland. Some possible explanations for this have been

offered by previous research. Among these explanations are that licensing

requirements for moped riding in Queensland are less stringent than in some (though

not all) other Australian jurisdictions, and that the climate is conducive to PTW

riding, thereby attracting commuters and tourists (Haworth, Greig et al., 2007;

Haworth, Nielson et al., 2008). Research question 1 seeks to confirm if this is the

case, as well as to explore other explanations for the previously observed increase in

moped and scooter use.

Research question 2: How does the usage of mopeds, scooters and

motorcycles differ?

Previous research has shown that moped, scooter and motorcycle usage

patterns differ, driven partly by different motivations among riders of each PTW type

(Christmas, Young et al., 2009; Harrison & Christie, 2003). It has also been shown

that these differences can have important implications for safety (Christmas, Young

et al., 2009; Haworth, Greig et al., 2009; Moskal, Martin et al., 2010). Motorcycles

are generally seen to be used more for recreational purposes and less for commuting


than mopeds (Broughton & Walker, 2009; Haworth, Greig et al., 2009). There are

exceptions to this generalisation, such as where high levels of recreational moped use

have been found among young riders in European countries with low minimum rider

age requirements (Moskal, Martin et al., 2010). The literature review identified little

research that specifically examined rider motivations and related usage patterns of

scooters which are not mopeds. As mentioned previously, research has tended to

group mopeds and larger scooters together due to conceptual similarities, or to

include larger scooters with motorcycles on the basis of legal definitions.

Additionally, most previous research has been of European and to a lesser extent US

origin, so is of limited relevance to Australia. Recent studies in Australia are few

and these have been based on low moped numbers despite substantially increased

usage. Research question 2 therefore seeks to further explore differences in usage of

mopeds, scooters and motorcycles, in terms of usage trends, usage patterns and rider

motivations.

Research question 3: How does the safety of mopeds, scooters and

motorcycles differ?

It is evident in the research literature that all PTW riders are vulnerable road

users. It is also apparent that some PTW riders are more or less vulnerable than

others and that their relative safety depends on a combination of behavioural, vehicle

and environmental factors. Differences in crash characteristics by PTW type have

been demonstrated and research findings on these appear generally consistent on

numerous variables (including crash type, contributing factors, trip purpose and rider

age and gender). However, there is a lack of consistency in research regarding

comparative crash rates and crash severity of moped and motorcycle crashes and,

again, most studies have not clearly separated larger scooters from the other PTW

types (ACEM, 2008b; Haworth & Nielson, 2008). Different research settings and

different methodologies have produced divergent results, but nonetheless there

appear to be differences in the relative safety of the different PTW types. Research

question 3 therefore further investigates the ways in which the safety of mopeds,

scooters and motorcycles differ.


Research question 4: Why does the safety of mopeds, scooters and

motorcycles differ?

As noted earlier, the safety of mopeds, scooters and motorcycles depends on

a combination of behavioural, vehicle and environmental factors. Six main

contributors to crash and injury risk for motorcycle riders were identified in the

literature review (Greig, Haworth et al., 2007), providing a framework for examining

the relative safety of moped, scooter and motorcycle use. Some of these risk factors

are roughly equally applicable to all PTW types, including inexperience or lack of

recent experience and driver failure to see motorcyclists. The other risk factors,

including risk taking, instability and braking difficulties, road surface and

environmental hazards and vulnerability to injury may be more or less prevalent in

moped and scooter crashes than motorcycle crashes. Although some research has

described the comparative safety of PTW types more or less in such terms, there is

once again little clear separation of mopeds, scooters and motorcycles, and only

limited international transferability of the research (Haworth & Nielson, 2008).

Regulatory differences also limit the transferability research. Research question 4

therefore seeks to further explore and articulate reasons for differences in the safety

of mopeds, scooters and motorcycles.

3.5 Research studies

The literature review summarised what is known from existing research

regarding the research questions listed above. Three separate studies were designed

to answer these questions within the scope of the research program. All studies were

geographically limited to Queensland for reasons outlined previously in sections 1.4

and 3.3. The research studies are not limited to answering only one research

question, and no research question is fully answered by any single study. The

relationship of the studies to the research questions is summarised below in Table

3.1. Greater detail about each of the studies and their relationship to the aims and

research questions is presented in the following sections.


Table 3.1 Relevance of the studies to the research aims and questions

Research aims

1. Develop a better understanding of moped and scooter usage trends and patterns

(relates to Research questions 1, 2 and 4)

2. Investigate factors underlying differences in moped, scooter and motorcycle

safety (relates to Research questions 2, 3 and 4)

Studies

Research questions

RQ1. Why has

moped and

scooter usage

increased?

RQ2. How

does the usage

of mopeds,

scooters and

motorcycles

differ?

RQ3. How

does the safety

of mopeds,

scooters and

motorcycles

differ?

RQ4. Why

does the safety

of mopeds,

scooters and

motorcycles

differ?

Study 1:

Observation of

PTW types in

inner city area

Monitoring

usage trends

Demonstrates

inner urban

moped/scooter

concentration

Vehicle

characteristics

& usage

patterns

Study 2:

Analysis of crash

& registration

data

Demonstrates

further

increased use

Reflects usage

patterns & rider

characteristics

Examines crash

rates, severity,

crash

characteristics

Compares rider

& vehicle

characteristics

& usage

Study 3a:

Focus groups

with moped &

scooter riders

Explores

motivations for

riding

Informs survey

development &

provides data

for comparison

Perspectives &

experiences of

riders

Rider

characteristics

perspectives &

experiences

Study 3b:

Moped &

Scooter Rider

Survey

Further

explores

motivations for

riding

Provides data

for comparison

Perspectives &

experiences of

riders

Rider

characteristics,

perspectives &

experiences

3.5.1 Study 1: Observation of powered two-wheeler types in an inner city area

Previous research indicates that Brisbane is an area of concentrated PTW use.

With limited exceptions, the review of the literature in Chapter Two identified a lack

of information specific to this type of location on the patterns and frequency of use of

various PTW types, including mopeds and scooters. The Queensland Government

has recently considered introducing mandatory motorcycle or moped licensing for

moped riders (Queensland Transport, 2008). A change in the numbers of mopeds

and scooters in use is expected if such changes are implemented, as are possible

changes in the characteristics of riders. In light of these possible changes and the

previously observed increase in moped and scooter use, it was thought useful to


better quantify and understand the current numbers and types of PTW in places

where there are known concentrations.

To achieve the objectives of Study 1, an unobtrusive observational survey

was conducted to obtain an approximate quantification of mopeds, scooters and

motorcycles within inner city Brisbane (Central Business District, or CBD) during

business hours (9am-5pm). This involved periodic seasonal counts, during weekdays

in summer and winter across two years, of all PTWs at thirteen designated parking

areas. Data collection was by visual inspection of vehicles on site. Variables of

interest in the study included vehicle location, make, model, year of manufacture and

engine capacity. On the basis of make and model details, the PTWs observed were

later coded in a database for PTW type.

While not providing a measure of exposure by distance travelled, the

observation study provided baseline data relating to frequency of use of different

PTW types, as well as monitoring trends over a two year period. The study also

examined the distribution of PTW types across different parking areas, the locations

of which may relate to motivations for moped and scooter use. Study 1 provided

input into Research questions 1, 2 and 4 (see Table 3.1). For Research questions 1

and 2 regarding increased moped and scooter usage, the study monitored current

trends in a known area of concentrated usage moped and scooter usage. The study

provided input to Research question 4 by examining vehicle characteristics and usage

patterns.

3.5.2 Study 2: Analysis of Queensland crash and registration data

Study 2 involved an in-depth analysis of Queensland PTW crash and

registration data covering a period of five years from July 2003 to June 2008

inclusive. In regard to the relative safety of different PTW types, the literature

review identified issues surrounding age and experience, motivations for use, design

and performance characteristics, vehicle and rider conspicuity, use of protective

clothing and risk-taking behaviour, among others. These issues were examined

within the framework of six main contributors to crash and injury risk drawn from

previous research (Greig, Haworth et al., 2007).

Previous research has compared moped and motorcycle usage, crashes and

related safety issues, but has not comprehensively examined mopeds, scooters and


motorcycles as separate PTW types. This is because they are generally not readily

identifiable in crash data files as separate types, with scooters typically grouped with

motorcycles. Therefore, the initial challenge of Study 2 was to incorporate vehicle

make and model details from registration data with the crash data so that PTWs

could be classified as mopeds, scooters or motorcycles. This essential process

facilitated the comparative analysis of the PTW types.

The study examines trends, differences and similarities between mopeds,

scooters and motorcycles in usage, crash rates, crash severity, crash characteristics

and location, contributing factors and rider demographics. Some of the research

literature notes tourist involvement in moped crashes in Queensland and this was

examined further in the current study. The main statistical analyses for this study are

performed using SPSS software, while MapInfo GIS (Geographic Information

System) software is also used to assist in data analysis and presentation. The study

also identifies methodological issues surrounding data quality.

Study 2 provided input into all four research questions. For Research

question 1, the crash and registration data provide evidence for increased moped and

scooter usage. For Research question 2, the different usage patterns and rider

characteristics for the PTW types were reflected in the crash data. For Research

question 3, the study examined crash rates, crash severity and crash characteristics to

determine how the safety of mopeds, scooters and motorcycles differed. For

Research question 4, the study compared rider characteristics, vehicle characteristics

and usage patterns to explore reasons for observed differences in the safety of

mopeds, scooters and motorcycles.

3.5.3 Study 3: Exploration of moped and scooter rider characteristics

Study 3 was a two-part study which explored a wide range of moped and

scooter rider characteristics. The first part of the study involved focus groups to

obtain qualitative information from moped and scooter riders and industry

representatives. This was used to guide the second part of the study which involved

developing and administering an online survey instrument for the collection and

subsequent analysis of quantitative data. Study 3 provides input into all four

research questions by exploring reasons for moped and scooter usage, differences in

travel patterns, demographic characteristics, crash experience, licensing and training,


protective clothing use, as well as self-reported riding skills and related knowledge.

3.5.3.1 Study 3a: Focus groups with moped and scooter riders

Focus groups are conducted in Study 3a to identify issues pertinent to moped

and scooter safety and the factors which motivate their use from the perspective of

riders. Specific groups of riders are targeted separately to allow comparison between

them, and to ensure comprehensive coverage of the relevant issues. The different

groups of riders targeted for this study include Brisbane city commuters, members of

online scooter forums, students, and industry stakeholders who are also scooter or

moped riders. The distinction between riders of mopeds and those of larger scooters

is considered important due to only the latter group requiring a motorcycle licence

and the focus group design is tailored accordingly. While it is likely that there is

some overlap of characteristics and motivations of riders across these groups, four

discrete focus groups are planned with an ideal representation of between six and

eight riders in each. Open-ended questions to guide focus group discussions were

developed by reference to a range of issues linking back to the six main contributors

to crash and injury risk drawn from previous research (Greig, Haworth et al., 2007):

inexperience or lack of recent experience; risk taking; driver failure to see

motorcyclists; instability and braking difficulties; road surface and environmental

hazards; and vulnerability to injury. In addition to these issues, other topics to be

specifically explored during the focus group sessions include motivations for moped

and scooter use, travel patterns, and attitudes and beliefs regarding licensing and

rider training.

Information gathered through focus groups was used to inform the

development of a questionnaire survey instrument. Although it was a qualitative

study designed to guide development of Study 3b, the qualitative findings of Study

3a are also of independent interest. Through covering a wide range of issues

regarding moped and scooter use and safety, it addressed all four research questions.

3.5.3.2 Study 3b: Queensland scooter and moped rider survey

The findings from the focus groups in Study 3a are used to inform

development of an online survey instrument which targets Queensland moped and


scooter riders for participation in Study 3b. The survey instrument is a questionnaire

containing 56 items related to moped and scooter use and safety. The questionnaire

development is also guided by the research literature, including that which identified

six main contributors to crash and injury risk (Greig, Haworth et al., 2007), and by

reference to previous questionnaires used for motorcycle and other road safety

research. Key Survey software used to deliver the survey and analysis was

performed in SPSS.

The survey is designed to collect data on the demographic, social,

motivational, attitudinal and other characteristics, including crash involvement and

licensing, of Queensland moped and scooter riders. These data can then be

compared with moped and scooter rider profiles from other jurisdictions, as well as

with motorcycle rider profiles, with a view to identifying usage patterns, rider

motivations and specific risk factors. Study 3b is a quantitative study which also

addresses all four research questions.

3.6 Chapter Three summary

This chapter has detailed the objectives of the research and the specific

research aims. The aims are 1) to develop a better understanding of moped and

scooter usage trends and patterns, and 2) to investigate factors underlying differences

in moped, scooter and motorcycle safety. The four research questions formulated to

achieve the research aims are:

RQ1: Why has moped and scooter usage increased?

RQ2: How does the usage of mopeds, scooters and motorcycles differ?

RQ3: How does the safety of mopeds, scooters and motorcycles differ?

RQ4: Why does the safety of mopeds, scooters and motorcycles differ?

The three studies designed to answer the research questions and thereby

achieve the overall research aims and objectives were then outlined, including:

Study 1: Observation of powered two-wheeler types in an inner city area.

Study 2: Analysis of crash and registration data.

Study 3: Exploration of moped and scooter rider characteristics, using focus


groups with moped and scooter riders, and a scooter and moped rider survey.

The relationship of the research aims, research questions and research studies

to one another was described, and a rationale provided for each of these research

components. The next chapter presents Study 1, an observational study of PTW

types in an inner city area, designed to monitor the extent of moped, scooter and

motorcycle usage, and to examine vehicle characteristics that may be relevant for

safety.


CHAPTER 4: STUDY 1 – AN OBSERVATION OF POWERED

TWO-WHEELER TYPES IN AN INNER CITY AREA

4.1 Introduction and rationale

The review of the literature in Chapter Two found only limited information

on the patterns and frequency of use of different PTW types, including mopeds and

scooters, in specific locations. Previous research indicates that the Brisbane inner

city area is one of concentrated PTW use relative to other Queensland locations

(Harrison & Christie, 2006; Haworth & Nielson, 2008). However, the amount of

moped and scooter use relative to that of other PTW types in this location has not

been thoroughly examined. The primary aim of Study 1 was therefore to improve

the understanding of PTW use by type in an inner city area.

Sales and registration data provide some evidence of trends in usage, but are

lacking in detail regarding actual usage trends and patterns in specific locations. The

last decade has seen substantial increases in PTW sales and use in developed

countries, although the trends are inconsistent across countries regarding PTW type.

Over this period, moped and scooter sales have grown at a faster rate than

motorcycle sales in some places, including Queensland, Australia. However, trends

have shifted since 2008 with PTW sales declining across developed countries due to

global economic circumstances (ACEM, 2010). In Australia, the recent decline in

moped and scooter sales has been generally greater than that for other PTW types

(FCAI, 2010). The Federal Chamber of Automotive Industries reported that mopeds

and scooters comprised approximately nine percent of new PTW sales in Australia in

2009, down from 11 percent on the previous year (FCAI, 2008, 2010).

Noting the increased use of mopeds in recent years, the Queensland

Government has considered proposals for introducing mandatory motorcycle or

moped licensing for moped riders (Queensland Transport, 2008). The current

regulations require moped riders to hold a car licence but not necessarily a

motorcycle licence, and this is thought to have encouraged moped use in the study

area. Regardless of the reasons for increased usage, there are now substantially

greater numbers of active PTW riders with little or no experience, education or

training regarding PTW use. From a regulatory perspective, there is potential to


address this issue by introducing a requirement for moped riders to obtain a PTW

licence. Such changes to current licensing requirements may result in reduced moped

usage and changes in the characteristics of riders, as well as potential increases in the

use of LC category scooters and motorcycles in Queensland. The current study

monitored trends in PTW use by type and was able to measure reductions or

increases in moped use relative to that of other PTW types.

Study 1 was an observational study providing baseline data relating to

frequency of use of different PTW types, as well as measuring usage trends over a

two year period. The study also examined the distribution of PTW types across

different parking areas, the locations of which may relate to motivations for moped

and scooter use. Study 1 provided input into Research questions 1, 2 and 4. For

Research questions 1 and 2 regarding increased moped and scooter usage, the study

monitored current trends in a known area of concentrated usage moped and scooter

usage. The study provided input to Research question 4 by examining vehicle

characteristics and usage patterns.

4.2 Study design and methods

An unobtrusive observational survey was used to obtain an approximate

quantification of different PTW types within the Brisbane inner city area during

business hours (9am-5pm on weekdays). This involved six-monthly counts during

summer and winter of all PTWs at 13 motorcycle parking areas across a 24 month

period. Data collection commenced in August 2008 and the last phase of data

collection took place in August 2010.

PTWs were observed at 13 designated motorcycle parking areas within the

Regulated Parking Zone 1 of the Brisbane inner city area (Brisbane City Council,

2011). The number of parking areas included in the study was limited to 13 due to

resource limitations (the initial survey in August 2008 indicated that 13 parking areas

could be confidently covered between 9am and 5pm by one researcher). However,

the number of spaces included in the study represented the majority of those

available in the Brisbane inner city area (Regulated Parking Zone 1). The location of

the 13 parking areas included in the study and the numbers of spaces provided are

presented in Table 4.1. With the exception of one parking area on university campus

grounds, the parking areas included in the study are identified on a map of ‘Brisbane


inner city motorcycle parking’ available on the Brisbane City Council (BCC) website

(Brisbane City Council, 2010) (see Appendix B1). The parking areas included in the

study were mostly off-street areas on or adjacent to footpaths, with the exception of

two areas providing on-street kerbside parking (sites 12 and 13). Five of the 17

parking areas illustrated on the BCC map were excluded from the study, including

Gipps Street (20 spaces), Elizabeth Street (10 spaces), Adelaide Street (10 spaces),

Hope Street (Southbank, 16 spaces) and Tribune Street (Southbank, 10 spaces).

These excluded areas were either not within the central city area (Southbank and

Gipps Street locations), or were areas with a small number of allocated spaces

(Elizabeth Street and Adelaide Street locations).

One parking area included in the study is controlled by Queensland

University of Technology (QUT) and is not identified on the BCC map (QUT,

Gardens Point Road, site 11). One BCC parking area was established between

August 2008 and February 2009, so was not included in the first data collection

phase but was included in subsequent phases (site 5). The decision to include this

site after the initial survey in August 2008 was based on its location (adjacent to

other study sites), size (20 spaces claimed), and increased efficiency of data

collection which permitted completion within the eight hour timeframe.

It can be seen in Table 4.1 that the number of marked spaces reported on the

BCC map was not always consistent with number of marked spaces actually present

at a given site. While the total number of spaces actually present increased between

August 2008 and August 2010 from 315 to 347, the latter figure still fell short of the

420 spaces reported to be available. While it appears that the addition and removal

of PTW parking spaces is an ongoing process undertaken by the BCC, the map was

not updated over the study period, with the exception of the addition of site 5.


Table 4.1 Parking areas included in observational study of PTW use

Site Location

Claimed

spaces

on BCC

map

Actual

spaces

Phase 1

Actual

spaces

Phase 5

1 Turbot St Bridge near Wharf St 35*

13 20

2 Turbot St, corner Wharf St 17 17

3 Turbot St, corner Wickham Terrace 40 28 28

4 Turbot St, between Edward St & Creek St 45 45 45

5 Turbot St, between Albert St & Edward St** 20 N/A 20

6 Turbot St, under Roma St (Pay & Display) 25 25 25

7 Turbot St, corner North Quay 55 32 32

8 North Quay, corner Ann St 100 75 75

9 William St, corner Elizabeth St 30 16 16

10 William St, between Margaret St & Alice St 25 27 32

11 Gardens Point Rd, QUT university campus N/A N/A N/A

12 Charlotte St, corner Albert St (Pay & Display) 25 25 25

13 Margaret St, corner Felix St 20 12 12

Total 420*** 315 347

*Sites 1 and 2 combined. **Established between August 2008 and February 2009, so not included in

first data collection phase. ***Excluding Gardens Point Rd.

Anticipating reduced PTW usage during wet weather, data collection was

only undertaken during fine weather so as to maintain consistent conditions. While

differences in usage of PTW types according to weather conditions are of interest

and relevance to moped and scooter safety, such a study was beyond the scope of this

program of research. Mondays, Fridays, public holidays and days adjoining public

holidays were avoided as Australian workers are more likely to take unscheduled

leave from work on these days.

Data were collected by visual inspection of vehicles on site. PTWs parked

within a marked space, or in a continuous line directly adjacent to a marked space

(but not necessarily within a marked space) in the selected areas were counted as

being within that area and so were included in the data. Variables recorded included

vehicle location, make, model, year of manufacture, ADR category and engine

capacity, where these could be determined without touching or handling the vehicles.

Information was obtained from registration labels, vehicle compliance plates and/or

manufacturer’s labels, decals and badges. Notes were taken for missing registration

labels and/or plates, and for vehicles registered interstate.

In the study area, a registration label is a self-adhesive label, display of which

is required by law to indicate currency of registration, registration number, and a


limited (and variable) amount of make and model information. A registration plate is

also required, being a metal plate fixed to the rear of PTWs, containing only the

registration number which must legible from a distance of 20 metres within an arc of

45 degrees from the plate surface (TMR, 2011b). A vehicle compliance plate is

usually fixed to the vehicle body prior to initial sale and registration, confirming

compliance with the relevant Australian Design Rules. The compliance plate

indicates date of manufacture, make and model details, Vehicle Identification

Number (VIN), and approval by the relevant federal authority (NSW Roads and

Traffic Authority, 1997).

PTWs were allocated to categories and sub-categories adapted from type

classifications within the US National Agenda for Motorcycle Safety (NAMS)

(Motorcycle Safety Foundation, 2000). The PTW categories and subcategories used

for the current study are presented in Table 4.2. Final allocation to categories took

place during the process of data entry after confirmation of the design and intended

purpose of particular models where possible. Sources used to confirm the PTW

make, model and relevant specifications included Bikez.com online motorcycle

catalogue, Google search engines, and manufacturers’ websites. Numerous issues of

the following periodical publications were also consulted for vehicle specifications:

Scooter magazine, Two Wheels magazine, Australian Road Rider magazine, and

Australian Motorcycle News magazine. This classification of PTW types provided

detailed information on PTW usage, including moped and scooter usage, which was

previously unavailable and may be useful in further research projects.

Some overlap of PTW types is inherent in the current market and difficulties

with type classification could not be avoided in some cases. For example, Honda’s

DN-01 has been described in one industry review as ‘two parts scooter; one part

cruiser; and one part sportbike’ (Duke, 2009). This is due to its continuously

variable transmission (CVT), typically characteristic of mopeds and scooters,

combined with a V-twin engine, chassis and other components more characteristic of

cruising and sport-oriented motorcycles. In the current study the DN-01 was defined

a ‘Cruiser’ in the primary and secondary subcategories. Another example is that of

motorcycle types often labelled ‘naked sport’, which may have a traditional (un-

faired) appearance combined with high performance components and specifications.

Such motorcycles include the Ducati Monster, Honda Hornet, Kawasaki Z1, Yamaha

FZN and Triumph Street/Speed Triple, among others. In most cases these have been


allocated to the ‘Sport’ secondary subcategory, however this classificatory system

remains imperfect. While some overlap of PTW types was unavoidable, it was

considered important to be consistent regarding PTW models which arguably could

be identified as belonging to more than one category.

Table 4.2 PTW type classification

Main category Primary subcategory Secondary subcategory

Moped or scooter

Moped Moped

Scooter Scooter

Scooter – 3 wheeled

Moped/Scooter? (unknown) Moped/Scooter? (unknown)

Motorcycle

Sport & Touring

Sport

Sport-touring

Touring

Traditional Traditional

Cruiser Cruiser

Off-road & Dual purpose Dual purpose

Enduro

Other

Motard

Postie (Honda CT)

Unknown

4.3 Results

4.3.1 PTW’s observed over the study period

Over the two year period spanning five data collection phases, approximately

36 percent of PTWs observed in inner city Brisbane motorcycle parking areas were

either mopeds (20.4%) or scooters (14.2%) (N = 2,642). Approximately 1.5 percent

could not be identified as either a LA category moped or LC category scooter.

Motorcycles represented the remaining 64 percent of PTWs observed. As shown in

Table 4.3, there was a progressive increase in the total number of PTWs observed

across data collection phases. This overall increase is attributed mostly to an

increase in the number of motorcycles rather than mopeds and/or scooters. The

number of motorcycles observed in August 2008 was 295, increasing to 395 in

August 2010, while the number of mopeds did not increase notably over the study


period (110 to 114). The overall number of scooters increased slightly from 68 to 88

over the same period. Scooter use appeared to increase in the latter half of the study

period. Although motorcycles increased slightly as a proportion of PTWs observed,

there was no significant difference in the distribution of PTW types by data

collection phase [ ² (8) = 5.06, p = .751].

Table 4.3 Mopeds, scooters and motorcycles observed in Brisbane CBD by

data collection phase

Collection

phase

PTW type

Moped Scooter Scooter/Moped Motorcycle Total

n % n % n % n % n

August 08 110 22.8 68 14.1 10 2.1 295 61.1 483

February 09 99 20.4 69 13.8 8 1.6 315 64.2 491

August 09 112 21.3 68 12.9 2 0.4 345 65.5 527

February 10 102 19.0 83 15.5 11 2.1 340 63.4 536

August 10 114 18.8 88 14.5 8 1.3 395 65.3 605

Total 537 20.4 376 14.2 39 1.5 1,690 64.0 2,642

No seasonal differences were evident regarding the total number of PTWs

observed or the proportions of types observed during summer and winter. Data

collection was only conducted during fine weather and it is possible if not likely that

reduced PTW usage would be observed during wet weather regardless of the season.

Table 4.4 shows the PTW types observed by secondary subcategory,

separating motorcycle types as well as scooter and moped types. Sport motorcycles

were the most common subcategory observed (27%), followed by mopeds (20%),

traditional motorcycles (14%) and two-wheeled scooters (14%). Table 4.4 also lists

information on engine cylinder capacity (or piston displacement) in cubic

centimetres (cc) for the PTW types. Moped engine cylinder capacities are limited to

50cc or less by ADR definition (Australian Government, 2008) and generally range

from 49 to 50cc in standard form. Four electric-powered mopeds were observed

across the study period, and as these vehicles do not use an internal combustion

engine they are required to comply only with the 50 km/h ADR speed restriction for

mopeds.

Engine cylinder capacities of LC/LE category scooters ranged from 100cc to

650cc and a large majority of LC scooters observed had engine capacities in the

lower end of this range. The mean engine cylinder capacity for LC/LE category

scooters was 179.5cc (n = 371) and was lower for 2-wheeled scooters (173.0, n =


360) than for those with 3 wheels (390.9, n = 11). Where engine size was known, 92

percent of scooters were in the 51-125cc (52.3%) and 126-260cc (39.6%) categories.

During the study period, three-wheeled scooters were only available with engine

cylinder capacities ranging from 250cc to 500 cc. Engine cylinder capacity could not

be determined for four LC category scooters.

The average engine cylinder capacity for motorcycles was 667.8cc (N =

1,665), ranging from 100cc to 2,300cc. Excluding mopeds and scooters, the lowest

mean engine cylinder capacities were found among Enduro (395.5cc, n = 39),

Motard (527.7cc, n = 20) and Traditional (567.2cc, n = 378) PTW types. The highest

mean engine cylinder capacities were observed among Touring (1,075.7, n = 35),

Sport-touring (785.6cc, n = 123) and Cruiser (726.3cc, n = 215) PTW types.

Table 4.4 PTW subcategory information for all PTW’s observed

Secondary subcategory n % Mean CC* CC Range*

Moped 537 20.3 50< N/A

Scooter – 2 wheeled 364 13.8 173.0 90 - 650

Scooter – 3 wheeled 11 0.4 390.9 250 - 500

Moped/Scooter? 39 1.5 N/A N/A

Sport 710 26.9 694.1 125 – 1400

Sport-touring 123 4.7 785.6 250 – 1400

Touring 35 1.3 1075.7 650 – 1500

Traditional 378 14.3 567.2 100 – 1400

Cruiser 215 8.1 726.3 125 – 2300

Dual purpose 141 5.3 716.6 175 – 1200

Enduro 39 1.5 395.5 250 - 690

Motard 20 0.8 527.7 200 - 690

Postie (Honda CT) 26 1.0 110.0 110

Unknown 4 0.2 N/A N/A

Total 2,642 100.0 468.9 <50 – 2300

*Approximate figures

Registration labels were not visible on 61 mopeds (11%), 24 scooters (6%),

11 moped/scooter unknowns (30%) and 40 motorcycles (2%). In some cases the

lack of visible registration labels prevented PTW type classification beyond that of

manufacturer and general style. Some PTWs were parked such that registration

labels could not be observed but may have been fitted (rear-end in to a surface for

example). Registration plates were fitted to all PTWs with the exception of one

moped and two motorcycles. There were 14 PTWs registered outside of Queensland,


including 4 mopeds and 10 motorcycles. Data on currency of registration were not

collected.

Examination of year of manufacture showed that mopeds and scooters were

newer on average compared to motorcycles. A one-way ANOVA test revealed a

significant difference between the age of motorcycles compared to mopeds and

scooters [F (2, 2346) = 132.17, p < .001]. Post-hoc tests showed that the average age

of mopeds (M = 3.48 years) and scooters (M = 2.99 years) was significantly different

from that of motorcycles (M = 7.06 years) but not different from each other. While

this finding almost certainly reflects the increased popularity of mopeds and scooters

in recent years, other factors potentially influencing this result are discussed below in

section 4.4 (Discussion). In addition to the increase in PTWs observed in the 13

selected designated parking areas, there was an apparent increase in PTWs parked in

other areas where no formal parking space was provided. These PTWs were not

included in the observational study and are mentioned here only as anecdotal

evidence in support of a perceived increase in pressure on PTW parking in the study

area.

4.3.2 PTW’s observed by geographic location

Aggregate data for all data collection phases show that there was a

statistically significant difference in the distribution of mopeds, scooters and

motorcycles by parking area location [ ² (24) = 82.97, p < .001]. These data are

presented in Table 4.5 and Figure 4.1. While Table 4.5 includes Scooter/Moped

(unknown) PTWs, these were excluded from statistical testing due to low numbers.

Generally, the sites in the northern part of the study area (sites 1 to 8) appear to be

less frequented by mopeds compared with sites in the southern inner city area. In

northern data collection sites (sites 1 to 8), mopeds represented approximately 18

percent of PTWs observed (N = 1,635). By comparison, in southern sites (sites 9 to

13) mopeds represented 25 percent of PTWs (N = 968). An opposite pattern was

observed for scooters, which accounted for around 16 percent and 12 percent of

observations in northern and southern sites respectively.

The new parking area (site 5) established between August 2008 and February

2009 and included in the study from the second data collection phase did not appear

to generate a significant increase in the number of PTWs observed. In the first


observation at this site (February 2009) 31 PTWs were counted, yet the total number

of PTWs observed across all sites increased by only eight over the initial data

collection phase in August 2008. The mean number of PTWs observed at this site

per collection phase was 35.5. This exceeds by about 10 the number of marked

spaces claimed for this site on the BCC map of motorcycle parking. It is possible

that this site absorbed some of the PTWs which were previously observed at other

sites that were occupied beyond capacity.

As noted above, the number of marked parking spaces actually provided at

some sites differed from the number claimed on the BCC map. The number of

spaces provided was sometimes less and sometimes more than the number claimed

by BCC for each site. The number of PTWs observed generally exceeded the

number of spaces claimed to be marked for the site on the BCC map, and often

exceeded the number of actual spaces provided. Both ‘pay and display’ sites (sites 6

and 12) had lower average occupancy than other sites and were frequented relatively

less by mopeds. As noted above, mopeds generally represented a lower proportion

of PTWs observed in northern sites than in southern sites. In particular, site 6 had

the lowest proportion of mopeds (11% of PTWs) relative to other sites, and the

lowest average occupancy rate (60%) overall.

Table 4.5 PTW’s observed in Brisbane CBD by type and location (aggregate)

Site

PTW type

Moped Scooter Scooter/Moped Motorcycle Total

n % n % n % n % n

1 16 18.0 19 21.3 1 1.1 53 59.6 89

2 12 14.5 22 26.5 0 0.0 49 59.0 83

3 31 17.8 31 17.8 0 0.0 112 64.4 174

4 45 18.4 23 9.4 2 0.8 175 71.4 245

5 24 16.9 26 18.3 1 0.7 91 64.1 142

6 8 10.7 8 10.7 0 0.0 59 78.7 75

7 45 15.8 34 11.9 3 1.1 203 71.2 285

8 111 20.0 101 18.2 7 1.3 337 60.6 556

9 37 31.9 14 12.1 7 6.0 58 50.0 116

10 51 21.2 40 16.6 6 2.5 144 59.8 241

11 110 25.5 40 8.9 11 2.5 275 63.1 436

12 22 18.6 14 11.9 1 0.8 81 68.6 118

13 25 30.5 4 4.9 0 0.0 53 64.6 82

Total 537 20.4 376 14.2 39 1.5 1,690 64.0 2,642


Figure 4.1 Aggregate PTW type distribution across Brisbane city parking areas


4.4 Discussion

Study 1 provided input into Research questions 1, 2 and 4. For Research

questions 1 and 2 regarding increased moped and scooter usage, the study monitored

current trends in a known area of concentrated moped and scooter usage. The study

provided input to Research question 4 by examining vehicle characteristics and usage

patterns. The results confirm that a large minority (36%) of PTWs in the Brisbane

inner city area over the study period were either mopeds (20%) or larger scooters

(14%). Sport motorcycles represented the largest secondary subcategory by PTW

type (27%), which is consistent with findings of the state-wide Queensland

motorbike usage survey in 2005 (Harrison & Christie, 2006).

Scooters and mopeds represented more than one third of PTWs observed in

inner city Brisbane over the study period, yet they comprised only nine percent of

new PTW sales nationally in 2009 (including off-road PTWs) (FCAI, 2010). The

concentration of mopeds and scooters in the Brisbane area therefore strongly

suggests that their use is primarily as an inner urban transport mode for commuting,

which is consistent with other research.

Of the PTWs that were identified as mopeds and scooters (N = 913), 59

percent were mopeds and 41 percent were scooters. This is roughly consistent with

recent Queensland sales data in which mopeds comprised 63 percent of new moped

and scooter sales in the twelve months to September 2009 (FCAI, 2009). Previous

research found that mopeds comprised 84 percent of new moped and scooter sales in

2005 (Haworth & Nielson, 2008). Scooter sales and use therefore appear to have

increased at a faster rate than that of mopeds despite the requirement for scooter

riders, but not moped riders, to hold a motorcycle licence.

Scooter riders appeared to represent approximately 14 percent of motorcycle

licence holders frequenting the study area, assuming a high rate of licensure on the

basis of previous studies. The Queensland motorbike usage survey in 2005 by

Harrison and Christie (2006) found that of survey respondents from the Brisbane

area, only four percent were scooter riders, while five percent were moped riders.

Although the Harrison and Christie sample was biased toward open licence holders

and probably underrepresented moped riders, it further suggests a proportional

increase in scooter and moped use relative to other PTW types in the current study.


Scooters and mopeds alike were found to be significantly younger than

motorcycles, arguably reflecting their increased popularity over the last decade. It is

also possible that mopeds and scooters are retired from service earlier than

motorcycles due to higher maintenance costs relative to replacement cost for ageing

vehicles. In addition, crash repair costs for mopeds and scooters may be a higher

proportion of their (lower) base cost than of many (more expensive) motorcycles,

thus leading them to be ‘written off’ (where repair costs exceed replacement cost).

Whether or not this is actually the case remains a potential topic for further research.

Scooters were marginally younger on average than mopeds in the current study,

again possibly reflecting recent sales trends, but the difference was not statistically

significant. While there was no significant increase observed in the proportion of

scooters relative to mopeds over two years in the current study, potential further

research will be able to use these baseline data to measure trends in future.

The PTW parking areas observed in inner city Brisbane were typically

occupied at or beyond capacity, with the exception of areas requiring payment for

parking. Despite provision of a new PTW parking area in September 2008, parking

pressure in the study area remains intense with PTW numbers continuing to increase

thereafter with no further provision of spaces. Given that PTW parking areas are

often filled to capacity, their use is now potentially constrained, raising the question

of whether the supply of parking areas should be increased in order to meet demand.

Current levels of enforcement regarding illegal parking are unknown and this is also

likely to have an impact on the need for further provision of designated parking areas

and spaces. If enforcement of parking regulations has been limited, then the current

study may have underestimated the extent of use and the increase in usage. On the

other hand, if regulations have been strongly enforced, then this may have served to

constrain a greater increase in usage than was actually observed. The extent to which

parking availability actually motivates PTW use in Brisbane is specifically explored

in Studies 3a and 3b (Chapter Six and Chapter Seven).

A relatively high proportion of PTWs observed in the Gardens Point campus

of Queensland University of Technology were mopeds (25.5%). This likely reflects

the popularity of mopeds among university students. Mopeds also represented a

relatively high proportion of PTWs at the four other sites which were closest to the

university. Larger (LC) scooters were relatively less frequent at these sites, while

motorcycles were proportionally represented relative to the aggregate data. This


supports the contention that low purchase and usage costs are a strong motivation for

moped use, which is further examined in Studies 3a and 3b.

Approximately 1.5 percent of PTWs observed were either mopeds or scooters

but could not be more accurately identified. Problems with distinguishing mopeds

from scooters related to a lack of model information on manufacturers’ labels and the

fact that vehicle compliance plates are not usually visible. On motorcycles,

compliance plates are usually fixed around the steering head or other parts of a

chassis where they are externally visible, simplifying identification. By contrast,

compliance plates on scooters and mopeds are almost always concealed by fairings

or outer body panels. There were also problems with a lack of detail on registration

labels, as well as a lack of visibility or complete absence of registration labels. This

suggests that it may also be difficult to distinguish some moped and scooter models

for the purpose of crash data collection and analysis, as well as enforcement of the

motorcycle licence requirement for scooter riders. Registration labels were less

likely to be visible on mopeds than on larger scooters or motorcycles, a result which

is likely due in part to their smaller size which makes it easier to manoeuvre them

into tight spaces. It cannot be reliably claimed that mopeds are less likely to have a

registration label attached, although observations suggest that this may be the case.

Previous research has provided some information on PTW use by type in the

Brisbane city area using crash data analysis and exposure surveys, but these studies

have been limited with regard to an objective measure of actual usage. The current

study has provided a reliable estimate of the amount of usage of each PTW type in

terms of frequency of use. The study confirms that moped and scooter use is

concentrated in the Brisbane city area. It also shows that motorcycles with large

engine capacities, including high performance sport motorcycles, are popular among

city commuters. The data can be used as baseline information for further monitoring

of PTW use by type in future research. In particular, this may be of use in the event

of changes to licensing and other regulations, changes within the PTW industry, and

changes in travel mode choice which may variously influence PTW use.


4.5 Limitations

Inclusion of all designated CBD parking areas was beyond the scope of this

study due to limited resources for data collection. However, the study is estimated to

have included the majority of PTWs parked in inner city Brisbane and the

distribution and proportions of PTW types is therefore argued to be representative.

As data were only collected during fine weather it was not possible to observe the

influence of wet weather on PTW use. Reduced PTW usage would be expected

during wet weather and this remains a potential area for further research.

A lack of visible identifiers meant that it was not possible to distinguish

mopeds from larger scooters in some cases during observation data collection. This

was the case for only 1.5 percent of all PTWs observed and is therefore considered a

minor limitation. It was not possible by external examination to reliably detect

moped performance modifications. This is a topic on which further research may be

warranted according to literature which suggests that modified mopeds are

overrepresented in crashes (ACEM, 2008b). Further, if mopeds had been modified

to increase performance they may no longer comply with the ADR definition of a

moped as an LA category vehicle.

The diversity of PTW designs and related characteristics results in

considerable overlap between PTW types among some models. For example, some

‘traditional’ models may have similar performance characteristics to many sport

motorcycles, but are categorised ‘traditional’ on the basis of minimal bodywork and

general appearance. Some dual purpose motorcycles may arguably meet

performance and design criteria for sport or sport-touring motorcycles, while the

difference between sport and sport-touring motorcycles themselves is also often

indistinct. References such as Bikez.com were used to help guide categorisation of

the PTW models observed, but all PTW classificatory systems have suffered from

this lack of clear distinction among many PTW models.


4.6 Chapter Four summary

This chapter has described Study 1, which was designed to assist in

answering Research questions 1, 2 and 4. The study addressed Research questions 1

and 2 by monitoring current trends in a known area of concentrated moped and

scooter usage. The study provided input to Research question 4 by examining

vehicle characteristics and usage patterns.

The Brisbane city study location represents an area of concentrated PTW use

relative to other areas of Queensland. The research may be relevant to other urban

areas where there has been an increase in moped and scooter usage from a

historically low base. An unobtrusive observational survey was repeated at six-

monthly intervals to explore the prevalence of moped and scooter use among other

PTWs and to identify any trends observable over a two year period. More than one

third of all PTWs observed were either mopeds or scooters. Approximately 60

percent of all mopeds and scooters observed were mopeds, which is roughly

consistent with recent sales data. The number of PTWs observed increased at each

six-monthly phase of data collection and the demand for PTW parking spaces may

exceed current supply despite an increase in the number of parking spaces over the

study period.

The next chapter describes Study 2, involving an in-depth analysis of PTW

crash and registration data covering a period of five years from July 2003 to June

2008 inclusive. In addition to exploring crash characteristics and circumstances,

Study 2 also examined moped, scooter and motorcycle usage patterns. While Study

2 included all of Queensland, the proportion of crashes by PTW type in inner

Brisbane was explored and the data compared with that regarding usage in Study 1.


CHAPTER 5: STUDY 2 – ANALYSIS OF CRASH AND

REGISTRATION DATA

5.1 Introduction

Riders of powered two-wheelers are collectively considered vulnerable road

users, but there is a need to better understand the qualitative differences between

moped, scooter and motorcycle riders in terms of safety. Greig, Haworth and

Wishart (2007) described six main contributors to crash and injury risk for

motorcycle riders, including: inexperience or lack of recent experience; risk taking;

driver failure to see motorcyclists; instability and braking difficulties; road surface

and environmental hazards; and vulnerability to injury. Other literature suggested

that these risk factors do not necessarily apply equally to moped, scooter and

motorcycle use, due to differences among the PTW types in usage patterns, rider

characteristics and motivations, and vehicle attributes.

There are a number of ways in which moped, scooter and motorcycle usage

differ that are likely to have some bearing on their relative safety. Previous research

in Australia and other developed countries has demonstrated that mopeds and

scooters are used more for urban commuting and less for recreation in comparison to

motorcycles. Research has also shown differences according to PTW type in riding

style, use of protective clothing, conspicuity and other factors which influence crash

and injury risk. Rider age and gender distributions generally differ according to

PTW type, as do licensing requirements in many jurisdictions. In Queensland, car

licence holders are permitted to operate a moped without any training, education,

theoretical knowledge, skills or practical testing specific to PTW riding. Riders of

motorcycles and larger scooters require a motorcycle licence to ride on public roads.

Analysis of crash and registration data provides information relevant to these

regulations and a wide range of other issues relevant to the research questions which

are briefly summarised in the following section (section 5.1.1).

As moped and scooter usage has increased in Queensland, so too has interest

and concern over the safety of moped and scooter riders. Previous research

identified a fourfold increase in Queensland moped crashes between 2001 and 2005

(Haworth & Nielson, 2008; Haworth, Greig & Nielson, 2009). That research also


compared moped and motorcycle crashes in Queensland, as described in detail in

Chapter Two. Building upon the earlier research, the current study examined crash

and registration data to further explore the usage and relative safety of mopeds and

motorcycles. Additionally, a critical feature of the current study was the creation of

a new data set allowing LC category scooters (above 50cc) to be separated from

motorcycles and mopeds to facilitate comparison of the three PTW types. This has

not been previously attempted in any comprehensive analysis of crash data reported

in the literature, as scooters which are not mopeds are usually grouped together with

motorcycles in crash data files. Due to differences in the characteristics, usage

patterns and crash risk of riders of different PTW types identified in other research, it

was expected that differences between moped and scooter riders as well as between

moped and motorcycle riders would be evident in the Queensland crash data.

5.1.1 Research questions

Study 2 provided input into all four research questions, with Research

questions 2, 3 and 4 as the primary concerns of the study. For Research question 1

regarding reasons for increased PTW usage, Study 2 sought evidence of increased

usage, which had been previously observed from 2001 to 2005 (Haworth & Nielson,

2008). For Research question 2 concerning differences in usage, the patterns of

usage as well as rider characteristics for the PTW types were explored in the crash

data. For Research question 3 regarding differences in safety, the study examined

crash rates, crash severity, crash characteristics and contributing factors to assess the

relative safety of mopeds, scooters and motorcycles. For Research question 4

concerning reasons for differences in the safety of mopeds, scooters and motorcycles,

the study explored various factors relating to risk, including the six main contributors

to crash and injury risk identified in the literature (Greig, Haworth & Wishart, 2007).

Addressing the research questions as such, Study 2 provided essential material for

achieving the research aims, which were to develop a better understanding of moped

and scooter usage trends and patterns, and to investigate factors underlying

differences in moped, scooter and motorcycle safety. The study therefore helps to

address the recognised knowledge gap regarding the relative safety of moped,

scooter and motorcycle use.



As noted previously, mopeds, scooters and motorcycles are separated in this

analysis to identify differences between the PTW types with regard to the variables

of interest. The analysis and discussion centres on describing crash rates, crash

severity, crash characteristics and circumstances, and riders involved. Key variables

for analysis therefore include crash severity, crash location, temporal characteristics,

crash types and configurations, contributing circumstances, and rider characteristics

and behaviour. Using the most recent data available, this description of crashes is

placed within the context of recent trends in PTW usage in the study area as

indicated by vehicle registrations.

In order to include the most recent data available, it was necessary to conduct

analysis on the basis of financial year (July-June inclusive) rather than calendar year

(January-December). Crash data were available for reported crashes which occurred

up until 30 June 2008, but were not available for the following six months to 31

December 2008. Registration data were also obtained for financial year periods so

that they would align with the crash data for analysis of crash rates by registered

PTW.

5.2.1 Acquisition of registration data

Registration data for PTWs on register in Queensland (including mopeds,

scooters and motorcycles) were sourced from the Queensland Department of

Transport and Main Roads (TMR) website (TMR, 2010). This covered financial

year periods from July 1922 to June 2009 (inclusive). Moped registration data for

July 2001 to June 2010 (inclusive) was sourced by data request to TMR in July 2010.

Mopeds are identifiable in registration data as they are classed as LA (or LB if three-

wheeled) category vehicles. As motorcycles and scooters are both classed as LC (or

LE if three-wheeled) category vehicles, they are not separated in the registration

data. Subsequently, the trends in registrations for these PTWs as distinct from one

another can only be inferred from sales data or by painstaking re-coding of make and

model to create motorcycle and scooter categories (where sufficient information

exists).


5.2.2 Acquisition, cleaning and coding of crash data

PTW type is not reliably captured in the official crash database. As the first

step in creating a new database for this study, TMR merged the crash data for all

PTWs in crashes for the five year period 1 July 2003 until 30 June 2008 with the

vehicle registration database using registration number as the matching variable. The

registration data provided additional information on make, model and body type for

the PTWs in crashes to enable the later identification of PTW type using the

augmented data.

Of those PTW types classified as ‘Unknown’ in the final augmented data file,

the make and model details were not available for a number of possible reasons

concerning the linkage of crash and registration data. These include that the

registration numbers were not recorded in crash reports, that the vehicle was

unregistered at the time of the crash, or the vehicle was no longer on the vehicle

register when the crash and registration data files were merged. Unknown PTW

types were involved in about 14 percent of all reported crashes, but the proportional

distribution of PTW types in those crashes cannot be reliably determined. However,

analysis of the crash data provides grounds to speculate that unknown PTW types

may be more likely to be motorcycles than mopeds or scooters, as will be discussed

in later sections of this chapter.

TMR supplied the augmented crash data in four files containing different

variables for the 8,608 reported Queensland PTW crashes for the five year period.

The separate files contained data on casualties (persons injured), controllers (road

user types and contributing circumstances), crashes (crash characteristics and

descriptions) and contributing circumstances. The files contained some common

variables including a reference number for each crash which enabled the files to be

linked or merged for analysis.

The data were processed using SPSS (version 17) software. The four data

files were merged using the common reference number allocated to each crash. Each

crash in the original files was coded for vehicle body type as either ‘MCYC’

(motorcycle), ‘MOPE’ (moped), ‘MQUA’ (four wheels), ‘MTRI’ (three wheels),

‘SCAR’ (side car motorcycle), ‘TQOR’ (three or four wheels, off-road) or

‘Unknown’. Examination of the data indicated numerous crashes where the coding

of vehicle body type was inconsistent with vehicle make and model details. The first


objective was therefore to rectify these inconsistencies where possible, and then to

classify each of the 8,608 crashes as involving one of the six PTW types identified in

Table 5.1: Motorcycle (LC category); Moped (LA category); Scooter (LC category);

Moped/Scooter (unknown); Unknown; and Other. A new variable called ‘PTW

type’ was created within an SPSS master file to facilitate this process.

Table 5.1 Reclassification of PTW types using vehicle make and model details

Vehicle

Body

Type

Reclassified PTW type according to make and model details

Moped

LA

Scooter

LC

Moped

or

Scooter

Motorcycle

LC

Unknown Other Total

MCYC 130 86 54 6,705 5 - 6,980

MOPE 411 8 4 - - - 423

MQUA - - - - - 1 1

MTRI - 1 - - - 8 9

SCAR - - - 6 - - 6

TQOR - - - - - 1 1

Unknown - - - - 1,188 - 1,188

Total 541 95 58 6,711 1,193 10 8,608

The reliability of subsequent crash data analysis relies heavily on the

accuracy of the original crash report and associated data entry. It is possible that

PTWs in some cases were incorrectly identified and/or incorrectly reported by

reporting officers, particularly in the case of mopeds (‘MOPE’) with ‘Unknown’

make and model details (some of these may have been LC category scooters).

However, every effort was made to include only reliably identified PTW types in

cases selected as valid for analysis.

Multiple information sources were used to identify PTW models as either

moped, scooter or motorcycle, including: Scooter Magazine, published biannually

with a comprehensive list of mopeds and scooters available new in Australia

(Bowdler, 2010); Bikez.com online motorcycle catalogue (Bikez.com, 2010),

claiming a list of over 17,000 PTWs manufactured since 1970; and make and model

information transferred from the Observation study of PTWs conducted as Study 2 of

this research. Several PTW manufacturers produce or have produced vehicles which

share chassis design, model names and other features, sometimes differing only in

engine cylinder capacity which subsequently determines their ADR category as

either LA moped or LC scooter (motorcycle).


The process used to classify PTW types began with confirming for each case

that vehicle body type and make and model details were consistent. Cases with

‘MOPE’ body type and ‘Unknown’ make and model details were assumed to be

mopeds and coded as such for analysis (‘Moped’). Cases with ‘MOPE’ body type

and make and/or model details indicating LC category motorcycle or scooter were

recoded accordingly (‘Motorcycle’ or ‘Scooter’). Cases with ‘MCYC’ body type

were then checked for make and model details to identify mopeds which may have

been misclassified, and (LC category) scooters which are considered motorcycles for

licensing and registration purposes. Cases where make and model details clearly

indicated a moped misclassified as ‘MCYC’ body type were subsequently recoded

‘Moped’ for analysis. Similarly, cases where make and model details clearly

indicated a (LC category) scooter with ‘MCYC’ body type were recoded as ‘Scooter’

for analysis. Cases in the crash dataset with ‘MCYC’ body type where it was unclear

from model details whether a PTW was a LA moped or LC scooter were recoded as

‘Moped/Scooter’ (unknown) and excluded from analysis.

This process revealed 411 mopeds which were correctly recorded as ‘MOPE’

body type, while a further 130 mopeds were incorrectly recorded as ‘MCYC’ body

type. Of the 95 LC category scooters identified, 86 were correctly recorded as

MCYC body type, 8 incorrectly recorded as ‘MOPE’ body type and 1 incorrectly

recorded as ‘MTRI’ body type. There were 55 vehicles recorded as ‘MCYC’ and 4

vehicles recorded as ‘MOPE’ that could not be confirmed as either LA moped or LC

scooter from make and model information. Models which are sold in both LA and

LC configurations with no indication of engine cylinder capacity (e.g. Piaggio ‘Zip’,

Bolwell ‘Shark’) have been coded for PTW type according to the vehicle body type

originally listed for that case if ‘MOPE’, or recoded for PTW type as

‘Moped/Scooter’ (unknown) if vehicle body type originally listed for that case was

‘MCYC’ (given the substantial number of mopeds reliably identified with MCYC

body type, it was not assumed these were LC category scooters). Where vehicle

make was known but vehicle model unknown (in all cases recorded as MCYC body

type), these were coded as ‘Motorcycle’ PTW type unless the vehicle manufacturer

was known to produce only scooters and mopeds and not motorcycles (e.g.

Vespa/Piaggio, Bolwell, TGB).

There was no information in the crash dataset for vehicle body type, make or

model in 1,193 (13.9%) cases and these cases were coded as ‘Unknown’. The


‘Moped/Scooter’ (unknown), ‘Unknown’ and ‘Other’ categories were excluded from

the main analysis to allow valid comparison between mopeds, scooters and

motorcycles which could be reliably identified and classified. A relatively large

proportion (30%) of ‘Unknown’ and ‘Other’ PTW types crashed in rural or remote

areas (‘Other’ regions), compared to ‘Scooter’ (17%) and ‘Moped’ (13%) types.

This was also the case for ‘Motorcycle’ types (26.2%), leading to the conclusion that

‘Unknown’ PTW types are likely to be motorcycles in a large majority of cases.

Exclusion of the ‘Unknown’ and ‘Other’ PTW types left a total of 7,347

valid crash cases allocated to one of three PTW types; Moped, Scooter and

Motorcycle. PTWs registered outside Queensland are listed in the crash data as

vehicle body type ‘unknown’ and therefore excluded from the main analysis. Note

also that due to the exclusion from the crash data of unknown PTW types registered

in Queensland, the crash rates calculated may underestimate the true values.

5.2.3 Data analysis

Most of the analyses performed in this study were crash-based. However,

there were slightly more PTW controllers involved than there were PTW crashes,

due to a small number of crashes involving multiple PTWs. Controller-based

analyses were therefore conducted for rider age, gender and licence characteristics.

To compare the overall safety of mopeds and motorcycles, moped and

motorcycle crash rates per 10,000 registered LA and LC category PTWs were

calculated for each financial year from 1 July 2003 to 30 June 2008. The average

crash rate over the entire five year period was also calculated by dividing the total

number of crashes by the total number of registrations (registration years). The

number of registration years in the five year period July 2003 – June 2008 for LA

mopeds and LC motorcycles and scooters was calculated by subtracting LA

registrations from the total motorcycles on register (which includes mopeds) for each

financial year period, then adding the totals for each year for LA and LC categories.

For an estimate of crash rates relative to exposure, the crash rate per million vehicle

kilometres travelled (VKT) was also calculated using distance travelled data from a

previous study of PTW usage in Queensland (Harrison & Christie, 2006). As noted

above, most of the analyses performed in this study were crash-based, including

those on crash rates. Although this discounts the small number of crashes which


involved multiple PTWs, it does not significantly impact the results or overall

findings of the study. The use of a quasi-induced exposure method to calculate crash

rates was discussed in the review of literature in Chapter Two. It would have been

possible to use this method in the current study to estimate crash rates. However, for

reasons previously discussed it was thought to offer no greater reliability than the

two methods which were ultimately employed and so was not utilised.

In this descriptive study, the main method used for crash data analysis was

crosstabulation incorporating Pearson’s Chi-square ( ²) tests for statistically

significant differences at the .05 level. The effect sizes relating to significant

differences were also estimated using Cramer’s V (Øc) calculations measuring

strength of association. In this study a Cramer’s V of 0.10 was considered a small

effect size, while 0.50 or greater was considered large, as suggested in the literature

(Aron & Aron, 1999). Chi-square and Cramer’s V results are reported in the text

associated with specific tables. Post-hoc tests were also conducted to obtain adjusted

standardised residual statistics. These statistics indicated the particular cells in which

observed frequencies were significantly lower or higher than expected. Adjusted

standard residuals outside + 2.0 and - 2.0 were considered significant (Haberman,

1978) and these percentages are bolded in the tables throughout this chapter.

To examine factors influencing crash severity, an ordered probit model was

used to control for a range of variables in addition to PTW type. The ordered probit

regression model accounts for the ordered nature of the dependent variable, in this

case crash severity, of which there are five levels in the current study (Fatal,

Hospitalisation, Medical treatment, Minor injury and Property damage only). Details

of an ordered probit model specification used to examine factors influencing

motorcycle crash severity are available in Quddus, Nolan and Chin (2002).

Variables included in the current ordered probit model were those considered likely

to have some influence on crash severity according to the literature (Zambon &

Hasselberg, 2006; Quddus, Noland & Chin, 2002), including speed zone, horizontal

alignment (curvature), day of week, time of day, and number of units involved.

Of primary interest were the characteristics and patterns (if any) observable in

moped crashes, including possible differences between moped and motorcycle

crashes. Differences between scooter crashes and those involving mopeds and

motorcycles were also examined, and tested for significance where the number of

scooter crashes sufficed for valid statistical analysis. For purely descriptive


purposes, in some cases frequency tables presented in this chapter contain more

groups and categories than were used for the Chi-square test on the particular

variable. Where data are provided in tables for descriptive purposes but excluded

from statistical analysis, this is noted in the text accompanying the relevant table.

New variables were created in the data file where it was appropriate for the

purpose of statistical or descriptive analysis to collapse categories within an existing

variable. The new variables included: Time of crash, where hour of crash was

collapsed into three-hour groupings; Crash location, where the Local Government

Areas (LGAs) and Statistical Local Areas (SLAs) containing moped crashes were

collapsed into six main geographic areas (see section 5.2.4); Day of crash, where the

day of week was collapsed into weekday (Monday – Friday) and weekend (Saturday

– Sunday) crashes; Road configuration, where categories were collapsed into

intersection and non-intersection configurations; Number of Units per crash

(vehicles, cyclists and pedestrians), where number of Units was collapsed into

categories of 1, 2, 3 and 4 or more Units; and Contributing circumstances, where

contributing circumstances cited were collapsed into like groupings as outlined

below.

Analyses of contributing circumstances and fault attribution were conducted

to identify the main factors in crash causation and the road user types (PTW rider or

other road user) deemed at fault in association with particular circumstances. A

binary logistic regression analysis was conducted to predict the odds of the PTW

being at fault in association with various crash and rider characteristics. Widely used

in statistical analyses in road safety and other fields, this model controlled for the

influence of other (independent) variables on the variable of interest when estimating

adjusted odds ratios. For examples of logistic regression used to examine fault in

motorcycle crashes and more detail on the method see Kim and Boski (2001) and

Lardelli-Claret, Jimenez-Moleon et al. (2005).

These analyses assist in the identification of areas which may be amenable to

interventions for reducing crash risk. The contributing circumstances were grouped

together for analysis as described in Table 5.2. The contributing circumstance

recorded as ‘alcohol/drug’ was excluded from analysis. This was cited in cases

where any level of alcohol or the presence of drugs was detected in a unit controller,

legal or otherwise. In cases where blood-alcohol concentration (determined by

testing) exceeded the legal limit for that controller, the contributing circumstance


‘drink driver/rider’ was also cited. ‘Drink driver/rider’ was therefore used as the

variable for analysis as this provided a more reliable indication of the contribution of

alcohol to the crash. ‘Alcohol/drug’ was cited in 452 (6.2%) of all cases, of which

318 (70.4%) also had ‘drink driver/rider’ recorded as a contributing circumstance.

There was no information in the data regarding the involvement or detection of licit

or illicit substances other than alcohol and this could therefore not be assessed.

Table 5.2 Grouping of contributing circumstances into like categories

Contributing circumstance

group Contributing circumstance originally cited

Speed Speeding driver/rider

Drink driver/rider Drink driver/rider

Alcohol/drug (excluded, see text above table)

Violation Fail to give way or stop

Disobey traffic light/sign

Illegal manoeuvre

Disobey road rules – other

Dangerous driving Dangerous driving

Inattention/distracted Distracted

Negligence

Inattention

Inexperience Inexperience

Vehicle defects Vehicle defects – mechanical

Vehicle defects – external

Fatigue-related Fatigue-related

Road condition Road surface

Road gradient

Road quality

Road - wet

Road - works

Road - other

Other Age – lack of perception (typically older road user)

Driver condition – other

Other

5.2.4 Mapping crash data

Urban areas comprise a small proportion of Queensland’s 1,730,648 km2 land

mass and most of the State’s population is concentrated in the southeast corner, with

smaller concentrations in regional centres to the north (particularly in coastal

regions). Previous research suggests that moped and scooter use is predominantly an

urban activity, while motorcycle use is generally more widespread and more likely to


encompass non-urban as well as urban areas. Differences in crash characteristics

between PTW types may therefore be partly related to location in terms of

environment, trip purpose and demographic characteristics of riders. For this reason,

it was useful to examine the distribution of PTW types by crash location. The

mapping of crash data in MapInfo facilitates spatial analysis including the

identification of areas and regions where concentrations of particular crash types can

be readily identified.

Data were exported from SPSS to MapInfo software for presentation

purposes and spatial analysis. All PTW crashes (excluding the ‘Other’ category of

PTWs) were mapped at the Local Government Area (LGA) level (a PTW crash was

reported in 115 LGAs, 42 of which included a moped crash). Moped crashes were

also mapped at the geographically smaller Statistical Local Area (SLA) level (a PTW

crash was reported in 434 SLAs, 92 of which included a moped crash). Maps

produced in MapInfo describing the spatial distribution of crashes across Queensland

for the study period are provided in the Appendices of this document.

The LGAs containing moped crashes were collapsed into six main

geographic areas as follows: Brisbane area; Gold Coast area; Townsville area; Cairns

area; Fraser/Coral Coast area; Sunshine Coast area. These areas are based on a

grouping of adjacent LGAs which contain the majority (87.6%) of moped crashes.

The remaining moped crashes (12.4%) were spread across 23 LGAs and grouped as

‘Other areas’. Although two southern Brisbane area LGAs (Logan City and Redland

Shire) are directly adjacent to the Gold Coast City LGA, the Gold Coast area was

defined as a separate region due to expected differences which would be observable

in the crash data. Similarly, possible differences were expected between Brisbane

and Sunshine Coast areas. Therefore, the Sunshine Coast area (including

Caboolture) was separated from the Brisbane area at the northern boundary of

Redcliffe City and Pine Rivers LGAs, these two LGAs containing the northernmost

portion of the greater Brisbane urban area.

There is evidence that mopeds are a popular type of rental vehicle among

tourists in some Queensland locations, particularly around traditional holiday

destinations such as the Gold Coast and parts of coastal north Queensland. Tourists

may represent an at-risk group of moped riders due to a number of possible factors

including lack of knowledge of local road rules, roadways and environmental

conditions, lack of protective clothing, and alcohol or drug impairment, among


others. It is therefore worth examining any differences regarding licence status and

place where licence (if any) was obtained, as a proxy measure for identifying

possible tourist involvement in moped crashes.

5.3 PTWs registered in Queensland

The observed fourfold increase in Queensland moped crashes between 2001

and 2005 (Haworth & Nielson, 2008) occurred during a period of unprecedented

sales growth in the moped and scooter sectors of the PTW market. While this

reflects a pattern seen with PTWs in general over the last decade, in Queensland

moped registrations and moped crashes both increased at a faster rate than was the

case for motorcycles. Mopeds comprise only a small proportion of PTWs registered

in Queensland.

The motorcycle and moped registration data supplied by TMR were analysed

to examine trends in registrations. The increase in Queensland PTW registrations

(including mopeds, scooters and motorcycles) from 2001-2009 is evident in Table

5.3. Over this nine year period, mopeds increased as a proportion of all PTWs from

around one percent in 2001 to almost nine percent in 2009, with the most rapid

period of growth from 2004 to 2007. While the number of all PTW registrations

doubled over this period, there was an almost fifteen-fold increase in moped

registrations.

Table 5.3 Queensland PTW registrations by type and year, June 2001-June

2009

Year All PTW

registrations*

All PTWs

Index

Moped

registrations

Moped

Index

Mopeds as %

of PTWs

2001 77,274 100 917 100 1.2

2002 81,278 105 1,605 175 2.0

2003 85,566 111 2,281 249 2.7

2004 92,174 119 3,822 417 4.1

2005 101,656 132 5,239 571 5.2

2006 115,870 150 8,275 902 7.1

2007 130,786 169 10,660 1,162 8.2

2008 145,513 188 12,573 1,371 8.6

2009 155,220 201 13,668 1,490 8.8

*Includes mopeds. Source: Queensland Department of Transport and Main Roads.


While Table 5.3 includes PTW registrations to June 2009, the moped

registration data sourced by data request in July 2010 indicates a slight decrease in

moped registrations in the most recent twelve-month period to June 2010. There

were 13,435 moped registrations in 2009-2010, down from 13,668 in 2008-2009.

Recent declines in moped sales explain some stabilisation of registrations, though the

following was also noted in personal communication with the Queensland

Department of Transport and Main Roads (TMR) (pers. comm. TMR, July 2010):

with the introduction of the Learner Approved Motorcycle Scheme

(LAMS) in July 2009 the makes and models of the motorcycles went

through a cleansing and the correct model descriptions were

included. This in turn took some of the ones that previously had body

code recorded as MOPE to move into the body type of MCYC and vice

versa.

Under present conditions and regulations, the current popularity of mopeds

and scooters in Queensland is likely to be sustained and may increase further. The

implications of this are unclear from safety and transport planning perspectives. In

order to build on the limited research conducted to date in Queensland and other

Australian jurisdictions, it is therefore important to conduct in-depth analyses of the

most recent available crash data. For clarification of PTW definitions used in this

study, refer to the relevant section of the introductory chapter (section 1.2).

5.4 Results

5.4.1 Trends in PTW crashes

Of the crashes where PTW type could be identified, 91.3 percent involved

motorcycles, 7.4 percent involved mopeds and 1.3 percent involved scooters (N =

7,347). The frequency of crashes by PTW type and financial year is presented in

Table 5.4. The total number of reported PTW crashes increased each year, from

1,456 in 2003/04 to 1,907 in 2007/08. There was a statistically significant difference

between the increase in moped, scooter and motorcycle crashes over time [ ² (8) =

32.09, p < .001], reflecting the larger proportional increases in moped and scooter


crashes than motorcycle crashes. Reported moped crashes increased by around 100

percent over the study period, with scooter crashes increasing at a similar rate, while

motorcycle crashes increased only moderately by comparison. As a result of moped

crashes increasing at a faster rate than motorcycle crashes, they comprise an

increasing proportion of all PTW crashes over the study period. Mopeds represented

around 9.2 percent of crash-involved PTWs in 2007/2008, up from 5.7 percent in

2003/2004.

The number of crashes of unknown PTW types also doubled over the five

years, a substantially larger increase than for all known PTW types combined (22%).

Unknown PTW types in Table 5.4 include PTWs which were known to be either a

LA moped or LC scooter but could not be more accurately identified. These were

defined as ‘Moped/Scooter’ (unknown) and constituted 58 (4.6%) of the 1,261

unknown PTW types overall. The increase in Moped/Scooter (unknown) crashes

was roughly consistent with the doubling in moped and scooter crashes across the

five year period. Also included in the Unknown column in Table 5.4 are the 10

‘Other’ PTW types identified previously in Table 5.2.

Table 5.4 Queensland PTW crashes by type and year, July 2003-June 2009

Year PTW type

Motorcycle Moped Scooter Valid total Unknown Total

03/04 n

%

1,210

93.4

74

5.7

12

0.9

1,296

100.0

160

(11.0)

1,456

(100.0)

04/05 n

%

1,328

93.3

85

6.0

10

0.7

1,423

100.0

226

(13.7)

1,649

(100.0)

05/06 n

%

1,382

90.1

130

8.5

21

1.4

1,533

100.0

246

(13.8)

1,779

(100.0)

06/07 n

%

1,384

91.6

106

7.0

21

1.4

1,511

100.0

306

(16.8)

1,817

(100.0)

07/08 n

%

1,407

88.8

146

9.2

31

2.0

1,584

100.0

323

(16.9)

1,907

(100.0)

Total

03/08

n

%

6,711

91.3

541

7.4

95

1.3

7,347

100.0

1,261

(14.6)

8,608

(100.0)

A small proportion of crashes (1.4%) involved multiple PTWs, the data for

which are presented below in Table 5.5. There were four crashes involving a moped

and a single motorcycle, and one crash involving a moped and a scooter. There were

91 crashes involving multiple motorcycles, and one crash involving a motorcycle and

an unknown PTW type.


Table 5.5 PTW crashes involving other PTWs

Crashes and PTWs

involved

PTW type

Moped Scooter Motorcycle Unknown Total

Crashes 541 95 6,711 1,261 8,608

PTWs in crashes 542 95 6,809 1,299 8,745

Other PTWs involved

None 537 94 6,619 1,235 8,485

1 moped - 1 4* - 1

1 scooter 1** - - - -

1 motorcycle 4 - 86 - 90

2 motorcycles - - 4 - 4

3 motorcycles - - 1 - 1

1 unknown - - 1 25 26

2 unknown - - - 1 1 * Not counted toward motorcycle crash total as they are already counted as moped crashes.

**Not counted toward moped crash total as this is already counted as a scooter crash.

5.4.1.1 Crash rates

The total number of moped crashes doubled from 2003/04 to 2007/08 and

moped crashes increased significantly as a proportion of all PTW crashes. Table 5.3

shows that the number of registered mopeds increased more than threefold over the

same period. The crash rates per 10,000 registrations by financial year over five

years for each PTW type are presented below in Table 5.6. The overall crash rate per

10,000 registration years across the five year period from July 2003 to June 2008 was

133.4 for LA mopeds and 124.8 for LC motorcycles and scooters.

Crash rates fell for both registration categories across the study period. The

rate of moped crashes per 10,000 registered vehicles declined by 40 percent from

193.6 to 116.1 over the five year period, while the rate of motorcycle and scooter

crashes declined only moderately by comparison (22%, from 138.3 to 108.21). LC

category scooter registrations, crashes and crash rates are subsumed within the LC

motorcycle data in Table 5.6.


Table 5.6 Crashes per 10,000 QLD registrations by financial year and

registration (ADR) category

Category Crash year

03/04 04/05 05/06 06/07 07/08 03-08

LA Moped Registrations 3,822 5,239 8,275 10,660 12,573 40,569

Crashes 74 85 130 106 146 541

Crashes/10,000

registrations 193.6 162.2 157.1 99.4 116.1 133.4

95% CI 149.5-

237.7

127.8-

196.7

130.1-

184.1

80.5-

118.4

97.3-

135.0

122.1-

144.6

LC Motorcycle

Registrations* 88,352 96,417 107,595 120,126 132,940 545,430

Crashes 1,222 1,338 1,403 1,405 1,438 6,806

Crashes/10,000

registrations 138.3 138.8 130.4 117.0 108.2 124.8

95% CI 130.6-

146.1

131.3-

146.2

123.6-

137.2

110.8-

123.1

102.6-

113.8

121.8-

127.7

*Motorcycles and scooters on register excluding LA moped registrations

A survey of PTW usage in Queensland was conducted by Harrison and

Christie (2006) in 2005. This survey collected data on distance travelled annually by

mopeds (n = 140), scooters (n = 88) and motorcycles (n = 2,975). As these data were

collected from across Queensland within the current study period, they provided an

opportunity to calculate police-reported crash rates per vehicle kilometre travelled

(VKT) for LA mopeds and LC motorcycles/scooters. These crash rates are presented

below in Table 5.7. The crash rates were 6.33 and 1.70 per million VKT for LA and

LC category vehicles respectively, a moped crash rate per VKT nearly four times

that of motorcycles and scooters.


Table 5.7 Crash rates per VKT for 5 years (using data from Harrison and

Christie, 2006)

Statistic ADR category

LA Moped LC Motorcycle

Registrations for 5 years 40,569 545,430

Mean VKT/year 2,107 7,327

Total moped km for 5 years 85,478,883 3,996,365,610

Crashes 541 6,806

Crashes/1million VKT 6.33 1.70

Median VKT/year 1,050 5,000

Number of survey respondents * 140 2,975

Range 15,000 70,000

*excludes 2 moped rider outliers who reported travelling > 32,000 km/year

5.4.2 Crash characteristics

5.4.2.1 Injured road user type

The casualty data indicated a total of 8,015 road users injured in the 7,347

crashes involving known PTW types (Table 5.8). Overall, approximately 89 percent

of injured persons were PTW riders, with PTW pillions comprising a further five

percent. The remaining six percent of injured persons were drivers and passengers of

other vehicles (4.4%), pedestrians (0.9%) and cyclists (0.5%). There was no

significant difference in the distribution of injured road user types by PTW type [ ²

(5) = 8.96, p = .111] (excluding scooters due to low numbers). The mean number of

persons injured per crash was 1.08 for moped and scooter crashes alike, and 1.09 for

motorcycle crashes.

Table 5.8 Injured road user type by PTW type for reported crashes, July 2003-

June 2008 (including fatally injured)

Road user

type*

PTW type

Moped Scooter Motorcycle Total

n % n % n % n %

PTW rider 523 89.9 93 90.3 6,508 88.8 7,124 88.9

PTW pillion 31 5.3 5 4.9 391 5.3 427 5.3

OV driver 14 2.4 1 1.0 273 3.7 288 3.6

OV passenger 1 0.2 2 1.9 61 0.8 64 0.8

Pedestrian 8 1.4 2 1.9 63 0.9 73 0.9

Cyclist 5 0.9 - 0.0 34 0.5 39 0.5

Total 582 100 103 100 7,330 100 8,015 100

*OV = Other vehicle


5.4.2.2 Crash severity

The following analysis is crash-based and as such does not describe the total

number of persons killed or injured. Of the 541 LA moped crashes, five resulted in a

single fatality and 242 resulted in hospitalisation of at least one road user. An

additional 288 crashes resulted in medical treatment (205) and minor injury (83) of at

least one person. Of the 6,806 LC motorcycle and scooter crashes, 231 resulted in at

least one fatality and 3,442 resulted in hospitalisation, while a further 2,992 resulted

in medical treatment (1,963) or minor injury (1,029). A small proportion (2%) of

crashes resulted in property damage only. Unfortunately the available data on crash

and injury severity are coarse and do not provide any information on the actual injury

types, required treatments and outcomes, or the duration of hospital admissions.

Overall, there was a statistically significant difference in crash severity

between the three PTW types, as evident in Table 5.9. Separating the three PTW

types, moped and scooter crashes were less likely than motorcycle crashes to result

in death or hospitalisation, and more likely to result in medical treatment [ ² (6) =

33.22, p < .001, Øc = .05] (excluding property damage only). Comparable

proportions of moped, scooter and motorcycle crashes resulted in minor injury and

property damage only.

Low numbers of fatal moped and scooter crashes preclude testing for

statistical significance of crash severity by rider age group. However, fatal moped

crashes (N = 5) involved riders aged 30-49 years (40%) and 75 years or over (60%).

The single fatal scooter crash involved a 30-39 year old rider. For fatal motorcycle

crashes (N = 230), 23.9 percent involved a younger rider (17-24 years), while only

3.9 percent involved an older rider (60 years or over).

Table 5.9 Crash severity by PTW type for report crashes, July 2003-June 2008

Crash severity

PTW type

Moped Scooter Motorcycle All PTWs

n % n % n % n %

Fatal 5 0.9 1 1.1 230 3.4 236 3.2

Hospitalisation 242 44.7 41 43.2 3,401 50.7 3,684 50.1

Medical treatment 205 37.9 38 40.0 1,925 28.7 2,168 29.5

Minor injury 83 15.3 15 15.8 1,014 15.1 1,112 15.1

Property damage 6 1.1 - 0.0 141 2.1 147 2.0

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0


The crash rates per 10,000 registration years by crash severity and PTW type

over five years are presented below in Table 5.10. The difference in crash severity

for mopeds compared to motorcycles as described above is reflected in the analysis

of crash rates by severity. Although scooters are included in the rates for

motorcycles and not separately described, the low number of scooter crashes has a

negligible influence on motorcycle crash rates. The confidence intervals for crash

rates by severity suggest that the clearest and most important difference between

mopeds and motorcycles is the greater likelihood of a fatal crash for motorcyclists.

There were 4.2 motorcyclist fatalities per 10,000 motorcycle registration years,

compared with 1.2 fatalities for mopeds. There was also a clear difference in the

greater likelihood of medical treatment for moped riders compared to motorcyclists,

with respective rates of 50.1 and 36.0 per 10,000 registration years.

Table 5.10 Crashes per 10,000 registration years by crash severity level and

registration category

PTW

Crash severity

Fatal Hospital-

isation

Medically

treated

Minor

injury

Property

damage

LA crashes/10,000

register years

(N = 40,569)

1.2 59.7 50.1 20.5 1.5

95% CI 0.2-2.3 52.1-67.2 43.6-57.4 16.1-24.9 0.3-2.7

LC crashes/10,000

register years

(N = 545,430)

4.2

63.1

36.0

18.9

2.6

95% CI 3.7-4.8 61.0-65.2 34.4-37.6 17.7-20.0 2.2-3.0

Non-use of helmets by PTW riders is a known contributor to increased injury

risk and injury severity. While there are no data in this study on specific injury types

and related outcomes, the vast majority of all PTW riders (99%) wore a helmet at the

time of the crash (N = 6,632, excluding cases where helmet use was unknown or not

applicable to the road user type in the casualties file). Moped riders wore a helmet in

97 percent of cases, compared with 100 percent of scooters riders and 99 percent of

motorcycle riders.

Crash severity is further examined using an ordered probit model following

the descriptive analyses of crash characteristics in Sections 5.4.2.3 to 5.4.2.6.


5.4.2.3 Crash location

As outlined in the introduction to this chapter, it is expected that the crash

data will reflect differences in usage by location, due to a proportionally greater use

of mopeds for commuting compared to motorcycles, and the use of mopeds by

tourists in holiday destinations. Examination of crash location by PTW type will

therefore help to answer the research questions concerning moped usage compared to

that of motorcycles.

As summarised in Table 5.11, around half of all PTW crashes (53%) occurred

in the Brisbane and Gold Coast areas. There were statistically significant differences

in crash location by PTW type [ ² (12) = 178.39, p < .001, Øc = .11]. Moped crashes

were more likely to occur in the Gold Coast region (18%) and less likely to occur in

Brisbane (33%) compared to motorcycle and scooter crashes. Crashes on the Gold

Coast represented 15 percent of cases involving scooters and 10.5 percent of

motorcycle crashes. Around 22 percent of moped crashes occurred in the northern

regional centres of Cairns and Townsville, while approximately 10 percent of

motorcycle and scooter crashes occurred in those areas. The Brisbane and Gold

Coast areas accounted for a relatively large proportion (66%) of LC scooter crashes.

Scooter crashes occurred in Brisbane in 52 percent of cases, compared with 33

percent and 43 percent of moped and motorcycle crashes respectively.

While PTW crashes involved more motorcycles than mopeds by a factor of

about 12.4 to 1 across all Queensland (excluding LC scooters), there was

considerable variation by geographic area. In particular, moped crashes were

proportionally high in the Fraser/Coral Coast, Cairns, Townsville and Gold Coast

areas, where there were between five and seven motorcycle crashes for every moped

crash reported. Moped crashes were proportionally low in other areas which mostly

comprise rural and remote locations. Maps of crash distribution for all PTW types

by Local Government Area (LGA) and for moped crashes by Statistical Local Area

(SLA) for selected areas are provided as Appendices.

In the Statistical Local Areas (SLAs) of Brisbane City (‘City – Inner’ and

‘City – Remainder’) over the 5 year study period there was a total of 124 reported

PTW crashes, of which 110 (88.7%) involved known PTW types. The known PTW

types included 9 (8.2%) mopeds, 3 (2.7%) scooters and 98 (89.1%) motorcycles (N =

110). Of the 14 cases involving unknown PTW types, 3 (21%) were either mopeds


or scooters (unknown), representing 2.4 percent of all reported crashes in the

Brisbane City SLAs (N = 124).

Table 5.11 Location characteristics of moped, scooter and motorcycle crashes

Location

PTW type

Moped Scooter Motorcycle *motorcycle/

moped ratio n % n % n %

Brisbane area 180 32.7 49 51.6 2,863 42.7 15.9

Gold Coast 99 18.3 14 14.7 705 10.5 7.2

Sunshine Coast 38 7.0 6 6.3 718 10.7 18.9

Cairns area 55 10.2 3 3.2 285 4.2 5.2

Townsville area 62 11.5 6 6.3 368 5.5 6.0

Fraser/Coral Coast 40 7.4 3 3.2 195 2.9 4.9

Other areas 67 12.4 14 14.7 1,577 23.5 23.5

Total 541 100.0 95 100.0 6,711 100.0 12.4

*Excludes scooters

5.4.2.4 Moped crash location and licence jurisdiction

For moped riders, the jurisdiction where their driver’s licence was issued is

compared with the crash location (Local Government Area) in Table 5.12. While the

number of moped crashes was too low to allow valid statistical analysis, there were

apparent differences regarding crash location and the place of licence issue.

Compared to Queensland as a whole, riders holding interstate licences were

overrepresented in the Gold Coast, Cairns and (to a lesser extent) Townsville areas,

while riders licensed in other countries were overrepresented in the Cairns area and

the Fraser/Coral Coast region. Riders with licence jurisdiction or status listed as

‘unknown’ were overrepresented in all of these areas and it appears highly likely that

at least some of these riders were not residents of Queensland. Again it must be

noted that a car licence is valid for moped riding in Queensland.

Of the ten crashes in the Gold Coast area where the moped controller was

listed as licensed interstate, five (50%) appeared to hold a NSW licence which was

valid for moped riding in Queensland. Given the close proximity of the Gold Coast

to the NSW border, it is possible that some or all of these riders are residents of the

general region and may not necessarily be tourists or irregular visitors.

In the Townsville/Thuringowa area in north Queensland, 17 of 62 moped

crashes (27.4%) occurred on Magnetic Island in the Townsville City Council LGA.

While these numbers are insufficient for valid testing of statistical significance, the


Magnetic Island crashes differ characteristically from others in the area. With the

exception of one ‘head on’ crash, all Magnetic Island cases were single vehicle ‘off

path’ crashes, compared with mainland crashes which involved more than one

vehicle in 57.8 percent (26) of cases (N = 45). Female moped controllers were

involved in 76.5 percent (13) of Magnetic Island cases (N = 17), compared with 35.5

percent (16) in mainland Townsville/Thuringowa (N = 45). All six Queensland-

licensed controllers in Magnetic Island crashes were female (licence state was

‘Unknown’ in 4 cases). Magnetic Island cases involved controllers licensed outside

of Queensland in at least 41.2 percent of cases (4 interstate, 3 international),

suggesting high tourist involvement, while mainland crashes in the area all involved

Queensland-licensed controllers. Alcohol does not appear to have contributed to

moped crashes in the area, with the exception of one case in the mainland Townsville

LGA involving a male learner rider aged 17-20 years detected with BAC 0.05 – 0.09

(prescribed BAC for Learner licence holders is 0.00).


Table 5.12 Moped crashes by LGA and place licensed, July 2003-June 2008

LGA Group & LGA Total

Place licensed

QLD Interstate Not Aust. Unknown*

n n % n % n % n %

Brisbane area

Brisbane City

Logan City

Redcliffe City

Ipswich City

Pine Rivers

Redland Shire

180 165 91.7 - 0.0 5 2.8 10 5.6

162

5

6

3

3

1

148

4

6

3

3

1

-

-

-

-

-

-

4

1

-

-

-

-

10

-

-

-

-

-

Gold Coast

Gold Coast City 99 65 65.6 10 10.1 5 5.0 19 19.2

99 65 10 5 19

Townsville area

Townsville City

Thuringowa City

62 48 77.4 4 6.5 3 4.8 7 11.3

51

11

37

11

4

-

3

-

7

-

Cairns area

Cairns City

Douglas Shire

55 36 65.4 4 7.3 7 12.7 8 14.5

49

6

34

2

3

1

6

1

6

2

Fraser/Coral Coast

Bundaberg City

Hervey Bay City

Miriam Vale Shire

Maryborough City

40 31 77.5 2 5.0 4 10.0 3 7.5

20

8

7

5

18

7

1

5

2

-

-

-

-

-

4

-

-

1

2

-

Sunshine Coast

Maroochy Shire

Caboolture Shire

Noosa Shire

Caloundra City

38 36 94.7 - 0.0 1 2.6 1 2.6

14

10

8

6

13

10

7

6

-

-

-

-

1

-

-

-

-

-

1

-

Other areas (23 LGAs) 67 59 88.1 3 4.5 0 0.0 5 7.5

Total 541 440 81.3 23 4.3 25 4.6 53 9.8

*The ‘Unknown’ column in this table includes some but not all cases where the ‘controller licence

type’ is listed as ‘Not known’, ‘Unlicensed’, ‘Never held a licence’, ‘Not applicable’ (where moped

was parked, stationary), ‘Cancelled; disqualified’ and ‘Inappropriate class’.

5.4.2.5 Temporal characteristics of crashes

The distributions of crashes by day of week and time of day for each PTW

type are presented in Table 5.13 and Table 5.14 respectively. Moped and scooter

crashes were more likely than motorcycle crashes to occur on weekdays (79-81%

compared with 69%) and the difference was statistically significant [ ² (2) = 27.85, p

< .001, Øc = .06]. There was also a statistically significant difference between


moped and motorcycle crash patterns by day of week, with moped crashes more

likely to occur on Wednesdays (18%) and motorcycle crashes less likely to occur on

Tuesdays (13%) or Wednesdays (14%) [ ² (12) = 34.90, p < .001, Øc = .05].

Table 5.13 Day of week for moped, scooter and motorcycle crashes

Day of week

PTW type


n % n % n % n %

Weekday 428 79.1 77 81.1 4,659 69.4 5,164 70.3

Weekend 113 20.9 18 18.9 2,052 30.6 2,183 29.7

Monday 66 12.2 13 13.7 827 12.3 906 12.3

Tuesday 83 15.3 17 17.9 857 12.8 957 13.0

Wednesday 97 17.9 16 16.8 956 14.2 1,069 14.6

Thursday 85 15.7 14 14.7 966 14.4 1,065 14.5

Friday 97 17.9 17 17.9 1,053 15.7 1,167 15.9

Saturday 63 11.6 13 13.7 1,019 15.2 1,095 14.9

Sunday 50 9.2 5 5.3 1,033 15.4 1,088 14.8

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Table 5.14 shows that for all PTW types, more than three quarters of all

crashes occurred during daylight hours (6am – 6pm). Overall, there was no apparent

difference between mopeds and motorcycles on this variable and while scooter

crashes appeared slightly less likely to occur at night, the difference was not

statistically significant [ ² (2) = 1.85, p = .397]. For each PTW type, the highest

frequency of crashes occurred in the 3pm – 6pm period. This was true on weekdays

when most crashes occurred, but there were different patterns on weekends.

Analysis of weekend crashes only found a statistically significant difference

between mopeds, scooters and motorcycles in daytime and night-time crash

involvement [ ² (2) = 6.45, p = .040, Øc = .05]. Moped crashes were more likely to

occur at night on weekends (29%) than on weekdays (21%). The reverse was true of

scooter crashes, with a smaller proportion of weekend crashes occurring at night

(11%) compared with weekday crashes (18%). For motorcycles, similar proportions

of crashes occurred at night-time on weekdays (23%) and weekends (20%). On

weekends, moped crashes peaked between 12pm and 3pm, while motorcycle crashes

were evenly distributed from 9am to 6pm. Weekend scooter crashes mainly occurred

from 12pm to 6pm.


Table 5.14 Time of day for moped, scooter and motorcycle crashes

Time of day

PTW type


n % n % n % n %

6am – 6pm 416 76.9 79 83.2 5,231 77.9 5,726 77.9

6pm – 6am 125 23.1 16 16.8 1,480 22.1 1,621 22.1

6am – 9am 68 12.6 19 20.0 948 14.1 1,035 14.1

9am – 12pm 84 15.5 18 18.9 1,106 16.5 1,208 16.4

12pm – 3pm 117 21.6 18 18.9 1,279 19.1 1,414 19.2

3pm – 6pm 147 27.2 24 25.3 1,898 28.3 2,069 28.2

6pm – 9pm 64 11.8 11 11.6 793 11.8 868 11.8

9pm – 12am 30 5.5 4 4.2 338 5.0 372 5.1

12am – 3am 8 1.5 1 1.1 143 2.1 152 2.1

3am – 6am 23 4.2 - 0.0 206 3.1 229 3.1

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Weekday

6am – 6pm 336 78.5 63 81.8 3,591 77.1 3,990 77.3

6pm – 6am 92 21.5 14 18.2 1,068 22.9 1,174 22.7

6am – 9am 62 14.5 17 22.1 788 16.9 867 16.8

9am – 12pm 68 15.9 15 19.5 578 12.4 661 12.8

12pm – 3pm 81 18.9 13 16.9 763 16.4 857 16.6

3pm – 6pm 125 29.2 18 23.4 1,462 31.4 1,605 31.1

6pm – 9pm 53 12.4 10 13.0 587 12.6 650 12.6

9pm – 12am 18 4.2 4 5.2 247 5.3 269 5.2

12am – 3am 3 0.7 - 0.0 77 1.7 80 1.5

3am – 6am 18 4.2 - 0.0 157 3.4 175 3.4

Total 428 100.0 77 100.0 4,659 100.0 5,164 100.0

Weekend

6am – 6pm 80 70.8 16 88.9 1,640 79.9 1,736 79.5

6pm – 6am 33 29.2 2 11.1 412 20.1 447 20.5

6am – 9am 6 5.3 2 11.1 160 7.8 168 7.7

9am – 12pm 16 14.2 3 16.7 528 25.7 547 25.1

12pm – 3pm 36 31.9 5 27.8 516 25.1 557 25.5

3pm – 6pm 22 19.5 6 33.3 436 21.2 464 21.3

6pm – 9pm 11 9.7 1 5.6 206 10.0 218 10.0

9pm – 12am 12 10.6 - 0.0 91 4.4 103 4.7

12am – 3am 5 4.4 1 5.6 66 3.2 54 2.5

3am – 6am 5 4.4 - 0.0 49 2.4 54 2.5

Total 113 100.0 18 100.0 2,052 100.0 2,183 100.0


5.4.2.6 Roadway characteristics

Moped and scooter crashes were more likely than motorcycle crashes to

occur in speed zones of 60 km/h or less and the difference was statistically

significant [ ² (8) = 131.85, p < .001, Øc = .09] (Table 5.15). A large majority of

moped and scooter crashes (90% and 88% respectively) occurred in speed zones up

to 60 km/h, compared with 70 percent for motorcycle crashes. Motorcycle crashes

occurred in speed zones of 80 km/h or more in 25 percent of cases, compared with

6.5 percent for both moped and scooter crashes.

It was thought that the observed difference in crash severity by PTW type

reported above (section 5.4.1.3) might reflect the speed zones in which crashes

occurred. To test whether this was the case, an analysis of crash severity by speed

zone and PTW type was conducted, results of which are presented in Table 5.15.

There was a statistically significant difference in crash severity between mopeds,

scooters and motorcycles in speed zones up to 60 km/h, in which around 90 percent

of moped and scooter crashes occurred [ ² (8) = 27.70, p < .001, Øc = .05]. In 60

km/h zones, moped crashes were less likely than motorcycle crashes to result in a

fatality (1.1% versus 2.6%) or hospitalisation (40.1% versus 47.3%). Moped crashes

were more likely than motorcycle crashes to result in medical treatment without

hospitalisation (42.4 percent versus 30.9%). There was no significant difference in

crash severity in 50 km/h zones, while the number of moped crashes in other speed

zones was too low for valid statistical analysis.

Mopeds and scooters were similar regarding the proportion of crashes

occurring at intersections (Table 5.15). Slightly more than half (53%) of all moped

and scooter crashes occurred at intersections, compared with 46 percent for

motorcycle crashes, a difference which was statistically significant [ ² (2) = 12.42, p

= .002, Øc = .04].

All three PTW types differed significantly with regard to roadway horizontal

alignment [ ² (2) = 46.78, p < .001, Øc = .08]. While most crashes occurred on

straight road sections for all PTW types (72%), this was more common for mopeds

(83%) and scooters (86%) than for motorcycles (71%).

Analysis of crashes on wet and dry sealed road surfaces revealed a

statistically significant difference between PTW types [ ² (2) = 11.90, p = .003, Øc =


.0] (excluding unknown cases). Moped crashes were more likely than motorcycle

crashes to occur on wet road surfaces, while scooter crashes were slightly less likely

(Table 5.15).

Table 5.15 Roadway characteristics of moped, scooter and motorcycle crashes

Characteristic

PTW type


n % n % n % n %

Speed zone km/h

0 – 50 136 25.1 17 17.9 1,034 15.4 1,187 16.2

60 349 64.5 67 70.5 3,632 54.1 4,048 55.1

70 21 3.9 5 5.3 389 5.8 415 5.6

80 – 90 21 3.9 4 4.2 720 10.7 745 10.1

100 – 110 14 2.6 2 2.1 936 13.9 952 13.0

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Crash severity in up to

60 km/h zone

Fatal 4 0.8 - 0.0 117 2.5 97 2.4

Hospitalisation 212 43.7 36 42.9 2,210 47.4 1,885 46.6

Medical treatment 191 39.4 34 40.5 1,434 30.7 1,299 32.1

Minor injury 73 15.1 14 16.7 796 17.1 674 16.7

Property damage 5 1.0 - 0.0 109 2.3 93 2.3

Total 485 100.0 84 100.0 4,666 100.0 5,235 100.0

Intersection

Yes 289 53.4 50 52.6 3,088 46.0 3,427 46.6

No 252 46.6 45 47.4 3,623 54.0 3,920 53.4

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Horizontal alignment

Straight 449 83.0 82 86.3 4,751 70.8 5,282 71.9

Curve 92 17.0 13 13.7 1,960 29.2 2,065 28.1

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Road surface*

Dry 469 87.3 89 93.7 6,053 91.6 6,611 91.3

Wet 68 12.7 6 6.3 557 8.4 631 8.7

Valid total 537 100.0 95 100.0 6,610 100.0 7,242 100.0

Unknown 4 - 101 105

*Sealed roads only

5.4.2.7 Crash type

Vehicles (units) involved in crashes

Crashes were coded according to the number of units involved, where a ‘unit’

is defined as any road user or vehicle involved in the actual collision. This includes


pedestrians and unoccupied (usually parked) vehicles as well as occupied vehicles,

whereby a collision with such a unit was coded as a multiple-unit crash. The

recording of units generally excludes road users who may have contributed to a crash

but were not involved in actual collision.

A summary of the number of units involved in crashes by PTW type is

presented below in Table 5.16. Comparing the three PTW types for multi-unit crash

involvement, scooters were the most likely to be involved in a multi-unit crash

(79%), followed by mopeds (71%), with motorcycles least likely to be involved in a

multi-unit crash (66%). The differences were statistically significant when all three

PTW types were analysed together [ ² (2) = 13.13, p = .001, Øc = .04] (N = 7,347),

and also when scooters were excluded to compare only mopeds with motorcycles [ ²

(1) = 6.20, p = .013, Øc = .03] (N = 7,252). The average number of units involved in

multi-unit crashes was 2.06 for mopeds, 2.05 for scooters and 2.08 for motorcycles.

Table 5.16 Number of units involved in moped, scooter and motorcycle crashes

Number

of units

PTW type


n % n % n % n %

Single 157 29.0 20 21.1 2,301 34.3 2,478 33.7

Multiple 384 71.0 75 78.9 4,410 65.7 4,869 66.3

1 unit 157 29.0 20 21.1 2,301 34.3 2,478 33.7

2 units 364 67.3 72 75.8 4,141 61.7 4,577 62.3

3 units 17 3.1 2 2.1 214 3.2 233 3.2

4> units 3 0.6 1 1.1 55 0.8 59 0.8

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

It was thought that the observed differences in crash severity by PTW type

reported in section 4.4.1.3 (Crash severity) may be a function of the number of units

involved in crashes. To test whether this was the case, an analysis of crash severity

by the number of units involved (single or multi-unit) was conducted. The results of

this analysis are presented below in Table 5.17. It was found that single vehicle

crashes were more severe on average than multi-unit crashes and the difference was

statistically significant [ ² (4) = 96.10, p < .001, Øc = .11]. While 33.7 percent of all

crashes involved only one vehicle, 43.6 percent of fatal crashes were single vehicle

crashes. Table 5.17 shows a consistent decline in severity in line with a decline in

the proportion of crashes which were single vehicle crashes.


Table 5.17 Number of units involved in crashes by crash severity

Crash severity

Number of units

Single vehicle Multi-vehicle

n % n %

Fatal 103 43.6 133 56.4

Hospitalisation 1,384 37.6 2,300 62.4

Medical treatment 692 31.9 1,476 68.1

Minor injury 272 24.5 840 75.5

Property damage only 27 18.4 120 81.6

Total 2,478 33.7 4,869 66.3

The types of other units involved in crashes with PTWs are presented in

Table 5.18. In multi-unit crashes, there was no difference by PTW type in the types

of other units involved [ ² (5) = 8.39, p = .136] (N = 5,072, excluding scooters and

multi-PTW crashes). Almost 90 percent of multi-unit PTW crashes involved a car,

van, utility or four-wheel drive (SUV). Trucks and buses were involved in about five

percent of multi-unit PTW crashes, while the remaining eight percent of crashes

involved pedestrians, animals, cyclists, other PTWs and other unit types (railway

stock, special purpose vehicles). Mopeds and scooters appeared slightly more likely

than motorcycles to collide with pedestrians, while motorcycles were slightly more

likely to collide with animals and other PTWs. Crashes with other PTWs occurred in

4 cases for mopeds, 96 cases for motorcycles and once for scooters.

Table 5.18 Road user types involved in multi-unit crashes with a PTW

Road user type

PTW type


n % n % n % n %

Car/Van/Utility 362 88.9 71 89.9 4,153 87.2 4,586 87.3

Truck/Bus 17 4.2 4 5.1 225 4.7 246 4.7

Pedestrian 8 2.0 2 2.5 63 1.3 73 1.4

Cyclist 5 1.2 - 0.0 34 0.7 39 0.7

Animal 6 1.5 1 1.3 161 3.4 168 3.2

Other PTW 4 1.0 1 1.3 96 2.0 101 1.9

Other 5 1.2 - 0.0 33 0.7 38 0.7

Total 407 100.0 79 100.0 4,765 100.0 5,251 100.0


Ordered probit model of severity

Following the initial analyses of crash severity, crash characteristics and rider

characteristics, an ordered probit model was used to further examine crash severity,

to specifically identify factors influencing crash severity for each PTW type. As

explained above in the methodology section on data analysis (Section 5.2.3), an

ordered probit regression model accounts for the ordered nature of the dependent

variable, in this case crash severity. Variables included in the ordered probit model

were those considered likely to have some influence on crash severity according to

the literature (Quddus et al., 2002; Zambon & Hasselberg, 2006), including speed

zone, horizontal alignment (curvature), day of week, time of day, and number of

units involved. The base (referent) groups for each variable were selected according

to their predominance in the overall data (see Tables 5.13 - 5.16 relating to relevant

crash characteristics). For example, as 55 percent of crashes occurred in 60 km/h

speed zones, this speed zone was selected as the base group for speed zone.

Similarly, as the majority of crashes occurred on straight road sections (72%), on

weekdays (70%), between the hours of 6am and 6pm (78%), and were multi-vehicle

crashes (66%), these were selected as the base groups for the respective variables.

The ordered probit model of severity was statistically significant for the entire

model for all PTWs [p <.001], and for mopeds [p = .048], scooters [p = .0045] and

motorcycles [p <.001] modelled separately. In the entire model for all PTWs there

was no statistically significant difference in crash severity by PTW type (moped

coefficient estimate -0.052, p = .285; scooter coefficient estimate -0.026, p = .813).

The three PTW types differed in terms of the particular factors which significantly

influenced crash severity. As shown in Table 5.19, when stratified by PTW type,

moped crashes were more severe between the hours of 6pm and 6am [p = .04] and in

speed zones of 90 km/h or more [p = .033]. Scooter crashes were more severe in 70

km/h speed zones [p = .016], though not in higher speed zones, and were less severe

on weekends [p = .002]. Motorcycle crashes were more severe in speed zones of 80

km/h or more [p <.001], on curves [p <.001], in single vehicle crashes [p <.001], on

weekends [p = .001] and between the hours of 6pm and 6am [p = .001].


Table 5.19 Parameter coefficient estimates of ordered probit model of severity

by PTW type

Variable Estimate Std.

error

z-

statistic p-value

95% CI

Lower Upper Moped

Speed zone (base 60km/h)

Up to 50 0.162 .113 1.43 .154 -0.060 0.384

70km/h -0.024 .250 -0.10 .922 -0.514 0.465

80km/h 0.019 .251 0.08 .939 -0.472 0.511

90km/h 0.693 .325 2.13 .033 0.056 1.331

On curve -0.014 .133 -0.11 .915 -0.276 0.247

Weekend -0.037 .119 -0.31 .754 -0.271 0.196

6pm-6am 0.240 .117 2.06 .040 0.011 0.469

Single vehicle 0.186 .110 1.68 .092 -0.030 0.403

Scooter


Up to 50 -0.092 .321 -0.29 .774 -0.721 0.537

70km/h 1.653 .686 2.41 .016 0.309 2.998

80km/h 0.408 .618 0.66 .509 -0.803 1.619

90km/h -0.823 .814 -1.01 .312 -2.418 0.773

On curve 0.165 .389 0.42 .673 -0.599 0.928

Weekend -0.935 .307 -3.04 .002 -1.538 -0.333

6pm-6am 0.554 .334 1.66 .097 -0.100 1.208

Single vehicle 0.228 .328 0.69 .487 -0.415 0.871

Motorcycle


Up to 50 0.003 .038 0.08 .934 -0.072 0.078

70km/h 0.082 .058 1.40 .161 -0.033 0.196

80km/h 0.239 .046 5.19 <.001 0.149 0.330

90km/h 0.374 .042 8.85 <.001 0.292 0.457

On curve 0.161 .033 4.95 <.001 0.097 0.225

Weekend 0.098 .030 3.22 .001 0.038 0.157

6pm-6am 0.104 .033 3.21 .001 0.041 0.168

Single vehicle 0.115 .032 3.64 <.001 0.053 0.177

Crash configuration

The crash configuration (termed ‘crash nature’ in the original data file) by

PTW type as recorded in the data is presented in Table 5.20. The six most prevalent

categories of crash configuration are presented individually, while the remaining five

categories were collapsed into the ‘other’ category. The difference in crash

configuration by PTW type was statistically significant [ ² (12) = 33.49, p = .001, Øc

= .05]. ‘Angle’ crashes comprised a large minority of cases for all PTW types, but

were most likely in scooter crashes (44%), followed by moped crashes (41%), with

motorcycle crashes least likely (35%). By contrast, ‘fall from vehicle’ crashes were

more prevalent in motorcycle crashes (23%) than in both moped and scooter crashes

(18%). Scooters were less likely than either mopeds or motorcycles to be involved in

‘hit object’ crashes, and more likely than either mopeds or motorcycles to be


involved in ‘sideswipe’ crashes. The differences in crash configuration by PTW type

are likely a consequence of the difference in number of units involved by PTW type

shown earlier (single vehicle or multi-vehicle). ‘Fall from vehicle’ and ‘hit object’

crashes constitute 99 percent of single vehicle crash configuration, consistent with

the higher proportion of motorcycle crashes with these crash configurations. ‘Angle’

crashes accounted for 53 percent of multi-unit crash configurations, explaining the

relatively high involvement of mopeds and scooters compared to motorcycles in that

type of crash.

Table 5.20 Crash configuration of moped, scooter and motorcycle crashes

Crash configuration

PTW type


n % n % n % n %

Angle 220 40.7 42 44.2 2,330 34.7 2,592 35.3

Fall from vehicle 98 18.1 17 17.9 1,570 23.4 1,685 22.9

Hit object 70 12.9 6 6.3 1,016 15.1 1,092 14.9

Rear end 74 13.7 12 12.6 815 12.2 901 12.3

Sideswipe 49 9.1 13 13.7 503 7.5 565 7.7

Head-on 4 0.7 2 2.1 148 2.2 154 2.1

Other* 26 4.8 3 3.2 329 4.9 358 4.9

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

*Includes ‘hit animal’, ‘hit parked vehicle’, ‘hit pedestrian’, ‘overturned’ and ‘other’

Crash group description

The variable ‘crash group description’ was derived from the original variable

provided in the data as ‘crash definitions for coding accidents (DCA) description’.

Table 5.21 summarises the reported crash group descriptions in order of frequency.

Analysis of crash type revealed a statistically significant difference between PTW

types [ ² (16) = 76.52, p < .001, Øc = .07] (N = 7,146) after excluding the two least

frequently cited crash types due to low numbers (there were no cases of scooter

involvement in ‘overtaking’ or ‘left or right turn’ crashes). For all PTWs, the most

common crash group descriptions were ‘same direction’ crashes, followed by

‘adjacent approach’ crashes (‘same direction’ crashes do not include ‘overtaking

crashes, which are coded separately in crash group description). Mopeds were more

likely than motorcycles to be involved in ‘adjacent approach’ (intersection), ‘off path

on straight’ and ‘manoeuvring’ crashes. Motorcycles were clearly more likely than

either mopeds or scooters to be involved in ‘off path on curve’ crashes. Scooters


were more likely than either mopeds or motorcycles to be involved in ‘same

direction’ crashes and did not appear consistently similar to either of the other PTW

types across the range of crash group descriptions listed.

Table 5.21 Crash group description of moped, scooter and motorcycle crashes

Crash group

description

PTW type


n % n % n % n %

Same direction 126 23.3 31 32.6 1,562 23.3 1,719 23.4

Adjacent approach 103 19.0 17 17.9 947 14.1 1,067 14.5

Off path on straight 88 16.3 10 10.5 836 12.5 934 12.7

Opposite approach 75 13.9 15 15.8 1,058 15.8 1,148 15.6

Manoeuvring 56 10.4 9 9.5 386 5.8 451 6.1

Off path on curve 38 7.0 6 6.3 1,032 15.4 1,076 14.6

On path 24 4.4 1 1.1 332 4.9 357 4.9

Passenger & misc. 15 2.8 4 4.2 308 4.6 327 4.5

Pedestrians 8 1.5 2 2.1 57 0.8 67 0.9

Overtaking 6 1.1 - 0.0 166 2.5 172 2.3

Left or right turn 2 0.4 - 0.0 27 0.4 29 0.4

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

In order to provide a greater level of detail on the most prevalent crash group

descriptions, breakdowns of the five most frequent moped crash group descriptions

listed in Table 5.21 are presented below. The most frequent crash group description

group for mopeds is described as ‘same direction’ crashes, a breakdown of which is

presented in Table 5.22. Rear-end crashes were the most frequent crash group

description within the ‘same direction’ group for all PTW types, comprising more

than half of all moped and motorcycle ‘same direction’ crashes. Within the ‘same

direction’ group, rear-end crashes were less likely for scooters (39%) compared to

mopeds (56%) and motorcycles (53%), though the difference was not statistically

significant. It should be noted that within the breakdown of crash configuration in

the previous section (Table 5.20), the proportion of scooter rear-end crashes is

comparable to that for both mopeds and motorcycles. The number of rear-end

crashes within the ‘same direction’ crash description group (n = 906) is slightly more

than the number of rear-end crashes recorded under ‘crash configuration’ (n = 901).

In 825 cases there was a direct match between the two variables, but in the remaining

cases only one variable had ‘rear-end’ recorded while the other variable provided an

alternative description (most frequently, ‘fall from vehicle’ was recorded in crash

configuration for 62 rear-end crashes).


Table 5.22 Breakdown of crash description – ‘Same direction’

‘Same direction’

group description

PTW type


n % n % n % n %

Rear-end 71 56.3 12 38.7 823 52.7 906 52.7

Lane changes 31 24.6 8 25.8 346 22.2 385 22.4

Parallel lanes turning 21 16.7 8 25.8 355 22.7 384 22.3

U-turn and Other 3 2.4 3 9.7 38 2.5 44 2.5

Total 126 100.0 31 100.0 1,562 100.0 1,719 100.0

‘Adjacent approach’ crashes were the second most frequent moped crash

group description (N = 1,067), all of which were described as ‘intersection from

adjacent approach’ crashes. Adjacent approach crashes accounted for similar

proportions of moped (19%) and scooter (18%) crashes, and a slightly smaller

proportion of motorcycle crashes (14%).

The results for ‘off path on straight’ crashes are presented below in Table

5.23, in which it is evident that ‘out of control on straight’ is the most prominent

crash description recorded for all PTW types within that group of crashes.

Compared to motorcycle crashes, moped crashes were less likely to be defined as

‘out of control on straight’, and more likely to be ‘off carriageway hit object’

crashes. The low number of scooter crashes within this grouping precludes any

reliable conclusions for that PTW type on this particular crash description.

Table 5.23 Breakdown of crash description – ‘Off path on straight’

‘Off path on straight’

group description

PTW type


n % n % n % n %

Out of control on straight 39 44.3 7 70.0 450 53.8 496 53.1

Off carriageway hit object 23 26.1 - 0.0 124 14.8 147 15.7

Off carriageway on straight 19 21.6 1 10.0 143 17.1 163 17.5

Other 7 8.0 2 20.0 119 14.2 128 13.7

Total 88 100.0 10 100.0 836 100.0 934 100.0

A breakdown of ‘opposite approach crashes’ is presented below in Table

5.24. All PTW types appear likely within this crash group description to be involved

in collision with another vehicle while turning, and mopeds are particularly

prominent in this crash group description. Compared to motorcycles and scooters,

mopeds are less likely to be involved in ‘head-on’ crashes, while for both mopeds


and scooters, ‘U-turn and Other’ crashes are infrequent compared with motorcycle

crashes.

Table 5.24 Breakdown of crash description – ‘Opposite approach’

‘Opposite approach’

group description

PTW type


n % n % n % n %

Opposing vehicles turning 68 90.7 11 73.3 741 70.0 820 71.4

Head-on 5 6.7 4 26.7 220 20.8 229 19.9

U-turn and Other 2 2.7 - 0.0 97 9.2 99 8.6

Total 75 100.0 15 100.0 1,058 100.0 1,148 100.0

The fifth most frequent moped crash description group is defined as

‘manoeuvring’ crashes, results for which are presented in Table 5.25.

‘Manoeuvring’ crashes involved a vehicle exiting a driveway in 60.7 percent of

moped crashes and 70.5 percent of motorcycle crashes. The types of crash

constituting ‘Other’ within this crash description grouping are not known, while ‘hit

parked vehicle’ crashes were relatively infrequent for all PTW types. As with ‘off

path on straight’ crashes, the low number of scooter crashes within this grouping

arguably precludes any reliable conclusions for that PTW type on this particular

crash description.

Table 5.25 Breakdown of crash description – ‘Manoeuvring’

5.4.3 PTW controlling characteristics

5.4.3.1 Age and gender

The age and gender characteristics of PTW riders in crashes are summarised

in Table 5.26. While motorcycle crashes involved a male rider (controller) in

‘Manoeuvring’ group

description

PTW type


n % n % n % n % Vehicle leaving

driveway 34 60.7 5 55.6 272 70.5 311 69.0

Other 17 30.4 3 33.3 84 21.8 104 23.1

Hit parked vehicle 5 8.9 1 11.1 30 7.8 36 8.0

Total 56 100.0 9 100.0 386 100.0 451 100.0


approximately 92 percent of cases, males comprised a substantially smaller

proportion of crashed moped riders at around 63 percent. The gender distribution for

crashed scooter riders fell between that for the moped and motorcycle crashes (78%

male). The differences in controller gender distribution by PTW type were

statistically significant, with Cramer’s V indicating a moderate effect size [ ² (2) =

519.35, p < .001, Øc = .26] (excluding unknown gender) (N = 7,408).

Medians of rider ages were calculated from the grouped age data provided.

The median age was lowest for moped riders (31.67 years), highest for scooter riders

(38.79 years) and intermediate for motorcycle riders (34.56 years). The age

distribution of crashed moped riders differed from that of both motorcycle and

scooter riders, and the differences were statistically significant [ ² (14) = 178.46, p <

.001, Øc = .11] (excluding unknown age and age below 17 years) (N = 7,374).

Moped crashes involved a higher proportion of riders under 25 years of age (31%)

than either motorcycle crashes (23%) or scooter crashes (9%). Scooter crashes

involved a relatively high proportion of older riders with 14 percent aged 60 years

older, compared with nine percent and three percent for moped and motorcycle riders

respectively.

As shown in Table 5.27, for moped crashes only (where age and gender were

known), female riders were more likely to be aged under 30 (54%) compared to male

riders (42%) (N = 536). Females were also less likely to be aged 60 years or over

(3.5%) compared to male moped riders (12%). The difference in age distribution by

gender for moped crashes was statistically significant when cases with age unknown

and gender unknown were excluded [ ² (7) = 17.44, p = .015]. There was no

significant difference in age by gender for motorcycle crashes, while scooter crash

numbers were too low to allow a valid statistical analysis of age distribution by

gender.


Table 5.26 Age and gender characteristics of PTW riders in crashes

Characteristic

PTW type

Moped Scooter Motorcycle All PTWs

n % n % n % n %

Gender

Male 344 63.1 75 78.1 6,284 92.3 6,703 90.0

Female 198 36.3 21 21.9 486 7.1 705 9.5

Unknown 3 0.6 - 0.0 37 0.5 40 0.5

Total 545 100.0 96 100.0 6,807 100.0 7,448 100.0

Age group

0-16 - 0.0 - 0.0 2 0.0 2 0.0

17-20 68 12.5 2 2.1 574 8.4 644 8.6

21-24 103 18.9 7 7.3 987 14.5 1,097 14.7

25-29 79 14.5 10 10.4 966 14.2 1,055 14.2

30-39 108 19.8 32 33.3 1,821 26.8 1,961 26.3

40-49 77 14.1 22 22.9 1,492 21.9 1,591 21.4

50-59 54 9.9 9 9.4 719 10.6 782 10.5

60-74 35 6.4 10 10.4 173 2.5 218 2.9

75> 12 2.2 3 3.1 11 0.2 26 0.3

Unknown 9 1.5 1 1.0 62 0.9 72 1.0

Total 545 100.0 96 100.0 6,807 100.0 7,448 100.0

Median age 31.7 38.8 34.6 34.5

Table 5.27 Moped rider age by gender crosstabulation

Age group

Moped rider gender

Male Female

n % n %

17-20 37 10.9 31 15.7

21-24 61 18.0 42 21.2

25-29 45 13.3 34 17.2

30-39 72 21.3 36 18.2

40-49 52 15.4 25 12.6

50-59 31 9.2 23 11.6

60-74 28 8.3 7 3.5

75 and over 12 3.6 - 0.0

Total 335 100.0 198 100.0

5.4.3.2 Licence characteristics

The licence characteristics of riders in moped, scooter and motorcycle crashes

are presented in Table 5.28. The difference between moped and motorcycle riders in

licence status was statistically significant after excluding cases where licence status

was ‘not known’ or ‘not applicable’ [ ² (8) = 139.17, p < .001, Øc = .14] (N =


7,206). Scooter riders were excluded from this analysis due to low numbers, but it

can be seen that they are similar to motorcycle riders in the proportion of riders

holding an open licence. Moped riders were less likely than motorcycle riders (and

probably scooter riders) to hold an open licence (issued in Australia), though it must

be noted that in the case of moped riders this refers to either a car or motorcycle

licence, where for other PTW riders it indicates possession of an open motorcycle

licence. Moped riders were slightly more likely to hold a provisional or restricted

licence, and to be unlicensed or not licensed in Australia. Scooter riders appeared

least likely to be unlicensed, though the numbers are too low to be reliable for

statistical tests.

There was also a statistically significant difference between moped and

motorcycle riders with regard to the place where their licence was issued (excluding

unknown cases) [ ² (4) = 223.25, p < .001, Øc = .18] (N = 7,163). Again, scooter

riders were excluded from this analysis due to low numbers. Moped riders were less

likely than motorcycle (and probably scooter) riders to hold a Queensland licence.

Moped riders licensed outside of Queensland were evenly split between interstate

and overseas licence holders. The ‘unknown’ category under ‘licence issued’ in

Table 5.28 includes some but not all unlicensed riders, as well as cases where licence

status is unknown. The licence jurisdiction of moped riders by crash location was

previously addressed in section 5.4.2.2 (Moped crash location and rider licence).


Table 5.28 Licence characteristics of PTW controllers in crashes

Licence characteristic

PTW type

Moped Scooter Motorcycle

n % n % n %

Licence status

Open/full 357 65.5 79 82.3 5,539 81.4

Provisional/restricted 51 9.4 5 5.2 519 7.6

Learner 29 5.3 5 5.2 292 4.3

Not licensed Australia 23 4.2 2 2.1 26 0.4

Unlicensed* 35 6.4 2 2.1 335 4.9

Not applicable 3 0.6 - 0.0 31 0.5

Not known 47 8.6 3 3.1 65 1.0

Total 545 100.0 96 100.0 6,807 100.0

Licence issued

Queensland 443 81.3 89 92.7 6,603 97.0

Interstate 23 4.2 2 2.1 41 0.6

Overseas 25 4.6 2 2.1 28 0.4

Unknown 54 9.9 3 3.1 135 2.0

Total 545 100.0 96 100.0 6,807 100.0

*Includes ‘Unlicensed’, ‘Cancelled; disqualified’, ‘Expired’, ‘Never held a licence’ and ‘Inappropriate

class’.

5.4.4 Fault attribution and contributing circumstances

Fault attribution and contributing circumstances were analysed to identify the

main factors in crash causation and the road user types deemed at fault in association

with particular circumstances. The contributing circumstances cited for crashes

should be viewed with caution. Contributing circumstances were clearly more likely

to be attributed in single vehicle crashes (99.5%) than in multi-vehicle crashes

(42.4%) [ ² (3) = 2283.15, p < .001, Øc = .56]. It is not known why no contributing

circumstance was reported for nearly 58 percent of multi-vehicle crashes. Where

multiple contributing circumstances are recorded for a single crash, the data are not

weighted to indicate which particular factors were more important contributors than

others. However, as is evident below, a large majority of cases do not have more

than one contributing circumstance attributed to each unit involved. Where this is

the case, the circumstance attributed to Unit 1 (most at fault) may be reasonably

assumed to have been the major contributing factor in most cases. As such, the

attribution of contributing circumstances is easier to interpret when the unit most at

fault is identified. There were very few contributing circumstances attributed to

PTWs when another unit was considered most at fault.


5.4.4.1 Attribution of fault

Considering all crashes for all known PTW types (N = 7,347), PTWs overall

were designated Unit 1 (most at fault) in 58 percent of cases and there was a

statistically significant difference between PTW types [ ² (2) = 11.22, p = .004, Øc =

.04] (Table 5.29). In single and multi-unit crashes combined, scooters were least

likely to be designated Unit 1 (45%), compared with mopeds (54%) and motorcycles

(59%). Although scooters were less likely than either mopeds or motorcycles to be

designated Unit 1, there was still a statistically significant difference between

mopeds and motorcycles when scooters were excluded from analysis [ ² (1) = 4.62,

p = .032] (N = 7,252).

PTWs were designated Unit 1 in all cases involving only one Unit, though

contributing circumstances were not always cited. PTWs were Unit 1 in 31 – 37

percent of multi-unit crashes, depending on the PTW type involved. Analysis of the

designation of the PTW as Unit 1 in all multi-unit crashes revealed no statistically

significant difference between mopeds, scooters and motorcycles [ ² (2) = 2.07, p =

.354] (N = 4,869).

Table 5.29 Attribution of Unit 1 (most at fault)

Unit 1 (most at fault)

PTW type


n % n % n % n %

All crashes

PTW 292 54.0 43 45.3 3,940 58.7 4,275 58.2

Other road user 249 46.0 52 54.7 2,771 41.3 3,072 41.8

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Multi-Unit crashes

PTW 135 35.2 23 30.7 1,639 37.2 1,797 36.9

Other road user 249 64.8 52 69.3 2,771 62.8 3,072 63.1

Total 384 100.0 75 100.0 4,412 100.0 4,869 100.0

Crash configuration and crash group description as a function of fault in multi-unit

crashes

While PTWs were Unit 1 in about 37 percent of multi-unit crashes overall,

particular multi-unit crash configurations and crash group descriptions yielded

different results when analysed as a function of fault. Tables 5.30 and 5.31 show that

some multi-unit crash types were more likely to be the fault of a PTW rider than


other crash types. There were also some differences between PTW types in the

proportions designated Unit 1 for particular configurations and descriptions, although

low numbers precluded valid statistical analyses on some of these variables.

For crash configuration (Table 5.30), a statistically significant difference was

found in ‘angle’ crashes, with moped riders more likely to be designated Unit 1

(30%) than motorcycle riders (21%) [ ² (2) = 9.06, p = .011] (N = 2,592). A

significant difference was also found in ‘rear end’ crashes, with moped riders less

likely to be designated Unit 1(39%) than motorcycle riders (56%) [ ² (2) = 7.43, p =

.024] (N = 901). In the analysis of rear-end crashes under the crash group description

‘same direction’ (n = 906), there was a larger difference between PTW types in the

attribution of fault [ ² (2) = 13.49, p = .001], with mopeds designated Unit 1 in 38

percent of rear-end crashes, compared with 50 percent and 60 percent for scooters

and motorcycles respectively. Put simply, mopeds were less likely to be the striking

vehicle in rear-end crashes than were scooters and motorcycles.

Overall, the multi-unit crash configurations most likely to be the fault of

PTW riders were ‘hit object’, ‘fall from vehicle’ and ‘head-on’ crashes, while ‘angle’

and ‘sideswipe’ crashes were the most likely to have another unit at fault.

Table 5.30 Proportions of crash configuration with PTW designated Unit 1

(multi-unit crashes)

Crash

configuration

PTW type


n % n % n % n %

Angle 66 30.0 11 26.2 498 21.4 575 22.2

Fall from vehicle 1 16.7 1 33.3 166 66.1 168 64.6

Hit object 4 80.0 - - 49 80.3 53 80.3

Rear end 29 39.2 6 50.0 453 55.6 488 54.2

Sideswipe 15 30.6 2 15.4 161 32.0 178 31.5

Head-on 2 50.0 2 100.0 95 64.2 99 64.3

Other* 18 69.2 1 33.3 217 71.9 236 71.3

*Includes ‘hit animal’, ‘hit parked vehicle’, ‘hit pedestrian’, ‘overturned’ and ‘other’

Table 5.31 shows the crash group descriptions for which a PTW was

designated Unit 1. Of the 10 crash group descriptions involving a Unit 1 motorcycle,

nine had also involved a Unit 1 moped in at least one case, while five had involved a

Unit 1 scooter in at least one case. In regard to some of these crash group

descriptions, mopeds and scooters were involved in a very low numbers of crashes.

The only crash group description to show a valid statistically significant difference


between PTW types was that for ‘manoeuvring’ crashes. Unit 1 in these crashes was

more likely to be a moped (49%) than a scooter (37.5%) or motorcycle (20%) [ ² (2)

= 20.54, p < .001]. Crash group descriptions in which PTWs overall were least

likely to be designated Unit 1 were ‘adjacent approach’, ‘manoeuvring’, ‘pedestrian’

and ‘opposite approach’ crashes.

Motorcycles were most likely to be designated Unit1 in ‘passenger and

miscellaneous’ (89%) and ‘overtaking’ crashes (85%). Mopeds and scooters were

not involved in crashes with those descriptions in sufficient numbers for valid

comparison with motorcycles.

Table 5.31 Proportions of crash group descriptions with PTW designated Unit

1 (multi-unit crashes)

Crash group

description

PTW type


n % n % n % n %

Same direction 46 36.5 10 32.3 699 44.8 755 43.9

Adjacent approach 26 25.2 3 17.6 167 17.6 196 18.4

Off path on straight 3 60.0 - - 13 54.2 16 55.2

Opposite approach 22 29.3 6 40.0 324 30.6 352 30.7

Manoeuvring 24 49.0 3 37.5 71 20.1 98 23.9

Off path on curve - - - - 5 55.6 5 55.6

On path 4 66.7 - - 42 79.2 46 78.0

Passenger & misc. 6 100.0 1 50.0 186 89.0 193 88.9

Pedestrians 3 37.5 - - 15 26.3 18 26.9

Overtaking 1 16.7 - - 117 84.8 118 81.9

5.4.4.2 Contributing circumstances attributed to a PTW

As evident in Table 5.32, there was no contributing circumstance attributed to

a PTW for 38 percent of all cases and there was a statistically significant difference

between PTW types in the number of circumstances cited [ ² (6) = 15.42, p = .017,

Øc = .03]. Moped and scooter crashes both appeared less likely than motorcycle

crashes to have at least one contributing circumstance attributed to a PTW, clearly

suggesting (as above) that mopeds and scooters are less likely than motorcycles to be

designated Unit 1. The number of circumstances cited appeared generally to increase

with crash severity and the overall greater severity of motorcycle crashes was shown

earlier.


Table 5.32 Number of contributing circumstances attributed to all PTWs

Number of

circumstances

PTW type


n % n % n % n %

None 239 44.2 45 47.4 2,534 37.8 2,818 38.4

1 212 39.2 40 42.1 3,093 46.1 3,345 45.5

2 80 14.8 9 9.5 935 13.9 1,024 13.9

3 or more 10 1.8 1 1.1 149 2.2 160 2.2

Total 541 100.0 95 100.0 6,711 100.0 7,347 100.0

Across all 7,347 cases there were 5,619 cases where at least one contributing

circumstance was attributed to a PTW and the number of circumstances cited ranged

from one to six. Where a PTW was designated Unit 1 and at least one contributing

circumstance was cited, the mean number of circumstances attributed to PTWs was

1.40 with a range of one to five (Table 5.33). On average, fewer contributing

circumstances were attributed to scooters (1.23) than to mopeds (1.43) or

motorcycles (1.40) when the PTW rider was designated Unit 1.

Table 5.33 Contributing circumstances (CCs) attributed to a Unit 1 PTW

PTW type Crashes (n) CCs (n) Range Mean SD

Moped 294 421 1 – 4 1.43 .624

Scooter 43 53 1 – 4 1.23 .571

Motorcycle 3,948 5,532 1 – 5 1.40 .663

Total 4,285 6,006 1 – 5 1.40 .659

The distribution of contributing circumstances attributed to a PTW in all

crashes is presented below in Table 5.34. This table does not indicate that a PTW

was necessarily most at fault (designated Unit 1) and as such is purely the

distribution of contributing circumstances attributed to the three PTW types.

‘Inattention/distracted/negligent’ represents the most frequently cited group of

contributing circumstances for mopeds and motorcycles, and is second only to

‘other’ circumstances in the case of scooters. ‘Inattention/distracted/negligent’

contains the commonly cited violation ‘undue care and attention’ and was cited in 16

percent, 17 percent and 18 percent of all moped, scooter and motorcycle crashes

respectively. This was a frequently cited contributing circumstance for at-fault

PTWs in single and multi-unit crashes alike (Table 5.33 and Table 5.32).

Overall, ‘road condition’ contributed more to motorcycle crashes (14.5%)

than to moped (10.0%) or scooter (4.2%) crashes. In single vehicle crashes, this was


deemed a contributing circumstance in about one third of moped and motorcycle

crashes alike, but was clearly less likely to be noted in scooter crashes.

Alcohol appears to have contributed to a small minority (<5%) of all crashes

where a PTW rider was at fault and was least observed in scooter crashes. In single

vehicle crashes, alcohol was more frequently attributed to a PTW rider in moped

crashes (11%) and motorcycle crashes (9%).

Scooter crashes attracted a relatively high proportion of ‘other’ circumstances

due to the inclusion in this group of ‘age; lack of perception, power or

concentration’. This contributing circumstance is typically attributed to older road

users, who comprised a relatively high proportion of scooter riders compared to

mopeds and motorcycles. ‘Inexperience’ is more frequently cited in moped crashes

(9%) than in motorcycle (5%) or scooter (4%) crashes due to a relatively high

involvement of young riders. In terms of factors which actually contributed to a

crash, the criteria for attribution of age-related circumstances other than age itself are

ambiguous and hence viewed with caution.

Table 5.34 Contributing circumstances attributed to a PTW (all crashes)

Contributing

circumstance

PTW type


n % n % n % n %

Speed-related 11 2.0 - 0.0 478 7.1 489 6.7

Drink driver 23 4.3 2 2.1 293 4.4 318 4.3

Violation 75 13.9 11 11.6 611 9.1 697 9.5

Inattention/distracted

/negligent 85 15.7 16 16.8 1,209 18.0 1,310 17.8

Dangerous driving 10 1.8 1 1.1 180 2.7 191 2.6

Fatigue-related 2 0.4 - 0.0 124 1.8 126 1.7

Inexperience 51 9.4 4 4.2 348 5.2 403 5.5

Road condition 54 10.0 4 4.2 970 14.5 1,028 14.0

Vehicle defects 3 0.6 1 1.1 103 1.5 107 1.5

Other 79 14.6 19 20.0 1,031 15.4 1,129 15.4

Total 393 58 5,347 5,798

Table 5.35 shows that compared with all crashes and single vehicle crashes,

multi-unit crashes were more likely to involve a violation of road rules by a PTW

rider where the PTW was designated Unit 1. In multi-unit crashes, violations were

attributed to at-fault riders in 54 percent, 48 percent and 35 percent of moped,

scooter and motorcycle crashes respectively. This mostly refers to right-of-way and

signal violations. Speeding violations and speed-related contributing circumstances


are separately recorded. In both single and multi-unit crashes, speed-related

circumstances were more likely to be attributed to motorcycle riders than to moped

or scooter riders.

Table 5.35 Contributing circumstances attributed to a PTW (multi-unit crashes,

PTW Unit 1)

Contributing

circumstance

PTW type


n % n % n % n %

Speed-related 5 3.7 - 0.0 148 9.0 153 8.5

Drink driver 5 3.7 1 4.3 65 4.0 71 4.0

Violation 73 54.1 11 47.8 571 34.8 655 36.4


/negligent 35 25.9 6 26.1 458 27.9 499 27.8


Fatigue-related - 0.0 - 0.0 8 0.5 8 0.4

Inexperience 21 15.6 3 13.0 156 9.5 180 10.0

Road condition 3 2.2 1 4.3 135 8.2 139 7.7

Vehicle defects - 0.0 - 0.0 17 1.0 17 0.9

Other 27 20.0 5 21.7 365 22.3 397 22.1

Total 179 28 2,082 2,289

Inexperience was more likely to be attributed to a PTW rider in moped

crashes than in scooter or motorcycle crashes. This was particularly the case in

single vehicle crashes, where inexperience was attributed to moped riders in 18.5

percent of cases, compared with 5.0 percent and 7.6 percent of scooter and

motorcycle riders respectively (Table 5.36).

Table 5.36 Contributing circumstances (single vehicle crashes)

Contributing

circumstance

PTW type


n % n % n % n %

Speed-related 6 3.8 - 0.0 284 12.3 290 11.7

Drink driver 17 10.8 - 0.0 214 9.3 231 9.3

Violation - 0.0 - 0.0 8 0.3 8 0.3


/negligent 50 31.8 10 50.0 738 32.1 798 32.2

Dangerous driving - 0.0 - 0.0 19 0.8 19 0.8

Fatigue-related 2 1.3 - 0.0 115 5.0 117 4.7

Inexperience 29 18.5 1 5.0 175 7.6 205 8.3

Road condition 51 32.5 3 15.0 820 35.6 874 35.3

Vehicle defects 3 1.9 1 5.0 85 3.7 89 3.6

Other 46 29.3 8 40.0 532 23.1 586 23.6

Total 204 23 2,990 3,217


5.4.4.3 Contributing circumstances attributed to other road users

In the 4,869 multi-unit crashes there were a total of 4,371 contributing

circumstances attributed to another road user (excluding animals and other PTWs).

The number of contributing circumstances attributed to other road users in all multi-

unit crashes is summarised in Table 5.37. The attribution of contributing

circumstances to another road user was similar for moped and motorcycle crashes.

Scooter crashes were more likely than crashes involving mopeds or motorcycles to

have at least one contributing circumstance attributed to another road user.

Table 5.37 Number of contributing circumstances attributed to other road users

in multi-unit crashes (whether Unit 1 or not)

Number of

circumstances

PTW type


n % n % n % n %

None 141 35.0 22 28.2 1,558 33.4 1,721 33.4

1 175 43.4 39 50.0 2,068 44.3 2,282 44.3

2 75 18.6 15 19.2 876 18.8 966 18.8

3 or more 12 2.9 2 2.6 167 3.6 181 3.6

Where another road user was designated Unit 1 and at least one contributing

circumstance was cited, the mean number of circumstances attributed to other road

users was 1.43 with a range of one to six (Table 5.38). On average, fewer

contributing circumstances were attributed to other road users in scooter crashes

(1.23) compared with moped (1.41) and motorcycle (1.40) crashes.

Table 5.38 Number of contributing circumstances attributed to other (Unit 1)

road users in multi-unit crashes (excluding animals, and other PTWs)

PTW type Crashes (n) CCs (n) Range Mean SD

Moped 247 349 1 – 5 1.41 .644

Scooter 52 70 1 – 3 1.35 .566

Motorcycle 2,762 3,952 1 – 6 1.43 .641

Total 3,061 4,371 1 – 6 1.43 .639

A summary of the contributing circumstances cited for other road users is

presented in Table 5.39. Where a contributing circumstance was attributed to

another road user, a right-of-way violation was reported in approximately three

quarters of cases for all PTW types. The majority of these violations involved a


failure to give way, illegal manoeuvres and failure to obey a traffic light or sign

(Table 5.40). The next most frequently cited group of contributing circumstances

was inattention/distraction/negligence, and particularly inattention which was more

often attributed in moped (24.3%) than in motorcycle (15.6%) or scooter (13.5%)

crashes. Inexperience was attributed to other road users in 12 percent of moped

crashes and 13 percent of motorcycle crashes, but only 4 percent of scooter crashes.

Table 5.39 Contributing circumstances attributed to other (Unit 1) road users

in multi-unit crashes (excluding crashes with animals and other PTWs)

Contributing

circumstance

PTW type


n % n % n % n %

Speed-related 1 0.4 - 0.0 14 0.5 15 0.5

Drink driver 5 2.0 1 1.9 49 1.8 55 1.8

Violation 179 72.5 38 73.1 2,145 77.7 2,362 77.2


/negligent 61 24.7 9 17.3 467 16.9 537 17.5


Fatigue-related - 0.0 - 0.0 1 0.0 1 0.0

Inexperience 29 11.7 2 3.8 369 13.4 400 13.1

Road condition 5 2.0 2 3.8 97 3.4 104 3.4

Vehicle defects 2 0.8 - 0.0 9 0.4 11 0.4

Other* 52 21.0 12 23.6 601 21.8 665 21.7

Total 342 66 3,860 4,268

*Includes ‘Age – lack of power/perception’ (typically older road user); ‘Driver condition – other’; and

‘Other’.

Table 5.40 Main circumstances attributed to other (Unit 1) road users in multi-

unit crashes (excluding crashes with animals and other PTWs)

Contributing circumstance

PTW type

Moped Scooter Motorcycle

n % n % n %

Violation 179 72.5 38 73.0 2,145 77.6

Fail to give way or stop 92 37.2 18 34.6 959 34.7

Illegal manoeuvre 73 29.6 19 36.5 1,022 37.0

Disobey traffic light/sign 13 5.3 1 1.9 150 5.4

Disobey road rules - other 1 0.4 - 0.0 14 0.5

Inattention/distracted/negligent 61 24.7 9 17.3 467 16.9

Inattention 60 24.3 7 13.5 430 15.6

Negligence 1 0.4 2 3.8 34 1.2

Distracted - 0.0 - 0.0 3 0.1


5.4.4.4 Logistic regression of fault in multi-vehicle crashes

The results of the binary logistic regression analysis considering attribution of

fault (Unit 1) in multi-vehicle PTW crashes as a whole are presented in Table 5.41.

The entire model was significant [ ² (19) = 1090.48, p < .001], explaining

approximately 28% of the variance in Unit 1 (most at fault) attribution [Nagelkerke

R² = 0.277]. There were significant effects of rider age, crash configuration, speed

zone, contributing circumstances (speed and alcohol), intersection crashes, time of

day (day vs. night) and day of week (weekday vs. weekend) and licence status

(licensed vs. unlicensed). Neither PTW type nor gender showed statistically

significant differences so could not be considered to predict fault in the overall

model.

Specifically, the statistically significant differences include that older rider

(60>) and younger rider (<25) crashes had 2.1 times and 1.4 times higher odds

respectively of the PTW being designated Unit 1 compared to those of riders aged 25

to 59 years. Angle crashes had 9.1 times lower odds of the PTW being designated

Unit 1 compared to all other crash configurations. Sideswipe crashes had 6.7 times

lower odds of the PTW being designated Unit 1 compared to all other crash

configurations. Fall from vehicle crashes had 1.6 times lower odds of the PTW

being designated Unit 1 compared to all other crash configurations. Rear end crashes

had 2.4 times lower odds of the PTW being designated Unit 1. Intersection crashes

had 1.2 times lower odds of the PTW being designated Unit 1 compared to non-

intersection crashes. Crashes in higher speed zones (80> km/h) had twice the odds of

the PTW being designated Unit 1 compared to crashes in 60 km/h zones. Crashes in

70 km/h zones had 1.3 times the odds of the PTW being designated Unit 1. Crashes

where excessive speed was cited as a contributing factor had 5.3 times higher odds of

the PTW being designated Unit 1 than if speed was not attributed. Crashes where

alcohol was cited as a contributing factor had 6 times higher odds of the PTW being

designated Unit 1 than if alcohol was not cited. Crashes at night (6pm-6am) had 1.4

times lower odds than daytime crashes of the PTW being designated Unit 1. Crashes

on weekends had 1.3 times higher odds of the PTW being designated Unit 1

compared to weekday crashes. Crashes involving unlicensed riders had

approximately twice odds of the PTW being designated Unit 1 compared to those

involving licensed riders.


Table 5.41 Entire logistic regression predicting odds of PTW being Unit 1

Variable Crude

OR OR* Wald

p-

value

95% CI**

Lower Upper

PTW type

Motorcycle (base) 1.00 1.00

Moped 0.92 1.11 0.697 .404 0.867 1.424

Scooter 0.75 0.99 0.001 .977 0.578 1.701

Age group

25-59 (base) 1.00 1.00

17-24 1.33 1.41 18.980 <.001 1.211 1.655

60 or over 1.80 2.07 14.871 <.001 1.429 2.988

Angle crash

No (base) 1.00 1.00

Yes 0.25 0.11 213.56 <.001 0.079 0.144

Fall from vehicle

No (base) 1.00 1.00

Yes 3.33 0.61 6.178 .013 0.418 0.902

Head on

No (base) 1.00 1.00

Yes 3.20 0.47 10.896 .001 0.303 0.738

Hit object

No (base) 1.00 1.00

Yes 7.14 1.96 2.713 .100 0.880 4.388

Rear end

No (base) 1.00 1.00

Yes 2.38 0.41 32.502 <.001 0.300 0.556

Sideswipe

No (base) 1.00 1.00

Yes 0.76 0.15 128.895 <.001 0.106 0.205

Intersection crash

No (base) 1.00 1.00

Yes 0.44 0.80 9.681 .002 0.689 0.919

Speed zone (km/h)

60 (base) 1.00 1.00

50 1.18 1.02 0.061 .804 0.851 1.231

70 1.52 1.34 4.395 .036 1.019 1.772

80> 2.94 2.04 55.969 <.001 1.694 2.462

Speed contributor

No (base) 1.00 1.00

Yes 6.25 5.34 78.420 <.000 3.684 7.733

Alcohol over limit

No (base) 1.00 1.00

Yes 7.69 5.95 34.540 <.001 3.284 10.79

Time

6am-6pm (base) 1.00 1.00

6pm-6am 0.90 0.71 14.014 <.001 0.598 0.852

Weekday/weekend

Weekday (base) 1.00 1.00

Weekend 1.59 1.28 9.563 .002 1.095 1.502

Valid licence

Yes (base) 1.00 1.00

No 2.63 2.07 18.422 <.001 1.487 2.895

*Adjusted for all variables in table. **CI’s for adjusted OR


The variables showing statistically significant differences presented above in

Table 5.41 were selected for inclusion in a regression analysis stratified by PTW

type. In this analysis, the results for motorcycles closely reflected those of PTWs as a

whole, due to motorcycles comprising a large majority (90%) of the overall sample.

For scooters, results of the stratified analysis were unreliable due to low numbers

which resulted in collinearity problems, incomplete equations and excessive

confidence intervals. The results of the model for scooters are therefore not reported.

For mopeds, there were significant effects of rider age, crash configuration and

whether or not crashes occurred at intersections, as presented below in Table 5.42.

Specifically, the statistically significant differences include that older rider

(60>) and younger rider (<25) crashes had 2.1 to 2.4 times higher odds of the PTW

being designated Unit 1 compared to those of riders aged 25 to 59 years. Angle

crashes had 5 times lower odds of the PTW being designated Unit 1 compared to all

other crash configurations. Rear end crashes had 3.8 times lower odds of the PTW

being designated Unit 1 compared to all other crash configurations. Sideswipe

crashes had 6.2 times lower odds of the PTW being designated Unit 1 compared to

all other crash configurations. Intersection crashes had 1.8 times lower odds of the

PTW being designated Unit 1 compared to non-intersection crashes.

Table 5.42 Logistic regression analysis predicting odds of moped being Unit 1

Moped variable Crude

OR OR* Wald p-value

95% CI**

Lower Upper

Age group

25-59 (base) 1.00 1.00

17-24 2.25 2.12 8.915 .003 1.306 3.624

60 or over 2.38 2.40 4.559 .033 1.075 5.377

Angle

No (base) 1.00 1.00

Yes 0.59 0.20 8.253 .004 0.070 0.604

Rear end

No (base) 1.00 1.00

Yes 1.24 0.26 5.428 .020 0.083 0.807

Sideswipe

No (base) 1.00 1.00

Yes 0.79 0.16 8.475 .004 0.046 0.549

Intersection crash

No (base) 1.00 1.00

Yes 0.53 0.57 4.803 .028 0.343 0.942

*Adjusted for all variables in table. **CI’s for adjusted OR


5.5 Discussion

5.5.1 Patterns of usage as indicated by crash data

Previous research which examined PTW crash data and usage supports the

view that the location, temporal and other characteristics of PTW crashes largely

reflect usage patterns (Christie, 2008). While the potential for bias in crash data has

also been noted (Wigan, 2000), Study 2 provided an opportunity to compare moped,

scooter and motorcycle usage using crash data. Usage patterns drawn from these and

other data are addressed in greater detail in Chapter Seven (Queensland scooter and

moped rider survey) and Chapter Eight (Discussion). The findings of the current

study regarding PTW usage patterns are generally consistent with findings in other

research. In terms of usage patterns, the main differences and similarities between

mopeds, scooters and motorcycles are outlined below.

Previous research has shown that moped and scooter use is predominantly an

urban activity which takes place mostly in low and moderate speed zones (ACEM,

2008; Christie, 2008; Haworth & Nielson, 2008). The current study reflects a similar

pattern for moped and scooter use in terms of location, with a large majority of

reported crashes occurring in cities and regional urban centres. Compared to

motorcycle crashes, relatively few moped and scooter crashes occurred in rural and

remote areas. The location of moped and scooter crashes suggests that they are used

more for commuting and less for recreation compared with motorcycles, though

there may be considerable recreational moped use among tourists in some locations.

Moped use for commuting and for recreation by tourists in some areas of Queensland

has been previously identified in research (Haworth & Nielson, 2008). While the

data show a relatively low proportion of moped crashes in Brisbane (33%) compared

to scooters (52%) and motorcycles (43%), proportionally higher moped use in

locations frequented by tourists (particularly Gold Coast, Cairns and Townsville)

probably accounts for much of this difference. However, much of the moped use in

these areas is probably by local residents according to the licence characteristics of

crashed riders.

Crashed moped and scooter riders are at lower risk of fatality or

hospitalisation than motorcycle riders due to lower travel speeds according to other

research (ACEM, 2008; Christie, 2008). While travel speed and impact speed were


not available in the crash data, speed zone was used as a general indicator of speed.

In the current study, crashes in speed zones up to 70 km/h constitute around 94

percent of reported moped and scooter crashes, compared with 75 percent of

motorcycle crashes. For all crashes where PTW type was known, in speed zones of

70 km/h or more, 63.4 percent of crashes resulted in fatality (5.4%) or hospitalisation

(58.0%). By comparison, 49.3 percent of crashes in speed zones up to 60 km/h

resulted in fatality (2.3%) or hospitalisation (47.0%). It therefore appears that crash

severity is generally related to travel speed. This is a possible influence on the

differences in crash rates by severity discussed below, although there are no

supporting data available on the amount of riding by speed zone for different PTW

types.

Approximately 80 percent of moped and scooter crashes occurred on

weekdays, compared with 70 percent of motorcycle crashes. These results are

consistent with other research findings that mopeds and scooters are used more for

commuting and less for recreation in comparison to motorcycles (Christie, 2008;

Haworth & Nielson, 2008), although this is not to suggest that all recreational riding

occurs on weekends. A high proportion of all PTW crashes (77% to 83%) occurred

during daylight hours and there was no significant difference between PTW types.

The analysis highlighted a difference between mopeds and other PTW types

regarding weekend crashes, where moped crashes were more likely to occur at night,

particularly between 9pm and midnight (10 of these 11 crashes were Brisbane or

Gold Coast crashes). While the number of such crashes is low, one possible

explanation is that mopeds are used in Brisbane as pizza delivery vehicles, some of

which may have been involved in crashes, though this cannot be confirmed.

Recreational moped use at night time on weekends is another possible explanation

for this difference.

Research has previously shown relatively high proportions of moped and

scooter riders to be female in comparison with motorcycle riders (ACEM, 2008;

Christie, 2008; Haworth & Nielson, 2008). The current study reports similar

findings, with females representing around 37 percent and 7 percent of crash-

involved moped and motorcycle riders respectively. The gender distribution of

crashed scooter riders fell between that for moped and motorcycle crashes, with

females representing 22 percent of crashed riders. This suggests that mopeds are

clearly more popular than motorcycles among females, and that scooters are also


more popular, though only some female riders are prepared to obtain a motorcycle

licence in order to ride a scooter with an engine capacity above 50cc. Research has

also suggested that male PTW riders travel greater distances annually than female

riders (Harrison & Christie, 2003; SWOV, 2006b) and this should be taken into

account when considering crash rates per distance travelled.

In terms of the licence characteristics of PTW users, it is important to note

that either a car or motorcycle licence is valid for moped riding in Queensland, while

for other PTWs a motorcycle licence is required. Moped riders were less likely than

both scooter and motorcycle riders to hold an open licence, and more likely to hold a

provisional or restricted licence, and to be unlicensed or not licensed in Australia.

This reflects two features of the demographic characteristics of riders. One is the

relatively young age of moped riders who have not been active riders or drivers long

enough to obtain an open licence. The other is the higher involvement of tourists in

moped than motorcycle crashes, as evidenced by the nine percent of moped riders

who were licensed overseas (4.6%) or interstate (4.3%). A further ten percent of

moped riders were recorded as ‘unknown’ with regard to the place of licence issue

and it is likely that at least some of these were also licensed outside Queensland.

Motorcycle riders were most likely to be licensed in the State of Queensland (97%),

while scooter riders were least likely to be unlicensed (2%).

5.5.2 Crash rates and related characteristics

The observed increase in moped, scooter and motorcycle crashes reflects a

continuation of the trend identified in previous study of Queensland moped crashes

from 2001 to 2005 (Haworth & Nielson, 2008), though the actual rate of increase

appears to have slowed somewhat (the current dataset temporally overlaps the earlier

data but they do not match directly as they report for financial years and calendar

years respectively). According to vehicle make and model details, approximately one

quarter of moped crashes are recorded in the data as motorcycle crashes (motorcycle

body type). As well as identifying the erroneously recorded mopeds, the data

cleaning process was also able to separate scooters from motorcycles and to identify

scooters recorded as mopeds. Such a process had not previously been undertaken to

separate the three PTW types. However, there was not enough information to

distinguish between moped and scooter models in all cases and there were many


other cases with little or no model information which led to their exclusion from the

main analysis.

The average crash rate per 10,000 registered vehicles over the study period

was slightly higher for mopeds (133.4) than for motorcycles and scooters combined

(124.8). Crash rates per registered vehicle fell for both registration categories (LA

mopeds and LC motorcycles/scooters) across the study period, but the decline in

(LC) motorcycle crash rates (22%) was less pronounced than that for mopeds (40%).

In 2003-4, the moped crash rate was statistically significantly higher than that for

motorcycles. Although the reason for the observed decline in moped crash rates per

10,000 registrations over the study period is not known, some possibilities are

considered later in this section.

The analysis identified differences between mopeds and motorcycles in crash

rates per 10,000 registration years as a function of crash severity. The order of

severity categories as reported were: fatal (most severe); hospitalisation; medically

treated; minor injury; and property damage only (least severe). In particular, there

was a lower rate of fatalities per 10,000 registration years in moped crashes (1.2)

compared with motorcycle crashes (4.2), which is statistically significant despite

relatively low numbers. As evident in the ordered probit model of severity, this

outcome likely reflects differences in crash characteristics such as speed zone, time

of day, day of week, horizontal alignment and number of vehicles involved, rather

than any differences between PTW types per se. While the ordered probit model did

not show a statistically significant difference in severity by PTW type, this may have

been due to relatively low numbers of moped and particularly scooter crashes. The

greater severity in association with crashes in higher speed zones, on curves, on

weekends, at night time and in single vehicle crashes reported in the current study for

all PTWs has also been shown in other research (Zambon & Hasselberg, 2006;

Haque, Chin & Huang, 2009; Quddus, Noland & Chin, 2002).

Although the difference in multi-vehicle crash involvement was not

statistically significant between mopeds and motorcycles, there was a statistically

significant difference in crash severity between single and multi-vehicle crashes,

with greater severity associated with single vehicle crashes. Part of this difference

can likely be explained by lower reporting rates of low severity crashes. The higher

rate per 10,000 registration years of crashes requiring medical treatment for mopeds

(50) compared with motorcycle crashes (36) may reflect the greater likelihood of


hospitalisation in motorcycle crashes. It may also partly relate to a greater reluctance

among moped riders to wear protective clothing, as identified in other research (de

Rome, Stanford & Wood, 2004), though there was no information in the crash data

on use of protective apparel other than helmets.

The main limitation of the crash rate estimates per registered vehicle is that

they do not account for actual PTW use in terms of distance travelled. As noted

previously, reliable exposure data are required to objectively determine crash risk for

the different PTW types relative to usage. Studies on PTW use suggest that average

annual distances travelled by mopeds are approximately one third to one half the

average distances covered by motorcycles (Harrison & Christie, 2003). This is

unsurprising and reflects the design characteristics and intended purposes of the

different PTW types. In light of the similar average crash rates per registered vehicle

for mopeds and motorcycles, and accepting that mopeds travel considerably less

distance on average, it is possible that the crash rate relative to actual exposure is

much higher for mopeds than for motorcycles. Crash rate calculations for LA

mopeds and LC motorcycles/scooters performed using Study 2 crash data and the

distance travelled estimates of Harrison and Christie (2006), suggested that this is

indeed the case: mopeds and motorcycles/scooters crashed at rates of 6.3 and 1.7 per

million VKT respectively. As LC scooters were not separated from motorcycles in

the registration data, their crash rates per distance travelled could not be calculated.

A more detailed analysis of the registration data is required to remedy this problem.

As stated above, the reason for the observed decline in moped crash rates per

10,000 registrations over the study period is not known. There have been changes to

motorcycle licensing in Queensland and changes in rider characteristics (particularly

age) which may have influenced declining motorcycle crash rates. As legislation

specifically governing moped use has not changed during or immediately prior to the

study period, this seems unlikely to have influenced a decline in moped crash rates.

In the absence of supporting evidence, the following speculative explanations for

declining crash rates may be considered. While there were certainly many more

mopeds in circulation in 2008 than in 2003, in light of other research (Harrison &

Christie, 2003; 2006) the crash data suggest that they do not travel as far on average

and that their use is now more discretionary than was previously the case. It is also

possible that people who have recently taken up moped riding purely for commuting

and mobility reasons (due to increasing traffic congestion, cost and parking


pressures) are on average more risk-averse than previous moped-riding populations.

In terms of all road users, modification and increased enforcement of road rules such

as speeding and drink driving laws may also have contributed to a reduction in PTW

crash rates.

The characteristics of crashes and the contributing circumstances differed for

moped, scooter and motorcycle crashes in many ways. In the current study, PTW

riders were considered most at fault in 58 percent of all crashes, with a statistically

significant difference between scooter (45%), moped (54%) and motorcycle riders

(59%). In multi-vehicle crashes, riders were considered most at fault 37 percent of

cases and although there was no statistically significant difference, scooter riders

again appeared least likely to be at fault (31%) compared with moped (35%) and

motorcycle riders (37%). These data suggest safer riding behaviour among scooter

riders than either moped or motorcycle riders. While moped and motorcycle riders

were relatively similar in the proportions of crashes where they were deemed most at

fault, the nature of some of the circumstances attributed differed among all three

PTW types. These are discussed below in section 5.5.3.

Compared to at fault (Unit 1) motorcycles in multi-vehicle crashes, at fault

mopeds were significantly over-involved in ‘manoeuvring’ crashes and ‘angle’

crashes, and under-involved in ‘rear-end’ crashes. The findings on ‘manoeuvring’

and ‘angle’ crashes are likely explained largely by the urban traffic environment in

which these crash characteristics are common and in which mopeds mostly operate.

They may also reflect poorer vehicle control skills of moped riders compared to

motorcycle riders, particularly at lower speeds, though there is no direct evidence to

support this in the current study. The lower involvement of at fault mopeds in ‘rear-

end’ crashes compared to motorcycles may be related to the limited performance of

mopeds. Limited moped performance may have at least two effects in this regard,

one being that mopeds travel slower than motorcycles behind other vehicles and

therefore have more time for braking when required. The other effect is that other

vehicles may travel more closely behind mopeds, leaving drivers less time for

braking when necessary.


5.5.3 Main contributors to crash and injury risk

Six main contributors to crash and injury risk for motorcyclists were

identified in previous research (Greig, et al., 2007), as: inexperience or lack of recent


difficulties; road surface and environmental hazards; and vulnerability to injury. As

noted previously, these risk factors do not necessarily apply equally to moped,

scooter and motorcycle use but they provide a useful framework for discussing the

results of the current study in terms of comparing the safety of mopeds, scooters and

motorcycles. The following six sections of this chapter are structured accordingly.

There is slight modification of the original terminology in the current section titles to

ensure inclusiveness of all the relevant issues identified in the crash data.

Specifically, driver failure to see motorcyclists is modified to other road users, and

instability and braking difficulties is modified to PTW control and riding skills.


Numerous studies have examined the role of inexperience and lack of recent

experience in PTW crashes, finding elevated crash risks in association with these

factors (ACEM, 2008; Haworth, Smith, Brumen & Pronk, 1997; Mullin, Jackson,

Langley & Norton, 2000). Inexperience with a particular vehicle or PTW type has

also been explored and found to be a crash risk factor in some studies (Mullin, et al.,

2000; Rutter & Quine, 1996). Rider age has often been interpreted as a key indicator

of rider experience, where young riders are generally less experienced, and

overrepresented in crashes. Increasing crashes among older riders in recent years

have also led to greater interest in lack of experience across a broader age range.

Rutter and Quine (1996) note that while PTW crash risk may relate to both age and

experience, only increasing age had a protective effect according to their study.

There were no data in the current study to directly determine the amount of

riding or driving experience accumulated by crash involved riders. Levels of

experience could therefore only be inferred from the data on rider age and licence

status. For motorcycle and scooter riders, some older riders may have been only

recently licensed or may have only recently returned from an extended break from

riding. Conversely, some younger riders may have had considerable off-road riding


experience (though this has not been found to have a protective effect in terms of

crash risk). As moped riders require only a car licence, there is no indication in the

crash data as to their riding experience. It can be reasonably assumed that riders

under 21 years of age in Queensland have only a few years of riding experience at

best, but it cannot be assumed that this not also the case for many older riders.

In the current study, moped riders tended to be younger while scooter riders

tended to be older compared with motorcycle riders. Moped riders are almost

universally found to be younger than motorcycle riders in other research. As noted

in previous research in Queensland (Haworth, et al., 2008), the European evidence of

higher risk among young moped riders is of limited relevance to Australia due to the

considerably younger age at which many European countries permit moped riding.

In the current study, while moped riders were the youngest riders on average, almost

70 percent of crashed moped riders were aged over 25 years. The higher proportion

of older scooter riders compared to moped riders may relate in part to licensing

requirements. It is possible that some younger riders who obtain a motorcycle

licence may be more likely to obtain a motorcycle than a scooter (or moped) due to

sensation-seeking motivations more prevalent in younger riders. Conversely, older

riders who hold a motorcycle licence may opt for a scooter rather than a motorcycle

due to the relative ease of use of scooters.

Moped riders in the current study were not only younger but also more likely

than scooter or motorcycle riders to hold a provisional or learner licence. This may

have been a car or motorcycle licence, but was likely only the former according to

previous research (Haworth, et al., 2008). By contrast, scooter riders were not only

older than moped or motorcycle riders, but were also most likely to hold an open

motorcycle licence. It is therefore tentatively concluded that moped riders have less

experience while scooter riders have at least as much experience as motorcycle riders

on average. As Study 2 was not able to examine the issue of riding experience in

greater depth due to the nature of the data, the topic is explored further in Study 3.

5.5.3.2 Risk taking

The literature clearly shows that sensation-seeking behaviours and associated

risk taking are most prevalent among younger riders and male riders (Noordzij,

Forke, Brendicke & Chinn, 2001; Rutter & Quine, 1996). There is also some


evidence of greater risk taking among recreational riders than commuters (Harrison

& Christie, 2003). However, the age and gender distributions as well as the patterns

of use observed in the current study show that the situation is more complex than

suggested by these generalisations. In particular, while moped riders were younger

than scooter or motorcycle riders, they were also more likely to be female and,

arguably, less likely to engage in recreational riding. While their age suggests that

moped riders may be the most predisposed to risk taking, the gender characteristics

and likely purpose of riding suggest otherwise. Scooter riders on the other hand were

the oldest on average, with a moderate proportion of female riders, and (probably)

relatively little recreational riding compared to motorcyclists. It might be expected

therefore that risk taking is least prevalent among scooter riders.

Evidence of deliberate risk taking by riders in the current study was found in

the attribution as contributing circumstances of excessive speed or exceeding

prescribed blood-alcohol concentrations (BAC). Unlicensed riding is also included

as a form of risk taking as previous research has found unlicensed riders to be

overrepresented in crashes. Although the practice may not be intrinsically risky,

unlicensed riding has been found to be associated with risky behaviours such as

speeding, impaired riding and non-use of helmets (Haworth, et al., 2009). Dangerous

driving was attributed to riders in a small number of crashes, but is not included here

due to lack of detail in the data regarding the particular circumstances and whether or

not the behaviour was deliberate.

Excessive speed was not attributed to a scooter rider in any crash, but was

attributed to riders in two percent and seven percent of all moped and motorcycle

crashes respectively. In single vehicle crashes, excessive speed was attributed to

riders in four percent of moped crashes and 12 percent of motorcycle crashes. The

proportionally greater attribution of speed to motorcycle than moped riders is

unsurprising given the design characteristics and performance restrictions applied to

mopeds. Previous research in Queensland found very similar results, also noting that

the lower involvement of mopeds in speeding may reflect their limited performance

rather than an intention to not speed (Haworth, et al., 2009) (scooters were not

separated from motorcycles in that study). With regard to speeding, the current

analysis suggests safer behaviour among scooter riders compared to both moped and

motorcycle riders.


Another finding that may relate to travel speed and risk taking is that for rear-

end crashes. As noted, mopeds were designated Unit 1 (most at fault) in 38 percent

of rear-end crashes, compared with 50 percent and 60 percent for scooters and

motorcycles respectively. According to the descriptive statistics, as the road user at

fault in rear-end crashes is usually the one who runs into the rear of another, it seems

that mopeds are more likely to be run into, while motorcycles are more likely to run

into another vehicle. The results of the logistic regression analysis of fault partly

confirm this, with moped riders less likely to be at fault than motorcycle riders in

rear-end crashes. However, with other factors accounted for in the model,

motorcycle riders also had 2.4 times lower odds of being most at fault in rear-end

crashes. That this result conflicts with the descriptive analysis is due to the influence

of the other variables included in the logistic regression model. In rear-end scooter

crashes fault was evenly split between riders and other road users, although low

numbers precluded a valid logistic regression analysis for scooters alone. The results

suggest that motorcycle riders and to a lesser extent scooter riders are more likely

than moped riders to follow too closely behind other vehicles, which could arguably

be termed a risky riding behaviour. This may relate again to the relatively limited

performance characteristics of mopeds which restrict their ability to keep up with

surrounding traffic.

Drink driving was attributed to riders in 4.3 percent and 4.4 percent of moped

and motorcycle crashes respectively and approximately three quarters of these were

single vehicle crashes for both PTW types. Drink driving was reported for relatively

few scooter riders (2.1%), further suggesting their safer riding behaviour, but this

may be unreliable due to low numbers. Similar findings for alcohol-related moped

and motorcycle crashes have been found in Queensland (Haworth, et al., 2009) and

Europe (ACEM, 2008), although once again scooters were included in the

motorcycle data in these studies. Logistic regression analysis of multi-vehicle

crashes showed that those involving an illegal BAC (.05>) for riders or other road

users were almost seven times more likely to have the PTW designated Unit 1 (most

at fault).

Where licence status was known (N = 7,239), unlicensed riding was more

prevalent in moped (7%) and motorcycle (5%) crashes than in scooter crashes (2%).

The proportions of unlicensed moped and motorcycle riders were similar to those

found in previous research in Queensland (Haworth, et al., 2009). Licence status was


more likely to be recorded as unknown in moped crashes and it possible that the

difference between PTW types in unlicensed riders was either more or less than that

revealed by the current analysis. As moped riders only require a valid car licence

while scooter and motorcycle riders require a motorcycle license, the analysis is of

limited use beyond the issue of licence validity. However, it can be tentatively

concluded that the lower rate of unlicensed riding among scooter riders supports the

previous indications of their lower propensity for risk taking.

5.5.3.3 Other Road Users

Crashes with other road users represented two thirds of all crashes involving

known PTW types, though they were most common in scooter crashes (79%),

followed by moped (71%) and motorcycle (66%) crashes. For all PTW types, almost

90 percent of crashes with other vehicles involved a passenger car or small

commercial vehicle. Approximately half of all crashes occurred at intersections and

these crashes were significantly more likely for mopeds and scooters (53%) than

motorcycles (46%). Logistic regression analysis of fault showed 1.25 and 1.75 times

lower odds of Unit 1 attribution to all PTWs and to mopeds respectively in

intersection crashes compared to those on other roadway configurations. These

findings are generally consistent with other research from Australia and also Europe,

however there are some differences.

The previous research on moped crashes in Queensland showed results most

comparable to the current study on multi-vehicle crashes (around 67% for both

mopeds and motorcycles) and intersection crashes (51% for mopeds, 48% for

motorcycles) (Haworth & Nielson, 2008). The 2008 report on scooter (including

moped) crashes in Victoria is the most recent relevant study from an Australian

jurisdiction, though mopeds represented only a small minority of crash-involved

scooters (which is likely a result of the Victorian requirement of a motorcycle licence

for moped riding) (Christie, 2008). The report showed around 60 percent of scooter

crashes involved another vehicle, lower than the proportion of multi-unit crashes in

the current study, while a higher proportion of crashes (62%) occurred at

intersections. The European MAIDS report showed the same pattern of more multi-

vehicle and more intersection crashes for L1 mopeds than L3 motorcycles/scooters,

but the overall percentages were somewhat higher, perhaps reflecting the higher


degree of urbanisation in Europe (ACEM, 2008).

In the current study, other road users were considered most at fault in almost

two thirds (63%) of all multi-unit crashes. In these crashes, right-of-way (ROW)

violations by other vehicles were the most commonly reported contributing

circumstance, as consistently found in other research (Hurt, Ouellet & Thom, 1981;

ACEM, 2008; Johnston, Brooks & Savage, 2008; Wells et al., 2004; Comelli,

Morandi, Magazzu, Bottazzi & Marinoni, 2008). In the current study, violations by

other road users included failure to give way, illegal manoeuvres and disobeying

traffic signals and other road rules, and were attributed in 77 percent of cases where

other road users were most at fault. In multi-vehicle crashes, inattention, distraction

or negligence were attributed to other road users in 25 percent of moped crashes and

17 percent of scooter and motorcycle crashes. Inexperience was attributed to other

road users in 12 percent of moped crashes, four percent of scooter crashes and 13

percent of motorcycle crashes, although this may have been attributed largely on the

basis of age alone. This lower attribution of inexperience in scooter crashes may be

due to unreliably low numbers. All of these contributing circumstances were also

those most commonly attributed to a PTW in multi-vehicle crashes, although they

were attributed in relatively small numbers as riders were not usually deemed to be at

fault.

The logistic regression model used to predict the odds of the PTW rider being

designated Unit 1 in multi-vehicle crashes showed that, compared with the respective

referent (base) groups, riders were more likely to be at fault in association with

certain crash and rider characteristics. These include hit object and higher speed

zone crashes, if excessive speed or alcohol was attributed, and if younger (<25) or

older (60>) riders were involved. Riders were less likely to be at fault in

intersection, night time, and rear-end crashes. Similar results were reported from a

multivariate probit analysis of fault in motorcycle crashes by Schneider, Savolainen

et al. (2012) and a logistic regression analysis by Kim and Boski (2001), with the

exception of rear-end crashes for which they each found higher odds of an at-fault

rider. Kim and Boski (2001) also reported that, generally, at-fault motorcycle riders

tended to be associated with risky riding behaviours, while at-fault drivers are

associated with conspicuity issues and lack of attention.

It could arguably be assumed in light of the literature that a large proportion

of crashes with other vehicles at fault would fall into the category of ‘looked but


failed to see’ (LBFTS) or ‘sorry mate I didn’t see you’ (SMIDSY) crashes (Brown,

2005; Broughton & Walker, 2009). Without the explanations from other road users

who were deemed to be at fault, this cannot be confirmed for the current study.

Nonetheless, it is probable that driver failure to see motorcyclists (or all PTW riders)

was a major contributor to crash risk in the current study as in other research (Greig,

et al., 2007). There were no data in the current study that could be used to compare

the conspicuity of mopeds, scooters and motorcycles or the riders involved in

crashes. However, more than three quarters of all reported crashes involving a

known PTW type occurred during daylight hours, suggesting that lighting conditions

were unlikely to be a significant factor in relation to conspicuity.

The findings on ROW violations, inattention and distraction among other

road users suggest that all riders need to be highly aware and able to anticipate and

respond to such hazards to avoid multi-vehicle collisions. International research

suggests that moped riders may be more at risk in such situations than scooter and

motorcycle riders. For example, the MAIDS report notes that moped crashes with

other vehicles were more likely than motorcycle crashes to involve failures of

perception, decision-making and execution by riders (ACEM, 2008). Considering

these findings, it is tempting to suggest that interventions aimed at improving hazard

perception and response of riders and other road users alike may help to reduce

crashes. Other relevant interventions discussed previously in the literature review

include increasing the conspicuity of PTWs and riders, and awareness campaigns

aimed at other road users.

5.5.3.4 PTW control and riding skills

As suggested in the literature, PTW control requires greater skills than car

driving due to the inherent instability and relatively poor braking performance of

single track vehicles. These issues have been referred to as instability and braking

difficulties in other research (Greig, et al., 2007). In some cases these factors may be

exacerbated for mopeds and scooters due to smaller wheel diameters, shorter

wheelbases and less advanced braking systems compared to motorcycles. Another

difference between the PTW types is the automatic transmission characteristic of

most mopeds and scooters, while most motorcycles have manual transmission

requiring clutch operation and gear changing. Scooter and motorcycle riders in


Queensland must demonstrate a basic level of riding competence before gaining an

appropriate licence. Moped riders in Queensland are not currently subject to this

requirement. The current study provided only limited data for assessing vehicle

control skills of moped, scooter and motorcycle riders, leading to the following

interpretations and tentative conclusions.

Considering crash group description and crash configuration, motorcycle

crashes were more likely than moped or scooter crashes to be ‘fall from vehicle’

crashes and ‘off path on curve’ crashes. This possibly reflects their greater

involvement in single vehicle crashes which may in turn relate to more use of

motorcycles for recreation outside of urban areas. The results contrast with previous

research in Queensland that found mopeds and motorcycles to be similarly involved

in ‘fall from vehicle’ crashes (Haworth, et al., 2008). While indicating that a rider

fell from a PTW in these crashes, it is unclear what actually caused then to fall,

though it would seem that instability and braking difficulties would broadly explain

many such cases according to other research (ACEM, 2008).

Perhaps the clearest indication of vehicle control skills in the current study is

the contribution of road conditions and wet roads to PTW crashes. These factors

were more likely to contribute to moped crashes than scooter crashes, suggesting

inferior riding skills of moped riders, particularly given the similar design and

performance characteristics of many mopeds and scooters (although larger scooters

generally have superior brakes and suspension to mopeds, there are relatively few of

these in use according to Study 1). Road conditions encountered by motorcycle

riders are likely to differ somewhat from those encountered by moped and scooter

riders due to the greater use of motorcycles outside urban areas. Road conditions and

environmental hazards are addressed further in the following section.

The greater involvement of mopeds and scooters in ‘manoeuvring’ crash

types, which mostly involved vehicles leaving driveways, is consistent with their

greater urban use compared to motorcycles. It may be that motorcycles are less

involved in this crash type for other reasons such as superior evasive skills and/or

braking performance, but there is no evidence for this. Given the similar

involvement of mopeds and scooters in such crashes, and the differences observed

between them on other variables, it would not appear that moped riders are less able

to anticipate and avoid such hazards.


5.5.3.5 Road surface and environmental hazards

Road conditions were attributed to a PTW more often in moped crashes

(10%) and motorcycle crashes (14%) than in scooter crashes (4%). The vast

majority of these were single vehicle crashes. The Australian and international

literature show that motorcycles are used proportionally more outside of urban areas

than mopeds or scooters (ACEM, 2008; Haworth & Nielson, 2008). Motorcycle

riders may therefore encounter poor or unpredictable road conditions more often than

moped or scooter riders. This may explain much of the observed difference between

motorcycle and scooter crashes regarding road conditions, but does not explain the

difference between mopeds and scooters which are each used predominantly in urban

environments. Further, wet road surfaces in particular contributed to a greater

proportion of moped crashes (13%) than scooter (6%) or motorcycle (8%) crashes.

PTW crashes on wet roads are likely to be associated with rider error according to

previous research (de Rome et al., 2010), but exposure to wet roads might be higher

for regular commuters than recreational riders who might avoid riding on wet days.

It has been suggested in the literature that mopeds and scooters may be more

susceptible than motorcycles to hazards such as potholes and rough surfaces, due to

smaller wheel diameters and limited suspension capabilities. Given the similar

design characteristics of many moped and scooter models, it might be expected that

they are similarly susceptible to such hazards. However, research has not clearly

articulated the actual contribution of these factors to PTW crashes. The findings of

the current study collectively suggest that poor road conditions present a greater

challenge to moped riders than scooter riders. Once more this suggests safer riding

by scooter riders, which in this instance may relate to a combination of more

experience and superior riding skills.


The greater vulnerability of PTW riders compared to car and other vehicle

occupants is well documented in the literature. The current study provided little

indication of the relative vulnerability of moped, scooter and motorcycle riders,

though some issues warrant discussion. Where helmet use was known, all scooter

riders and nearly all moped and motorcycle riders (97% and 99%) complied with the


existing mandatory helmet use laws. While descriptive analyses suggested

significant differences in crash severity by PTW type, the ordered probit model

which accounted for the influence of other variables showed no statistically

significant difference. This remains an issue for further exploration, but the ordered

probit model suggests that differences in severity by PTW type are related to the

various crash characteristics rather than the PTW type itself. Motorcycle riders may

be somewhat more vulnerable than moped and scooter riders and this is likely related

to travel speed as well as greater involvement in single vehicle crashes which were

found to be more severe. More detailed injury data is required to more accurately

assess vulnerability to injury. As noted in previous research, moped and scooter

riders may be less likely to wear protective clothing than motorcycle riders (de

Rome, et al., 2004). Consequently, hospitalisation of moped and scooter riders may

result more from lacerations and less from fractures compared to motorcycle riders

(Haworth, et al., 2008). Issues regarding vulnerability and protective clothing use

among moped and scooter riders are specifically explored in Study 3.


In seeking answers to the four research questions, Study 2 compared moped,

scooter and motorcycle crashes to identify similarities and differences which are

relevant to PTW safety and also usage. As well as analysing crash data from a five

year period the study also examined registration data. The data revealed differences

according to PTW type in crash rates, crash characteristics and circumstances, rider

characteristics and some usage patterns. Some similarities were also observed

regarding crash characteristics, crash circumstances and usage patterns.

Research question 1: Why has moped and scooter usage increased? The

crash and registration data indicate that PTW use has increased substantially as

whole in the study area, and that moped use has increased significantly relative to

motorcycle use over the study period. Scooter use also appears to have increased

according to the crash data, although the number of reported scooter crashes remains

relatively small. As such, Study 2 has demonstrated an ongoing increase in moped

and scooter usage, beyond that which has previously been observed for mopeds in

the study area (Haworth & Nielson, 2008). While sales and registration data indicate


growth in PTW ownership from 2001 to 2009, the increased crash numbers provide

evidence for an increase in actual usage. The increase in PTW usage, and

particularly moped usage, observed in this study and in earlier research has been

substantial and sustained over the last ten years. The confirmation of this trend

justifies an exploration in Study 3 of the underlying motivations for increased moped

and scooter usage.


motorcycles differ? The crash data suggest that compared with motorcycle use, there

is more moped and scooter use by females, more use in urban areas, on lower speed

roads, and on weekdays. These findings are consistent with other research

suggesting more use for commuting and less recreational use of mopeds and scooters

compared with motorcycles. Moped riders were younger on average than

motorcycle riders, while scooter riders tended to be older. It may be that some of the

usage patterns observed reflect risk as well as usage, an issue which is explored

further in the third and final study of this program of research.


motorcycles differ? The PTW types differed regarding crash rates, crash

characteristics and contributing circumstances. Compared to motorcycles and

scooters, mopeds had a slightly higher crash rate per registered vehicle and a

substantially higher crash rate per distance travelled, although questions remain

regarding the reliability of these denominator data. Crash rates per registered vehicle

declined for mopeds and also for scooters and motorcycles over the study period,

though this decline was greater for mopeds. Crash rates were not calculated

separately for scooters as they were not separated from motorcycles in the

registration data and this requires further exploration. However, sales data supported

by the findings of Study 1 suggest that scooters are likely to be underrepresented in

the crash data relative to mopeds and motorcycles. While moped crash rates were

higher, the severity of moped and also scooter crashes was lower overall than that of

motorcycles. Differences in crash severity were associated more with crash and rider

characteristics than with PTW type.

The majority of all crashes occurred in urban areas of southeast Queensland,

though a relatively high proportion of moped crashes occurred in regional urban


centres of coastal Queensland, while there were relatively more motorcycle crashes

in rural and remote areas. Moped and scooter crashes were more likely to occur at

intersections and in lower speed zones than motorcycle crashes, while moped crashes

were also more likely to occur on wet roads. Considering all of the study findings

collectively, scooter use therefore appeared to be safer than either moped or

motorcycle use.


motorcycles differ? Study 2 provided some possible explanations for the observed

differences in the safety of mopeds, scooters and motorcycles, though the analysis

also raised some further questions. As noted by Noordzij et al (2001, p.4), ‘a

statistical relationship may be found between moped/motorcycle characteristics and

accident rate. But it is the rider motivation or riding style, rather than the vehicle

characteristics which can explain this relation’. Rider characteristics, behaviours and

travel patterns are likely to explain much of the observed difference, as demonstrated

in the ordered probit and binary logistic regression model results.

The literature suggests that younger riders have relatively higher crash risk

and moped riders were found to be younger than either motorcycle or scooter riders

in the current study. The younger age and higher crash rates of moped riders

suggests that on average they are relatively inexperienced riders and may have

difficulty perceiving and responding to hazards. Compared to scooters, mopeds and

motorcycles were both more likely to crash on wet roads and to have road conditions

attributed in single vehicle crashes, tentatively supporting this contention

(motorcycle riders were also younger than scooter riders). It seems likely given the

Queensland regulations that most crashed moped riders did not hold a motorcycle

licence and that few had undertaken any rider training or education, though for the

current study this is speculative and is therefore explored specifically in Study 3. It

must also be remembered that evaluations of rider training programs have been

generally inconclusive with regard to safety benefits.

Analysis of contributing circumstances in crashes found that of the three

PTW types, scooter riders were least likely to engage in speeding, driving while

impaired, or unlicensed riding. The engagement of moped and motorcycle riders in

impaired riding and unlicensed riding was comparable. Motorcycle crashes were

more likely to be speed-related than moped as well as scooter crashes. Scooter riders


were also least likely to be considered at fault in crashes generally and were least

likely to be involved in single vehicle crashes, which were found to be of generally

higher severity. Single vehicle crashes were more likely to involve motorcycles than

scooters, as well as mopeds to a lesser extent. This factor combined with

proportionally greater motorcycle use on weekends likely reflects more recreational

motorcycle use, a riding purpose which has been associated with higher crash risk in

the literature.

There was insufficient detail in the crash data to determine the role of vehicle

factors in crashes (independent of riders), such as those related to PTW design and

performance. However, the performance characteristics of mopeds and scooters

arguably provide fewer opportunities than motorcycles for some sensation-seeking

behaviours. Some literature also argues that limited moped performance may result

in higher crash risk where they are unable to conform to the speed of surrounding

traffic. PTW performance characteristics remain a topic for further exploration.

5.6 Limitations of Study 2

Study 2 involved an analysis of crash and registration data in the Australian

State of Queensland, which has seen a notable increase in moped and scooter use

over the last decade. As such, the environmental, socio-cultural, economic and

regulatory conditions of Queensland must be taken into account when considering

the relevance of the study to other locations. In particular, the regulation which

permits moped riding for car licence holders is likely a major factor influencing

moped use.

The study has analysed only those crashes which were reported by police and

subsequently entered into the crash data files provided by the Queensland

Department of Transport and Main Roads. It can be safely assumed that reported

crashes are generally more severe than unreported crashes, though it is expected that

many unreported crashes also result in injury requiring treatment (Haworth, 2003).

The proportion of reported crashes relative to unreported crashes is unknown. It

cannot be concluded from the current analysis that the characteristics and

circumstances surrounding reported crashes are entirely the same as those of

unreported crashes.


Of the total reported PTW crashes over the five year study period, the PTW

type involved was reliably ascertained in 85 percent of cases (N = 8,608). Of these,

8.7 percent of crashes involved a moped or scooter (N = 7,347). In the 15 percent of

cases where PTW type could not be reliably identified, 4.6 percent (58) involved a

PTW type that was either a moped or scooter but could not be further identified (N =

1,261). This was due to insufficient information in the original data regarding

vehicle make, model and/or body type. These cases were excluded from the main

analysis. As cases involving unknown PTW types occurred more in rural and remote

areas than in urban areas, it seems possible that they may be biased toward

motorcycles.

The exclusion of unknown PTW types from the main analysis results in

underestimates of crash rates per registered vehicle. On the other hand, registration

data do not provide a measure of actual usage and some registered PTWs may

receive little or no use on public roads. Further, LC/LE category scooters are not

separated from LC/LE motorcycles in the registration data and it was therefore only

possible to compare LA moped crash rates per registered vehicle with those of all

LC/LE category PTWs combined.

Crash rates per kilometre travelled were calculated for each of the PTW types

using exposure data obtained from previous research (Harrison & Christie, 2006).

While the data for distance travelled were obtained in the middle of the current study

period, these crash rate estimates are of limited reliability due to low numbers of

moped and scooter riders among survey respondents. There is also a potential self-

selection bias among survey participants, as well as potential for unreliability in self-

report data.

The attribution of contributing circumstances relies on the availability of

evidence and on the judgement of investigating officers after the event. As noted in

previous studies, objective information is not always available and the different

perceptions of reporting officers regarding mopeds and motorcycles may influence

their judgements (Haworth, et al., 2009). In some cases there were no contributory

factors reported and in other cases the reported factors may not be entirely accurate.

It is not known why no contributing circumstance was reported for nearly 58 percent

of multi-vehicle crashes but virtually all single vehicle crashes. Where multiple

contributing circumstances are recorded for a single crash the data are not weighted

to indicate which particular factors were more prevalent contributors than others.


Attribution of ‘inexperience’ and ‘age – lack of perception’ as contributing

circumstances in crashes must be viewed with caution. These contributing

circumstances are often attributed on the basis of age, to younger and older road

users respectively, rather than on an objective assessment of inexperience or lack of

perception as a causal factor in crashes.

Analysis of alcohol and drug involvement in crashes relied on a positive test

result for exceeding prescribed blood alcohol concentrations (BAC). This almost

certainly underestimates the extent of alcohol or drug involvement in reported

crashes. The detected presence of any alcohol or drugs was also reported as a

separate contributing circumstance, and was always reported where there was a

positive BAC test, though may also have been reported when no test was conducted.

This variable was considered less reliable and therefore likely to have resulted in an

overestimate of alcohol or drug involvement if included.

The data provided on crash severity was coarse and contained no information

on injury types, duration of hospital admissions, medical treatments or outcomes

(other than fatalities). Therefore, while the overall crash severity could be compared

for the PTW types, the absence of any medical data meant that the nature, types and

severity of injury could not be compared. Linking of the crash data with hospital

data would be required to overcome this limitation.

With regard to crash location and the mapping of crashes, the data only

provided information to the level of Statistical Local Area. It was not possible to

identify the actual roadways or specific locations of crashes, which would have

enabled a more detailed analysis of crash concentrations by PTW type. This remains

another potential topic for further research.

5.7 Chapter Five summary

This chapter has described Study 2, which involved analysis and discussion

of Queensland PTW crash and registration data to compare the safety and usage of

mopeds, scooters and motorcycles. Specific areas which were examined included

increased PTW usage, crash rates, crash severity, crash characteristics, rider

characteristics, contributory factors and other road users involved. Study 2 provided

information relevant to licensing and training requirements, speed restrictions and

other regulations which apply to moped riding. Study 2 was also instrumental in


helping to identify areas on which may be amenable to potential interventions.

Riding skills, hazard perception and response, and general road user awareness of

PTWs were identified as key issues.

The study has provided essential information relating to the four research

questions. For Research question 1 regarding reasons for increased PTW usage,

Study 2 provided evidence of increased usage beyond that which had been

previously reported. For Research question 2 concerning differences in usage,

patterns of usage as well as rider characteristics for the PTW types were reflected in

the crash data. For Research question 3 regarding differences in safety, the study

examined crash rates, crash severity, crash characteristics and contributing factors to

assess the relative safety of mopeds, scooters and motorcycles. For Research

question 4 concerning reasons for differences in the safety of mopeds, scooters and

motorcycles, the study explored various factors relating to risk, including the six

main contributors to crash and injury risk identified in previous research (Greig, et

al., 2007). Factors associated with crash severity and fault attribution were examined

using ordered probit and binary logistic regression models respectively.

Addressing the research questions as such, Study 2 provided essential

material for achieving the research aims, which were to develop a better

understanding of moped and scooter usage trends and patterns, and to investigate

factors underlying differences in moped, scooter and motorcycle safety. The study

therefore helps to address the recognised knowledge gap regarding the relative safety

of moped, scooter and motorcycle use.

The next two chapters present Study 3, designed to gather information from

riders regarding moped and scooter safety and usage. Chapter Six presents Study 3a,

in which a series of focus group discussions were held with moped and scooter riders

to identify key issues regarding moped and scooter safety from the perspective of

riders. This is followed by a presentation of Study 3b in Chapter Seven, which

involved development and administration of a survey of moped and scooter riders.


CHAPTER 6: STUDY 3A – FOCUS GROUPS WITH MOPED

AND SCOOTER RIDERS

6.1 Introduction

The published research reviewed earlier in this thesis has identified a need for

better understanding of moped and scooter use and safety, particularly in places

where the use of these PTWs has increased substantially (Haworth, Nielson et al.,

2008). Recent international literature also suggests a specific need for more

qualitative research on this topic, arguing that statistical analyses may fail to address

social and cultural factors influencing PTW use and safety (Nja & Nesvag, 2007).

The research reported in this chapter attempts to meet these needs through Study 3a,

a qualitative component of the current research program into moped and scooter

safety.

Focus groups were conducted to identify issues pertinent to moped and

scooter safety, and the factors which motivate their use, from the perspective of

riders. Specific groups of riders were engaged separately to allow some comparisons

between them, and to ensure comprehensive coverage of the relevant issues. The

different groups of riders included inner city commuters, members of online scooter

forums, students, and industry stakeholders who were also scooter or moped riders.

A distinction between riders of mopeds and those of larger scooters was considered

important due to the differences in crash characteristics observed in Study 2, as well

as differences in licensing requirements for the two PTW types. The focus group

design was tailored accordingly. While there is some overlap in the characteristics

and motivations of riders across these groups, four discrete focus groups were

planned with an ideal representation of six to eight riders in each.

Open-ended questions to guide focus group discussions were developed by

reference to a range of issues linking back to the six main contributors to crash and

injury risk drawn from previous PTW safety research (Greig, Haworth et al., 2007)

(see section 6.2.4 below). In addition to these issues, other topics specifically

explored during the focus group sessions included motivations for moped and scooter

use, travel patterns, and attitudes and beliefs regarding licensing and rider training.

Through covering a wide range of issues regarding moped and scooter use and


safety, Study 3a addressed all four research questions.

In relation to Research questions 1 and 2, Study 3a explored motivations for

moped and scooter use, as well as similarities and differences in usage of the PTW

types. Study 3a also addressed Research questions 3 and 4 by examining a wide

range of safety-related issues to identify similarities and differences among moped

and scooter riders that might impact their overall and relative safety.

Finally, information gathered through the focus groups was used to inform

the development of a survey instrument for use in Study 3b. Although Study 3a was

a qualitative study designed to guide development of Study 3b, the qualitative

findings of Study 3a are also of independent interest. This chapter begins with a

description of the methods and then proceeds to outline and discuss the findings,

implications and limitations of the study.

6.2 Methods

6.2.1 Setting

Previous research has shown that approximately 30 percent of moped crashes

in Queensland occur in the city of Brisbane, reflecting the amount of moped and

scooter usage there relative to other Queensland areas (Haworth, et al., 2008). Study

2 found similar results regarding the location of moped crashes, 33 percent of which

occurred in Brisbane. An even greater concentration of scooter crashes (52%)

occurred in Brisbane (Study 2). It was therefore decided that it would be valid as

well as practical to sample participants from Brisbane as a major hub of moped and

scooter use, though some participants also regularly rode mopeds or scooters outside

of Brisbane. The four focus groups were held in a library meeting room at

Queensland University of Technology’s Kelvin Grove campus, situated in inner

urban Brisbane. One focus group was held on each evening of March 11, 17, 24 and

25, 2009. Each focus group was planned to run for a maximum duration of two

hours.


6.2.2 Selection criteria, recruitment and participation

Specific groups of riders were targeted separately for this study to allow

comparisons between them, and to ensure comprehensive coverage of the relevant

issues. The study design sought to include Brisbane city commuters, members of

online scooter forums, students, and key stakeholders with industry links who were

also active moped or scooter riders. To effectively explore differences between

moped riders and those of larger scooters, focus group design was tailored

accordingly. The first group was intended to engage industry representatives as key

stakeholders, whose knowledge and opinions may have differed from other riders.

The second group was to include only LA moped riders, the third group only LC

scooter riders, and the fourth and final group to include both moped and scooter

riders (depending on recruiting success). While the earlier evidence indicated that

there would be some overlap of characteristics, opinions and motivations of riders

across these groups, it was decided that sufficient differences might emerge that

would prove useful to the future study and possible recommendations.

Consequently, four discrete focus groups were planned with an ideal representation

of between six and eight riders in each.

The study was not designed to include motorcycle riders who were not active

moped or scooter riders. Compared with mopeds and scooters, there is a relative

abundance of research literature covering motorcycle use and safety. As relevant

information regarding motorcycle use in the study area could be obtained from

previous research, focus group design excluded motorcycle only riders. However,

some focus group participants did have experience riding motorcycles as well as

mopeds or scooters.

Three methods were used to recruit participants for this study. A message

was posted on an online forum for scooter enthusiasts, following a written request for

approval by the forum moderator on 2 February 2009 which was subsequently

granted. During data collection for Study 1 on 26 February 2009, 58 A4 colour

flyers were placed on mopeds and scooters in Brisbane inner city designated parking

areas (see Chapter Four). The flyers invited regular riders to participate in the focus

groups and were attached to mopeds and larger scooters in roughly equal

proportions. Direct contact was also made with managerial staff at Brisbane moped

and scooter retail outlets in order to recruit industry representatives. These industry-


based participants were also active moped or scooter riders.

All participants were provided an information sheet describing the research,

were asked to sign an appropriate consent form, and were offered AUD$50 each to

compensate for their time. This study was confirmed as meeting the requirements of

the National Statement on Ethical Conduct in Human Research by the Queensland

University of Technology Human Research Ethics Committee on 15 December 2008

(approval number 0800000955).

6.2.3 Data collection

A digital voice recorder was used for data collection and the recorded

discussions were later transcribed by a professional transcription service. Written

notes were also taken during focus group discussions to provide clarification on

some of the issues covered. Reiteration and paraphrasing of comments and

discussion points was also used by the researcher to clarify the meaning of some

statements or discussions that might otherwise have been misinterpreted. As the age

of all participants was not adequately recorded at the time of participation, follow-up

communications were made by email and telephone to verify the age of participants.

6.2.4 Guiding questions

The focus group discussions were guided by a series of open-ended questions

designed to explore topics identified in the research literature. Key issues included

the previously identified main contributors to crash and injury risk for motorcyclists

(Greig, et al., 2007): inexperience or lack of recent experience; risk taking; driver

failure to see motorcyclists; instability and braking difficulties; road surface and

environmental hazards; and vulnerability to injury. In addition to these and other

issues that were raised as relevant to safety (including crash experience and rider

licensing and training), discussions also covered usage patterns, motivations for

moped and scooter use, and alternative transport modes.

Care was taken where possible to avoid the use of leading questions which

may have produced unnecessarily biased responses. For example, to introduce the

topic of risk taking for discussion, the guiding question used was ‘what kinds of

things do you consider risky in terms of riding?’ This was considered less likely to


introduce or exacerbate bias (due to socially desirable responses for example) than a

question such as ‘do you consider speeding to be risky?’ A complete list of guiding

questions developed for the focus groups is provided in Appendix B. The questions

were often modified prior to delivery as deemed appropriate for the particular

context. Many questions were answered in the course of general discussion so did

not need to be asked directly. The guiding questions for industry representatives did

not differ from those used for other groups.

6.2.5 Analysis

The data were analysed by reference to the key issues through the guiding

questions discussed in the previous section. Keyword searches of the focus group

transcripts were used to identify statements relevant to these issues. These issues

were grouped into broad themes reflected in the structure of the following results

section (section 6.3). The themes include: purpose of riding and trip characteristics;

motivations for riding; perceived hazards and safety awareness; vehicle control

skills; road positioning; deliberate risk taking protective clothing and conspicuity;

and attitudes and beliefs about licensing, training and education. As a qualitative

exploratory study, the objective was to extract the meaning of statements and general

discussion rather than to quantify references to the individual issues. It has been

noted that themes can be identified as expressions made with frequency,

extensiveness, or intensity (Krueger, 1998). It has also been argued that frequency

should not be taken as an indicator of importance and that critical findings might be

mentioned only once (Krueger, 2006). Accepting this for identifying themes in the

current study, priority was given to the extensiveness and intensity of statements

rather than the frequency.

With the differences between moped and scooter riders central to the overall

research aims, the stated views and experiences of moped riders were compared with

those of scooter riders. Analysis was continued only to the point where all specific

topics relevant to the key issues had been identified. Bearing in mind that the Study

3b survey would be necessarily limited in the number of items included, and that it

would also be informed substantially by the relevant literature, the Study 3a objective

was in large part to gather information specific to the study area. A greater depth of

analysis would have been possible, potentially using software designed for


qualitative data analysis such as NVivo. However, as the study was exploratory and

not intended to test or develop theory, it was not considered necessary to adopt such

methods.

6.3 Results

6.3.1 Participation

A total of 28 people responded to the request for focus group participants, of

which 23 actually participated. The 58 flyers distributed in parking areas generated

responses from 21 riders, 17 of whom participated in focus groups (a response rate of

29%). The online forum message generated four responses, three of whom

participated. The remaining three participants were recruited by direct contact at

retail scooter outlets in Brisbane.

The 23 focus group participants included 16 males (70%) and seven females

(30%), with an age range of 20 to 61 years and a mean age of 38.1 years. Moped

riders were generally younger (mean age 31.3 years) and less experienced, with

seven of 11 participants having ridden at least occasionally for between one and five

years. Of the scooter riders (mean age 44.4 years), seven out of 12 participants had

been riding some form of PTW for 20 years or more. Female riders were distributed

more or less evenly across groups, with the exception of the first group containing

industry-involved riders who were all male. The gender distribution was roughly

consistent with that seen in moped crashes Queensland.

As intended in the study design, the first three groups were each of a different

composition in terms of experience levels, motorcycle industry knowledge, and/or

type of scooter usually ridden (LA moped or LC scooter). Participants included

eleven LA moped riders (48%) and twelve LC scooter riders (52%).

The first group included two managers of retail scooter/moped outlets, a

motorcycle rider trainer and two riders with extensive scooter riding experience. All

five participants in this group were male and were current scooter (4) or moped (1)

riders. The second group comprised of six participants who were exclusively moped

riders, including three males and three females. The third group also included six

participants, four male and two female, all of whom were LC category scooter riders.

The fourth and final group of six participants were mostly LA category moped riders,


four male and two female, with one LC category scooter rider.

6.3.2 Topics on moped and scooter usage

6.3.2.1 Purpose of riding and trip characteristics

Most participants (~85%) were regular city commuters who travelled less

than 100 kilometres per week on average. As well as commuting, riders frequently

used mopeds and scooters for short trips to shops and entertainment destinations such

as cinemas and cafes. Some participants rode occasionally for recreation, although

none used their scooter or moped predominantly for that purpose and some did not

ride at all for recreation. Most participants also had regular access to a car and had

acquired their moped or scooter as a second vehicle. In some cases a moped was

shared among family and friends if the use of such was thought more convenient

than a car or public transport.

For us, I am married and we didn't want to buy another car. So this was just

a way to have a second car without really having a second car (female moped

rider).

Two riders of larger scooters rode substantially greater distances in rural as

well as urban areas. One rider had recently toured his 250cc scooter from Brisbane

to Tasmania and back, a distance of some 5,000 kilometres or more. Another

participant travelled around 1,000 kilometres per week on PTWs, mostly commuting.

This participant alternated between a motorcycle and a three-wheeled 500cc scooter

for his daily commute.

6.3.2.2 Motivations for riding

When asked about their motivations for riding, respondents typically

mentioned multiple factors including cost, time-efficiency, practicality and

enjoyment. Environmental considerations such as fuel consumption, exhaust

emissions and traffic congestion were mentioned as a secondary factor by some

participants, but these were not a primary motivating factor for any participant.

There was some sense of a scooter culture or social scene to which a minority of


participants were attracted and in which some participated. However, a desire for

social inclusion through participation in rider clubs or organisations was not

expressed by most participants.

Cost

Cost as discussed by participants generally included vehicle purchase cost,

the cost of fuel and the cost of parking in the city, which combined together were

invariably seen to be cheaper than using public transport. Most participants stated

that using ‘pay-and-display’ parking areas in inner city Brisbane defeated the

purpose of riding a scooter or moped and these parking areas were therefore

generally avoided. Several participants claimed that against the cost of public

transport they would recoup their total investment in a moped within two years:

I actually leave home an hour later than what I was originally doing on the

bus… $25/30 a week in public transport, now it's $3 a week (male 125cc

scooter rider).

I find it's cheaper than public transport… it's actually cheaper for me to scoot

into work every day than catch a City Cat [ferry] (male moped rider).

Practicality and convenience

Some participants also sometimes rode or had in the past ridden motorcycles,

but mopeds and scooters were thought more practical than motorcycles for their

storage space and manoeuvrability in parking areas. Scooters and mopeds were seen

as time-efficient, both for ease of parking and also for negotiating congested traffic.

They were also widely considered an easy option for short trips to local shops and

services:

…it's great if you want to duck down and get a few things… Finding a park is

not an issue… (female moped rider).

…I was actually thinking of upgrading my scooter next year to a bigger

scooter, like a 300 or something... The reason I was thinking of a scooter is

because I looked at motorcycles, I was looking at a 500 or 600 cc, and one of

the issues was that I found all of a sudden storage capacity disappeared …the

new scooters that are coming out have great storage facilities (male 125cc

scooter rider).


Enjoyment

Moped and scooter riders alike claimed to enjoy riding and no participant

said that that they did not enjoy it overall, despite noting certain instances where they

felt uncomfortable. Participants often spoke of enjoying a sense of freedom,

heightened awareness and engagement with the world outside while riding. Open

roads with low traffic volumes were seen as the most enjoyable road and traffic

environment. Several participants disliked riding in bad weather and would avoid it

if possible, but one respondent claimed that wet weather provided an opportunity for

yet more enjoyment:

I really enjoy it but mainly because I am from England and you wouldn't ever

be able to do it for enjoyment in England. I enjoy it in the sunshine (male

scooter rider).

It's a lot more fun than sitting on a bus (male moped rider 1).

Except when it rains (male moped rider 2).

Ah, that can be fun too (male moped rider 1).

Environmental motivations

Environment and conservation were mentioned as motivating factors by a

minority of participants, but never as primary factors. Perhaps unsurprisingly, some

industry representatives seemed the most enthusiastic in discussing the

environmental benefits of PTW use. Otherwise, the topic was mostly only

mentioned in passing and was clearly considered less important than issues such as

personal mobility, cost and practicality in overall discussions:

So economical, absolutely, and environmentally friendly (male moped rider).

Yeah, you are reducing your footprint, which is a significant factor now

(female moped rider).

Those who did note environmental considerations as a motivating factor

alluded to fuel consumption and vehicle emissions particularly, but also traffic

congestion as an environmental problem.


6.3.3 Topics concerning safety

6.3.3.1 Perceived hazards and safety awareness

Most participants appeared optimistic regarding hazard perception and safety

awareness in general. There was widespread recognition among participants of their

greater vulnerability relative to car occupants, in terms of both injury risk and crash

risk in certain situations. At the same time, a commonly expressed view was that

risk can be sufficiently negated with experience over time, as captured in the

following statement:

You become very aware of how to be safe, I think (male moped rider).

Some riders claimed that the official 50 km/h top speed for mopeds is

hazardous in 60 km/h speed zones because they are unable to keep up with traffic

flows. There was general agreement on this point and higher speed zones were

largely avoided by moped riders:

I find that limitation very restrictive. I don’t think it’s very safe, personally

…if people could travel at 65, the going rate for a 60 zone most mornings, or

60…because that’s the speed limit, it would be safer. Going at the speed of

the traffic is often safer than going under the speed limit (male scooter rider).

Power increases for mopeds were therefore generally seen as advantageous in

terms of safety. Some moped riders claimed that their moped was capable of speeds

up to 70 km/h without any performance modifications. There was some knowledge

expressed of how to modify a moped to increase engine power output, such as fitting

aftermarket exhaust systems or removing ‘restrictors’. Some riders had spoken to

mechanics about carrying out such work, and at least one mechanic had refused to

modify a moped as it is illegal to do so.

Other road users

Participants universally perceived other road users, particularly four-wheel

drives (SUVs), utilities, trucks and buses, as the primary hazard and threat to safety.

The key issues raised in relation to other vehicles were conspicuity, driver distraction

and vigilance, lane positioning and proximity to other vehicles, and ability to keep up


with traffic flows.

The risk of colliding with pedestrians while accessing parking emerged as a

concern for some riders, particularly where high kerbs prevent easy access to

available space for most PTWs, and particularly mopeds and scooters with small

wheels and low ground clearance:

You have got to find somewhere to get up on a kerb (female moped rider).

You have to go up where the footpath is, so you have to actually watch out for

pedestrians as well when you are trying to park, trying to get up somewhere

(male scooter rider).

Aggressive behaviour by other road users was an issue raised by numerous

participants. More than one participant claimed to have been physically and verbally

harassed by aggressive car drivers:

I don't drive extremely or anything, but this guy… gosh, he was in a rage. I

hadn't - he wasn't even close to me when I had done it… He got up close and

then started trying to push into me and yelling abuse at me. That was really

scary (female moped rider).

I worry about cars pulling out of the lane into your lane… people pull out

really quickly and if you are coming down that side, they don't look in their

mirrors. They don't see a scooter. Like, that's the most dangerous time (male

moped rider).

It was frequently acknowledged that PTWs can be difficult to see among

other traffic and some riders took personal responsibility for maximising their

conspicuity. Some riders implied that this was a greater problem for small scooters

and mopeds than for larger PTWs. This topic is addressed further below in section

6.3.3.3.

Environmental hazards

Environmental hazards as defined here include poor road surfaces and related

infrastructure, and poor weather conditions. Adverse environmental conditions were

said to have contributed to a small minority of crashes in which participants were

involved. Poor and contaminated road surfaces were mentioned frequently as a


cause for concern, yet riders did not generally seek to avoid routes on which these

hazards were known to exist. Potholes and steel manhole covers were mentioned

more often than any other road surface hazard, while painted markings were also

mentioned by some participants. Some participants stated that they would prefer to

take a known route with familiar significant hazards to one that may have been in

better condition but was ultimately unknown. Several participants mentioned the

particular vulnerability of mopeds and small scooters to poor road surfaces in terms

of their small wheels, (perceived) limited braking capabilities and suspension:

…there was one part of the road where it was literally like someone had a

bucket and just scooped up part of the road and if the front tyre had gone

over that, I would have crashed… (male moped rider).

There's some serious ditches in Brisbane, really savage holes (female moped

rider).

As noted above, several participants disliked riding in rain and high winds

and would avoid it if possible. One rider noted that other traffic, and particularly

heavy vehicles, can exacerbate hazards produced by poor weather conditions:

There's an off-spray as well when it's raining. You get a big truck come past

you when it's raining, you don't stand a chance because you can't see… it will

knock you off your bike... They don't slow down as they come past you


6.3.3.2 Riding skills and behaviour

Vehicle control skills

The relative ease of use of mopeds and scooters compared to motorcycles

was clearly attractive to many participants. The general perception was that mopeds

and scooters are easy to ride, and this appeared to relate mostly to their automatic

transmission and light weight.

Some riders had limited knowledge of how their vehicle might perform under

certain circumstances or through certain actions, and some appeared to have been

misinformed, leading to the adoption of unsafe riding practices. Braking skills in

particular were the most prominent issue with regard to vehicle control. Several


young and inexperienced moped riders expressed a fear of using their front brake,

due to something they had heard from other riders.

…the girl I bought my bike off said ‘don't touch it (front brake), or you will

come off your bike… All right, never touch that’ (female moped rider).

Another rider talked of crashing due to braking problems, although his

statement suggests he may have avoided crashing had he maintained a safer distance

to the vehicle in front:

…braking. I fall twice already because of that… the car in front of you

stop(s) and then the wheels just lock and you fall and there's nothing you can

do about it… I never hit another car, but I fall because of it (male moped

rider).

Road positioning

Knowledge and beliefs about lane positioning in traffic varied considerably

among participants. Riders with more experience expressed greater awareness of the

importance of lane positioning, as well as of maintaining appropriate buffer zones

between themselves and other vehicles. It was widely acknowledged that positioning

in traffic impacts upon rider conspicuity and that, theoretically at least, a rider should

‘own’ their lane to deter other vehicles from encroaching on their space:

Even up the hill, I reckon, I always ride in the centre of the lane. Whereas if

you go in the gutter, you just get pushed in the gutter more (male moped

rider).

I always try and stay fairly wide in my lane, so that even if they do pull out,

I'm as far away from them as possible. Also, then I'm not in their blind spot

as well, I think (male moped rider).

In contrast, some moped riders said that they sometimes felt safer riding in

bicycle lanes and would occasionally do so despite awareness that this is not legal in

Australia. One participant thought that mopeds should be permitted on designated

bicycle and walking tracks which are often isolated from roadways in Brisbane,

though this view was not supported by others.


You should be able to drive on the bicycle lane, I think. I did that once. I got

a fine, $60. I say to the cop, ‘Well, it was dangerous for me to be on that

road, it was an 80 (km/h zone), so I was trying to go slower on the bicycle

lane’ (male moped rider).

Deliberate risk-taking

Some participants admitted occasional deliberate risk-taking, most commonly

speeding and lane splitting through traffic. Some riders considered filtering through

stationary traffic somewhat risky, but most participants seemed to consider it safe.

Following too closely behind other vehicles was also mentioned several times, with

the general disclaimer that this was often difficult to avoid in heavy traffic. These

actions were mostly considered to represent low and acceptable risks to be taken as

part of a daily commute. As mentioned, on-road bicycle lanes had been used by

some participants, with claims that this was safer for them in some situations (and a

hint of acknowledgement of the greater risk to pedestrians and cyclists).

Speeding behaviours were often placed in the context of surrounding traffic

flows, with the view that it is often safest to travel at similar speeds to surrounding

traffic. Some mopeds were said to be capable of speeds well in excess of the 50

km/h to which they are ostensibly limited. One moped rider noted that his 50cc

vehicle is also available with a more powerful engine in an otherwise identical

model, therefore claiming that his moped was safe at much higher speeds than the

legal limit of 50 km/h. Some male riders admitted to having engaged in ‘stupid’

behaviour in the past, best described as sensation-seeking:

Before I got a scooter, I had hired one… with a group of friends… there was

definitely competition between us and we were going down hills, seeing who

could get there fastest, trying to overtake each other… stupid stuff… if you

were to ride in groups, especially boys, there would be a bit of stupidity and

competition (male moped rider).

Queensland regulations concerning lane splitting and filtering through traffic2

were poorly understood. As mentioned above, lane splitting through moving traffic

was generally seen as risky, though some riders were willing to do it. By contrast,

2 Technically neither lane splitting or filtering are permitted in the study area in most cases, though

there is generally little enforcement.


filtering through stationary traffic was generally seen as safe and was practised at

least occasionally by most riders. One moped rider also suggested that filtering was

particularly risky for mopeds due to their limited acceleration capabilities:

Only problem when you filter on a 50 (moped), you can get to the front of the

traffic but then the lights go green. The motorbike has the power to get away

but a 50 doesn't, so you have to start again (male moped rider).

I lane split… It'd take me five times as long to get home if I didn't (male

moped rider).

…when the traffic has stopped, I trundle down the middle. I wouldn't do it if

the traffic is moving. That's too risky (male moped rider).

Crash experience

Numerous participants talked of involvement in low-speed crashes, one of

which was said to result in serious injury. Participants had been involved in crashes

with other vehicles and in single vehicle crashes, including some where riders

claimed to be avoiding another road user. Crashes apparently occurred mostly due to

other vehicles failing to yield, and to poor vehicle handling skills and road

positioning on the part of riders. As noted above, braking skills were clearly an issue

for some moped riders. Also noted above, poor weather conditions were said to have

contributed to a small number of crashes in which participants were involved.

Some riders who had never crashed were unsure whether or not they would

ride again after such an event, while some who had crashed and sustained injury

were not deterred from riding as a result:

…the bus pulls out (of a tunnel) and I was flying along and I had to stop to let

the bus out and the other bloke (behind) didn't stop. He just hit me, hit me

pretty hard (male moped rider).

… if I ever got hit I probably wouldn't get back on a scooter... probably

wouldn't want to take that risk again (female moped rider).


6.3.3.3 Protective clothing and conspicuity

Decisions on whether or not to use protective clothing appeared to hinge to

varying degrees on convenience, comfort (particularly in very warm or cold

weather), perceptions of onlookers regarding image, and self-perceived crash and

injury risk. The cost of protective clothing was also seen to be prohibitive by some

participants, although it was evident from responses to such claims that certain items

are more affordable than some riders believe. Some participants admitted to

prioritising fashion and immediate comfort over protection, while others who did

wear protective clothing appeared sometimes to be self-conscious about being

‘overdressed’:

I think a lot of people think you’re a bit silly... I’ve got a jacket with the

shoulder pads and elbow pads and people think, ‘oh, that’s a bit of overkill

isn’t it?’ (female scooter rider).

I am probably the worst person. I am that person on the bike that wears the

singlet and shorts and thongs every day …you want to feel the sun on your

skin... If I was riding a bike and I was doing some serious speed, then of

course I would buy protective gear, because you would look stupid… some

people on scooters go way overboard on protective gear. They are not going

fast enough to rip an arm off... (female moped rider).

I see some people just wearing thongs on a scooter. I always wear shoes

because I don't want to lose a toe or a foot (male moped rider).

The issue of conspicuity in traffic was raised by several participants. As

noted previously, participants recognised that PTWs can be difficult to see in traffic

and some riders implied that this was a greater problem for small scooters and

mopeds than for larger PTWs. One rider suggested that high visibility clothing was

more important than clothing with impact and abrasion protection, due to its

perceived value in preventing collisions with other vehicles which thus negates the

need for protective clothing:

…(high visibility jackets) should be compulsory …they do it for truck drivers,

why don’t they do it for motorbike drivers? …if someone’s going to hit you


with the safety gear, you’re dead anyway …so it’s more important to be seen

(male scooter rider).

Opinions differed on this point however, with some riders claiming that

drivers failed to see them simply because they just failed to look, and that high

visibility clothing was therefore of little if any value.

6.3.3.4 Attitudes and beliefs about licensing, training and education

There were differences of opinion between participants with motorcycle

licences and those without regarding a potential mandatory PTW licence requirement

for moped riders. Unsurprisingly, moped riders who did not hold a motorcycle

licence generally endorsed the status quo, implying that introduction of a specific

moped (or motorcycle) rider licence was unnecessary and would clearly deter some

new moped riders. The point was also made that no amount of instruction or

education will deter some individuals from unsafe and risky behaviour:

… they really got it right for the 50cc riders, that you don't need to do a

licence or a training program because if you got your car licence you know

the road rules... if you weren't feeling comfortable you could go and get

training. But to make it compulsory would be unfair (male moped rider).

I think if you are going to get a test on it, you might as well get a real bike


…there wouldn't be many left on the road, if they made that compulsory. I

think that's why they are so popular because they are not much effort (female

moped rider).

To be honest, if you were going to be a nutter on a bike, whether you have a

test or not, you are going to be a nutter on a bike (male moped rider).

Moped riders were somewhat more open to the concept of training and

education than they were to mandatory licensing as a general tool to improve rider

safety. However, actual uptake would depend on the delivery format and required

time commitment, among other factors. It also appeared that some riders were

largely unaware of where or how to access such services and material:


If there was something to go to …I would definitely go to it (female moped

rider 1).

…if there was a video, yeah, I would watch it, but if I had to go to some sort

of program, I wouldn't (female moped rider 2).

I wouldn't go and book in for a whole day or anything …if it was just a short

hour, two hours or something like that, yeah, I would be prepared to pay

(female moped rider 1).

Riders of scooters and mopeds alike spoke more favourably of rider training

and education programs than they did of a mandatory PTW licence for moped riders.

This was true for both trained and untrained riders although trained riders stressed

that the value of training and education is only fully recognised over time and after

course completion:

I did a Ride Smart course… I actually did learn a stack… Not specific to

scooter riding… but some hints/tips… it was really valuable and I use it every

day (male moped rider).

The general discussion reflected a perception among moped riders in

particular that little skill is required to ride a moped and that the safety benefits of

education, training and skills testing would be negligible. Some participants were

evidently uninterested in improving their safety by even the simplest means:

I'm not sure, my rego (registration renewal notice) came the other day and

they sent me out a booklet, which I chucked in the trash, but it could have

actually been to do with scooter safety (male moped rider).

6.4 Discussion

The Study 3a focus groups provided an invaluable qualitative foundation for

the development of Study 3b, involving a larger quantitative survey of moped and

scooter riders in the broader study area of Queensland. Despite the small sample

size, the current study supported the view that moped riders differ from riders of

larger scooters in terms of experience, knowledge, attitudes and beliefs regarding

rider safety. However, there was also a diversity of views among participants


independent of PTW type. Diversity of characteristics among motorcyclists has been

noted in previous research (Broughton & Walker, 2009; Wong, Chung, & Huang,

2010) and the current study shows that Brisbane moped and scooter riders are

likewise a heterogeneous group. This has implications for intervention development

and implementation, as riders are likely to differ substantially in both their needs and

their receptivity to particular programs (Wong, et al., 2010).

As might be expected of a heterogeneous sample, there was considerable

diversity of knowledge, attitude, belief and opinion expressed by focus group

participants. However, a number of themes drew consistent if not universal

responses. A brief excerpt of discussion from the fourth and final focus group

captures two key points which were arguably most prominent in the overall study.

First, the overall cost of moped and scooter travel was a primary motivating factor

for all participants. This refers to time as well as financial cost, as perceived (if not

accurately estimated) by participants. Second, risk was recognised as essentially

inherent in the activity of riding and was something to be accepted. Risk was

something actively managed by some participants yet hardly managed at all by

others, and this appeared to be an area where LA moped and LC scooter riders

differed in this study.

I think it works out, all up, even with depreciation and all this sort of caper,

with insurance, rego, this, that and the other, it's ten bucks a week …half (the

cost of) public transport, come and go as you please. Sure, you might take

your life in your hands every day, but, you know… (male moped rider 1).

You do some days (male moped rider 2).

…at the end of the day you push that to the back of your mind (male moped

rider 1).

6.4.1 PTW usage

When asked about their motivations for scooter or moped riding, responses

invariably included some combination of the following: cost (vehicle purchase, fuel

efficiency, parking, maintenance); practicality and ease of use (storage space,

automatic transmission, light weight, manoeuvrability); time efficiency (negotiating


traffic, parking); and enjoyment. Approximately 85 percent of participants were

regular commuters, all of whom agreed that competition for motorcycle parking

spaces in the Brisbane CBD becomes intense after about 8am on weekdays. They

were generally reluctant to use a ‘pay and display’ parking area, claiming such would

defeat the purpose of riding a scooter or moped, although paid PTW parking is a

small fraction (~20%) of the cost of city car parking (Brisbane City Council, 2011).

Cost comparisons of moped and scooter use were made against public

transport as a cheap alternative to cars. The cost estimates given for moped use were

not generally detailed or necessarily inclusive of registration and insurance fees,

vehicle purchase, maintenance and servicing, or other incidental expenses. However,

approximate calculations indicate that moped use in Brisbane may be substantially

cheaper than public transport, but only after initial purchase costs are recouped over

a period of two or three years. These findings assist in answering research question

1, as discussed below in section 6.4.3.

Some participants mentioned environmental considerations as a secondary

motivation, but personal mobility and cost always appeared more important than

altruistic motives. When purported environmental benefits of moped and scooter use

were mentioned, the perceived environmental impact was compared against that of

cars rather than public transport, walking or cycling. As noted above, some industry

representatives seemed the most enthusiastic in discussing environmental benefits of

PTW use, arguably reflecting their position as key stakeholders.

Some participants appeared quite passionate about their scooter or moped

riding, while others seemed to view it simply as a means of transport. However,

nobody said that they did not enjoy riding or that they were uncomfortable with it. A

sense of freedom, heightened awareness and engagement with the outside world

while riding are feelings that moped and scooter riders seemingly share with

motorcyclists generally. However, their affinity with motorcyclists on other levels

was variable. Mopeds and scooters were preferred to motorcycles for various

reasons including lower purchase and running costs, ease of use, superior storage

space and, in the case of mopeds, less stringent licensing requirements. Moped and

scooter riders found other motorcyclists in their presence to be protective and

comforting at times, yet aggressive and intimidating at others.

The shared use of mopeds among friends and family members mentioned by

some participants suggests that a single moped may be used at least occasionally by


multiple riders. It is arguably less likely that motorcycles are shared among multiple

riders in this way due to licensing requirements. Moped use in these situations and

also more generally is arguably encouraged by the regulation permitting moped

riding with a valid car licence as the minimum requirement. Moped riders mostly

opposed the idea of a compulsory motorcycle or moped licence for moped riding and

a minority of them said that they would not get one if such legislation was

introduced. These findings assist in answering research question 2, as discussed

below in section 6.4.3.



identified in previous research (Greig, et al., 2007): inexperience or lack of recent


difficulties; road surface and environmental hazards; and vulnerability to injury. As

in the previous chapter, the following sections of this chapter are structured

according to these main contributors to crash and injury risk with slight modification

of the original terminology in the section titles to ensure coverage of all relevant

issues. Specifically, driver failure to see motorcyclists is addressed under the section

on other road users (section 6.4.2.3), and instability and braking difficulties

addressed under PTW control and riding skills (section 6.4.2.4). These findings

assist in answering research questions 3 and 4, as discussed below in section 6.4.3.


Previous research has found elevated crash risks among inexperienced riders,

those lacking recent experience, and those inexperienced with a particular PTW type

(ACEM, 2008; Haworth, Smith et al., 1997; Mullin, Jackson et al., 2000; Rutter &

Quine, 1996). It appears that most young riders have relatively little riding

experience. However, the growth in moped, scooter and motorcycle use appears to

be driven partly by those taking up riding at an older age, or those returning to riding

after an extended break. Inexperience is therefore considered a potential risk factor

for riders of all ages. Having said this, in the current study (with a small sample of

riders), safety awareness and adoption of safe riding practices appeared to increase


with age and experience. The older and more experienced participants tended to be

riders of LC category (over 50cc) scooters, all of whom claimed to hold a motorcycle

licence and may have attended training. In contrast to the younger and less

experienced riders, they were more likely to value (and usually wear) protective

clothing, had substantially greater knowledge of hazards, mechanics and physical

dynamics with regard to vehicle performance, and perceived rider training to be

beneficial.

In the current study, those who had undertaken rider training tended to value

it highly, while those who had not generally thought it unnecessary. On the whole

participants seemed to derive their safety awareness largely from personal

experience, which can only be gained through riding, or otherwise anecdotally.

Moped riders were largely unaware of where to source information on PTW safety

and none of them appeared to have actively sought such information as a means to

compensate for inexperience. If any such information was to be sought, interactive,

online, video and other easily accessible formats were usually preferred.

Despite the lack of positive and relevant evaluations, training and education

programs remain the most popular means of attempting to compensate for

inexperience. One of the problems with designing and delivering effective rider

training arguably lies with the heterogeneity of the riding population (Haworth &

Rowden, 2010). Heterogeneity was a feature of the small sample in the current

study, with some riders disinterested in vehicle performance and handling, while

others seemed interested and also knowledgeable about these issues. The former

might be at greater risk even if they have a ‘safe’ attitude; skills training as well as

education might benefit these riders. The latter are potentially at greater risk if their

attitude is ‘unsafe’ and these riders may benefit from education, although none of

these self selected participants showed a strong propensity for risk taking (see section

6.4.2.2 below).

The current study indicated that the issue of inexperience should be further

explored in the survey of moped and scooter riders in Study 3b. Specifically, the

survey should examine the amount of riding experience, frequency of moped or

scooter use, distances travelled and participation in rider training.


6.4.2.2 Risk taking

The literature suggests that risk taking on PTWs is generally more prevalent

among young riders, male riders and recreational riders. In some European

countries, recreational moped use by riders as young as 14 years of age has been

associated with risky riding behaviours and subsequently high crash and injury rates

(Aare & Holst, 2003; SWOV, 2009). Given the older age at which moped riding is

permitted in the current study area (17 years), and that riders are also permitted to

drive a car at that age, less recreational moped use and less risky riding among young

riders might be expected in the current research.

As noted in the previous section, a strong propensity for risky riding

behaviours was not evident among participants in the current study and some

participants were arguably risk-averse. While some participants may have sought to

offer socially desirable responses, this general finding appears to relate to the age,

gender and motivations of participants for riding. The youngest participant in the

current study was 20 years of age and the mean ages were 31 and 44 years for moped

and scooter riders respectively. Almost one third of participants were female and the

vast majority of participants were commuting riders who did not ride for recreational

purposes.

In some instances participants did admit to and discuss risk taking for

enjoyment, which mostly involved speeding. The clearest example of this involved a

group of male riders who had hired mopeds while on holiday. However, activities

that might be defined as risky, including speeding, lane splitting and closely

following other vehicles, usually took place within the context of commuting and

negotiating traffic. It was widely believed that an ability to keep up with traffic

flows was important, even if this meant slightly exceeding speed limits sometimes.

As such, several participants argued that speed restrictions on mopeds should be

increased to 60-65 km/h. These and other issues regarding risk taking are explored

further through the survey of moped and scooter riders in the following study

reported in chapter 7.


6.4.2.3 Other road users

Driver failure to see PTW riders has been identified as a main contributor to

PTW crash and injury risk in previous research (ACEM, 2008; Broughton & Walker,

2009; Huang & Preston, 2004). In the current study, participants universally

perceived other road users, in particular larger vehicles, as the primary hazard and

threat to their safety. All participants seemed able to recount instances where they

had not been seen by other road users, which had produced predictably hazardous

situations. Additionally, it was also noted that other road users do not always yield

to a PTW with right-of-way, despite the other road user having seen the PTW in

advance. In extreme cases as described by participants, other road users can become

physically aggressive and intimidating to the point where right-of-way becomes a

question of size and mass.

Participants all accepted that other road users represent a substantial hazard,

but some appeared more proactive than others in managing or trying to manage this

hazard. It was generally acknowledged that defensive riding techniques are useful

for reducing risk of collision with other vehicles, though not all riders seemed to

know of specific techniques. Some riders went to great lengths to maximise their

conspicuity, while others thought this a fruitless exercise as other road users ‘don’t

look anyway’.

Overall, the problem of other road users for PTW riders is clearly

documented in the research literature. Driver failure to see PTWs is a stated cause of

a large proportion of PTW crashes with other vehicles, and this seemed well

understood by focus group participants in the current study. Issues regarding other

road users, including defensive riding techniques and PTW conspicuity, are explored

further through the survey of moped and scooter riders.


As suggested in the literature, PTW control requires greater skills than car



difficulties in other research (Greig, et al., 2007). In some cases these factors may be

exacerbated for mopeds and scooters due to smaller wheel diameters, shorter


wheelbases, less advanced braking systems and limited suspension capabilities

compared to motorcycles. Most participants in the current study did not seem to be

aware of these potential differences between PTW types, or the impact they might

have on safety.

The reluctance of some untrained riders to use front brakes is of concern

given that front brakes provide most of the potential stopping power of PTWs,

including mopeds and scooters (Broughton & Walker, 2009; Corno, Savaresi et al.,

2008). In the absence of combined braking systems, which do not feature on most

current moped models, sole reliance on the rear wheel for braking and deceleration is

potentially hazardous. There was some indication in Study 2 of poorer vehicle

control skills among moped riders compared with scooter riders, which may have

related in part to braking skills. Confidence with brake application was therefore

included as a specific question in the survey of moped and scooter riders in Study 3b.

Some participants referred to rider training as a valuable source of

information and skills development, and these were mostly LC scooter riders holding

motorcycle licences. Moped riders generally seemed to believe that they do not

need training or education as they do not travel at high speed, and that holding a car

licence provides them with sufficient knowledge of the relevant road rules.

However, if moped riders could be encouraged or indeed required to undertake

training and/or education they would generally accept it as beneficial, but only after

having done it. While training would be expected to improve the skills of some

riders, the extent to which mandatory licensing and training might actually result in

safer riding practices remains unclear.


As mentioned above, the literature suggests that mopeds and scooters may be

more susceptible than motorcycles to hazards such as potholes and rough surfaces,

due to smaller wheel diameters, less advanced braking systems and limited

suspension capabilities. In the current study, poor road surfaces appeared to be the

greatest perceived hazard after other vehicles. While poor and contaminated road

surfaces were mentioned frequently as a cause for concern, riders did not generally

avoid routes on which these hazards were known to exist. Although improved road

surfaces were generally desired, participants seemed to accept poor surfaces as an


inherent hazard to be managed, along with other vehicles. Other environmental

hazards such as rain and wind appeared not to concern most participants greatly.

They were largely prepared to ride in poor weather conditions if necessary, except

perhaps under extreme circumstances.

A particular hazard identified in the current study but not in the

literature was that of poor access to some designated parking areas in inner city

Brisbane. Having to climb gutters and ride along footpaths or sidewalks (technically

an offence in the study area), some participants noted that this was hazardous to

riders and pedestrians alike. Although the frequency and likely severity of crashes in

such situations is probably low, the issue may warrant further exploration.

The current study did not identify any clear differences between moped and

scooter riders regarding road surface and environmental hazards. However, it was

tentatively concluded in Study 2 that poor road conditions and wet road surfaces

present greater problems for moped riders than scooter riders (comparison with

motorcyclists was confounded by different usage patterns). The topic of road surface

and environmental hazards is therefore further explored in Study 3b.


The greater vulnerability to injury of PTW riders than car and other vehicle

occupants is well documented in the literature and participants in the current study

were generally aware of this greater vulnerability. However, while some took active

steps to reduce their crash risk through rider training and/or their injury risk through

use of protective clothing, others did not do so. There was a sense among moped

riders particularly that they do not travel fast enough to warrant any substantial

investment in reducing their crash risk or their vulnerability to injury.

The use and characteristics of protective clothing have been identified in the

literature as important for rider safety. Protective clothing is known to reduce

severity of non-fatal injuries (de Rome et al., 2011). Some focus group participants

appeared to underestimate the potential benefits of protecting clothing, which may

have related to low perceived crash risk, injury risk, or both. Other participants

appeared relatively cautious, though they did not necessarily use clothing specifically

designed for the purpose. Appearance and image seemed to be an important factor

for many participants in deciding what to wear while riding. The warm climate in


the study area also appeared to discourage some riders from using protective

clothing.

Other research has found lower use of protective clothing among moped and

scooter riders compared to motorcycle riders (de Rome, 2006a; Christie, 2008). In

the current study, protective clothing use was more prevalent among scooter than

moped riders, although no conclusions can be drawn due to the small number of

participants. Use of protective clothing is explored in specific and detailed questions

in the Study 3b survey of moped and scooter riders.


In seeking answers to the four research questions, this study explored rider

perspectives on issues relevant to PTW safety and also usage, identifying key issues

for further exploration. The study suggested important differences between moped

and scooter riders with regard to safety and also usage, although the qualitative

nature of the study and small sample of participants precludes any strong

comparative conclusions.

Research question 1: Why has moped and scooter usage increased? Study 2

demonstrated an ongoing increase in moped and scooter usage, beyond that which

has previously been observed for mopeds in the study area (Haworth & Nielson,

2008). Having confirmed this trend, in the current study, the main motivations for

moped or scooter use were related to cost and convenience, where PTW use was seen

to offer greater overall value than either car use or public transport. In particular,

ability to move through traffic, availability of parking and low purchase and running

costs were the major motivating factors. In the study area, the regulations that permit

moped riding for car licence holders may also have encouraged moped use,

particularly as other motivating factors became more important to participants.

Study 3b further explores the factors underlying increased moped and scooter usage

among a larger sample of riders.


motorcycles differ? The current study involved a small sample of mainly city

commuters and excluded motorcycle-only riders. As an exploratory study, it was not

expected that conclusions would be drawn on different usage patterns of mopeds and


scooters. Indeed in most respects there was little difference in the use of mopeds and

scooters in the current study in terms of where and when they were used, or for what

purpose. However, moped riders were generally younger and less experienced than

scooter riders, who also tended to travel further, sometimes for recreation. Usage

patterns of moped and scooters are explored further in Study 3b.


motorcycles differ? Study 2 suggested that scooter riders are safer than both moped

and motorcycle riders. Given the small number of participants in Study 3a, it was

not possible to objectively compare the safety of mopeds and scooters and thereby

support or challenge the Study 2 findings. However, responses of participants in

general discussion suggested that of the riders present, scooter riders showed greater

safety awareness than moped riders. This is not to say that they were necessarily

more risk-averse, but that they invested more in safety, perhaps out of experience

and/or the requirement for them to hold a motorcycle licence (which may have

exposed them to rider training and education). Specific questions regarding crash

involvement, usage patterns and rider behaviour, among others, are included in the

survey in Study 3b to further explore the relative safety of moped and scooter riders.


motorcycles differ? Study 2 provided some possible explanations for the observed

differences in the safety of mopeds, scooters and motorcycles, though the analysis

also raised some further questions. Noordzij et al (2001: 4) stated that ‘it is the rider

motivation or riding style, rather than the vehicle characteristics which can explain

(the) relation... between moped/motorcycle characteristics and accident rate’.

However, in the current study, the moped and scooter rider motivations and riding

style were very similar (as are the vehicle characteristics to a large extent), so this

does not explain differences observed in Study 2. The qualitative data has suggested

some differences between moped and scooter riders in age, experience, riding skills

and knowledge on safety-related issues which may impact their relative safety.

These tentative findings are consistent with those in Study 2 on rider characteristics,

with the exception of knowledge on safety-related issues (which were not examined

in Study 2). The survey explores these issues further using a larger sample of moped

and scooter riders.


6.4.4 Limitations of Study 3a

Given the sample size of 23 participants and the qualitative nature of the

study, only limited generalisations can be made and these should be in specific

reference to moped and scooter riders in an inner city area. Inner Brisbane represents

the major hub of moped and scooter activity in Queensland so it was thought

essential to canvass the views of these riders as a starting point, although this may

have introduced a bias toward city commuters. As stated above, the focus groups

were used to inform development of a questionnaire targeting a larger sample of

scooter and moped riders across the broader study area of Queensland.

Differences in licensing requirements mean that much of the data may not be

transferrable to other jurisdictions, particularly those which require a motorcycle

licence for moped riding.

There is potential for some sampling bias in that the study may have attracted

disproportionate numbers of riders with a particular interest in safety and/or attracted

those who may have been largely motivated by the financial incentive offered. It is

possible that some participants may have been inclined toward socially desirable

responses. However, as a whole they seemed to be open and relaxed in discussing

potentially sensitive topics such as personal engagement in deliberate risk taking.

6.5 Chapter Six summary

This chapter has presented the findings from an exploratory focus group

study with Brisbane moped and scooter riders. The study has helped to achieve the

overall research aims, which were to develop a better understanding of moped and

scooter usage trends and patterns, and to investigate factors underlying differences in

moped, scooter and motorcycle safety. More specifically, the study has helped to

highlight issues relevant to all four research questions, as summarised above. These

issues were identified as requiring further exploration in Study 3b, involving a state-

wide survey of moped and scooter riders.

Participants offered a wide range of views and perspectives on PTW use and

safety and the overall sample was heterogeneous in terms of age, gender and other

characteristics. Moped and scooter riders generally differed in regard to their

knowledge and practice of safe and risky riding behaviours. In particular, scooter


riders universally valued rider training while most moped riders had not undertaken

any training. Scooter riders appeared to have greater theoretical knowledge than

moped riders, some of which may have translated to safer riding behaviour and

higher skill levels. As scooter riders were older on average than moped riders (44

years versus 31 years), some of the differences observed may relate to experience.

Scooter and moped riders shared motivations for PTW use, with low cost and

convenient commuting the primary motivation and purpose.

The next chapter describes Study 3b, a survey of Queensland moped and

scooter riders. As mentioned previously, Study 3a was designed to assist

development of the survey instrument for Study 3b. In particular, through focusing

on issues pertinent to riders, as well as the general tone of discussions, Study 3a

provided insight into key topics for further exploration and the ways in which these

should be approached in the Study 3b survey. The current study also provided

insight into recruitment methods which could be employed in Study 3b. The key

issues addressed in the current chapter describing Study 3a are carried through to

Study 3b, as described in Chapter Seven.


CHAPTER 7: STUDY 3B – QUEENSLAND SCOOTER AND

MOPED RIDER SURVEY

7.1 Introduction

The main aims of this program of research were to develop a better

understanding of moped and scooter usage, and to investigate factors underlying

differences in moped, scooter and motorcycle safety. Differences in crash and injury

risk for these PTW types have been shown in some research from other jurisdictions,

but the situation in the study area is not well understood to date (Haworth, Greig et

al., 2009; Haworth, Nielson et al., 2008). The analysis of crash and registration data

in Study 2 revealed significant differences between mopeds, scooters and

motorcycles in usage, crash involvement and crash characteristics. In Study 3a,

exploratory focus group discussions also suggested differences between moped and

scooter riders which may impact their relative safety. Study 3a and the review of

literature were subsequently used to inform development of a survey questionnaire

for Study 3b, titled The Queensland Scooter and Moped Rider Survey 2010.

This chapter presents Study 3b, describing and discussing the rationale,

methods, results and implications of The Queensland Scooter and Moped Rider

Survey 2010. The survey was designed to collect information on the demographic,

social, motivational, attitudinal and other characteristics, including crash

involvement and licensing, of Queensland moped and scooter riders. The findings of

the study could then be interpreted in light of the findings of Studies 1, 2 and 3a, and

compared with other research findings, including the six main contributors to crash

and injury risk for motorcyclists (Greig, Haworth et al., 2007).

Study 3b was designed as a quantitative study to address all four research

questions in a similar manner to Study 3a. For Research questions 1 and 2, Study 3b

explored motivations for moped and scooter use, as well as similarities and

differences in usage of the PTW types. Specifically, Study 3b examined self-

reported travel patterns, including distance travelled, frequency of riding, roadway

types used, trip purpose and other information on usage. Study 3b addressed

Research questions 3 and 4 by examining a wide range of safety-related issues that

might impact the overall and relative safety moped and scooter riders. These issues


include motivations for riding, knowledge, beliefs and opinions relevant for safety,

riding behaviour, riding skills and experience, licensing and training, and crash

involvement.


7.2.1 Survey content and delivery

The Queensland Scooter and Moped Rider Survey 2010 (hereafter referred to

as ‘the Survey’) was developed with input from the exploratory focus group

discussions in Study 3a, as well as from relevant material in the literature review.

The questionnaire contained a total of 57 questions grouped in five main sections and

was designed to take approximately fifteen minutes to complete. A hard-copy of the

questionnaire which was used for postal returns is provided in Appendix C. The five

sections into which questions were grouped reflect common themes on which the

PTW safety literature most often focuses, including:

Motivations for riding and travel patterns

Rider licensing and training

Crash involvement3

General approach to riding and riding practices

Demographic characteristics

The five sections appeared in the Survey in the above order, after

consideration of the most appropriate sequence for participants. Care was taken to

ensure a logical structure and to avoid early presentation of potentially sensitive

questions. The Survey did not aim to specifically explore illegal riding behaviours

and direct questions about personal engagement in illegal riding were therefore

avoided. The questionnaire did include specific questions about self-rated riding

skill and perceived risk associated with a range of riding scenarios.

The Survey was developed and delivered using Key Survey software and was

available on the CARRS-Q (Queensland University of Technology) website for a

period of 15 weeks, from 9 February until 30 May 2010. Paper copies of the

3 A 'crash' was defined as ‘any event where you have had a collision with any object or other road

user, or have fallen from your scooter or moped while moving’.


questionnaire generated by Key Survey in PDF format were also available by request

over this period. The Survey was open to anyone who had ridden a scooter or moped

at least monthly in Queensland during the three months prior to questionnaire

completion.

7.2.2 Recruitment and participation

A range of recruitment strategies were employed to attract participants

throughout Queensland, though recruitment focused mainly on larger urban areas. A

media release was prepared by QUT Marketing and Communications in consultation

with the researchers and distributed to appropriate outlets on 16 February 2010. The

survey was promoted in print media between 18 February and 10 March by

Queensland’s largest selling daily newspaper (the Courier Mail), three regional

newspapers (the Gympie Times, Queensland Times, and Bayside Star) and one of the

largest national motorcycle magazines (Australian Motor Cycle News). A link to the

survey was provided on the homepage of the CARRS-Q website. The survey was

also advertised at online scooter forums including Scooteroo and Scooter

Community. While the Survey was designed primarily for online delivery, paper

copies of the same questionnaire with reply-paid envelopes were also supplied upon

request.

In addition to the media release and online advertisements, the survey was

advertised through the distribution of DL-size colour flyers at a range of locations,

events and businesses throughout Queensland (Figure 7.1).


Figure 7.1: Flyer used for recruitment of survey participants

During data collection for the observational study (Study 1) on 23 February

2010, survey flyers were placed on approximately 200 mopeds and scooters in inner

city Brisbane parking areas (102 mopeds, 83 scooters and 11 moped/scooter

unknowns). Flyers were placed on 28 mopeds and scooters at the Kelvin Grove

campus of Queensland University of Technology between 15 March and 4 May

2010, and on approximately 20 parked mopeds and scooters sighted incidentally at

various locations around Brisbane. Flyers were also distributed at the Gold Coast

Australian Motorcycle Expo on 20 February (placed on parked mopeds and scooters

and left on counter at Motorcycle Network stand) and at the Australian National

Scooter Rally in Stanthorpe on 17 April. Further flyers were distributed to moped

and scooter retail and service outlets in Brisbane, Cairns, Rockhampton and Mackay.

An information page describing the research was supplied at the beginning of

the online survey. This information included an assurance of confidentiality and

anonymity with regard to responses and personal information supplied by

individuals. Participants completing hard copy versions of the questionnaire were

supplied the same information on a separate sheet of paper. Completion of the

questionnaire was taken to indicate consent to participate in the research. An

incentive to participate in the research was provided in the form of a prize to be

awarded to one randomly drawn participant. The item offered and supplied was a

DriRider Climate Control Pro motorcycle jacket valued at AU$350. The study was

confirmed as meeting the requirements of the National Statement on Ethical Conduct

in Human Research by the Queensland University of Technology Human Research


Ethics Committee on 11 December 2009 (approval number 0900001230).

As with Study 3a, the current study excluded motorcycle riders who were not

also active moped or scooter riders. As there is a relative abundance of research

literature covering motorcycle use and safety, relevant information regarding

motorcycle use in the study area could be obtained from previous research.

However, the survey did include questions about motorcycle riding experience and

current use of PTW types other than mopeds or scooters.

7.2.3 Data processing and analysis

Data from the online survey were transferred from Key Survey to SPSS

(version 17) software, after which data from completed paper questionnaires were

added to the SPSS data file for processing and analysis. At the beginning of the

survey, participants were asked whether they rode an LA category moped or LC

category scooter. Participants were asked to select which category of vehicle they

rode most often in the event that they rode both categories of PTW. The comparative

analysis of moped and scooter riders was subsequently based on the PTW type used

most often.

Pearson’s Chi Square tests were performed to identify statistically significant

differences at the .05 level between the responses of moped riders and scooter riders.

For questions regarding distance travelled annually and average single trip distance,

the median and mean scores of moped and scooter riders were compared. A Mann-

Whitney U test was performed on average annual distance travelled to identify any

significant difference between mopeds and scooters.

Some of the original variables were recoded into new variables so as to

strengthen the power for statistical analyses. This was necessary due to a lower

number of participants than was anticipated during survey design. Where

respondents were asked to rate perceived risk levels on a scale of one (risky) to five

(safe), answers were recoded into a three point scale (risky; neutral; safe). Similarly,

where respondents rated the importance of factors influencing particular actions,

beliefs or behaviours on a scale of one (unimportant) to five (important), answers

were recoded into a three point scale (unimportant; neutral; important). Respondents

were asked how often they wore a range of protective and non-protective clothing

items while riding, using a scale of one (never) to five (always), responses for which


were recoded into a three point scale (never or rarely; sometimes; often or always).

Key words appearing in self-reported crash descriptions were used to code

crashes into six main crash types, including: Single vehicle lost control; Single

vehicle lost control avoiding other vehicle; Two vehicle – other vehicle failed to give

way; Two vehicle – other vehicle rear-ended scooter; Two vehicle – scooter rear-

ended other vehicle; and Two vehicle – aggressive other vehicle driver.

7.3 Results

A total of 198 people responded to the survey, six of whom did not satisfy the

inclusion criterion of having ridden a scooter or moped at least monthly in

Queensland over the previous three months (Question 1). This left a total of 192

valid survey completions (including partial completions). A total of nine (4.7%)

respondents reported riding both mopeds and scooters, in which case they were

coded as either moped or scooter riders according to the PTW type used most often.

With those who rode both PTW types separated on this basis, there were 153 survey

completions by LC scooter riders (79.7%) and 39 from LA moped riders (20.3%).

The vast majority of respondents (186) completed the survey online, with

only six returning completed paper copies of the survey. All six paper copies were

returned by mail from moped and scooter retail and service outlets in north

Queensland who agreed to assist in promoting the study to their clients.

Before proceeding to examine the differences between moped and scooter

riders, a summary of findings are reported here for all respondents as a whole

(comprised mostly of scooter riders).

Approximately 80 percent of respondents were Brisbane residents and about

70 percent of respondents were born in Australia. Commuting for work or study was

the main purpose of about two thirds of moped and scooter trips. Two thirds of

respondents rode daily or almost daily, while nine percent rode less than once per

week. The most important factors influencing moped or scooter use were availability

of parking, practicality and ease of use, enjoyment and fuel costs.

Males comprised approximately three quarters of survey respondents and

more than 90 percent of respondents were aged 25 years or older. Approximately 72

percent of respondents had been riding mopeds or scooters for five years or less and

40 percent had been riding for two years or less. Almost two thirds of respondents


(65%) had completed some form of rider training.

Crash involvement in the last five years was reported by 29 percent of

respondents, the majority (64%) of whom reported requiring no professional medical

treatment. High levels of perceived risk were associated with not being able to keep

up with surrounding traffic, lane splitting, riding in the far left of lanes and in bicycle

lanes, non-use of protective clothing and emergency braking.

7.3.1 Characteristics of respondents

7.3.1.1 Demographic characteristics

The age groups and gender characteristics of respondents are presented in

Table 7.1. The mean age of respondents was 38.6 and 44.2 years for moped and

scooter riders respectively, and an independent samples t-test revealed that the

difference was statistically significant [t (190) = -2.50, p = .013]. Approximately 23

percent of moped riders were aged 29 years or younger (with 10% below 25 years),

compared with 12 percent of scooter riders (with 5% under 25). By contrast, 34

percent of scooter riders were aged 50 years or older, compared with 18 percent of

moped riders. Males comprised approximately three quarters of survey respondents

and there was no difference in the gender distribution of moped riders compared to

scooter riders.

Table 7.1 Age and gender of respondents

Characteristic Moped Scooter

n % n %

Age group

16-19 1 2.6 2 1.3

20-24 3 7.7 6 3.9

25-29 5 12.8 11 7.2

30-39 12 30.8 44 28.8

40-49 11 28.2 38 24.8

50-59 4 10.3 34 22.2

60 or over 3 7.7 18 11.8

Valid total 39 100.0 153 100.0

Gender

Male 29 74.4 112 73.2

Female 10 25.6 41 26.8

Valid total 39 100.0 153 100.0


Survey respondents provided their postcode to identify their usual place of

residence. Approximately 80 percent of respondents lived in Brisbane, with the

remaining 20 percent divided between South Eastern and Central Coast/Northern

regions (Table 7.2). There was a similar pattern for moped and scooter riders

regarding place of residence. In two cases, scooter riders provided a postcode

indicating a usual place of residence outside Queensland. It remains possible that

these riders satisfied selection criteria of having ridden a moped or scooter at least

monthly over a three month period in Queensland prior to survey completion and

they were subsequently included in the study.

Table 7.2 Place of residence of respondents

Queensland region Moped Scooter

n % n %

Brisbane 30 76.9 122 79.7

South Eastern (excluding Brisbane) 3 7.7 15 9.8

Central Coast and Northern 6 15.4 14 9.2

Other* - 0.0 2 1.3

Valid Total 39 100.0 153 100.0 *Postcode indicating a usual place of residence outside Queensland

The individual weekly income levels and occupations of respondents are

presented below in Table 7.3. Approximately half of all respondents earned average

or above average incomes according to Australian income estimates for full-time

workers (ABS, 2010). There were slightly more moped riders (26%) than scooter

riders (20%) on lower income levels (less than AU$600 per week), but the difference

was not statistically significant. Respondents were asked to report their employment

status according to the options listed in Table 7.3. Multiple responses to this

question were enabled and Table 7.3 therefore reports the numbers and percentages

of participants who selected each individual option. Two thirds of moped and

scooter riders alike reported working in full-time employment. Approximately 13

percent of moped riders and 12 percent of scooter riders reported being students.

Permanent part-time employment was reported more frequently by scooter riders

(9%) than moped riders (3%). By contrast, casual employment was reported more

frequently by moped riders (13%) than scooter riders (5%). Low proportions of

moped and scooter riders alike reported being unemployed (<3%), retired (<7.5%) or

engaged in ‘other’ occupations (<6%).


Table 7.3 Weekly individual income and employment status of respondents


n % n %

Weekly income before tax (AU$)

Up to 399 5 12.8 18 11.9

400-599 5 12.8 12 7.9

600-799 2 5.1 16 10.6

800-999 4 10.3 13 8.6

1,000-1,299 4 10.3 20 13.2

1,300 or more 15 38.5 49 32.5

Not sure/Rather not say 4 10.3 23 15.2

Valid Total 39 100.0 151 100.0

Missing - 2

Employment status*

Full-time employment 26 66.7 101 66.0

Full-time home duties - 0.0 5 3.3

Part-time permanent employment 1 2.6 14 9.1

Casual employment 5 12.8 8 5.2

Student 5 12.8 18 11.8

Unemployed 1 2.6 2 1.3

Retired 2 5.1 11 7.2

Other 2 5.1 9 5.9

*Multiple responses were enabled for this question

Other general characteristics of respondents are presented below in Table 7.4.

A large majority of respondents were either married or living with a partner. This

was the case for 82 percent and 74 percent of moped and scooter riders respectively.

Similarly, the majority of both moped riders (61%) and scooter riders (71%) had no

children under the age of 16 years.

Approximately 70 percent of respondents were born in Australia and there

were no significant differences between moped and scooter riders by country of

origin [ ² (2) = 2.348, p = .309]. Scooter riders appeared more likely than moped

riders to report a history of moped or scooter riding in their family, though the

difference was not statistically significant [ ² (1) = 2.98, p = .084] (excluding ‘don’t

know’).


Table 7.4 General demographic characteristics of respondents


n % n %

Marital status

Single 6 15.4 21 13.8

Married/living with partner 32 82.1 113 74.3

Have partner but don’t live with them - 0.0 3 2.0

Divorced/separated/widowed 1 2.6 10 6.6

Rather not say/Other - 0.0 5 3.3

Valid total 39 100.0 152 100.0

Missing - 1

Children under 16 years of age

Yes 15 38.5 44 29.1

No 24 61.5 107 70.9

Valid total 39 100.0 151 100.0

Missing - 2

Country of origin

Australia 27 69.2 110 71.9

New Zealand 5 12.8 9 5.9

Other 7 17.9 34 22.2

Valid total 39 100.0 153 100.0

Family history of moped/scooter riding

Yes 10 25.6 65 42.5

No 27 69.2 88 57.5

Don’t know 2 5.1 - 0.0

Valid total 39 100.0 153 100.0

7.3.1.2 Participation in scooter clubs and online forums

A large minority (40%) of respondents were members of scooter or

motorcycle clubs or organisations and/or users of online forums. Forum use or club

membership was more prevalent among scooter riders (45.4%) than moped riders

(20.5%) and the difference was statistically significant [ ² (1) = 7.99, p = .005].

Respondents reported use or membership of 26 local and international clubs and

forums. Australian clubs and organisations included the Brisbane Canetoads,

Brisbane Lambretta Club, Ducati Owners Club Queensland, Maxitag (scooter)

Tourers, Motorcycle Riders Association Queensland (MRAQ), Motorcycling

Australia, Ulysses Club and the Virago Owners Club Queensland. About half of the

clubs and forums mentioned were not specifically aimed at scooter or moped riders

but were aimed at motorcyclists specifically or PTW riders generally.


7.3.1.3 Rider licensing, experience and training involvement

The licence characteristics of respondents are presented in Table 7.5. A large

majority (>92%) of moped and scooter riders alike held an open car licence and there

was no statistically significant difference on this variable. There was a statistically

significant difference regarding the proportion of riders who held some form of

motorcycle licence, with scooter riders more likely to hold a motorcycle licence than

moped riders (95.4% versus 34.2%) [ ² (3) = 95.24, p < .001]. Due to low numbers,

this test required collapsing together of the two probationary categories (RE and RE

A) and the two open licence categories (R and R A) in order to obtain a valid result.

While a motorcycle licence is required for LC scooter riding in Queensland,

approximately five percent of scooter riders reported holding no valid motorcycle

licence.

All riders who reported holding a car learner licence (n = 8) also reported

holding a probationary or open motorcycle licence, including two moped riders and

six scooter riders. All of these riders were aged 22 years or older so may have

obtained motorcycle licences prior to recently introduced legislation requiring a car

licence to be held for one year prior to obtaining a motorcycle licence. They may

also have obtained a motorcycle licence in a jurisdiction other than Queensland

where a car licence was not a prerequisite for motorcycle licensure.

Of the respondents who held a probationary motorcycle licence, 40 percent of

moped riders (n = 2) and 37 percent of scooter riders (n = 13) held licences with an

automatic-only condition (RE A class licence). Open motorcycle licenses were held

by 7.9 percent and 69.3 percent of moped and scooter riders respectively. No moped

riders with open motorcycle licences held licences with an automatic-only condition.

Of scooter riders with open motorcycle licences (N = 106), 26.4 percent held licences

with an automatic-only condition.

Licence suspension or cancellation with the last five years was reported by

two percent (n = 3) of scooter riders and no moped riders. None of the scooter riders

who reported a licence suspension or cancellation with the last five years reported

not holding a valid licence at the time of survey completion.


Table 7.5 Licence characteristics of respondents

Licence/s held and recent history of

sanctions

Moped Scooter

n % n %

Car licence held

Learner 2 5.3 6 3.9

Probationary (P1 or P2) 1 2.6 3 2.0

Open 35 92.1 143 94.1

None - 0.0 - -

Valid total 38 100.0 152 100.0

Missing 1 1

Motorcycle licence held

Learner 5 13.2 5 3.3

Probationary (RE) 3 7.9 22 14.4

Probationary – automatic only (RE A) 2 5.3 13 8.5

Open (R) 3 7.9 78 51.0

Open – automatic only (R A) - 0.0 28 18.3

None 25 65.8 7 4.6

Valid total 38 100.0 153 100.0

Missing 1 -

Licence suspended/cancelled last 5 years

Yes - 0.0 3 2.0

No 39 100.0 150 98.0

Valid total 39 100.0 153 100.0

Table 7.6 summarises the riding experience and rider training involvement of

respondents. Scooter riders appeared slightly more experienced than moped riders,

with a smaller proportion of new riders and a larger proportion of riders with more

than 10 years experience, though the difference was not statistically significant [ ²

(4) = 4.16, p = .385]. A large majority of moped (81.5%) and scooter (72%) riders

had been riding for five years or less, with approximately 60 percent in both groups

having ridden for between one and five years.

Participants were also asked if they had ever ridden a motorcycle on road

and, if so, how long ago they first did so. One third of moped riders (n = 13) and

almost two thirds of scooter riders (n = 96) reported having ridden a motorcycle on

road. Of those, moped riders reported having first ridden a motorcycle more recently

than scooter riders. Approximately 54 percent of moped riders who had ever ridden

a motorcycle on road first did so in the last five years, compared with 29 percent of

scooter riders. By contrast, 55 percent of scooter riders who had ever ridden a

motorcycle on road first did so more than twenty years ago, compared with 31

percent of moped riders.


Approximately two thirds (n = 122) of survey respondents reported having

undertaken rider training. Scooter riders were significantly more likely than moped

riders to have undertaken some type of formal riding training [ ² (1) = 22.69, p <

.001]. While approximately 70 percent of moped riders had taken no rider training,

72 percent of scooter riders had taken either pre-licence training (49%), post-licence

training (14%), or both (9%).

Of the twelve moped riders who reported having undertaken training, three

(25%) reported riding other PTW types, including scooters (1) and sport motorcycles

(2). Of the 27 moped riders who reported no training, one reported also riding a

scooter and one reported riding a sport-touring motorcycle.

Table 7.6 Riding experience* and training undertaken


n % n %

Riding experience*

Less than 1 year 7 18.4 19 13.8

1-2 years 10 26.3 41 27.0

3-5 years 14 36.8 45 31.2

6-10 years 6 15.8 27 17.5

More than 10 years 1 2.6 19 10.6

Valid total 38 100.0 151 100.0

Missing 1 2

Training undertaken

None 27 69.2 43 28.1

Q-Ride pre-licence training 7 17.9 65 42.5

Other pre-licence training 3 7.7 10 6.5

Post-licence training 1 2.6 21 13.7

Pre- and post-licence training 1 2.6 14 9.2

Valid total 39 100.0 153 100.0 *Excluding breaks of 1 year or more

The riders who reported undertaking formal rider training were asked to rate

the usefulness of the training on a scale of one (not at all useful) to five (very useful).

Approximately 83 percent of respondents considered rider training to be very useful,

while none reported that training was not at all useful.

7.3.2 Moped and scooter characteristics

Participants were asked about the engine cylinder capacity of their moped or

scooter and, for moped riders only, whether their moped had been modified to


improve its performance. Performance-related modifications were reported by

approximately 13 percent (n = 5) of moped riders. One moped rider reported an

engine cylinder capacity of 70cc, indicating that their moped (as reported) was

technically a scooter. Moped engine cylinder capacities of 49-50cc were reported by

97.3 percent (37) of moped riders. One moped rider and five scooter riders did not

report an engine cylinder capacity.

Engine cylinder capacities reported by scooter riders ranged from 50cc to

650cc. Of respondents who identified themselves as scooter riders (N = 153), 8.3

percent reported engine cylinder capacities of 50cc or less. At least some of these

participants were probably moped riders, though this could not be reliably

determined without more detailed information on vehicle specifications.

Moderate engine sizes were reported by most scooter riders, with 68 percent

reporting engine cylinder capacities from 100cc to 300cc. The most frequently

reported engine cylinder capacities were within this range, with 21.8 percent

reporting 250cc engines, 15.6 percent 125cc engines and a further 12.2 percent

reporting 200cc engines. Scooter engine sizes between 400cc and 650cc (commonly

termed ‘maxi’ scooters) were reported by 21.7 percent of scooter riders.

7.3.3 Travel patterns

The reported frequency of moped and scooter use is presented below in Table

7.7. Most respondents reported frequent moped or scooter use. Daily or almost daily

use was reported by 68 percent of moped riders and 65 percent of scooter riders. A

further 18 percent of moped riders and 16 percent of scooter riders reported riding at

least three times per week.

Table 7.7 Riding frequency

How often do you ride? Moped Scooter

n % n %

Daily or almost daily 26 68.4 96 64.9

At least 3 times per week 7 18.4 24 16.2

Once or twice per week - 0.0 17 11.5

2 or 3 times per month 3 7.9 10 6.8

Once per month or less 2 5.3 1 0.7

Valid total 38 100.0 148 100.0

Missing 1 5


Scooters were reported to travel considerably further than mopeds in terms of

both annual distance travelled and average single trip distance (Table 7.8). The

median distance travelled annually by mopeds (3,000 km) was 60 percent of that

travelled by scooters (5,000 km), while the median single trip distance for mopeds

(10 km) was half that of scooters (20 km). A Mann-Whitney U test was performed

on average annual distance travelled and indicated a statistically significant

difference between mopeds and scooters [z = 3.60, p < .001].

Table 7.8 Distance travelled by moped and scooter riders

Statistic Moped Scooter

Kilometres travelled in last year*

Mean 3,188.24 7,186.23

Median 3,000 5,000

Range 15 – 8,000 10 – 40,000

SD 2,126.01 6,992.68

Std. Error of Mean 349.51 570.95

N 37 150

Average Km travelled in single trip**

Mean 11.89 46.09

Median 10.0 20.0

Range 1 - 64 2 - 700

SD 12.27 84.27

Std. Error of Mean 2.02 6.86

N 37 151

*Excludes missing data and 1 moped outlier with claimed annual distance of 56,000 km

** Excludes missing data and 1 moped outlier with claimed average trip distance of 510 km

The distances travelled were generally proportional to engine size. Annual

travel in excess of 10,000 kilometres was reported by only 3.5 percent of 50-125cc

riders (N = 86), compared with 19 percent of 126-260cc scooter riders (N = 63), 42

percent of 261-400cc scooter riders (N = 12) and 50 percent of 401-650cc scooter

riders (N = 20). A large majority of 50cc (mostly moped) riders (83.7%) reported

travelling no more than 5,000 kilometres per year.

The proportions of riding by speed zone and weekday/weekend are presented

in Table 7.9. Travel in speed zones of 60 km/h or less accounted for about 91

percent and 63 percent of moped and scooter travel respectively. Both mopeds and

scooters were used predominantly on weekdays, though weekend use accounted for a

higher proportion of scooter use (28%) than moped use (19%).


Table 7.9 Proportion (mean %) of riding by speed zone and weekday/weekend


Speed zone (km/h)

40-50 34.23 19.79

60 56.54 43.20

70-80 7.69 21.88

100-110 1.54 15.14

Valid total 100.0 100.0

N 39 153

Weekday/weekend

Weekday (Monday-Friday) 80.64 71.52

Weekend (Saturday-Sunday) 19.36 28.48

Valid total 100.0 100.0

N 39 153

In addition to moped and scooter use, participants were asked how often they

used other transport modes, including cars, motorcycles, bicycles, public transport

and walking. While a range of other transport modes were used on a regular basis by

many respondents, approximately half (49.7%) never used a bicycle, more than one

third (36.6%) never used public transport and one fifth (20.1%) never walked as a

means of transport.

7.3.4 Trip purpose and motivations for riding

Respondents were asked to report the proportions of riding they did for

particular purposes, the results of which are presented in Table 7.10. Commuting for

work or study was the main purpose of 64 percent of moped trips and 57 percent of

scooter trips. Recreation accounted for a similar proportion of moped trips (15%)

and scooter trips (11%), while shopping accounted for a larger proportion of scooter

trips (25%) than moped trips (13%).

Table 7.10 Riding purpose as a mean proportion of usage

Riding purpose Moped Scooter Total

Commuting for work/study 63.95 57.07 58.48

Recreation 15.00 11.01 11.82

Shopping 12.77 24.99 22.49

Working* 3.97 3.20 3.36

Other 4.31 3.74 3.85

Total 100.0 100.0 100.0

* Deliveries, courier, marketing or training


Respondents were asked to rate the importance of a range of factors which a)

influenced their decision to use a moped or scooter generally, and b) may influence

decisions to use a moped or scooter for particular journey. Factors influencing

decisions to use a moped or scooter generally are presented in Table 7.11. The

factors most frequently rated as important by all respondents included practicality

(89%), availability of parking (83%), ease of use (83%), fuel costs (77%) and

enjoyment (77%). There were statistically significant differences between moped

and scooter riders regarding the importance of enjoyment [ ² (2) = 8.96, p = .011]

and licensing regulations [ ² (2) = 13.50, p = .001]. Enjoyment was rated as

important by 60 percent of moped riders, compared with 82 percent of scooter riders.

Licensing regulations were important for 56 percent of moped riders, compared with

28 percent of scooter riders. Compared with scooter riders, moped riders also

appeared somewhat more influenced by vehicle purchase costs, and less influenced

by environmental considerations, but the differences were not statistically significant.

Table 7.11 Importance of factors influencing moped and scooter use generally

Factor

Rating

1 = Unimportant, 2 = Neutral, 3 = Important

Moped Scooter

1 2 3 1 2 3

Parking availability n 3 3 32 10 16 121

% 7.9 7.9 84.2 6.8 10.9 82.3

Practicality n 1 3 35 5 11 134

% 2.6 7.7 89.7 3.3 7.3 89.3

Ease of use n 2 2 35 7 21 121

% 5.1 5.1 89.7 4.7 14.1 81.2

Enjoyment* n 7 8 23 9 18 122

% 18.4 21.1 60.5 6.0 12.1 81.9

Fuel costs n 5 2 31 19 17 115

% 13.2 5.3 81.6 12.6 11.3 76.2

Traffic congestion n 5 6 27 13 27 108

% 13.2 15.8 71.1 8.8 18.2 73.0

Vehicle purchase costs n 4 5 29 17 41 92

% 10.5 13.2 76.3 11.3 27.3 61.3

Vehicle maintenance costs n 4 9 22 18 43 86

% 11.4 25.7 62.9 12.2 29.3 58.5

Poor public transport options n 11 10 17 44 40 65

% 28.9 26.3 44.7 29.5 26.8 43.6

Environmental considerations n 9 11 18 27 34 86

% 23.7 28.9 47.4 18.4 23.1 58.5

Licensing regulations* n 3 13 20 52 57 42

% 8.3 36.1 55.6 34.4 37.7 27.8

*p <.05


The factors influencing decisions to use a moped or scooter for a particular

journey are presented in Table 7.12. The most influential factor appeared to be

whether or not riders needed to carry goods such as shopping or luggage. This was

rated as important by approximately 67 percent of all respondents. Other factors

rated as important by a majority of all respondents were weather conditions (66%)

and distance (62%). There were statistically significant differences between moped

and scooter riders regarding the importance of distance [ ² (2) = 11.93, p = .003],

speed zones [ ² (2) = 18.10, p < .001] and gradient (steepness of hills) [ ² (2) =

10.33, p = .006].

Table 7.12 Importance of factors influencing moped and scooter use for a

particular journey

Factor

Rating

1 = Important, 2 = Neutral, 3 = Unimportant

Moped Scooter

1 2 3 1 2 3

Distance* n 2 4 33 47 19 87

% 5.1 10.3 84.6 30.7 12.4 56.9

Speed zones* n 4 6 28 69 25 58

% 10.5 15.8 73.7 45.4 16.4 38.2

Time of day n 10 12 16 69 27 55

% 26.3 31.6 42.1 45.7 17.9 36.4

Weather conditions n 3 5 30 29 27 95

% 7.9 13.2 78.9 19.2 17.9 62.9

Need to carry passenger n 12 9 17 58 32 63

% 31.6 23.7 44.7 37.9 20.9 41.2

Need to carry goods n 6 8 25 33 16 102

% 15.4 20.5 64.1 21.9 10.6 67.5

Road surface type n 12 8 18 44 42 66

% 31.6 21.1 47.4 28.9 27.6 43.4

Road condition n 11 6 21 45 43 62

% 28.9 15.8 55.3 30.0 28.7 41.3

Gradient* n 13 9 16 80 44 27

% 34.2 23.7 42.1 53.0 29.1 17.9

Physical wellbeing n 17 10 11 42 38 71

% 44.7 26.3 28.9 27.8 25.2 47.0

Emotional wellbeing n 20 11 7 54 47 50

% 52.6 28.9 18.4 35.8 31.1 33.1

*p <.05

Participants were also asked if they had any chronic injury which influenced

their decision to ride a moped or scooter rather than a motorcycle. Such a condition

was reported by approximately seven percent of participants (N = 184), all of whom


were scooter riders, although details of injury were not requested in the survey.

7.3.5 Risk perception and risk management

7.3.5.1 Self-perceived riding skill and risk appraisal

Respondents rated their self-perceived level of riding skill on a scale of one

(beginner) to five (expert), results of which are presented in Table 7.13. The

majority of all riders (58%) rated themselves ‘competent’, while 27 percent of moped

riders and 40 percent of scooter riders rated themselves ‘advanced’ or ‘expert’ riders.

The mean self-rated riding skill level was 3.18 for moped riders and 3.43 for scooter

riders. This difference fell short of statistical significance at the .05 level in an

independent samples t-test [t (186) = -1.90, p = .058].

Table 7.13 Self rated level of riding skill

Skill level Moped Scooter

n % n %

Beginner 1 2.7 - 0.0

Basic 2 5.4 6 4.0

Competent 24 64.9 85 56.3

Advanced 9 24.3 49 32.5

Expert 1 2.7 11 7.3

Valid total 37 100.0 151 100.0

Missing 2 2

Participants were asked to indicate the level of risk they perceived to be

associated with nineteen different riding scenarios (Table 7.14). Overall, the highest

levels of perceived risk related to not being able to keep up with surrounding traffic,

lane splitting, riding in the far left of lanes and in bicycle lanes, non-use of protective

clothing and emergency braking. Other scenarios rated risky by a majority of both

moped and scooter riders included filtering, riding in wet weather, struggling to

climb hills and accelerating down hills more rapidly than usual.

There were statistically significant differences between moped and scooter

riders in the perceived risk associated with four of the riding scenarios. Compared

with scooter riders, moped riders perceived greater risk in riding on unfamiliar roads

[ ² (2) = 9.80, p = .007], riding at night [ ² (2) = 7.23, p = .027], riding with pillions


[ ² (2) = 6.83, p = .033] and riding among larger PTWs (motorcycles) [ ² (2) =

15.01, p = .001]. Compared with scooter riders, moped riders also perceived greater

risk in riding with a group of mopeds or scooters, though the difference was not

statistically significant [ ² (2) = 5.00, p = .082].

Table 7.14 Perceived risk associated with riding scenarios

Riding scenario

Rating

1 = Risky, 2 = Neutral, 3 = Safe

Moped Scooter

1 2 3 1 2 3

Speeding to keep up with traffic n 9 12 18 35 32 86

% 23.1 30.8 46.2 22.9 20.9 56.2

Inability to keep up with traffic n 31 4 4 128 15 8

% 79.5 10.3 10.3 84.8 9.9 5.3

Lane splitting (in moving traffic) n 32 4 3 136 13 4

% 82.1 10.3 7.7 88.9 8.5 2.6

Filtering (in stationary traffic) n 23 7 9 82 21 49

% 59.0 17.9 23.1 53.9 13.8 32.2

Riding in wet weather n 30 6 3 93 27 33

% 76.9 15.4 7.7 60.8 17.6 21.6

Riding at night* n 23 10 6 58 41 54

% 59.0 25.6 15.4 37.9 26.8 35.3

Riding with a pillion passenger* n 21 13 5 50 61 42

% 53.8 33.3 12.8 32.7 39.9 27.5

Riding in scooter/moped groups n 8 19 12 28 48 75

% 20.5 48.7 30.8 18.5 31.8 49.7

Riding among larger motorcycles* n 17 14 8 26 54 72

% 43.6 35.9 20.5 17.1 35.5 47.4

Riding in light traffic n 2 9 28 10 18 124

% 5.1 23.1 71.8 6.6 11.8 81.6

Riding in heavy/congested traffic n 16 7 16 59 26 66

% 41.0 17.9 41.0 39.1 17.2 43.7

Ride on lane edge to let traffic past n 31 3 5 120 18 13

% 79.5 7.7 12.8 79.5 11.9 8.6

Using cycle lanes to let traffic past n 31 4 4 129 13 9

% 79.5 10.3 10.3 85.4 8.6 6.0

Riding without protective clothing n 31 6 2 134 13 6

% 79.5 15.4 5.1 87.6 8.5 3.9

Struggling to climb up steep hills n 27 8 4 116 29 6

% 69.2 20.5 10.3 76.8 19.2 4.0

Accelerating rapidly down hills n 23 12 4 111 29 11

% 59.0 30.8 10.3 73.5 19.2 7.3

Braking hard so as to stop quickly n 29 7 3 98 25 28

% 74.4 17.9 7.7 64.9 16.6 18.5

Approaching intersections n 16 12 11 55 53 43

% 41.0 30.8 28.2 36.4 35.1 28.5

Riding on unfamiliar roads* n 22 10 7 45 64 43

% 56.4 25.6 17.9 29.6 42.1 28.3

*p <.05


7.3.5.2 Accessing moped and scooter safety resources

Respondents were asked where and how often they look to find information

about moped and scooter safety, results of which are presented in Table 7.15.

Scooter riders reported looking for safety-related information more frequently than

moped riders and the difference was statistically significant [ ² (4) = 9.90, p = .042].

Almost 20 percent of moped riders reported never looking for such information,

compared with about 12 percent of scooter riders. The main sources of information

reported by respondents were talking to other riders (55%), use of online forums

(46%) and scooter or motorcycle magazines (39%). Government websites were a

source of information for about one quarter of respondents.

Table 7.15 Accessing safety-related information and resources


n % n %

How often do you look for safety-related information

At least monthly 3 9.7 36 28.1

Every two or three months 3 9.7 22 17.2

Several times per year 9 29.0 36 28.1

At least once per year 10 32.3 18 14.1

Never 6 19.4 16 12.5

Valid total 31 100.0 128 100.0

Missing 8 25

Where do you look for safety-related information*

Talk to other riders 20 51.3 86 56.2

Scooter or motorcycle magazines 5 12.8 71 46.4

Online forums 14 35.9 71 46.4

Government websites 9 23.1 37 24.2

Other websites (non-government) 6 15.4 35 22.9

Newspapers 7 17.9 15 9.8

Other 3 7.7 11 7.2

Not applicable – I don’t look for such information 9 23.1 26 17.0

*Multiple responses were enabled for this question

7.3.5.3 Use of protective clothing

Questions on the use of protective clothing were included in the section on

approaches to riding and practices employed while doing so. The reported usage of

various items of clothing while riding, protective or otherwise is presented in Table

7.16 (upper body and eyes) and 7.17 (lower body and feet). Statistically significant

differences between moped and scooter riders were observed regarding use of about


half of all clothing items listed. Specifically, scooter riders were significantly more

likely than moped riders to often or always wear fabric or mesh motorcycle jackets

(p < .001), motorcycle gloves (p < .001), motorcycle boots (p = .003), Kevlar-

reinforced motorcycle jeans (p = .007), and helmet visors or goggles (p = .002).

Scooter riders were less likely than moped riders to wear (as outer clothing

items while riding) short sleeves (p = .002), shorts, skirts or dresses (p = .027), or

open shoes (thongs and sandals) (p = .039).

Table 7.16 Frequency of use of upper body clothing items while riding

Items

Frequency of use

1 = Never/rarely, 2 = Sometimes, 3 = Often/always

Moped Scooter

1 2 3 1 2 3

Fabric motorcycle jacket* n 23 5 11 28 20 103

% 59.0 12.8 28.2 18.5 13.2 68.2

Leather motorcycle jacket n 30 2 6 115 14 17

% 78.9 5.3 15.8 78.8 9.6 11.6

Other jacket (non-m’cycle) n 22 9 7 101 27 17

% 57.9 23.7 18.4 69.7 18.6 11.7

Short sleeves* n 18 5 16 103 24 23

% 46.2 12.8 41.0 68.7 16.0 15.3

Motorcycle gloves* n 18 5 14 24 14 113

% 48.6 13.5 37.8 15.9 9.3 74.8

Non-motorcycle gloves n 30 5 3 122 9 16

% 78.9 13.2 7.9 83.0 6.1 10.9

Visor/motorcycle goggles* n 9 2 26 9 3 136

% 24.3 5.4 70.3 6.1 2.0 91.9

Optical or sunglasses n 9 3 27 24 21 104

% 23.1 7.7 69.2 16.1 14.1 69.8

*p <.05


Table 7.17 Frequency of use of lower body clothing items while riding

Items

Frequency of use

1 = Never/rarely, 2 = Sometimes, 3 = Often/always

Moped Scooter

1 2 3 1 2 3

Jeans/other (non-m/cycle)

long pants n 5 12 22 26 36 91

% 12.8 30.8 56.4 17.0 23.5 59.5

Kevlar reinforced jeans* n 36 - 1 111 15 25

% 97.3 0.0 2.7 73.5 9.9 16.6

Leather motorcycle pants n 37 1 1 141 5 4

% 94.9 2.6 2.6 94.0 3.3 2.7

Other motorcycle pants n 38 - 1 122 13 14

% 97.4 0.0 2.6 81.9 8.7 9.4

Shorts, skirts or dresses* n 20 6 12 113 12 26

% 52.6 15.8 31.6 74.8 7.9 17.2

Motorcycle boots* n 35 1 2 96 12 42

% 92.1 2.6 5.3 64.0 8.0 28.0

Other boots or enclosed shoes n 7 4 28 23 21 104

% 17.9 10.3 71.8 15.5 14.2 70.3

Open shoes (thongs, sandals)* n 26 5 7 125 16 9

% 68.4 13.2 18.4 83.3 10.7 6.0

*p <.05

Further items included in the questionnaire but not presented in the above

tables covered use of high visibility clothing and wearing of office or business

clothes while riding. There were no significant differences between moped and

scooter riders on these variables. About half of moped (51%) and scooter (47%)

riders alike reported never using high visibility clothing items. Office or business

clothes were worn often or always while riding by 37 percent of moped riders and 29

percent of scooter riders.

The factors influencing decisions on what to wear while riding are presented

below in Table 7.18. Comfort, fit and ventilation were three most important factors

influencing choice of clothing, with large majorities of both moped riders (~80%)

and scooter riders (~93%) rating these as important considerations. About half of

both moped (51%) and scooter (54%) riders considered visibility (conspicuity) an

important factor influencing choice of clothing. On some other variables,

statistically significant differences were observed between moped and scooter riders.

Compared with moped riders, scooter riders placed greater importance on impact

protection (p < .001), abrasion resistance (p < .001), quality of manufacture (p =

.015) and fit (p = .020). While moped riders rated cost and appearance as slightly

more important compared to scooter riders, the differences were not statistically

significant.


Table 7.18 Rating of factors influencing choice of clothing

Factor

Rating

1 = Unimportant, 2 = Neutral, 3 = Important

Moped Scooter

1 2 3 1 2 3

Cost n 13 5 19 46 45 62

% 35.1 13.5 51.4 30.1 29.4 40.5

Appearance (fashion) n 15 12 11 71 42 40

% 39.5 31.6 28.9 46.4 27.5 26.1

Visibility (conspicuity) n 9 10 20 21 48 82

% 23.1 25.6 51.3 13.9 31.8 54.3

Brand/reputation n 19 7 11 58 38 56

% 51.4 18.9 29.7 38.2 25.0 36.8

Comfort n 2 5 31 3 8 142

% 5.3 13.2 81.6 2.0 5.2 92.8

Fit* n 1 7 30 2 8 143

% 2.6 18.4 78.9 1.3 5.2 93.5

Abrasion resistance* n 3 11 24 4 15 133

% 7.9 28.9 63.2 2.6 9.9 87.5

Impact protection* n 4 14 20 8 18 125

% 10.5 36.8 52.6 5.3 11.9 82.8

Ventilation n 2 5 31 4 7 141

% 5.3 13.2 81.6 2.6 4.6 92.8

Insulation n 3 7 28 13 17 122

% 7.9 18.4 73.7 8.6 11.2 80.3

Water resistance n 4 5 29 10 14 129

% 10.5 13.2 76.3 6.5 9.2 84.3

Quality manufacture* n 4 8 26 3 20 130

% 10.5 21.1 68.4 2.0 13.1 85.0

*p <.05

7.3.5.4 Potential future introduction of a PTW licence for moped riders

Introduction of a motorcycle licence or other PTW licence for moped riders is

among a range of countermeasures recently considered by the Queensland

Government for improving rider safety (Queensland Transport, 2008). Survey

respondents who were moped riders and did not hold a motorcycle licence were

asked what they would do in the event that such a licence was introduced. Their

responses are presented below in Table 7.19. Most respondents (79%) indicated that

they would obtain the necessary licence, though a minority of these would only do so

if rider training was not mandatory. More than one third of respondents reported that

they would obtain the required licence and change to a larger scooter (27%) or a

motorcycle (9%). Approximately 21 percent (n = 7) reported that they would stop

riding a moped. These data exclude three respondents who selected an ‘Other’


option, including two riders who already held a motorcycle licence (although the

question was not directed at these riders) and one rider who had undertaken training

but did not hold a motorcycle licence.

Table 7.19 Response to introduction of a PTW licence for moped riders

If a PTW licence was introduced for mopeds I would… Moped

n %

Get a licence if I didn’t have to do a training course 6 18.2

Get a licence even if I had to do a training course 8 24.2

Get a licence and change to a larger scooter 9 27.3

Get a licence and change to a motorcycle 3 9.1

Keep riding without a (new) licence - 0.0

Stop riding a moped 7 21.2

Valid Total 33 100.0

Missing 6

7.3.6 Crash involvement

With regard to crash involvement, a 'crash' was defined as ‘any event where

you have had a collision with any object or other road user, or have fallen from your

scooter or moped while moving’. Participants who reported crash involvement were

asked to specify how many crashes they had been involved in the previous five years,

and to report in more detail on the crash they considered most serious if involved in

multiple crashes.

Approximately 19 percent of moped riders and 31 percent of scooter riders

(30% of all respondents) reported having been involved in a crash in the last five

years (Table 7.20). The number of crashes was not specified by 78 percent of all

respondents, while one moped rider and 10 scooter riders reported having been

involved in multiple crashes. A total of eight moped crashes and 63 scooter crashes

were reported.

The number of moped crashes reported in detail (N = 7) was too low to draw

any conclusions, while the number of scooter crashes (N = 484) also limits reliability

of analysis. The difference in crash involvement between moped and scooter riders

was not statistically significant [ ² (1) = 2.25, p = .134]. Respondents reported their

own level of injury as riders (Table 7.20) and were also asked if anyone else was

4 Including only the ‘most serious’ crash if multiple crashes were reported.


injured (to their knowledge). The vast majority of self-reported moped (86%) and

scooter (62%) crashes resulted in no injury requiring professional treatment.

Admission to hospital for at least 24 hours was reported for 8.4 percent of scooter

crashes and no moped crashes. Another person was reported injured in four scooter

crashes (8.3%) and no moped crashes.

Table 7.20 Self-reported crash involvement and injury severity

Crash involvement and medical treatment Moped Scooter

n % n %

Crashed in last 5 years

Yes 7 18.9 48 31.4

No 30 81.1 105 68.6

Valid total 37 100.0 153 100.0

Missing 2 -

Total crashes reported 8 63

Level of injury sustained by rider*

No injury requiring professional treatment 6 85.7 30 62.4

Medically treated by local doctor only - 0.0 6 12.5

Treated at hospital but not admitted 1 14.3 8 16.7

Admitted to hospital for 24 hours or less - 0.0 1 2.1

Admitted to hospital for more than 24 hours - 0.0 2 4.2

Admitted to hospital ICU (intensive care) - 0.0 1 2.1

Valid total 7 100.0 48 100.0 *Riders reporting multiple crashes were asked to report on the most serious crash

Table 7.21 presents the number of vehicles involved in crashes, police

attendance at crash scenes and amount of damage to mopeds and scooters, as

reported by respondents. Approximately half of all self-reported scooter crashes

were single vehicle crashes, compared with 86 percent of moped crashes. Police

were thought not to have attended the vast majority (81.5%) of self-reported crashes.

Where involvement of other vehicles and police attendance was known (n = 52),

approximately 21 percent of multi-vehicle crashes and 11 percent of single vehicle

crashes were attended by police according to survey respondents (numbers were

insufficient for valid statistical analysis). A large proportion of mopeds (86%) and

scooters (54%) sustained only minor damage or no obvious damage.


Table 7.21 Vehicle involvement, vehicle damage and police attendance


n % n %

Was another vehicle involved?

Yes 1 14.3 23 47.9

No 6 85.7 24 50.0

Don’t know - 0.0 1 2.1

Valid total 7 100.0 48 100.0

Did police attend the crash scene?

Yes - 0.0 8 17.0

No 7 100.0 37 78.7

Don’t know - 0.0 2 4.3

Valid total 7 100.0 47 100.0

Missing - 1

Amount of damage to moped or scooter

No obvious damage 2 28.6 10 20.8

Minor repairs required 4 57.1 16 33.3

Moderate repairs required 1 14.3 11 22.9

Extensive repairs required - 0.0 3 6.3

Irreparable (write off) - 0.0 8 16.7

Valid total 7 100.0 48 100.0

The speed zones and road conditions (dry or wet) in which crashes occurred

are presented in Table 7.22. Self-reported crashes occurred predominantly in speed

zones up to 60 km/h for both mopeds (71%) and scooters (77%). Approximately

three quarters of scooter crashes and almost half of moped crashes were reported to

have occurred on dry roads.

Table 7.22 Road characteristics in self-reported crashes


n % n %

Speed zone (Km/h)

40-50 4 57.1 22 45.8

60 1 14.3 15 31.3

70-80 1 14.3 4 8.3

100-110 - 0.0 3 6.3

Not sure 1 14.3 4 8.3

Valid total 7 100.0 48 100.0

Dry or wet road

Dry 4 57.1 35 74.5

Wet 3 42.9 12 25.5

Valid total 7 100.0 47 100.0

Missing - 1


More than 80 percent of all self-reported crashes occurred on weekdays

during daylight hours (Table 7.23). All self-reported moped crashes occurred in

daylight hours, mostly during morning. About one fifth (20.1%) of scooter crashes

occurred after 6pm (twilight or night time), with the rest distributed fairly evenly

between morning (33%) and afternoon (46%).

Table 7.23 Temporal characteristics in self-reported crashes


n % n %

Weekday or weekend

Weekday 6 85.7 40 85.1

Weekend 1 14.3 7 14.9

Valid total 7 100.0 47 100.0

Missing - 1

Time of day

6am-12pm 6 85.7 16 33.4

12pm-6pm 1 14.3 22 45.8

6pm-12am - 0.0 10 20.8

12am-6am - 0.0 - 0.0

Valid total 7 100.0 48 100.0

Respondents who reported being involved in a crash were asked to describe

briefly in their own words what happened. Key words appearing in crash

descriptions were used to code crashes into the crash types presented below in Table

7.24. Single vehicle crashes in which a PTW rider lost control of the vehicle

comprised about 86 percent of moped crashes and 54 percent of scooter crashes. The

remaining 46 percent of scooter crashes and 14 percent of moped crashes appeared

mostly to involve right of way violations by other vehicle drivers.

Table 7.24 Self-reported crash description (coded)

Coded crash description Moped Scooter

n % n %

Single vehicle lost control 5 71.4 21 43.8

Single vehicle lost control avoiding OV* 1 14.3 5 10.4

Two vehicle – OV* failed to give way 1 14.3 6 12.5

Two vehicle – OV* rear-ended scooter - 0.0 9 18.8

Two vehicle – Scooter rear-ended OV* - 0.0 2 4.2

Two vehicle – aggressive OV* driver - 0.0 2 4.2

No memory/no comment - 0.0 3 6.3

Valid total 7 100.0 48 100.0 *Other vehicle


The licence characteristics of respondents who reported crash involvement

are presented below in Table 7.25. It must be noted that these were the licence

characteristics at the time of survey completion, which may differ from the licence

characteristics at the time of the reported crash. All crashed moped riders held an

open car licence valid for moped riding and one also held a probationary motorcycle

licence. An open motorcycle licence was held by two thirds of crashed scooter

riders. Probationary and learner motorcycle licences were held by about six percent

and 19 percent of crashed scooter riders, while eight percent reported holding no

valid licence for LC scooter riding.

Table 7.25 Licence characteristics of crash-involved riders

Licences Moped Scooter

n % n %

Car licence held

Learner - 0.0 3 6.3

Probationary (P1 or P2) - 0.0 1 2.1

Open 7 100.0 44 91.7

None - 0.0 - 0.0

Valid total 7 100.0 48 100.0

Motorcycle licence held

Learner - 0.0 3 6.3

Probationary (RE or RE A) 1 14.3 9 18.9

Open (R or R A) - 0.0 32 66.7

None 6 85.7 4 8.3

Valid total 7 100.0 48 100.0

The age, gender and training involvement of crash-involved riders are

presented in Table 7.26. The age distribution of crash-involved riders generally

reflected that of respondents as a whole. However, no young moped riders (aged 16-

24) reported a crash, while young scooter riders represented 8.3 percent of crash-

involved riders compared with 5.2 percent of participants. Approximately two thirds

(67%) of crash–involved scooter riders were male, again generally reflecting the

characteristics of respondents (73% male). The proportion of crash-involved scooter

riders who had not undertaken rider training (25%) was similar to the proportion of

untrained scooter riders among the entire sample (28%).


Table 7.26 Age, gender and training involvement of crash-involved riders


n % n %

Age group

16-24 - 0.0 4 8.3

25-39 3 42.8 21 43.7

40-49 2 28.6 7 14.6

50 or over 2 28.6 16 33.3

Valid total 7 100.0 48 100.0

Gender

Male 4 57.1 32 66.7

Female 3 42.9 16 33.3

Valid total 7 100.0 48 100.0

Training undertaken

None 6 85.7 12 25.0

Q-Ride pre-licence training 1 14.3 18 37.5

Other pre-licence training - 0.0 4 8.3

Post-licence training - 0.0 9 18.8

Pre- and post-licence training - 0.0 5 10.4

Valid total 7 100.0 48 100.0

The data on crash involvement and annual distance travelled (exposure)

provided an opportunity to calculate self-reported crash rates per million vehicle

kilometres travelled (VKT) for mopeds and scooters. The exposure data also

allowed an estimate of police-reported crash rates per million VKT for mopeds using

crash and registration data from Study 2. As noted in Study 2, estimation of police-

reported scooter crash rates is not possible because of the lack of registration data to

scale survey data up to the state-wide level of the crash data (scooter registrations are

not separated from motorcycle registrations).

The crash rates based on self-reported survey data are presented below in

Tables 7.27 and 7.28. The self-reported crash rates per million VKT were similar for

mopeds (12.9) and scooters (11.5) based on a mean distance travelled. Using a

median distance travelled, the crash rate per million VKT was slightly higher for

scooters (16.5) than for mopeds (13.7). For mopeds only, combining Study 2 crash

and registration data with exposure data reported in the current study, the police-

reported crash rate was 4.2 per million VKT.


Table 7.27 Self-reported crash rates per million vehicle kilometres travelled

Statistic Moped Scooter

Number of respondents 39 153

Mean VKT per year* 3,188 7,186

Total VKT for 5 years 621,660 5,497,290

Total crashes for 5 years 8 63

Crashes per million VKT (based on mean) 12.9 11.5

Median VKT per year* 3,000 5,000

Total VKT for 5 years 585,000 3,825,000

Total crashes for 5 years 8 63

Crashes per million VKT (based on median) 13.7 16.5 *Excludes missing data and 1 moped outlier with claimed annual distance of 56,000 km

Table 7.28 Police-reported crash rates per million vehicle kilometres travelled

Statistic Moped

Registrations 40,569

Mean VKT/year 3,188

Total moped kilometres for 5 years 129,333,972

Crashes 541

Crashes per million VKT (based on mean) 4.183

7.3.7 Comments on moped and scooter safety and transport planning

Respondents were invited to comment on aspects of moped and scooter use

which concerned them in regard to safety and transport planning. Comments were

made by 65 percent of respondents (n = 124), including approximately 61 percent of

moped riders and 68 percent of scooter riders. Of those who responded to this

question, 26 percent (n = 32) were concerned about lack of awareness of PTWs and

carelessness among other vehicle drivers. Provision of parking and access to parking

to parking areas was raised by 16 percent (n = 20) of respondents, who generally

thought that more parking should be provided for PTWs. Some respondents

suggested that difficult access to parking areas was potentially hazardous for riders.

Approximately 15 percent (n = 19) of respondents suggested that moped

riders should be required to undergo training and/or hold a motorcycle or other PTW

licence. These were predominantly scooter riders who already held a motorcycle

licence, with the exception of one moped rider who held a motorcycle learner

licence. Smaller proportions of respondents to this question also perceived poor road

conditions (8%), non-use of protective clothing (6.5%), risky riding behaviour (6%)


and low maximum speed of mopeds (5%) to be hazardous for riders.

7.4 Discussion

The specific aims of this thesis were to develop a better understanding of

moped and scooter usage trends and patterns, and to investigate factors leading to

differences in moped, scooter and motorcycle safety. As Study 3b, the Queensland

Scooter and Moped Rider Survey 2010 was designed to assist this examination of

moped and scooter use and the factors which influence rider safety, building on

Studies 1, 2 and 3a. As with Study 3a involving focus groups, the current study

addressed all four research questions. For Research questions 1 and 2, Study 3b

explored motivations for moped and scooter use, as well as similarities and

differences in usage of the PTW types. Specifically, Study 3b examined self-

reported travel patterns, including distance travelled, frequency of riding, roadway

types used, trip purpose and other information on usage. Study 3b addressed

Research questions 3 and 4 by examining a wide range of safety-related issues that

might impact the overall and relative safety moped and scooter riders. These issues

include motivations for riding, knowledge, beliefs and opinions relevant for safety,

riding skills and behaviour, experience, licensing and training, and crash

involvement. .

The overall response to the Survey was lower than expected and this was

particularly so in regard to moped riders. This occurred despite a range of strategies

being implemented to advertise and encourage participation, including online

promotion, print media articles in major publications and distribution of

approximately 500 flyers at numerous locations. While Study 1 found that mopeds

significantly outnumbered scooters in inner city Brisbane (where several hundred

flyers were distributed), less than one quarter of respondents to the current survey

identified themselves as moped riders. However, while this represents a limitation in

some ways, it is argued here that this may also represent an important finding in

itself. Specifically, moped riders in particular appear difficult to recruit for

participation in road safety research, which raises questions regarding their general

interest in rider safety.

The low participation of moped riders in particular suggests that many are

generally uninterested in engaging in communication about safety issues. Several


findings of the survey support this contention. Among these differences were levels

of participation in scooter or motorcycle clubs and online forums, which was much

more likely for scooter riders (45.4%) than moped riders (20.5%). It is likely that a

substantial proportion of all survey participants were recruited through the two online

forums in which the study was advertised, and these participants were most likely

scooter enthusiasts. This appears to partly explain the relatively low participation of

moped riders and suggests a bias in the sample toward scooter enthusiasts. Scooter

enthusiasts and recreational riders may therefore be overrepresented due to the

methods used to recruit participants. Some research suggests that recreational riders

may have higher average crash risk than commuting riders, and that their crashes are

more likely to involve only one vehicle (Broughton & Walker, 2009; Harrison &

Christie, 2003; Blackman, Cheffins et al., 2009). Somewhat paradoxically, the

current study suggests that enthusiasts and recreational riders may also be more

attentive with regard to safety issues and may be more active in risk management.

7.4.1 PTW usage

In the current study, moped and scooter riders differed in some aspects of

PTW usage, including travel patterns and some demographic characteristics, yet they

were similar in other ways. Moped and scooter riders in the current study each seem

to differ somewhat from motorcycle riders as described in other research, particularly

in terms of age, gender and riding experience. However, it has been noted previously

(Haworth & Rowden, 2010) that PTW riders on the whole are a heterogeneous group

and that it is therefore difficult to generalise about their characteristics and

behaviours. The current study leads to a similar conclusion about moped and scooter

riders overall, though their main motivations for PTW use are clearly similar. In

Study 3a, when asked about their motivations for scooter or moped riding, responses

invariably included some combination of cost, practicality, ease of use, time

efficiency and enjoyment. Approximately 85 percent of participants were regular

commuters. The findings of Study 3b regarding usage were largely consistent with

those of Study 3a, though enjoyment was significantly less important for moped than

scooter riders, perhaps reflecting a bias in the sample toward scooter riding

enthusiasts. More than 80 percent of moped and scooter riders alike considered

parking availability, practicality and ease of use to be an important motivating factor.


Licensing regulations were a stronger motivating factor for moped use than

scooter use, suggesting that the lack of a motorcycle licence requirement encourages

moped use. The inexperience of moped riders and (to a lesser degree) scooter riders

in the current study relative to motorcyclists in other research likely reflects the

increased use of mopeds and scooters in Queensland in recent years, as discussed

further below.

The findings of this study regarding distances travelled show that mopeds

travel considerably less on average than scooters, with mean annual distance

travelled of approximately 3,200 and 7,200 km respectively. This is consistent with

the findings of Harrison and Christie (2006). Although mopeds and scooters were

reported to travel 15 to 20 percent further annually in the current study than in

Harrison and Christie (2006), both studies found that mopeds travelled considerably

less distance than scooters. Harrison and Christie (2006) were able to obtain

odometer readings over 12 months from a sample including 145 moped owners,

which should arguably produce more reliable results than the current study which

relied on simple estimates from a smaller sample.

7.4.2 Crash involvement

The current study collected information about a small number of moped and

scooter crashes and the statistical power of comparisons was therefore constrained.

The slightly greater proportion of scooter riders than moped riders who had crashed

in the last five years (19% versus 31%) may be influenced by the smaller proportion

of moped riders who had been riding for more than five years. While Study 2 found

moped and scooter crashes to be less severe than motorcycle crashes, 45 percent of

police-reported moped crashes resulted in hospitalisation. By contrast, about 22

percent of moped and scooter crashes reported in the Survey resulted in hospital

treatment (including non-admissions). Police were thought to have attended none of

the moped crashes and approximately 17 percent of scooter crashes reported by

survey respondents (or 15% of all crashes). While reliability is limited due to the

low number of crash-involved survey respondents, the proportion of moped and

scooter crashes which are reported by police is likely less than one fifth of all crashes

which occur.


Underreporting of crashes by police appeared to be more prevalent with

single vehicle crashes than those involving more than one vehicle. The majority of

moped (86%) and scooter (54%) crashes were single vehicle crashes, although five

(10%) scooter riders attributed crash causation to other vehicles. In contrast to the

current study, analysis of police-reported crash data in Study 2 found that 29 percent

of moped and 21 percent of scooter crashes involved only one vehicle. As noted

previously, the analysis of self-reported crashes in the current study is limited by low

numbers, particularly for mopeds. However, the higher rate of single vehicle crashes

compared to Study 2 may reflect some bias in both the survey sample and the police-

reported data, including that single vehicle crashes are less likely to be reported to or

by police.

For moped and scooter crashes combined (N = 55), most were reported to

have resulted from loss of control by riders (58%), or other vehicles failing to give

way (29%, including rear-end crashes). These general crash characteristics are

typical of those found in Study 2 and in other research examining PTW crashes,

although the overall findings on crash involvement in the current study must be

viewed with caution due to low numbers and potential self-report bias.



identified in previous research (Greig, Haworth et al., 2007): inexperience or lack of

recent experience; risk taking; driver failure to see motorcyclists; instability and

braking difficulties; road surface and environmental hazards; and vulnerability to

injury. As in the previous chapter, the following sections of this chapter are

structured according to these main contributors to crash and injury risk with slight

modification of the original terminology in the section titles to ensure coverage of all

relevant issues. Specifically, driver failure to see motorcyclists is modified to other

road users, and instability and braking difficulties is modified to PTW control and

riding skills. The survey found that moped riders differ significantly from scooter

riders in some ways with regard to safety and risk management.



Previous research has found elevated crash risks among inexperienced riders,

those lacking recent experience, and those inexperienced with a particular PTW type

(ACEM, 2008a; Haworth, Smith et al., 1997; Mullin, Jackson et al., 2000; Rutter &

Quine, 1996). In Study 3a, safety awareness and adoption of safe riding practices

appeared to increase with age and experience, and the older and more experienced

participants tended to be scooter riders who held a motorcycle licence. Study 3a also

suggested that moped riders were largely uninterested in sourcing information on

PTW safety as a means to compensate for inexperience. These findings indicated

that the issue of inexperience should be further explored in the survey of moped and

scooter riders in Study 3b. Consequently, the current study examined the amount of

riding experience, frequency of moped or scooter use, distances travelled and

participation in rider training. The study also examined access of information

sources relating to PTW safety.

With scooter riders approximately six years older on average than moped

riders in the current study, it is unsurprising that they were also slightly more

experienced riders. However, the vast majority of moped and scooter riders alike

had been riding mopeds or scooters for no more than five years and more than 40

percent had no more than two years riding experience. This likely reflects the

increase in moped and scooter sales and usage observed in recent years that was also

evident in Study 2. However, the findings are also consistent with those found five

years earlier (Harrison & Christie, 2006), which indicates that increased usage has

been largely driven by new riders taking up moped and scooter use over the last

decade.

The amount of riding experience as measured by the number of years riding

is not a true measure of exposure and the distances travelled over a given timeframe

must also be considered. The lower mean kilometres travelled annually by moped

riders than scooter and also motorcycle riders indicates that moped riders will

accumulate less experience over a given timeframe than riders of other PTW types.

The findings on years of riding mopeds and scooters do not reveal all PTW

riding experience. A third of moped riders reported motorcycle riding experience

and over half of those had gained such experience in the last five years. A greater

proportion of scooter riders had ridden motorcycles (63%), with over half of those


having first done so more than twenty years ago. While it is not known how much

motorcycle riding experience was held in total, it seems likely that scooter riders

would hold more experience than moped riders in riding other PTW types, as well as

mopeds or scooters.

One potential means to compensate for inexperience is participation in rider

training. In Study 3a, participants who had undertaken rider training (mostly scooter

riders) tended to value it highly, while those who had not generally thought it

unnecessary (mostly moped riders). The greater engagement of scooter riders than

moped riders in training may be driven largely by the motorcycle licence

requirement which guides the former toward participation in pre-licence training,

which half of scooter riders in the current study had undertaken. While moped rider

training was generally supported as a potential intervention, some survey respondents

were strongly opposed to a mandatory PTW licence for moped riders.

Focus group participants in Study 3a seemed to derive their safety awareness

largely from personal experience, which can only be gained through riding, as well

as through talking to other riders. This finding was consistent with other research

with motorcycle riders (Natalier, 2001), and was strongly supported by findings of

the current study. Talking to other riders (including use of online forums) was the

primary source of information for moped and scooter riders alike. That scooter

riders were more likely than moped riders to seek safety-related information possibly

reflects a bias in the sample toward scooter riding enthusiasts. It may also reflect

some influence of the licensing and training system, through which they have been

exposed to safety-related information.

7.4.3.2 Risk taking

As noted in the previous chapter, a strong propensity for risky riding

behaviours was not generally evident among Study 3a focus group participants, and

some participants were arguably risk-averse. With risk taking more prevalent among

young, male and recreational riders according to other research, this finding in Study

3a may relate to the age, gender and motivations of participants for riding. The mean

ages of focus group participants were 31 and 44 years for moped and scooter riders

respectively, almost one third were female, and the vast majority were commuters

who did not ride for recreational purposes. On these three characteristics, survey


respondents in Study 3b were similar, with mean ages of 37 and 44 years for moped

and scooter riders respectively, over one quarter female, and recreation accounting

for 12 percent of riding purpose on average. On the basis of age, gender and

motivations for riding, a strong propensity for risk taking appears unlikely in the

current survey sample overall. This is possibly supported by the finding that no

moped riders and only two percent of scooter riders reported licence suspension or

cancellation in the last five years. However, 4.6 percent of scooter riders reported

holding no valid licence for scooter riding.

The finding in Study 2 that six percent of crashed moped riders were

effectively unlicensed suggests a higher crash risk among unlicensed riders, as found

in other research. In the current study, of respondents who reported a crash in the

past five years, eight percent of scooter riders reported holding no valid licence for

LC scooter riding at the time of survey completion. This is higher than the

proportion of all participating scooter riders who reported holding no valid licence

(4.6%). While the number of crash-involved scooter riders was low (48), the data

suggest that unlicensed riders may be overrepresented in crashes, a finding that is

consistent with other research (Haworth, Smith et al., 1997).

As noted in section 7.2 (Study design and methods), Study 3b did not aim to

specifically explore illegal riding behaviour and direct questions about personal

engagement in illegal riding were therefore avoided. As a result, the Survey

contained only limited questions concerning risk taking (although risky riding

behaviours are not necessarily illegal). However, the questionnaire did include

specific questions about perceived risk associated with a range of riding scenarios,

including speeding, lane splitting and filtering through traffic, lane positioning and

use of bicycle lanes.

In Study 3a many participants felt that an ability to keep up with traffic flows

was important, even if this meant slightly exceeding speed limits sometimes.

Consistent with this, in Study 3b only 23 percent of moped and scooter riders alike

considered this to be risky. In both studies, several participants argued that speed

restrictions on mopeds should be increased to 60-65 km/h to allow them to keep up

with traffic. However, rapidly accelerating down hills was considered risky by 59

percent of moped riders compared with 73 percent of scooter riders. This finding is

interesting in that the situation represents one where mopeds may easily exceed their

maximum permitted speed of 50 km/h, yet moped riders found it less risky than


scooter riders (although the difference was not statistically significant). In Study 2,

while speed was attributed to riders in only a small proportion of moped crashes

(2%), it was not attributed to scooter riders in any crash.

Lane splitting in moving traffic was widely regarded as risky in Study 3b,

with 82 and 89 percent of moped and scooter riders respectively rating the behaviour

as risky. Filtering in stationary traffic was seen as somewhat safer, though 59 and 54

percent of moped and scooter riders respectively still rated the behaviour as risky.

Other behaviours rated as risky by approximately 80 percent of moped and scooter

riders alike were riding in the far left of lanes to let traffic past, and riding in bicycle

lanes. All of these findings are generally consistent with those of Study 3a.

Survey participants were asked to rate the perceived risk associated with

these behaviours and scenarios, but were not asked about whether or not they

engaged in them. It remains possible if not likely that some riders engaged lane

splitting, filtering and speeding despite rating those behaviours as risky. These

questions may also have drawn some socially desirable responses.

7.4.3.3 Other road users

Driver failure to see PTW riders has been identified as a main contributor to

PTW crash and injury risk in previous research (Comelli, Morandi et al., 2008;

Huang & Preston, 2004; Ivers, Wells, Blows, Liu, Stevenson, Lo, & Norton, 2003).

In Study 3a, participants universally perceived other road users, in particular larger

vehicles, as the primary hazard and threat to their safety. The extent of active

management of this general risk factor, through defensive riding techniques and

maximising conspicuity in particular, was varied among focus group participants.

Issues regarding other road users, including defensive riding techniques and PTW

conspicuity, were therefore explored further in the current study.

There was general agreement in the current study that the behaviour and

awareness of other vehicle drivers is a key concern for PTW riders, with numerous

survey respondents commenting on this particular issue. Specific comments related

to not being seen by other road users, while others referred to a lack of care among

other vehicle drivers, as well as impatience and aggressiveness. Some respondents

advocated separating PTWs from other motorised traffic, including allowing mopeds

to share bicycle lanes (though this was not supported by the majority of participants).


The Study 3b findings were largely consistent with those of Study 3a, with the risk

associated with other road users widely acknowledged, and a wide range of views on

potential countermeasures.

Roughly 40 percent of moped and scooter riders alike considered riding in

heavy traffic to be risky, while riding in light traffic was considered risky by about

six percent of respondents. This reflects the perceived risk that riders associate with

other vehicles, a perception that appears valid in light of research examining multi-

vehicle PTW crashes. As a means to increase their conspicuity and lessen the risk of

not being seen by other road users, about half of respondents considered the use of

high visibility clothing to be important. Some participants in Study 3a suggested that

there is little value in bright or reflective clothing as drivers often simply fail to look.

This may be reflected in the current study where visibility was considered

unimportant by 21 percent and 14 percent of moped and scooter riders respectively.


As noted in previous chapters, PTW control requires greater skills than car



difficulties in other research (Greig, Haworth et al., 2007). While the ability to

control a PTW is clearly important, it is also argued that effective hazard perception

and response is a necessary skill for rider safety. In Study 3a, focus groups provided

qualitative data suggesting that moped riders may be at somewhat greater risk than

scooter riders due to some combination of poorer braking skills, hazard perception

and response and road positioning. Study 3b has provided quantitative data which

generally supports these findings.

PTW control and riding skills could not be measured objectively in the

current study, but the Survey did provide useful information on rider perceptions of

their own skills and abilities. Moped and scooter riders were similar in their self-

rated level of riding skill, with over 90 percent in both groups rating themselves

competent, advanced or expert riders. Slightly higher proportions of scooter riders

considered themselves advanced or expert, but there was no statistically significant

difference. This finding is generally consistent with other research reporting on self-

rated driving and riding skill.


It was suggested in Study 3a that the reluctance of some untrained riders to

use front brakes is of concern given that front brakes provide most of the potential

stopping power of PTWs, including mopeds and scooters (Broughton & Walker,

2009; Corno, Savaresi et al., 2008). Study 2 also provided some indication of poorer

vehicle control skills among moped riders compared with scooter riders, which may

have related in part to braking skills. Confidence with brake application was

therefore included as a specific question in Study 3b, with results suggesting greater

confidence in braking among scooter riders than moped riders.

In addition to braking, scooter riders were more confident than moped riders

in riding on unfamiliar roads, riding in wet weather and riding at night. The finding

regarding wet weather is supported by Study 2 which found that moped riders were

more likely to crash on wet roads than scoter riders. The risks associated with these

scenarios are arguably related to hazard perception and response as well as the ability

to control a PTW.

In Study 3a, scooter riders holding motorcycle licences referred to rider

training as a valuable source of information and skills development. Moped riders

in Study 3a generally expressed the view that they do not need rider training or

education as they do not travel at high speed. These beliefs were strongly reflected

in the findings of Study 3b. When invited to comment on any issue relevant to

moped or scooter use and safety, 12 percent of scooter riders suggested that moped

riders should be required to undergo training and/or hold a motorcycle or other PTW

licence. When specifically asked, less than one quarter of moped riders reported that

they would undertake ride training if required for moped riding. As noted

previously, training would be expected to improve the skills of some riders, but the

extent to which it might actually result in safer riding remains unclear.


As mentioned previously, the literature suggests that mopeds and scooters

may be more susceptible than motorcycles to hazards such as potholes and rough

surfaces, due to smaller wheel diameters, less advanced braking systems and limited

suspension capabilities. In Study 3a, poor road surfaces appeared to be the greatest

perceived hazard after other vehicles. Study 3a did not identify any clear differences

between moped and scooter riders regarding road surface and environmental hazards.


However, it was tentatively concluded in Study 2 that poor road conditions and wet

road surfaces present greater problems for moped riders than scooter riders

(comparison with motorcyclists was confounded by different usage patterns). The

topic of road surface and environmental hazards was therefore further explored in

Study 3b.

A particular hazard identified in Study 3a but not in the literature was that of

poor access to some designated parking areas in inner city Brisbane. Having to

climb gutters and ride along footpaths or sidewalks (technically an offence in the

study area), some participants noted that this was hazardous to riders and pedestrians

alike. This issue was not addressed in specific questions in the current study, but

several survey participants mentioned this as a concern when invited to comment on

any issue relating to moped and scooter use and safety. As suggested in Study 3a,

this particular issue may warrant further exploration.


The greater vulnerability of PTW riders compared to car and other vehicle

occupants is well documented in the literature and the use and characteristics of

protective clothing have been identified as important for rider safety (de Rome, Ivers,

Fitzharris, Du, Haworth, Heritier, & Richardson, 2011; de Rome & Stanford, 2006).

Although of little benefit in high impact crashes, protective clothing is known to

reduce the severity of non-fatal injuries.

In Study 3a, some focus group participants took active steps to reduce their

injury risk through use of protective clothing, while others did not do so.

Appearance and image seemed to be an important factor for many participants in

deciding what to wear while riding. The warm climate in Queensland also appeared

to discourage some riders from using protective clothing. In Study 3a, while

protective clothing use was more prevalent among scooter than moped riders, no

conclusions could be drawn due to the small number of participants. Use of

protective clothing was therefore explored further in the current study.

The current Study has found scooter riders to value protective clothing more

highly, and to use it more often, compared to moped riders. Scooter riders were

significantly more likely than moped riders to wear a motorcycle jacket, gloves,

reinforced jeans and motorcycle boots. Scooter riders were significantly less likely


to wear short sleeves, open shoes (including thongs and sandals) and/or shorts, skirts

or dresses. Additionally, key factors in protective clothing characteristics, including

fit, abrasion resistance, impact protection and quality of manufacture were

significantly more likely to be rated as important by scooter riders than moped riders.

Other research has also found lower rates of protective clothing use among

moped and scooter riders in comparison to motorcyclists (de Rome, Stanford, &

Wood, 2004). In the current study, while use and knowledge of protective clothing

was more apparent among scooter riders than moped riders, not all scooter riders

seemed to consider it of high importance. The differences observed between moped

and scooter riders regarding protective clothing may relate to differences in

knowledge and awareness, with moped riders in this sample less experienced, less

likely to have undertaken training and less likely to seek road safety information. It

is also possible that moped riders perceive a lower injury risk (though not necessarily

a lower crash risk) compared to scooter riders due to the (50 km/h) speed restrictions

applied to mopeds. The view that protective clothing was less important for moped

than scooter or motorcycle riders was expressed by some focus group participants in

Study 3a.


Following on from Study 3a, the current study assisted in answering the four

research questions that are central to the overall program of research. Study 3

excluded motorcycle-only riders as the aim was to focus on comparing moped and

scooter use and safety. As with Study 3a, the current study suggested important

differences between moped and scooter riders with regard to safety and also usage.

Research question 1: Why has moped and scooter usage increased? An

increase in moped and scooter usage was evident in Study 2, continuing a trend

observed for mopeds in the study area from 2001 to 2005 (Haworth & Nielson,

2008). According to focus group participants in Study 3a, mopeds and scooters

provided low cost and convenient mobility compared to car use and public transport.

In particular, traffic congestion, availability of parking and low cost were the major

motivating factors. In the current study, these factors were rated as important by

more than three quarters of survey respondents. That large proportions of moped and


scooter riders (81% and 73% respectively) had less than five years riding experience

suggests that alternative transport modes have become comparatively less attractive

in recent years, stimulating increased moped and scooter use. With most moped and

scooter riders alike aged 30 years or older (as in Study 2), and most respondents

earning average or above average incomes, it does not appear that mopeds and

scooters appeal more to young people or to those on low incomes. Study 3a also

found that licensing requirements for moped riding may have encouraged moped use,

particularly as other motivating factors became more important to participants.

Study 3b supports this finding, with 56 percent of moped riders considering licensing

regulations an important motivating factor. However, only 15 percent of moped

riders indicated that they would stop riding a moped (or other PTW) if a PTW

licence was introduced for moped riding.


motorcycles differ? As suggested in Study 3a, the current study found similarities in

the use of mopeds and scooters in terms of where and when they were used, and for

what purpose. Study 3 excluded motorcycle-only riders as the aim was to focus on

comparing moped and scooter use and safety. Mopeds and scooters alike were used

primarily for commuting, on weekdays, and in lower speed zones. Scooters were

used in higher speed zones more frequently than mopeds, yet only 15 percent of

reported scooter usage occurred in 100-110 km/h zones. The greater use of scooters

for recreation suggested in Study 3a was not reflected in the current survey findings.

Although weekend use accounted for more scooter than moped use, shopping (much

of which may have occurred on weekends) accounted for a greater proportion of

scooter use than did recreational riding.

Three quarters of respondents were male, with no difference in PTW type by

gender. As suggested in Study 3a, in the current study moped riders were generally

younger and less experienced than scooter riders, though only 10 percent of moped

riders were under 25 years of age. More than three quarters of survey respondents

were Brisbane region residents, with no difference between PTW types, but this

result may be heavily influenced by sampling bias (Study 2 suggested different usage

patterns in terms of location, with statistically significant differences between PTW

types). Scooter riders tended to travel roughly twice the distance of moped riders on

average, both annually in single journeys. This is consistent with previous research


findings in the study area (Harrison & Christie, 2006). Scooter or motorcycle club

membership and use of online forums was more common among scooter riders than

moped riders, suggesting possible bias in the sample toward scooter riders who are

enthusiasts.


motorcycles differ? Given the findings of Study 2 and Study 3a, it might be expected

that a lower crash rate would be observed among scooter riders than moped riders,

but this was not the case for self-reported crashes in the current study. It is possible

given the small sample size that the self-reported crash involvement was less

representative of riders as whole than some of the other characteristics reported by

respondents. In other words, an inclination to safer behaviour among scooter riders

compared with moped riders is generally evident in Studies 2 and 3, including in

specific aspects of the current study. The notable exception is the greater crash

involvement of scooter riders in Study 3b, but this particular result is unreliable in

the small survey sample. Unreliability of the result notwithstanding, it is the only

finding in the overall program of research that conflicts with the bulk of other

evidence, warranting further research into comparative PTW crash rates (as noted in

Chapter Five, a lack of registration data for scooters prevented estimation of police-

reported scooter crash rates in Study 2). Nonetheless, while the current study has not

definitively answered research question 3, it has provided strong support for an

overall conclusion that scooter riders are generally safer than moped riders.


motorcycles differ? In the qualitative component of Study 3 involving 23 focus

group participants, Study 3a found that scooter riders showed greater safety

awareness and invested more in safety than moped riders. Possible explanations for

this included more riding experience among scooter riders, greater knowledge about

vehicle handling and performance, and the requirement for them to hold a

motorcycle licence (which may have exposed them to rider training and education).

Following on from Study 3a, the survey results showed that compared to moped

riders, scooter riders are older and more experienced on average, access safety-

related information more often and are more likely to use protective clothing. The

current study also indicated that most scooter riders (72%) do undertake some rider


training while most moped riders (69%) do not. Additionally, some scooter riders

suggested that moped riders should be required to undertake some rider training

and/or hold a motorcycle or other PTW licence. Inability to keep up with

surrounding traffic was considered risky by most participants and some moped riders

suggested that this could be addressed by raising the maximum permitted moped

speed from 50 km/h to 60 km/h. All of these findings are consistent with those of

Study 3a.

7.4.5 Limitations

There are inherent limitations with regard to the reliability of self-reported

data and these limitations are typical of survey questionnaires such as used for this

study. The detail and depth of information gathered in the Survey was also limited

by consideration of the estimated time that participants would be willing to spend

completing it.

The overall response to the survey was lower than expected and the small

number of participants therefore limited the power available for statistical analysis.

This was particularly so regarding the limited number of responses from moped

riders, although it is argued above that this limitation may also represent an

important finding in itself (that moped riders are uninterested in safety issues). It

appears that the sample was somewhat biased toward scooter enthusiasts.

Of respondents who identified themselves as scooter riders (N = 153), 8.3

percent reported engine cylinder capacities of 50cc or less. At least some of these

participants were probably moped riders, though this could not be reliably

determined without more detailed information on vehicle specifications.

The Survey did not include a question asking where or how participants

found out about the Survey. On reflection, such a question would have enabled

evaluation of recruitment methods and thus been useful for future research.

However, it appears from other information gathered that neither print media articles

nor flyer distribution were highly successful recruitment methods. A large

proportion of respondents were members of scooter or motorcycle clubs and/or users

of online forums (on two of which the survey was advertised), suggesting that the

sample may be somewhat biased toward enthusiasts and not representative of all

moped and scooter riders.


Participation in the Survey was limited to regular moped or scooter riders in

the State of Queensland. The findings are most relevant to other jurisdictions where

moped riding is permitted on a car licence, and where moped and scooter usage has

increased substantially from a low base.

7.5 Chapter Seven summary

This chapter has described Study 3b, in which a questionnaire survey

instrument was used to gather information from riders on moped and scooter usage

and safety. The development of Study 3b was guided by focus group findings

obtained in Study 3a and by previous research and literature presented in Chapter

Two. The study assisted in answering all four research questions.

The findings of Study 3b generally aligned in many respects with those in

Study 3a (and to some extent Study 2). While participants had similar usage

patterns, common motivations and perceived other vehicles as a major hazard,

moped riders appeared somewhat less safe than scooter riders. In both studies,

moped riders on the whole were younger and less experienced, less likely to have

undertaken rider training, to use or value protective clothing, or access safety

resources compared to scooter riders.

Strong comparative conclusions were precluded by a small sample size,

highlighting the difficulties in recruiting survey participants. This result in itself

suggests that many moped and scooter riders, and particularly moped riders, may

have little interest in PTW safety issues.

The next chapter (Chapter Eight) concludes this thesis by discussing

collectively the rationale, results and implications of the four studies conducted in the

overall program of research. The key similarities and differences between the three

PTW types and their riders with regard to safety and related issues are highlighted,

and the implications considered. Potential measures to improve rider safety are

discussed, limitations of the research are described and the chapter concludes by

identifying issues to be addressed in future research.


CHAPTER 8: DISCUSSION

8.1 Introduction

This thesis examined the use of mopeds and motor scooters and the factors

which influence rider safety. As mopeds and scooters have historically comprised

only a small proportion of road traffic in Australia, they have not been a major

concern there for PTW safety research. A relatively recent increase in moped and

scooter use in Australia, and Queensland in particular, has generated increased

concern about their safety compared to motorcycles, and also compared to each

other. Despite some improvements in rider safety relative to exposure in recent

decades, increased PTW usage has been accompanied by increases in PTW rider

deaths and injuries. As the vulnerability of PTW riders in crashes has long been

recognised, there is a considerable body of research literature on the safety and usage

of PTWs in developed countries. However, most research into PTW safety has

focused on motorcycles and has not addressed moped and scooter use in depth.

Further, the bulk of the research that has focused on moped and scooter safety

originates from Europe where these vehicles have been traditionally popular. Some

of this research is relevant to the Australian context, but socioeconomic, cultural,

legislative and environmental differences between Australia and elsewhere limit the

transferability of findings.

The relevant literature identified gaps in knowledge of the safety of moped

and scooter use in Australia, and this research was conducted with two main aims:

RA1. To develop better knowledge and understanding of moped and scooter usage

trends and patterns.

RA2. To determine the factors leading to differences in moped, scooter and

motorcycle safety.

It was expected that achieving these aims would identify potential areas for

improving safety for moped and scooter riders, thereby informing relevant

countermeasures. Four research questions were developed to inform these aims to

guide the program of research:


RQ1: Why has moped and scooter usage increased?

RQ2: How does the usage of mopeds, scooters and motorcycles differ?

RQ3: How does the safety of mopeds, scooters and motorcycles differ?

RQ4: Why does the usage of mopeds, scooters and motorcycles differ?

Three studies were designed to answer the research questions and thereby

achieve the overall research aims. Study 1 involved an observation of PTW types in

an inner city area, measuring trends over a period of two years, addressing research

questions 1, 2 and (to a limited extent) 4. An analysis of Transport Department crash

and registration data was conducted in Study 2, covering a period of five years,

addressing research questions 2, 3, 4 and (to a limited extent) 1. Study 3 involved an

exploration of moped and scooter rider characteristics, behaviours and experiences,

addressing all four research questions. Study 3a was a qualitative study involving

focus groups with moped and scooter riders. A scooter and moped rider survey was

developed and administered for Study 3b to examine and quantify the issues raised in

the focus groups across a wider sample.

8.2 Review of findings

8.2.1 RQ1: Why has moped and scooter usage increased?

This program of research explored reasons for the observed increase in

moped and scooter usage, seeking to develop better knowledge and understanding of

moped and scooter usage trends and patterns. There has been a substantial increase

in PTW use in the study area over the last decade, with moped use increasing at a

faster rate than motorcycle use. All Queensland PTW registrations doubled between

2001 and 2009 (from 77,274 to 155,220), while mopeds increased as a proportion of

Queensland registered PTWs from 1.2 percent to 8.8 percent over the same period.

The relatively recent increase in moped and scooter usage is reflected in the younger

mean age of mopeds (3.5 years) and scooters (3 years) compared to motorcycles (7

years) observed in Study 1.


Increased traffic congestion and increasing pressure on transport-related

infrastructure and services has resulted from sustained population growth in many

cities, including Brisbane. In particular, the availability of parking is constrained

while its cost has increased, as have travel times for many private vehicle users.

Additionally, rising fuel costs have also contributed to increases in the overall cost of

private motor vehicle use. This common pattern has been observed internationally as

well as locally and is well documented in the literature. In some locations, most

notably in Europe, mopeds and scooters have been traditionally popular as a means

of urban transport, such that their use has even declined in some places due to

alternative mode choices (including motorcycles). In other places where mopeds and

scooters have been historically less popular, including the current study area, their

use has increased as more people reject car use as the preferred option for many

journeys, while their needs may not be met by public transport, cycling or walking.

An increase in moped and scooter usage was demonstrated in Study 2 of the

current program of research, continuing a trend previously observed for mopeds in

the study area (Haworth & Nielson, 2008). Having confirmed this ongoing trend,

Study 3 found that the main motivations for moped or scooter use were related to

cost and convenience, where PTW use was seen to offer greater overall value than

either car use or public transport. In particular, ability to move through traffic,

availability of free (or low cost) parking, and low purchase and running costs were

the major motivating factors for moped and scooter use. In the study area, it is likely

that the moderate licensing requirements for moped riding have also encouraged

moped use, particularly as other motivating factors became more important to

participants.

Studies 1 and 3 in the current research identified growing demand for more

designated PTW parking spaces, despite some recent increases in supply. Access to

free parking is a key factor motivating moped and scooter use, and moped riders

appear less willing to pay for parking than motorcycle riders according to the

findings of Study 1. Although the overall cost of moped use (and scooter use to

varying degrees) is low relative to that of other private motorised vehicles, Study 3

found that moped use does not appear to attract low income earners in particular.

This is despite findings that low cost is a primary motivating factor for moped use.

While moped and scooter use has increased substantially, there have been

comparatively moderate increases in motorcycle use in the study area over the last


decade. According to other research this increase is associated with greater use of

motorcycles for recreation as well as commuting. This is a key difference underlying

increases in moped, scooter and motorcycle use, as mopeds and scooters are not

typically used as recreational vehicles in the study area according to Study 3. This

also highlights a difference in moped use between Australia and some European

countries where mopeds are often used for recreation, particularly by young riders.

Changes to motorcycle and moped rider licensing requirements have potential

to influence the popularity and use of different PTW types. In Spain, for example,

new legislation allowing car licence holders to ride light motorcycles (up to 125cc)

resulted in an increase in light motorcycle use and a reduction in moped use (with

apparent negative safety outcomes) (Albalate & Fernandez-Villadangos, 2009;

Puerto, Ballbé, Albalate, & Fernández, 2009). In Queensland, new legislation

introduced in 2007 requires motorcycle riders to hold a car licence for one year prior

to application for a motorcycle licence. This may have encouraged moped use

among those who held a (provisional) car licence but were as yet unable to obtain a

motorcycle licence. In such cases, moped use may be temporary and continue only

until a motorcycle licence can be obtained. The current research was unable to

determine the impact of this new requirement, but Study 3 did find that licensing

requirements were a stronger motivating factor for moped use than scooter use.

The Queensland licensing requirements are often referred to in the marketing

of mopeds though the extent to which this has encouraged moped use remains

unknown. The PTW industry also tends to claim social and environmental benefits

of PTW use in alleviating traffic congestion and reducing fuel consumption and

emissions. Industry representatives argue that PTWs are ‘part of the solution’ to

such environmental problems (Bowdler, 2008). While personal mobility needs

appear to be the major motivating factor underlying increased moped and scooter

use, environmental considerations are likely to be a secondary motivation for some

users according to Study 3. Marketing strategies and arguments employed within the

PTW industry may therefore also have contributed to the increase in moped and

scooter use.

Over the two year study period, the overall number of PTWs observed in

Study 1 increased, but mopeds and scooters did not increase as a proportion of

PTWs. This is consistent with sales data which show a decline in moped sales

relative to motorcycle sales in the last two years as a result of the recent global


economic downturn. The moped market thus appears to be more volatile than the

motorcycle market, being prone to greater fluctuations depending on economic

circumstances. Moped use may often be more discretionary than the use of other

transport modes, a suggestion supported by Study 3 findings that mopeds are often

not the primary household vehicle. The increase in moped and scooter sales and

usage, as well as the more recent decline in sales, appears to be at least partly related

to economic conditions.

8.2.2 RQ2: How does the usage of mopeds, scooters and motorcycles differ?

Study 1 found that mopeds represented about 20 percent of PTWs used in

inner city Brisbane, while scooters comprised a further 14 percent. As mopeds

comprised about 9 percent of Queensland registered PTWs in 2009, Study 1 shows

that moped use is concentrated in urban areas. This was supported by Study 2 and

Study 3b, both of which found about 90 percent of moped use to occur in speed

zones up to 60 km/h. This is to be expected for mopeds given the ADR maximum

legal speed restriction of 50 km/h. Although not separated from motorcycles in the

registration data, scooter use also appears to be concentrated in urban areas and

lower speed zones according to Study 2 and Study 3b, despite performance

characteristics allowing many scooters to travel at highway speeds.

The Brisbane and Gold Coast areas accounted for a large proportion of all

crashes analysed in Study 2 and these data largely reflect usage patterns. Scooter

crashes occurred in Brisbane in about 52 percent of cases, compared with 33 percent

and 43 percent of moped and motorcycle crashes respectively. Crashes on the Gold

Coast represented around 15 percent of cases involving scooters, compared with 18

percent for mopeds and 10 percent for motorcycles. The crash data also suggest

greater use of mopeds in tourist areas, including the Gold Coast, Townsville and

Cairns areas in particular, though not all of this use is actually by tourists.

Study 1 showed consistency of usage over the two year study period, with no

variation by PTW type according to season. This confirms that the climate in the

study area is conducive to year-round riding, as suggested in other research.

Study 2 found that motorcycles are used relatively more outside of urban

areas. Usage patterns evident in Study 2 also indicate that a greater proportion of

motorcycle use (31%) occurs on weekends, compared with moped and scooter use


(~20%). Further, the current research also found a greater proportion of motorcycle

use (25%) occurs in high speed zones (80-110 km/h), compared with moped and

scooter use (~6%). These findings are consistent with local and international

literature indicating that motorcycles are used more for recreational purposes and less

for commuting compared with mopeds and scooters. However, the literature often

overlooks the reality that ‘recreation’ and ‘commuting’ are not the only purposes of

PTW use. PTWs may be used as general transport for purposes which do not clearly

fit either of these definitions. For example, in Study 3b, 13 percent of reported

moped trips and 25 percent of scooter trips were for the purpose of shopping.

Study 3b found that scooters travelled further per year on average than

mopeds, but not that they were used more for recreational riding. This is despite

some indication of a bias toward enthusiasts among scooter riding survey

participants. The small sample size arguably limits the reliability of the survey data

and many conclusions drawn from Study 3b are only tentative. However, the self-

reported distances travelled by mopeds and scooters were roughly consistent with

that reported in other research and were substantially less than the distances travelled

by motorcycles (Harrison & Christie, 2006).

According to the crash data analysis in Study 2, the median age of moped

riders (32 years) was less than that of motorcycle (35 years) and scooter (39 years)

riders. Earlier crash data analysis in the study area also found moped riders to be

younger than motorcycle riders, though scooters were not separated from

motorcycles in that study (Haworth, Nielson et al., 2008). Study 3b also found

moped riders to be younger than scooter riders, though they were each slightly older

on average than in Study 2. This was also the case in the survey by Harrison and

Christie (2006), suggesting that the survey samples were biased toward older riders

and/or that the crash data overrepresented younger riders. Harrison and Christie

(2006) also found significant differences in the age of motorcycle riders according to

PTW type, which should be taken into account when considering average motorcycle

rider age. For example, riders of sport and off-road motorcycles were younger than

moped and scooter riders, while riders of touring motorcycles were older.

Queensland moped riders are typically older than European riders. Moped

riding is permitted in most European countries from sixteen years of age, and from

fourteen years of age in some countries. In contrast to Queensland, moped riding is

generally permitted at a younger age than car driving in European countries, an


approach which has been criticised in some of the literature (ERSO, 2006). Moped

riders under 25 years of age comprise a large proportion of moped rider fatalities in

some European countries, and the majority of fatalities in some countries, though the

crash involvement of young riders varies considerably across Europe. Across twenty

European countries in 2005, riders younger than 25 years comprised half of all

moped rider fatalities. Study 2 of the current research found approximately 31

percent of crashed moped riders were aged below 25 years, but all five fatal moped

crashes involved older riders.

Moped and scooter riding attracts a higher proportion of female riders than

motorcycle riding. This is consistently observed across all developed countries. In

Study 2 of the current research, males represented about 92 percent of crashed

motorcycle riders and 63 percent of crashed moped riders, while scooter riders fell

between mopeds and motorcycles in the proportion of male riders (78 percent). This

suggests that scooters may appeal to females in a similar way to mopeds, but that

only some females are prepared to obtain a motorcycle licence. In survey data

collected for Study 3b, males represented about 73 percent of respondents and there

was no difference between moped riders and scooter riders in gender distribution.

The crash data from Study 2 is consistent with earlier research in Queensland

regarding gender distribution of moped and motorcycle riders (with scooter riders

included as motorcycle riders) (Haworth, Nielson et al., 2008). According to

Christie (2008), there is no reason to assume that females are overrepresented in

scooter crash data, which suggests in turn that they are probably underrepresented in

the Study 3b survey data.

Moped usage by gender is similar in Queensland to other Australian and

international jurisdictions, with males generally comprising between 60 percent and

80 percent of riders. For example, in the MAIDS control sample of 921 riders from

France, Germany, Italy, the Netherlands and Spain, 77 percent of moped riders were

male (ACEM, 2008b). In slightly more recent crash data from Barcelona, males

comprised 65 percent of moped riders (Perez, Mari-Dell'Olmo et al., 2009).

Motorcycle riders in Study 2 of the current research were more likely to hold

a Queensland licence5 (97%) than either moped riders (81%) or scooter riders (93%).

The relatively high proportion of crash-involved moped riders licensed outside of

5 For motorcycle and scooter riders this refers to a motorcycle licence, while moped riders most likely

held a car licence but not a motorcycle licence.


Queensland suggests greater tourist use of mopeds than of scooters or motorcycles.

Similar proportions of moped riders (6%) and motorcycle riders (5%) in crash data

were unlicensed, compared with scooter riders who were least likely to be unlicensed

(2%), though this result is somewhat unreliable due to the low number of scooter

riders. The survey conducted for Study 3b showed a different result with regard to

unlicensed scooter riders. Of all survey respondents who were scooter riders (n =

153), 4.6 percent reported not holding the required motorcycle licence. Of scooter

riders who reported a crash within the last five years (n = 48), 8.3 percent reported

not holding the required motorcycle licence. This suggests that the survey sample

was not representative of scooter riders in the study area.

Riders and industry often do not distinguish between LA mopeds and LC

scooters conceptually, typically referring to both PTW types as ‘scooters’. For

example, Scooter magazine, an offshoot of a major Australian motorcycling

magazine Two Wheels, includes both mopeds and scooters in its regular ‘scooter

listing’ and reviews of currently available new models (Anonymous, 2010).

However, while there is relatively little variation in moped design and performance

characteristics, many of which are shared by smaller LC scooters, there is

considerable variation in the characteristics of LC scooters as a whole. For example,

a Vespa LX 50 (LA) moped is virtually identical to a Vespa LX 125 (LC) scooter in

all but engine design and performance, while a Suzuki Burgman 650 (LC) scooter

shares as much in common with medium capacity touring motorcycles (weight, ABS

brakes, maximum speed) as it does with mopeds (automatic transmission, step-

through chassis, smaller diameter wheels). The diversity of scooters in use in the

study area (as indicated by engine size) was demonstrated in Study 1 and Study 3b.

As with motorcycles, the type of scooter chosen will likely depend on a wider range

of rider objectives and motivations than is the case with mopeds. This is important

to consider as rider motivations and objectives have been shown to influence crash

risk in other research (Harrison & Christie, 2006; Sexton, Baughan et al., 2004).

8.2.3 RQ3: How does the safety of mopeds, scooters and motorcycles differ?

According to Study 2, over five years to June 2008, police-reported crash

rates per 10,000 registration years were slightly higher for LA mopeds (133) than for

LC motorcycles and scooters combined (125). Crash rates per registered vehicle fell


for both LA and LC categories over the study period, though the rates declined more

sharply for mopeds. The descriptive analysis identified differences between mopeds

and motorcycles in crash rates per registered vehicle as a function of crash severity,

with moped crashes generally less severe compared with motorcycle crashes (1.2

fatalities per 10,000 moped registration years compared with 4.2 for motorcycles)6.

Study 2 also found that police-reported scooter crashes were less severe than

motorcycle crashes, though crash rates per registered vehicle or as a function of

severity could not be calculated. A more complex analysis of severity using an

ordered probit model to control for other factors suggested that severity was

influenced by crash characteristics more than by PTW type per se. Differences in

crash severity therefore relate strongly to differences in crash characteristics and

circumstances. In particular, moped and scooter crashes occurred less in high speed

zones compared to motorcycle crashes and this appears to be the main difference in

relation to crash severity (although motorcycle crashes were more severe regardless

of speed zone). Another difference is that motorcycle crashes were most likely to

involve a single vehicle (34%), and scooter crashes least likely (21%), with greater

severity associated with single vehicle crashes.

As noted previously, reliable data on distance travelled (exposure) are

required to accurately assess differences in crash rates between mopeds, scooters and

motorcycles. In the examination of self-reported crashes and distanced travelled in

Study 3b, scooters were separated from mopeds, while motorcycles were not

included. Study 3b showed that the self-reported crash rate per million VKT (based

on mean) was similar for mopeds (12.9) and scooters (11.5). This is generally

consistent with the Queensland Motorbike Usage Survey (Harrison & Christie,

2006), although the crash rates were more similar for mopeds and scooters in the

current research. Applying the same exposure data for police reported crash rates

there were 4.2 police-reported moped crashes per million VKT (scooter crash rates

could not be reported due to lack of separation from motorcycles in registration

data). Using Queensland Motorbike Usage Survey exposure data, the police-

reported crash rate was higher (6.3 per million VKT) as the annual distance travelled

was lower than reported in Study 3b. Whichever exposure estimate is used for

mopeds, the moped crash rate per million VKT appears much higher than that of

6 This analysis did not separate scooters from motorcycles.


motorcycles (1.7)7. According to these findings, most moped and scooter crashes are

not reported to police. This is very likely also the case with motorcycle crashes, and

crash rates based on police-reported crashes therefore do not reflect actual crash rates

per PTW type, as noted in previous research (Haworth, 2003).

The Brisbane and Gold Coast areas accounted for a large proportion of all

crashes analysed in Study 2 and this largely reflects usage patterns. Tourist

involvement seems to feature more in moped crashes than in scooter or motorcycle

crashes according to licensing and geographic information in Study 2, though the

absolute numbers do not suggest a particular problem regarding tourists. Many of

the moped crashes in areas frequented by tourists probably involve local residents.

As was the case with moped crashes, scooter crashes occurred on weekdays

in around 80 percent of cases, compared with 69 percent for motorcycle crashes,

again reflecting relatively more use of scooters and mopeds for commuting. Scooters

were slightly more likely than the other PTW types to crash in daylight hours, with

approximately 83 percent of scooter crashes occurring between 6am and 6pm,

compared with around 77 percent for mopeds and motorcycles.

There are numerous apparent differences between the PTW types regarding

crash type, crash configuration, fault attribution and contributing circumstances. As

noted above, scooters were significantly more likely than the other PTW types to be

involved in a multi-vehicle crash. Scooters also appeared less likely than either

mopeds or motorcycles to be designated Unit 1 (most at fault) in both single and

multi-vehicle crashes, though the difference was not statistically significant. Moped

and scooter crashes were both more likely than motorcycle crashes to occur at

intersections, and while ‘angle’ crashes comprised a large minority of cases for all

PTW types, they were most likely in scooter crashes (44%). Scooter crashes were

clearly less likely than those involving either mopeds or motorcycles to be ‘hit

object’ crashes, and more likely than either moped or motorcycle crashes to be

‘sideswipe’ crashes.

A number of findings in the overall program of research suggest that scooter

use is safer than that of mopeds or motorcycles, despite some questions remaining

over crash rates. These findings include the high proportion of scooters observed in

Study 1 (14% of all PTWs) against the relatively low proportion of scooters in

7 Using Queensland Motorbike Usage Survey data.


Brisbane city crashes observed in Study 2 (2.7% of PTWs). While 1.3 percent of all

crashes in Study 2 involved scooters, scooter riders represented 2.7 percent of

respondents in the Queensland Motorbike Usage Survey (Harrison & Christie, 2006).

These data as well as recent sales data (presented in Chapter Two) suggest that

scooters are underrepresented in crashes, but a more detailed analysis of registration

data is needed (separating scooters from motorcycles) in order to confirm this.

Other findings outlined earlier in this section were that scooter crashes were

less severe than motorcycle crashes, as were moped crashes, with the differences

largely attributable to crash characteristics. Moped use appears safer than

motorcycle use with regard to crash severity, but this is tempered by the finding that

mopeds are more likely to crash according to the available data. It has thus been

difficult to disentangle the issues which determine the relative safety of the PTW

types. The next section explores the specific reasons why scooter use appears safer

than either moped or motorcycle use, and further highlights the complexity of

comparing the three PTW types.

8.2.4 RQ4: Why does the safety of mopeds, scooters and motorcycles differ?

Study 2 found that moped crash rates per registered vehicle declined more

rapidly than those of motorcycles over the study period. It is possible that new

moped riders are more risk averse than those in previous periods and may therefore

crash less, despite inexperience being a known risk factor for PTW crashes. It is also

possible that, while there are now more registered mopeds, they are used less

frequently and travel relatively fewer kilometres than those in previous periods. A

further possibility is that recreational riding, associated with higher risk than

commuting in the literature, has lead to the increase in motorcycle use (but not

moped use).

Inexperience or lack of recent experience was noted in the literature as one of

the main contributors to PTW crash and injury risk (Greig, Haworth et al., 2007). In

Study 2, inexperience was attributed more to riders in moped crashes (9%) than in

motorcycle (5%) or scooter (4%) crashes. In some cases this may be attributed by

police solely on the basis of rider age, and moped riders were younger on average

than scooter and motorcycle riders. However, in Study 3, safety awareness and safe

riding practices appeared to increase with age and experience. The older and more


experienced participants tended to be scooter riders who held a motorcycle licence

and most had also attended rider training. Study 3 also suggested that moped riders

were largely uninterested in sourcing information on PTW safety as a means to

compensate for inexperience. The Study 3 findings suggest that the attribution by

police of inexperience in Study 2 may be accurate with regard to comparing moped

and scooter riders. Study 2 suggests that motorcycle riders are more similar to

scooter riders than moped riders regarding experience, though they were not included

in Study 3 to support this suggestion.

Another of the main contributors to PTW crash and injury risk identified in

the literature was risk taking (Greig, Haworth et al., 2007). Risk taking is noted in

the literature as being more prevalent among younger riders and also male riders.

Age characteristics of crashed riders in Study 2 showed that less than 10 percent of

scooter riders were aged below 25 years, compared with 31 and 23 percent of moped

and motorcycle riders respectively. Scooter riders were also less likely to be male

(78%) than motorcycle riders (92%), though more likely than moped riders (63%).

In light of the literature, these characteristics suggest that scooter riders may be least

likely to engage in risk taking. Other findings in Study 2 support this, including that

scooter riders were least likely to be considered at fault in police-reported crashes,

least likely to be unlicensed, least likely to ride (or crash) while impaired by alcohol,

and least likely to speed. Ultimately it appears that scooter riders are more compliant

with road rules than either moped or motorcycle riders.

The differences between moped and motorcycle riders with regard to risk

taking are more difficult to discern. In terms of age and gender, the younger age of

moped riders may be countered somewhat by the higher proportion of female riders.

Their lower involvement in speed-related crashes compared to motorcycle riders may

relate to limited moped performance as much as an intention to not speed. They

were similarly likely to ride unlicensed and ride while impaired by alcohol. As

moped riders engage more in commuting and less in recreational riding than

motorcycle riders, it remains possible that they are less inclined toward risk taking,

though the current program of research could not determine this.

Studies 2 and 3 suggested that PTW control and riding skills, encompassing

instability and braking difficulties and road surface and environmental hazards

(Greig, Haworth et al., 2007) for the purpose of this discussion, may represent a

greater crash risk for moped riders than scooter riders. In Study 2, moped riders


were more likely than scooter riders to crash on wet roads and as a result of poor

road conditions. Motorcycle riders were statistically more similar to moped riders

than scooter riders on these variables, but they are arguably not directly comparable

due to more motorcycle use outside of urban areas (therefore encountering different

road-based hazards). In Study 3a, some moped riders demonstrated poor

understanding of brake operation and vehicle performance, while scooter riders had a

better grasp of basic principles of PTW control. In Study 3b, scooter riders reported

generally greater confidence with vehicle control than moped riders. These findings

are unsurprising, as compared with moped riders, scooter riders were generally more

experienced, held a motorcycle licence and had undertaken rider training. In these

respects scooter riders are more similar to motorcycle riders than moped riders. Like

scooter riders, motorcycle riders would be expected to have superior riding skills to

moped riders. However, as motorcycle riders were not included in Study 3 they

could only be directly compared in crash data, which was of limited use for

comparing rider skills.

It is suggested in the literature that mopeds and scooters may be more

susceptible than motorcycles to road-related hazards such as potholes and uneven

surfaces, due to smaller wheels and shorter wheelbases. It is also suggested that

mopeds and scooters may be less safe than motorcycles due to less advanced braking

technology. These suggestions do not hold clearly given the differences observed

between moped and scooter crash characteristics and circumstances. While larger

scooters are typically superior to mopeds in performance and handling, the majority

of scooters in use are of smaller engine capacity and thus technologically comparable

to mopeds.

A further contributor to rider crash and injury risk indentified in the literature

is driver failure to see motorcyclists (Greig, Haworth et al., 2007), a topic on which

there is an abundance of research literature. This was discussed in the previous

chapter in the section titled ‘Other road users’ (section 7.4.3.3), as while rider

conspicuity is important there are also other risks associated with sharing the road

with other vehicles. Studies 2 and 3 of the current program of research provided

further evidence that this is a major concern for all PTW riders and not just

motorcyclists. However, there is no clear evidence that it is more or less of a

concern for riders of particular PTW types, with scooter and moped riders alike

frequently noting in Study 3 that hazards such as failure to see, inattention and


aggressive behaviour among other road users is a major concern.

Some literature has suggested that mopeds and scooters may be harder to see

than motorcycles due to being smaller, but no supporting data are presented (de

Rome, 2006). Research in Europe found that conspicuity-related crashes increased

with PTW engine size, suggesting greater involvement of motorcycles in such

crashes, but the difference is likely to be related to clothing choice rather than PTW

type (Comelli, Morandi et al., 2008). Differences between mopeds, scooters and

motorcycles in single and multi-vehicle crash involvement suggest that the issue may

be more important for scooter riders and moped riders. However, this likely relates

more to the urban traffic environment in which mopeds and scooters mostly operate,

rather than a failure of drivers to see mopeds and scooters less than motorcycles. It is

possible according to Study 3 that scooter riders may be better at hazard perception

and responding than moped riders, due to more experience and participation in rider

training, though there are no data available to confirm this.

Vulnerability to injury was identified in the literature as a main contributor to

injury risk for motorcycle riders (Greig, Haworth et al., 2007), as it also is for moped

and scooter riders. Helmets are arguably the most effective item of apparel for

reducing injury severity among riders, and were used by almost all on-road riders in

the study area according to Study 2. The vulnerability of PTW riders to injury can

also be moderated by the use of protective clothing with abrasion resistance and/or

impact protection, including boots, jackets, gloves and pants (de Rome, Ivers et al.,

2011). However, the use of such items can vary among riders of different PTW

types and is related partly to motivations for riding and riding purpose according to

the literature. Some research suggests that using protective clothing can encourage

risk taking, while non-use can lead to more cautious riding, though this was drawn

from a sample of mostly older recreational riders (Watson, Tunnicliff et al., 2007).

Research in Australia has found more use of protective clothing among

motorcycle riders than scooter or moped riders (Christie, 2008). In Study 3 of the

current research, a difference was also found between moped and scooter riders, with

scooter riders generally more likely to use protective clothing. A perception among

moped riders that they do not travel fast enough to warrant the inconvenience and

cost of using protective clothing appears to underlie this difference (and discounts

the potential consequences of a crash at moderate speed). However, there were no

detailed injury data available in the current research to assess the implications of this


finding. Given the apparent bias in Study 3b toward scooter riders who were

enthusiasts (who may be more likely to use protective clothing than non-enthusiasts),

the difference between moped and scooter riders may have been exaggerated.

Moreover, in Study 3a opinions on this topic varied widely independent of PTW

type. This is a topic that appears to warrant further research.

8.3 Implications of the research

The research has identified the key factors motivating increased moped and

scooter usage, providing answers to research question 1. These factors include low

cost, convenience and practicality, ease of use, enjoyment, and perceived

environmental benefits. The key motivating factors in the study area align closely

with those seen internationally. Overall, mopeds and scooters appear preferable to

car use and public transport for particular journeys, providing relatively cheap and

efficient mobility. Mopeds and scooters are predominantly used for commuting in

the study area, as is also the case internationally. In some European countries there

appears to be more recreational moped use, particularly among young people, which

is likely a result of the younger age at which Europeans are generally permitted to

ride mopeds.

An examination of moped, scooter and motorcycle use provided answers to

research question 2. Compared to motorcycle use in the study area, there was more

moped and scooter use in urban areas, more use by females and less weekend use.

There was also proportionally more moped use in tourist areas, though much of this

use was probably by local residents. Moped riders were typically younger than

motorcycle riders, while scooter riders were older. In most respects (other than rider

age) moped use was similar to scooter use. These findings of the current research are

also generally consistent with those found internationally, although other research

has often not separated scooters from motorcycles.

Moped use has increased less rapidly in other Australian jurisdictions where

their use requires a motorcycle licence, suggesting that Queensland licensing

requirements encourage moped use. A change in legislation to require a motorcycle

or other PTW licence for moped use may result in reduced moped usage but may

also increase scooter and motorcycle use. Under current conditions it appears likely

that moped and scooter usage may increase further, or at least be sustained at current


levels in the foreseeable future. The implications of this are that the use of mopeds

and scooters, as distinct from motorcycles, should be monitored more closely in

future and that their use should be integrated specifically in strategic transport

planning. This would include consideration of appropriate infrastructure and

regulations that recognise the benefits of moped and scooter use, including increased

mobility, reduced congestion, fuel consumption and emissions, as well as the safety

implications. To that end, broader consideration of PTWs in strategic transport

planning has been advocated in Australia (Haworth, 2006; Wigan, 2000), as well as

internationally (ACEM, 2008a). In the study area, previous and current strategies

and programs targeting PTWs have focused almost entirely on safety issues, and

mopeds and scooters have received little specific attention as distinct from

motorcycles.

The research found key differences in the safety of mopeds, scooters and

motorcycles. For reasons outlined above in sections 8.2.3 and 8.2.4, it was argued

that scooter riders generally exhibit safer behaviour than both moped riders and

motorcycle riders in the study area. It was suggested that the safer behaviour of

scooter riders leads to their lower crash involvement (despite self-reported survey

data indicating otherwise in Study 3), though more reliable exposure data are

required to confirm the crash rates of scooter riders in the study area. It was also

argued that moped riders are somewhat safer than motorcyclists in terms of crash

severity, though not in terms of crash rates.

With regard to the differences between moped and scooter riders, factors

underlying safer behaviour of scooter riders include that they are older, more

experienced, and less inclined toward risk taking. The literature suggests that

differences in crash and injury risk among moped, scooter and motorcycle riders

relate more to rider characteristics, motivations and behaviour than to specific

characteristics of the particular PTW types. This is consistent with the findings of

the ordered probit model of severity used in the current program of research. The

one possible exception to this is that the 50 km/h maximum speed restriction applied

to mopeds may play some role in moderating crash severity for moped riders.

However, scooter crash severity was similar to that of mopeds despite the fact that

scooters are not subject to the same performance restrictions.

As all PTW riders are considered vulnerable road users, safety-oriented

countermeasures often aim to improve the safety of riders of all PTW types.


However, the current research focused mostly on moped and scooter use and safety

as there is already a substantial body of literature regarding motorcycle safety. A

crucial implication of the current research is that scooter riders provide a potential

safety benchmark that should conceivably be achievable for their moped riding

counterparts. This is not to suggest that the safety of scooter riders cannot or should

not also be improved, but that the safety of moped riders may be of higher priority,

particularly in light of the increased moped usage observed.

8.3.1 Implications for policy and planning

During the course of this research, questions emerged regarding what

constitutes an acceptable level of risk for moped (and other PTW) riders, and

whether current regulations governing moped use are adequate to ensure rider safety.

With regard to this problem, ‘a fundamental discussion concerning risk acceptance in

a risky society, and the questions of what is a reasonable and responsible expectation

of risk reduction,’ is required (SWOV, 2006a, p. 22). According to the literature,

such a discussion is yet to eventuate in a formal context wherein the full range of

relevant stakeholders is engaged at an international level. However, the lack of such

discussion to date does not preclude the identification of aspects of moped and

scooter use that may be amenable to safety-oriented countermeasures or

interventions. To that end, the findings of this research highlight differences in the

use and safety of mopeds, scooters and motorcycles, and the factors which underlie

those differences.

The aim of this thesis was not to offer policy recommendations, but to

provide an assessment of moped and scooter use and safety that may be used to guide

policy and countermeasure development. One potential countermeasure considered

in recent years by the Queensland Government is the introduction of a mandatory

PTW licence for moped riders. This proposal appears to have strong support among

scooter riders who hold motorcycle licences. Some comments on this potential

countermeasure are provided below to conclude this final discussion.

Despite the lack of empirical evidence in support of moped rider training and

PTW rider training generally, the belief that training (and licensing) at least

potentially improves rider safety is widely held among researchers, government

departments, (trained) riders and the PTW industry (ACEM, 2010; Bowdler, 2011;


Schoon, 2004). In regard to the current research, some moped riders appear deficient

in areas that could potentially be addressed by rider training, such as vehicle control

skills and hazard perception and response. However, given that scooter and

motorcycle riders are trained and licensed under the same system in the study area, it

appears that the relative safety of scooter riders may be attributable to factors other

than training and licensing. As noted previously, psychological and social factors

which influence the behaviour of riders likely underlie some of the differences

observed in the safety of moped, scooter and motorcycle riders (Watson, Tunnicliff

et al., 2007).

At present, psychological and social factors influencing safety are largely

beyond the scope of training programs to address, though recent research suggests

that the application of more appropriate teaching and learning principles may assist

in this area (Rowden, Watson et al., 2007; Rowden, Watson, Wishart, & Schonfeld,

2009). A particular challenge concerns the heterogeneity of the moped, scooter and

motorcycle riding populations, which has been observed in the current research and

also consistently in other studies. Aside from methodological problems associated

with delivering effective rider training, its potential to improve rider safety is also

limited where training is not a required component of licensing systems, or indeed

where a PTW licence is not required at all. Previous research has argued that

compulsory rider training is more effective than voluntary training which is known to

attract low participation rates. However, it can also be argued that training may not

be essential where an adequate level of competency can be demonstrated through

practical testing procedures. These issues present a considerable challenge for policy

makers and planners, particularly in the absence of more conclusive rider training

program evaluations.

The contribution of other vehicle drivers to PTW crashes is well documented

and the current program of research has further highlighted the importance of related

issues such as PTW conspicuity and driver inattention. Given that mopeds and

scooters are used predominantly in urban areas (which largely explains their high

involvement in multi-vehicle crashes), these issues will remain important and may

become more so in the event of further increases in moped and scooter use. As

mentioned above, rider training may improve the hazard perception and responding

of riders, thereby helping to reduce multi-vehicle crashes. Nonetheless, the

behaviour of other road users also remains an important policy consideration.


The use and knowledge of protective clothing could be improved among

PTW riders generally, and particularly among moped riders according to the current

program of research. Mandatory use of protective clothing other than helmets is not

generally recommended in the research literature, though financial incentives such as

tax and insurance rebates and concessions have been advocated. There are also

various other avenues through which riders could be encouraged to dress

appropriately to reduce potential injury severity. Once again, this is a problem that

could be addressed through an educational component in rider training programs,

participation in which would likely be maximised through requirement of a

motorcycle (or other PTW licence) for moped riders.

8.4 Strengths and limitations of the research

The specific strengths and limitations of the three studies conducted in this

research program were addressed in the relevant chapters of this thesis. Presented

below is a discussion of the main strengths and limitations of the overall program of

research.

This thesis explored differences in the use and safety of mopeds, scooters and

motorcycles through complementary quantitative and qualitative research methods.

Published research to date has not comprehensively compared mopeds, scooters and

motorcycles by simultaneously examining observational data (Study 1), crash and

registration data (Study 2), and focus group and rider survey data (Study 3). In

respect of moped and scooter usage, the three studies conducted represent a process

of triangulation whereby the results of each study were supported by those of the

other studies. In regard to their relative safety, the separation of scooters from

motorcycles in crash data was a novel undertaking that revealed important

differences between mopeds, scooters and motorcycles that had not been previously

identified. The difficulty in distinguishing mopeds from scooters in some cases was

a limitation with regard to data analysis, but also highlights potential problems for

reporting and enforcement.

The research was geographically limited to the study area of Queensland,

Australia. Queensland has seen a more rapid increase in moped use than other

Australian jurisdictions, some of which require a motorcycle licence for moped

riding, in contrast to the study area. Nonetheless, the research may be relevant to


other places where moped and scooter use has increased from traditionally low

levels, particularly including places where moped riding is permitted on a car

licence.

The lack of separation of motorcycles from scooters in registration data

prevented an estimation of scooter crash rates per registered vehicle. This may have

influenced the estimation of motorcycle crash rates, though the impact is likely to be

minimal given the low number of scooters relative to motorcycles as indicated by

sales and also crash data.

Despite the employment of a range of recruitment methods, it was difficult to

recruit participants for Study 3b. While the small survey sample size presents

statistical limitations and potential biases, it also indicates that moped riders in

particular are difficult to recruit, suggesting that they may lack interest in safety

issues. It is thought that participants responding to the survey in Study 3b as scooter

riders may have been biased toward enthusiasts. The calculation of crash rates per

distance travelled is not entirely reliable due to small sample sizes from which data

on distance travelled were drawn. The survey sample is not assumed to be

representative of riders in the study area.

8.5 Potential topics for further research

Further research is required regarding the effectiveness of rider training

systems in reducing crash risk, including identification of specific components which

have positive and negative effects on safety. Numerous evaluations of rider training

to date have been inconclusive, suggesting a need for more effective evaluation

frameworks and methodologies.

Reliable exposure data for moped and scooter use in Queensland and other

Australian jurisdictions are still lacking. A suitable recruitment methodology for

attracting greater numbers of moped and scooter riders as survey participants is

needed to address this lack of data. Moped riders appear particularly difficult to

recruit for survey participation.

Lane splitting and filtering in traffic was considered risky by some survey

and focus group participants, but there is no information in official crash data to

determine the actual risk associated with these practices. As these practices appear

to be common among moped and scooter riders, as well as motorcycle riders, future


research could aim to study these behaviours with a view to determining objective

crash risks.

There is evidence that the use of mopeds by tourists in Queensland

contributes to a small proportion of moped crashes. As a road safety problem, this

may have been overstated in some of the literature. There is little data available

beyond information on licensing such as analysed in Study 2 to determine the extent

of this perceived problem. Future research could explore the use of mopeds by

tourists and the attendant safety implications, as well as the potential impact on

moped hire companies of the introduction of a mandatory PTW licence for moped

riding.

Continued observation of PTW types used in the Brisbane inner city area

(Study 1) would provide ongoing objective information on PTW usage that is not

available in registration, sales, crash or survey data. As it is already established, the

process of recording PTW types and basic characteristics is simple and efficient and

would therefore require only a minimal input of resources.

Few evaluations have been conducted to assess the effectiveness and

feasibility of education and awareness campaigns targeting other road users.

Recognition that other road users represent a significant safety hazard for PTW riders

has led to the development and implementation of many such campaigns and

programs, but there is little information on what actually works to improve rider

safety in this area. This is therefore a final suggested topic for further research.

8.6 Concluding remarks

Moped and scooter use has increased at a much faster rate than motorcycle

use in Australia in recent years. This is particularly evident in the State of

Queensland where moped riding is permitted for car licence holders and a

motorcycle licence is not required. As might be expected in light of this increased

usage, increasing numbers of moped crashes have also been observed, leading to

greater interest in the relative safety of mopeds, scooters and motorcycles. The aims

of this program of research were to achieve a better understanding of moped and

scooter usage trends and patterns, and to explore differences in moped, scooter and

motorcycle safety. While an extensive body of knowledge exists on motorcycle

safety, moped and scooter safety has received comparatively little focused research


attention. No other research to date has thoroughly examined the differences and

similarities between mopeds and larger scooters, or between larger scooters and

motorcycles, in relation to usage and safety.

The current program of research involved three complementary studies,

including an observational study of PTW use, an analysis of crash and registration

data, and an exploration of moped and scooter rider characteristics, travel patterns,

beliefs, attitudes and experiences. The main findings include that mopeds and

scooters are similar in many aspects of usage, including usage patterns and

motivations, but scooter riders appear to be safer than moped riders due to a

combination of superior skills, safer riding behaviour and greater experience. The

requirement for scooter riders but not moped riders to hold a motorcycle licence

likely explains at least some of this difference.

It was also found that scooter riders are safer than motorcycle riders, despite

both being subject to the same licensing requirements which encourage participation

in rider training. This suggests that safer attitudes and motivations rather than

superior skills and knowledge underlie the differences between scooter and

motorcycle riders. Moped and scooter crashes were generally less severe than

motorcycle crashes and this was related to crash characteristics rather than to PTW

type per se, such as greater involvement of motorcycles in higher speed zone crashes.

While moped crashes were found to be of lower severity than motorcycle crashes,

the crash rates per registered vehicle and per vehicle kilometre travelled were higher

for mopeds than for motorcycles and scooters combined (scooter crash rates could

not be estimated separately). Thus, while moped riders are more likely to crash than

motorcycle riders, they are somewhat less likely to sustain severe injuries.

It is reasonable to expect that mopeds and scooters will remain popular in

Queensland in future and that their usage may further increase, along with that of

motorcycles. This research therefore has some important practical implications

regarding moped and scooter use and safety. Future policy and planning should

consider potential options for encouraging moped riders to acquire better riding skills

and greater safety awareness, such as is evident among scooter riders. While rider

training and licensing appears an obvious potential countermeasure, the effectiveness

of rider training has not been established and other options should also be strongly

considered. Such options might include rider education and safety promotion, while

interventions could also target other road users and urban infrastructure.


Future research could further address the effectiveness of rider training and

licensing through program evaluations, the need for more detailed and reliable data

(particularly crash and exposure data), protective clothing use, risks associated with

lane splitting and filtering, and tourist use of mopeds. Some of this research would

likely be relevant to motorcycle use and safety, as well as that of mopeds and

scooters.


REFERENCES

Aare, M., & Holst, H. (2003). Injuries from motorcycle and moped crashes in

Sweden from 1987 to 1999. Injury Control and Safety Promotion, 10(3), 131

– 138.

ABS. (2008). Survey of motor vehicle use Australia: 12 months ended October 2007

(No. 9208.0). Canberra: Australian Bureau of Statistics.

ABS. (2009). Motor Vehicle Census, Australia (No. 9309.0). Canberra: Australian

Bureau of Statistics.

ACEA. (2010). European automobile industry report 09/10. Brussels: European

Automobile Manufacturers Association.

ACEM. (2008a). MAIDS: In-depth Investigations of Accidents Involving Powered

Two Wheelers - Final Report 2.0. Brussels: Association of European

Motorcycle Manufacturers.

ACEM. (2008b). Green paper on urban transport: Towards a new culture for urban

mobility. Brussels: Association of European Motorcycle Manufacturers.

ACEM. (2010a). ACEM Report. Brussels: Association of European Motorcycle

Manufacturers.

ACEM. (2010b). News from the motorcycle industry in Europe (ACEM newsletter

#22). Brussels: Association of European Motorcycle Manufacturers.

Albalate, D., & Fernandez-Villadangos, L. (2009). Exploring determinants of urban

motorcycle accident severity: the case of Barcelona. Barcelona: University of

Barcelona.

AMCN. (2010, 3 -16 March). State of the world: US leads bike sales plummet.

Australian Motorcycle News, 59 (17), 8.

Ameratunga, S., Hijar, M., & Norton, R. (2006). Road-traffic injuries: confronting

disparities to address a global-health problem. The Lancet, 367(9521), 1533-

1540.

Andrea, D. (2006). Development of a strategic motorcycle safety program in

Victoria, Australia. Paper presented at the International Motorcycle Safety

Conference (IMSC), Long Beach, California.

Anonymous. (2010, June). Scooter listing. Scooter, 24, 64-76.

Antonio, P., & Matos, M. (2008). An evaluation of the Portuguese moped rider

training program. In L. Dorn (Ed.), Driver Behaviour and Training (Vol. 3,

pp. 399 - 413). Aldershot: Ashgate.

ATSB. (2002). Monograph 12 - Motorcycle rider age and risk of fatal injury.

Canberra: Australian Transport Safety Bureau.

ATSB. (2007). Road Deaths Australia: 2006 Statistical Summary (June 2007,

DOTARS 50249). Canberra: Australian Transport Safety Bureau.

Australian Government. (2008). Third edition ADRs - Applicability Summary Two

and Three Wheeled Vehicles. Retrieved 29 June, 2008, from

http://www.infrastructure.gov.au/roads/motor/design/pdf/ADR_Applicability

_Summary-L-Group.pdf

Bellaby, P., & Lawrenson, D. (2001). Approaches to the risk of riding motorcycles:

reflections on the problem of reconciling statistical risk assessment and

motorcyclists' own reasons for riding. The Sociological Review, 49(3), 368-

388.

http://www.infrastructure.gov.au/roads/motor/design/pdf/ADR_Applicability_Summary-L-Group.pdf

http://www.infrastructure.gov.au/roads/motor/design/pdf/ADR_Applicability_Summary-L-Group.pdf


Berg, F., Rucker, P., Gartner, M., Konig, J., Grzebieta, R., & Zou, R. (2005).

Motorcycle impacts to roadside barriers - real-world accident studies, crash

tests and simulations carried out in Germany and Australia (No. 05-0095).

Melbourne: Monash University.

BITRE. (2010). Road deaths Australia: July 2010. Canberra: Bureau of

Infrastructure, Transport and Regional Economics, Department of

Infrastructure, Transport, Regional Development and Local Government.

Blackman, R., Cheffins, T., Veitch, C., & O'Connor, T. (2009). At work or play: A

comparison of private property vehicle crashes with those occurring on public

roads in north Queensland. Australian Journal of Rural Health, 17(4), 189-

194.

Blackman, R., Steinhardt, D., & Veitch, C. (2009). Fatal motorcycle crashes in north

Queensland: characteristics and potential interventions. Paper presented at

the 10th National Rural Health Conference, Cairns, Queensland.

Blackman, R., Veitch, C., & Steinhardt, D. (2008). Non-fatal motorcycle crashes on

public roads in north Queensland. Paper presented at the 2008 Australasian

Road Safety Research, Policing and Education Conference, Adelaide, South

Australia.

Bostrom, L., Wladis, A., & Nilsson, B. (2002). Injured moped riders who required

admission to hospital in Sweden from 1987 to 1994. European Journal of

Surgery, 168(6), 360-365.

Bowdler, J. (2008). Editorial: Happy days are here. Scooter, 21, 3.

Bowdler, J. (2011a). Do as I say, not as I do. Two Wheels, 02/11, 113-114.

Bowdler, J. (2011b). Things to come. Two Wheels, 02/11, 54-60.

Brisbane City Council. (2010). Inner city motorcycle parking. Retrieved 5 May,

2011, from

http://www.brisbane.qld.gov.au/2010%20Library/2009%20PDF%20and%20

Docs/3.Traffic%20and%20Transport/3.1%20Parking/2009_motorcycle_parki

ng_map.pdf

Brisbane City Council. (2011). Regulated parking fees. Retrieved 27 April, 2011,

from http://www.brisbane.qld.gov.au/traffic-transport/parking/parking-

meters/Regulatedparkingfees/index.htm

Broughton, P., & Walker, L. (2009). Motorcycling and Leisure: Understanding the

Recreational PTW Rider. Farnham: Ashgate Publishing.

Brown, C. V. R., Hejl, K., Bui, E., Tips, G., & Coopwood, B. (2011). Risk factors for

riding and crashing a motorcycle unhelmeted. The Journal of Emergency

Medicine, 44(6), 441-446.

Brown, I. (2005). Review of the ‘Looked but Failed to See’ Accident Causation

Factor (No. 60). London: Department for Transport.

Buche, T., Williams, S., & Ochs, R. (2010). MSF RETS: A system designed to

succeed. Paper presented at the International Conference on the Safety and

Mobility of Vulnerable Road Users, Jerusalem, Israel. Retrieved 24 August,

2010, from http://www.msf-usa.org/vru/MSF_RETS-

A_System_Designed_to_Succeed.pdf

Canada Safety Council. (2009). Safety and the motorcycle rider. Retrieved 17 July,

2009, from http://www.safety-council.org/info/traffic/mtp.html

Chang, H.-L., & Wu, S.-C. (2008). Exploring the vehicle dependence behind mode

choice: Evidence of motorcycle dependence in Taipei. Transportation

Research Part A: Policy and Practice, 42(2), 307-320.

http://www.brisbane.qld.gov.au/2010%20Library/2009%20PDF%20and%20Docs/3.Traffic%20and%20Transport/3.1%20Parking/2009_motorcycle_parking_map.pdf



http://www.brisbane.qld.gov.au/traffic-transport/parking/parking-meters/Regulatedparkingfees/index.htm

http://www.brisbane.qld.gov.au/traffic-transport/parking/parking-meters/Regulatedparkingfees/index.htm

http://www.msf-usa.org/vru/MSF_RETS-A_System_Designed_to_Succeed.pdf

http://www.msf-usa.org/vru/MSF_RETS-A_System_Designed_to_Succeed.pdf

http://www.safety-council.org/info/traffic/mtp.html


Christie, R. (2008). Analysis of involvement of scooters in crashes and their common

crash characteristics. Report produced for VicRoads Road Safety & Network

Access Division, Melbourne, Victoria.

Christie, R., & Newland, R. (2001). Motorcyclist fatality and motorcycle sales

patterns in Australia. Paper presented at the 2001 Australasian Road Safety

Research, Policing and Education Conference, Melbourne, Victoria.

Christmas, S., Young, D., Cookson, R., & Cuerden, R. (2009). Passion,

Performance, Practicality: Motorcyclists’ Motivations and Attitudes to Safety

(No. TRL-PPR442). London: Transportation Research Laboratory.

Clarke, H., & Hawkins, A. (2006). Economic framework for Melbourne traffic

planning. Agenda, 13(1), 63-80.

Comelli, M., Morandi, A., Magazzu, D., Bottazzi, M., & Marinoni, A. (2008).

Brightly coloured motorcycles and brightly coloured motorcycle helmets

reduce the odds of a specific category of road accidents: a case-control study.

Biomedical Statistics and Clinical Epidemiology, 2(1), 71-78.

Constant, A., & Lagarde, E. (2010). Protecting vulnerable road users from injury.

PLoS Medicine, 7(3), 1-4.

Corno, M., Savaresi, S. M., Tanelli, M., & Fabbri, L. (2008). On optimal motorcycle

braking. Control Engineering Practice, 16(6), 644-657.

Cossalter, V., Doria, A., Lot, R., Ruffo, N., & Salvador, M. (2003). Dynamic

properties of motorcycle and scooter tires: Measurement and comparison.

Vehicle System Dynamics, 39(5), 329 – 352.

Coxon, I. (2002). Journey to work, buzz or bore? A phenomenological, ethnographic

study of motor scooter riders in Sydney. Paper presented at the 25th

Australasian Transport Research Forum, Canberra, ACT.

de Groot, J., & Steg, L. (2006). Impact of transport pricing on quality of life,

acceptability, and intentions to reduce car use: An exploratory study in five

European countries. Journal of Transport Geography, 14(6), 463-470.

de Lapparent, M. (2006). Empirical Bayesian analysis of accident severity for

motorcyclists in large French urban areas. Accident Analysis & Prevention,

38(2), 260-268.

de Rome, L. (2006a). The injury reduction benefits of motorcycle protective clothing.

Paper presented at the NTSB Motorcycle Safety Forum, Washington, D.C.

de Rome, L. (2006b). Linking the silos: planning for motorcycle safety. Paper

presented at the International Motorcycle Safety Conference (IMSC), Long

Beach, California.

de Rome, L., Ivers, R., Haworth, N., Heritier, S., Fitzharris, M., & Du, W. (2010). A

survey of novice riders and their riding experience prior to licensing.

Transportation Research Record, 2194, 75-81.

de Rome, L., Ivers, R., Fitzharris, M., Du, W., Haworth, N., Heritier, S., et al. (2011).

Motorcycle protective clothing: Protection from injury or just the weather?

Accident Analysis & Prevention, 43(6), 1893-1900.

Delhomme, P., Chappé, J., Grenier, K., Pinto, M., & Martha, C. (2010). Reducing

air-pollution: A new argument for getting drivers to abide by the speed limit?


Department for Transport. (2004). Compendium of motorcycling statistics 2004.

London: Department for Transport.

Department for Transport. (2009). Compendium of motorcycling statistics 2009.

London: Department for Transport.


Department for Transport. (2010). Reported Road Casualties Great Britain: 2009

Annual Report. London: Department for Transport.

DeYoung, D. J., Peck, R. C., & Helander, C. J. (1997). Estimating the exposure and

fatal crash rates of suspended/revoked and unlicensed drivers in California.

Accident Analysis and Prevention, 29(1), 17-23.

DITRDLG. (2009). Road deaths Australia: 2008 statistical summary (Road Safety

Report No. 4). Canberra: Department of Infrastructure, Transport, Regional

Development and Local Government.

Duke, K. (2009). 2009 Honda DN-01 review: A marriage of scooter and motorcycle

(16 March 2009). Retrieved 29 July, 2010, from

http://www.motorcycle.com/manufacturer/2009-honda-dn01-review-quick-

ride-88080.html

Elliott, M. A., Baughan, C. J., Broughton, J., Chinn, B., Grayson, G. B., Knowles, J.,

et al. (2003). Motorcycle safety: a scoping study (TRL581). London:

Transport Research Laboratory.

ERSO. (2006). Powered two wheelers. Retrieved 22 July, 2009, from

http://www.erso.eu/knowledge/Fixed/45_PoweredTwoWheelers/powered%2

0two%20wheelers.pdf

Evangelou, S. (2003). The control and stability analysis of two-wheeled road

vehicles. Doctoral dissertation. University of London, London.

Evans, L. (1978). Improving the moped's status and safety. Journal of Traffic Safety

Education, 25(3), 10-30.

Evans, L. (2004). Traffic Safety. Bloomfield Hills: Science Serving Society.

Faberi, M., Martuzzi, M., & Pirrami, F. (2004). Assessing the health impact and

social costs of mopeds: Feasibility study in Rome. Copenhagen: World Health

Organisation.

Factor, R., Mahalel, D., & Yair, G. (2007). The social accident: A theoretical model

and a research agenda for studying the influence of social and cultural

characteristics on motor vehicle accidents. Accident Analysis and Prevention,

39(5), 914-921.

Fajans, J. (2000). Steering in bicycles and motorcycles. American Journal of Physics,

68(7), 654-659.

FCAI. (2008). A record year for motorcycle sales (17 January 2008). Retrieved 30

July, 2009, from http://www.fcai.com.au/news/2008/all/153/a-record-year-

for-motorcycle-sales-

FCAI. (2009a). Half yearly motorcycle sales result. Retrieved 30 July, 2009, from

http://www.fcai.com.au/news/2009/all/217/half-yearly-motorcycle-sales-

result

FCAI. (2009b). Motorcycle tracker: competitive position report by class. Excel file

provided upon request by the Australian Federal Chamber of Automotive

Industries, Canberra.

FCAI. (2010). Solid result for motorcycle sales in 2009. Retrieved 16 March, 2010,

from http://www.fcai.com.au/news/all/all/235/solid-result-for-motorcycle-

sales-in-2009

Fellows, N. T., & Pitfield, D. E. (2000). An economic and operational evaluation of

urban car-sharing. Transportation Research Part D: Transport and

Environment, 5(1), 1-10.

FEMA. (2007). A European agenda for motorcycle safety. Brussels: Federation of

European Motorcyclists Associations.

http://www.motorcycle.com/manufacturer/2009-honda-dn01-review-quick-ride-88080.html

http://www.motorcycle.com/manufacturer/2009-honda-dn01-review-quick-ride-88080.html

http://www.erso.eu/knowledge/Fixed/45_PoweredTwoWheelers/powered%20two%20wheelers.pdf

http://www.erso.eu/knowledge/Fixed/45_PoweredTwoWheelers/powered%20two%20wheelers.pdf

http://www.fcai.com.au/news/2008/all/153/a-record-year-for-motorcycle-sales-

http://www.fcai.com.au/news/2008/all/153/a-record-year-for-motorcycle-sales-

http://www.fcai.com.au/news/2009/all/217/half-yearly-motorcycle-sales-result

http://www.fcai.com.au/news/2009/all/217/half-yearly-motorcycle-sales-result

http://www.fcai.com.au/news/all/all/235/solid-result-for-motorcycle-sales-in-2009

http://www.fcai.com.au/news/all/all/235/solid-result-for-motorcycle-sales-in-2009


FEMA. (2008). Press release: FEMA President opens the world’s first Vision Zero

motorcycle road. Brussels: Federation of European Motorcyclists

Associations.

FEMA. (2009). Position statement: European Road Safety Observatory publication

on powered two wheelers. Brussels: Federation of European Motorcyclists

Associations.

Ferguson, E. (1997). The rise and fall of the American carpool: 1970–1990.

Transportation, 24(4), 349-376.

Gabler, H. C. (2007). The risk of fatality in motorcycle crashes with roadside

barriers. Paper presented at the 86th

Annual Meeting of the Transportation

Research Board, Washington, D.C.

Gershon, P., & Shinar, D. (2010). Motorcycle conspicuity and visibility under

various environmental conditions. Paper presented at the International

Conference on the Safety and Mobility of Vulnerable Road Users, Jerusalem,

Israel.

Goldenbeld, C., Twisk, D., & de Craen, S. (2004). Short and long term effects of

moped rider training: A field experiment. Transportation Research Part F:

Traffic Psychology and Behaviour, 7(1), 1-16.

Greig, K., Haworth, N., & Wishart, D. (2007). Identifying programs to reduce road

trauma to ACT motorcyclists. Brisbane: Centre for Accident Research and

Road Safety – Queensland.

Haque, M. M., Chin, H. C., Debnath, A. K. (2012). An investigation on multi-vehicle

motorcycle crashes using log-linear models. Safety Science 50(2), 352-362.

Harrison, W., & Christie, R. (2003). Exposure study by motorcycle make and type:

Final report. Motor Accidents Authority, New South Wales.

Harrison, W., & Christie, R. (2005). Exposure survey of motorcyclists in New South

Wales. Accident Analysis & Prevention, 37(3), 441-451.

Harrison, W., & Christie, R. (2006). Queensland motorbike usage survey 2005:

Stage 1 interim report. Unpublished report to Queensland Transport.

Haworth, N. (2003). How valid are motorcycle safety data? Paper presented at the

2003 Australasian Road Safety Research, Policing and Education

Conference, Sydney, New South Wales.

Haworth, N. (2006). Integrating policy approaches for vulnerable road users. Paper

presented at the 29th Australasian Transport Research Forum, Gold Coast,

Queensland.

Haworth, N., de Rome, L., Varnsverry, P., & Rowden, P. (2007). Motorcycle

protective clothing: Are stars better than standards? Paper presented at the

2007 Australasian Road Safety Research, Policing and Education

Conference, Melbourne, Victoria.

Haworth, N., Greig, K., & Nielson, A. (2009). A comparison of risk taking in moped

and motorcycle crashes. Transportation Research Record, 2140, 182-187.

Haworth, N., Greig, K., & Wishart, D. (2007). Motorcycle and scooter training and

licensing (No. RSD-0367). Brisbane: Centre for Accident Research and Road

Safety – Queensland.

Haworth, N., Greig, K., & Wishart, D. (2008). Moped and motor scooter licensing

and training: Current approaches and future challenges. Paper presented at

the 2008 Australasian Road Safety Research, Policing and Education

Conference, Adelaide, South Australia.

Haworth, N., & Mulvihill, C. (2005). Review of motorcycle licensing and training

(No. 240). Melbourne: Monash University Accident Research Centre.


Haworth, N., & Mulvihill, C. (2006). A comparison of hazard perception and

responding in car drivers and motorcyclists. Paper presented at the

International Motorcycle Safety Conference (IMSC), Long Beach, California.

Haworth, N., Mulvihill, C., & Clark, B. (2006). Motorbike safety in Queensland

technical paper (unreleased report). Melbourne: Monash University Accident

Research Centre.

Haworth, N., Mulvihill, C., & Rowden, P. (2006). Teaching old dogs new tricks?

Training and older motorcyclists. Paper presented at the 2006 Australasian

Road Safety Research, Policing and Education Conference, Gold Coast,

Queensland.

Haworth, N., Mulvihill, C., & Symmons, M. (2002). Motorcycling after 30 (No.

192). Melbourne: Monash University Accident Research Centre.

Haworth, N., Mulvihill, C., Wallace, P., Symmons, M., & Regan, M. (2005). Hazard

perception and responding by motorcyclists – Summary of background,

literature review and training methods (No. 234). Melbourne: Monash

University Accident Research Centre.

Haworth, N., & Nielson, A. (2008). Motor scooters and mopeds: Are increasing sales

translating into increasing crashes? Transportation Research Record, 2074,

69-76.

Haworth, N., Nielson, A., & Greig, K. (2008). Moped crashes in Queensland.

Journal of the Australasian College of Road Safety, 19(3), 31-37.

Haworth, N., & Rodney, G. Psychosocial and licensing factors influencing the use of

mopeds and motor scooters for commuting. Unpublished research, Centre for

Accident Research and Road Safety – Queensland.

Haworth, N., & Rowden, P. (2006). Fatigue in motorcycle crashes: Is there an

issue? Paper presented at the 2006 Australasian Road Safety Research,

Policing and Education Conference, Gold Coast, Queensland.

Haworth, N., & Rowden, P. (2010). Challenges in improving the safety of learner

motorcyclists. Paper presented at the 20th Canadian Multidisciplinary Road

Safety Conference, Niagara Falls, Ontario.

Haworth, N., & Schulze, M. T. (1996). Motorcycle crash countermeasures:

Literature review and implementation workshop (No. 87). Melbourne:

Monash University Accident Research Centre.

Haworth, N., Smith, R., Brumen, I., & Pronk, N. (1997). Case-control study of

motorcycle crashes (No. CR 174). Canberra: Federal Office of Road Safety.

Hedlund, J. (2011). Motorcyclist traffic fatalities by state - 2010 preliminary data.

Washington, D.C.: Governors Highway Safety Association.

Huang, B., & Preston, J. (2004). A literature review on motorcycle collisions: Final

report. Oxford: University of Oxford Transport Studies Unit.

Hurt, H. H., Ouellet, J. V., & Thom, D. R. (1981). Motorcycle accident cause factors

and identification of countermeasures, Final report (No. DOT-HS-F-01160).

Los Angeles: University of Southern California Los Angeles.

Ibrahim, S. A. K., Radin, U. R. S., Habshah, M., Kassim, H., Stevenson, M., &

Hariza, A. (2006). Mode choice model for vulnerable motorcyclists in

Malaysia. Traffic Injury Prevention, 7(2), 150 – 154.

IRTAD. (2010). IRTAD database, June 2010 - fatalities by road use. Publication.

International Transport Forum. Retrieved 23 August, 2010, from

http://internationaltransportforum.org/irtad/pdf/roaduse.pdf

Jacobson, M. Z. (2008). On the causal link between carbon dioxide and air pollution

mortality. Geophysical Research Letters, 35(3), L03809.

http://internationaltransportforum.org/irtad/pdf/roaduse.pdf


Jamson, S., & Chorlton, K. (2009). The changing nature of motorcycling: Patterns of

use and rider characteristics. Transportation Research Part F: Traffic

Psychology and Behaviour, 12(4), 335-346.

Johnston, P., Brooks, C., & Savage, H. (2008). Monograph 20 - Fatal and serious

road crashes involving motorcyclists. Canberra: Australian Government

Department of Infrastructure, Transport, Regional Development and Local

Planning.

Kaiser, F. G., Frick, J., & Stoll-Kleemann, S. (2001). Accuracy of self-reports:

Validating the general ecological behavior scale. Diagnostica, 47(2), 88-95.

Kennedy, R. (2007). Scooters on campus: Responding to the sudden growth in use of

a "new" vehicle at the University of Wisconsin-Madison. Paper presented at

the 86th

Annual Meeting of the Transportation Research Board, Washington,

D.C.

Kim, K., & Boski, (2001). Finding fault in motorcycle crashes in Hawaii:

Environmental, temporal, spatial, and human factors. Transportation

Research Record 1779(01-2295), 182-188.

Kim, K., & Levine, N. (1996). Using GIS to improve highway safety. Computers,

Environment and Urban Systems, 20(4-5), 289-302.

Kim, K., Pant, P., & Yamashita, E. (2010). Accidents and accessibility: Measuring

the influences of demographic and land use variables in Honolulu, Hawaii.

Paper presented at the 89th

Annual Meeting of the Transportation Research

Board, Washington, D.C.

Kim, K., Takeyama, D., & Nitz, L. (1995). Moped safety in Honolulu, Hawaii.

Journal of Safety Research, 26(3), 177-185.

Koornstra, M., Lynam, D., Nilsson, G., Noordzij, P., Petterson, H., Wegman, F., et

al. (2002). SUNflower: A comparative study of the development of road safety

in Sweden, the United Kingdom, and the Netherlands. Leidscendam: SWOV

Institute for Road Safety Research.

Kopjar, B. (1999). Moped injuries among adolescents: A significant forgotten

problem? Accident Analysis & Prevention, 31(5), 473-478.

Krueger, R. A. (1998). Moderating Focus Groups: Focus Group Kit 4. Thousand

Oaks: Sage Publications.

Krueger, R. A. (2006). Analysing focus group interviews. Journal of Wound, Ostomy

and Continence Nursing, 33(5), 478-481.

Lajunen, T., & Summala, H. (2003). Can we trust self-reports of driving? Effects of

impression management on driver behaviour questionnaire responses.

Transportation Research Part F: Traffic Psychology and Behaviour, 6(2), 97-

107.

Langley, J., Mullin, B., Jackson, R., & Norton, R. (2000). Motorcycle engine size

and risk of moderate to fatal injury from a motorcycle crash. Accident

Analysis and Prevention, 32(5), 659-663.

Lardelli-Claret, P., Jimenez-Moleon, J. J., de Dios Luna-del-Castillo, J., Garcia-

Martin, M., Bueno-Cavanillas, A., & Galvez-Vargas, R. (2005). Driver

dependent factors and the risk of causing a collision for two wheeled motor

vehicles. Injury Prevention, 11(4), 225-231.

Lateef, F. (2002). Riding motorcycles: Is it a lower limb hazard? Singapore Medical

Journal, 43(11), 566-569.


Lee, J., Chong, C., & Gitano, H. (2010). Analysis of motorcycle fuel consumption in

Malaysia. Paper presented at the 2010 Small Engine Technology Conference

and Exposition. Retrieved 11 May, 20011, from http://papers.sae.org/2010-

32-0048/

Lin, M.-R., & Kraus, J. F. (2009). A review of risk factors and patterns of motorcycle

injuries. Accident Analysis & Prevention, 41(4), 710-722.

Litman, T. (2003). Integrating public health objectives in transportation decision-

making. American Journal of Health Promotion, 18(1), 103-108.

Liu, B. C., Ivers, R., Norton, R., Boufous, S., Blows, S., & Lo, S. K. (2008). Helmets

for preventing injury in motorcycle riders. Cochrane Database of Systematic

Reviews(1) No. CD004333.

Mackett, R. L., & Ahern, A. (2000). Potential for mode transfer of short trips:

Report on the analysis of the survey results. London: University College

London.

Madson, B. (2010). Motorcycle sales down 40.8% says MIC. Retrieved 16 March,

2010, from http://www.motorcycle-usa.com/2/5588/Motorcycle-

Article/Motorcycle-Sales-Down-40-8--Says-MIC.aspx

MAG UK. (2006). How close is too close? Concerning car collisions and

motorcycles. Rugby: Motorcycle Action Group UK.

Matzsch, T., & Karlsson, B. (1986). Moped and motorcycle accidents: Similarities

and discrepancies. Journal of Trauma, 26(66), 538-543.

McHugh, T., & Stinson, E. (1984). Moped injuries. Annals of Emergency Medicine,

13(1), 35-39.

MIC. (2011). U.S Motorcycle sales rise during the first quarter - economical two-

wheelers lead the increase. Irvine: Motorcycle Industry Council.

Mihailovic, B. (2010). Who's vulnerable? Australian Motorcycle News, 60(3), 99.

Morris, C. (2009). Bureau of Transportation Statistics Special Report: Motorcycle

trends in the United States. Washington, D.C.: U.S. Department of

Transportation.

Moskal, A., Martin, J.-L., & Laumon, B. (2010). Risk factors for injury accidents

among moped and motorcycle riders. Accident Analysis & Prevention, In

Press, Corrected Proof.

Moskal, A., Martin, J., Lenguerrand, E., & Laumon, B. (2007). Injuries among

motorized two-wheelers in relation to vehicle and crash characteristics in

Rhone, France. Lyon: University of Lyon.

Motorcycle and Moped Industry Council. (2009). 2008 Motorcycle, scooter and all-

terrain vehicle annual industry statistics report. Ontario: Motorcycle and

Moped Industry Council.

Motorcycle Safety Foundation. (2000). National Agenda for Motorcycle Safety

(NAMS). Irvine, CA: Motorcycle Safety Foundation.

Motorcycle Safety Foundation. (2010). Scooter School. Retrieved 22 January, 2011,

from http://www.msf-usa.org/scooterschool.cfm

Muller, A. (1982). An evaluation of the effectiveness of motor cycle daytime

headlight laws. American Journal of Public Health, 72(10), 1136-1141.

Mullin, B., Jackson, R., Langley, J., & Norton, R. (2000). Increasing age and

experience: Are both protective against motorcycle injury? A case-control

study. Injury Prevention, 6(1), 32-35.

http://papers.sae.org/2010-32-0048/

http://papers.sae.org/2010-32-0048/

http://www.motorcycle-usa.com/2/5588/Motorcycle-Article/Motorcycle-Sales-Down-40-8--Says-MIC.aspx

http://www.motorcycle-usa.com/2/5588/Motorcycle-Article/Motorcycle-Sales-Down-40-8--Says-MIC.aspx

http://www.msf-usa.org/scooterschool.cfm


Musso, A., Vuchic, V. R., Bruun, E., & Corazza, M. V. (2010). A research agenda

for public policy towards motorized two-wheelers in urban transport. Paper

presented at the 89th Annual Meeting of the Transportation Research Board,

Washington, D.C.

Myers, J. S., & Ridout, J. S. (2010). The use of low speed vehicles to achieve energy

consumption reductions. Paper presented at the 89th Annual Meeting of the

Transportation Research Board, Washington, D.C.

Naci, H., Chisholm, D., & Baker, T. D. (2009). Distribution of road traffic deaths by

road user group: A global comparison. Injury Prevention, 15(1), 55-59.

Natalier, K. (2001). Motorcyclists' Interpretations of risk and hazard. Journal of

Sociology, 37(1), 65-80.

NHTSA. (2007). Traffic safety facts 2006 data: Motorcycles. Washington, D.C.:

National Highway Traffic Safety Administration.

NHTSA. (2009). 2008 Traffic safety annual assessment - Highlights (No. DOT HS

811 172). Washington D.C.: National Highway Traffic Safety Administration

NHTSA. (2010). Highlights of 2009 motor vehicle crashes. Washington D.C.:

National Highway and Traffic Administration.

Nja, O., & Nesvag, S. M. (2007). Traffic behaviour among adolescents using mopeds

and light motorcycles. Journal of Safety Research, 38(4), 481-492.

Noordzij, P., Forke, E., Brendicke, R., & Chinn, B. (2001). Integration of needs of

moped and motorcycle riders into safety measures (No. D-2001-5).

Leidschendam: SWOV Institute for Road Safety Research.

NSW Roads and Traffic Authority. (1997). Motor vehicle compliance plates.

Retrieved 10 May, 2011, from

http://studentweb.usq.edu.au/home/q1121625/NSW%20vehicle%20guide%2

0lines/vsi19.pdf

Ogilvie, D., Egan, M., Hamilton, V., & Petticrew, M. (2004). Promoting walking and

cycling as an alternative to using cars: Systematic review. British Medical

Journal, 329(7469), 763-768.

Otte, D., Willeke, H., Chinn, B., Doyle, D., & Schuller, E. (1998). Impact

mechanisms of helmet protected heads in motorcycle accidents – accident

study of COST 327. Paper presented at the Safety Environment Future II:

1998 International Motorcycle Conference (IFZ No.8).

Paine, M., Paine, D., Haley, J., & Cockfield, S. (2005). Daytime running lights for

motorcycles. Paper presented at the 19th International Technical Conference

on the Enhanced Safety of Vehicles, Washington D.C.

Paulozzi, L. J. (2005). The role of sales of new motorcycles in a recent increase in

motorcycle mortality rates. Journal of Safety Research, 36(4), 361-364.

Perez, K., Mari-Dell'Olmo, M., Borrell, C., M, N., Villalbi, J., Santamarina, E., et al.

(2009). Road injuries and relaxed licensing requirements for driving light

motorcycles in Spain: A time series analysis. Bulletin of the World Health

Organisation, 87(7), 497-504.

Potts, I., Garets, S., Smith, T., Pfefer, R., Neuman, T. R., Slack, K. L., et al. (2008).

A guide for addressing collisions involving motorcycles. Washington, D.C.:

Transportation Research Board.

PTUA. (2010). Common myths about urban transport. Retrieved 4 July, 2011, from

http://www.ptua.org.au/myths/

Puerto, L., Ballbé, A., Albalate, D., & Fernández, L. (2009). La accidentalidad de las

motocicletas en zona urbana: Barcelona 200 - 2007. Barcelona: University of

Barcelona.

http://studentweb.usq.edu.au/home/q1121625/NSW%20vehicle%20guide%20lines/vsi19.pdf

http://studentweb.usq.edu.au/home/q1121625/NSW%20vehicle%20guide%20lines/vsi19.pdf

http://www.ptua.org.au/myths/


Purvis, B. (2010, 31 March - 13 April). Electric blues. Australian Motorcycle News,

59(19), 56-60.

Quddus, M. A., Noland, R. B., & Chin, H. C. (2002). An analysis of motorcycle

injury and vehicle damage severity using ordered probit models. Journal of

Safety Research 33(4), 445-462.

Queensland Transport. (2008). Motorbike safety in Queensland - consultation paper.

Retrieved 29 June, 2009, from

http://www.transport.qld.gov.au/resources/file/ebc13c098517a69/Pdf_motorb

ike_safety_consultation_paper_v3.pdf

Reeder, A. I., Alsop, J. C., Langley, J. D., & Wagenaar, A. C. (1999). An evaluation

of the general effect of the New Zealand graduated driver licensing system on

motorcycle traffic crash hospitalisations. Accident Analysis & Prevention,

31(6), 651-661.

Round, M. (2010). Cheap and green. Scooter, 23.

Rowden, P., Watson, B., & Haworth, N. (2007). What can riders tell us about

motorcycle rider training? A view from the other side of the fence. Paper

presented at the 2007 Australasian Road Safety Research, Policing and

Education Conference, Melbourne, Victoria.

Rowden, P., Watson, B., Wishart, D., & Schonfeld, C. (2009). Changing motorcycle

rider safety attitudes and motives for risk taking: Process evaluation of a

rider training intervention. Paper presented at the 2009 Australasian Road

Safety Research, Policing and Education Conference, Sydney, New South

Wales.

Rutter, D. R., & Quine, L. (1996). Age and experience in motorcycling safety.


SafetyNet. (2009). Powered Two Wheelers Web text. Retrieved 9 March, 2010, from

http://ec.europa.eu/transport/road_safety/specialist/knowledge/pdf/powered_t

wo_wheelers.pdf

Savolainen, P., & Mannering, F. (2007). Probabilistic models of motorcyclists' injury

severities in single- and multi-vehicle crashes. Accident Analysis &

Prevention, 39(5), 955-963.

Schneider, W., Savolainen, P., Van Boxel, D. & Beverley, R. (2012). Examination of

factors determining fault in two-vehicle motorcycle crashes. Accident

Analysis & Prevention, 45(2012), 669-676.

Schoon, C. (2004). Traffic legislation and safety in Europe concerning the moped

and the A1 category (125 cc) motorcycle. Leidschendam: SWOV Institute for

Road Safety Research.

Schulze, H., & Koßmann, I. (2010). The role of safety research in road safety

management. Safety Science, 48(9), 1160-1166.

Sexton, B., Baughan, C., Elliott, M., & Maycock, G. (2004). The accident risk of

motorcyclists (No. TRL607). London: Transport Research Laboratory.

Sexton, B., Hamilton, K., Baughan, C., Stradling, S., & Broughton, P. (2006). Risk

and motorcyclists in Scotland. Edinburgh: Scottish Executive.

Shattuck, C., & Peterson, E. (2005). Scooters: Red Eyes, Whitewalls and Blue

Smoke. Golden: Speck Press.

Sheehan, M., Siskind, V., Turner, R., Veitch, C., O’Connor, T., Steinhardt, D., et al.

(2008). Rural and Remote Road Safety Research Project: Final Report

(Monograph 4) Brisbane: Centre for Accident Research and Road Safety –

Queensland.

http://www.transport.qld.gov.au/resources/file/ebc13c098517a69/Pdf_motorbike_safety_consultation_paper_v3.pdf

http://www.transport.qld.gov.au/resources/file/ebc13c098517a69/Pdf_motorbike_safety_consultation_paper_v3.pdf

http://ec.europa.eu/transport/road_safety/specialist/knowledge/pdf/powered_two_wheelers.pdf

http://ec.europa.eu/transport/road_safety/specialist/knowledge/pdf/powered_two_wheelers.pdf


South Gloucestershire Council. (2008). The 'sorry mate I didn't see you' campaign.

Retrieved 23 October, 2008, from http://www.smidsy.co.uk/

Steg, L. (2005). Car use: Lust and must. Instrumental, symbolic and affective

motives for car use. Transportation Research Part A: Policy and Practice,

39(2-3), 147-162.

Steg, L., Geurs, K., & Ras, M. (2001). The effects of motivational factors on car use:

A multidisciplinary modelling approach. Transportation Research Part A:

Policy and Practice, 35(9), 789-806.

Steg, L., & Tertoolen, G. (1999). Sustainable transport policy: The contribution from

Behavioural Scientists. Public Money and Management, 19(1), 63-69.

Steg, L., & van Brussel, A. (2009). Accidents, aberrant behaviours, and speeding of

young moped riders. Transportation Research Part F: Traffic Psychology

and Behaviour, 12, 503-511.

SWOV. (2006a). Advancing Sustainable Safety: National Road Safety Outlook for

2005 - 2020. Leidschendam: SWOV Institute for Road Safety Research.

SWOV. (2006b). Fact sheet: Young moped riders. Leidschendam: SWOV Institute

for Road Safety Research.

SWOV. (2009). Moped and light-moped riders. Leidschendam: SWOV Institute for

Road Safety Research.

Teoh, E. R., & Campbell, M. (2010). Role of motorcycle type in fatal motorcycle

crashes. Journal of Safety Research, 41(6), 507-512.

TMR. (2009). Queensland motorcycle safety strategy. Retrieved 24 September,

2009, from http://www.tmr.qld.gov.au/~/media/15a8d250-8918-4067-b136-

513da8c6412c/pdf_motorcycle_safety_strategy_2009_2012_complete_may0

9.pdf

TMR. (2010). Motorcycles on register in Queensland as at 30 June 1922 to 2009.

Retrieved 2 September, 2010, from

http://www.tmr.qld.gov.au/~/media/safety/transport-and-road-

statistics/registration/pdf_stats_motorcycles_on_register_queensland.pdf

TMR. (2011a). Queensland road toll weekly report (675) (No. 675). Brisbane:

Queensland Department of Transport and Main Roads.

TMR. (2011b). Number plates. Retrieved 11 June, 2011, from

http://www.tmr.qld.gov.au/Registration/Number-plates.aspx

Tomaskovic-Devey, D., Pfaff Wright, C., Czaja, R., & Miller, K. (2006). Self-reports

of police speeding stops by race: Results from the North Carolina Reverse

Record Check Survey. Journal of Quantitative Criminology, 22(4), 279.

Tubre, A. H., Bell, S. T., Arthur, W., Edwards, B. D., Tubre, T. C., & Day, E. A.

(2005). Convergence of self-report and archival crash involvement data: a

two-year longitudinal follow-up. Human Factors, 47(2), 303-313.

Tunnicliff, D. (2006). Psychosocial factors contributing to motorcyclists' intended

riding style: An application of an extended version of the theory of planned

behaviour. Unpublished Masters dissertation, Queensland University of

Technology, Brisbane.

Umar, R. S. R. (2006). Motorcycle safety programmes in Malaysia: How effective

are they? International Journal of Injury Control and Safety Promotion,

13(2), 71-79.

Vanlaar, W., Mayhew, D., Marcoux, K., Wets, G., Brijs, T., & Shope, J. (2009). An

evaluation of graduated driver licensing programs in North America using a

meta-analytic approach. Accident Analysis & Prevention, 41(5), 1104-1111.

http://www.smidsy.co.uk/

http://www.tmr.qld.gov.au/~/media/15a8d250-8918-4067-b136-513da8c6412c/pdf_motorcycle_safety_strategy_2009_2012_complete_may09.pdf



http://www.tmr.qld.gov.au/~/media/safety/transport-and-road-statistics/registration/pdf_stats_motorcycles_on_register_queensland.pdf

http://www.tmr.qld.gov.au/~/media/safety/transport-and-road-statistics/registration/pdf_stats_motorcycles_on_register_queensland.pdf

http://www.tmr.qld.gov.au/Registration/Number-plates.aspx


Vick, M. (2006). Poststructuralist theory and methodology: A complementary

approach to road safety research. Paper presented at the 2006 Australasian

Road Safety Research, Policing and Education Conference, Gold Coast,

Queensland.

Watson, B. (2004). The psychosocial characteristics and on-road behaviour of

unlicensed drivers. Unpublished Doctoral dissertation, Queensland

University of Technology, Brisbane.

Watson, B., & Steinhardt, D. (2006). A comparison of the crash Involvement of

unlicensed motorcycle riders and unlicensed drivers in Queensland. Paper

presented at the 2006 Australasian Road Safety Research, Policing and

Education Conference, Gold Coast, Queensland.

Watson, B., & Steinhardt, D. (2007). The long-term crash involvement of unlicensed

drivers and riders in Queensland, Australia. Paper presented at the

International Council on Alcohol, Drugs, and Traffic Safety (ICADTS),

Seattle, WA.

Watson, B., Tunnicliff, D., White, K., Schonfeld, C., & Wishart, D. (2007).

Psychological and social factors influencing motorcycle rider intentions and

behaviour (No. RSRG 2007-04). Canberra: Australian Transport Safety

Bureau.

Wegman, F., & Aarts, L. (2006). Advancing Sustainable Safety: National Road

Safety Outlook for 2005-2020. Leidschendam: SWOV Institute for Road

Safety Research.

Weinstein, N. (1993). Testing four competing theories of health-protective

behaviour. Health Psychology, 12(4), 324-333.

Wells, S., Mullin, B., Norton, R., Langley, J., Connor, J., Lay-Yee, R., et al. (2004).

Motorcycle rider conspicuity and crash related injury: Case-control study.

British Medical Journal, 328(7444), BMJ 328:857.

Wigan, M. R. (2000). Motorcycle transport: Powered two wheelers in Victoria. (No.

2000-1-1). Melbourne: Victorian Motorcycle Advisory Council.

Wigan, M. R., & Carter, A. J. (1980). Mopeds and the Australian user profile. Paper

presented at the International Motorcycle Safety Conference (IMSC),

Washington, D.C.

Wong, J.-T., Chung, Y.-S., & Huang, S.-H. (2010). Determinants behind young

motorcyclists' risky riding behavior. Accident Analysis & Prevention, 42(1),

275-281.

World Health Organization. (2009). Global status report on road safety: Time for

action. Geneva: World Health Organization.

Yannis, G., Golias, J., & Papadimitriou, E. (2005). Driver age and vehicle engine

size effects on fault and severity in young motorcyclists accidents. Accident

Analysis & Prevention, 37(2), 327-333.

Zador, P. (1985). Motorcycle headlight-use laws and fatal motorcycle crashes in the

US, 1975-83. American Journal of Public Health, 75(5), 543-546.


APPENDICES

Appendix A1 Moped rider licensing in European countries .................................... 276

Appendix B1 Brisbane City Council designated motorcycle parking areas ............ 277

Appendix C1 Queensland PTW crash distribution, July 2003-June 2008 ............... 278

Appendix C2 Southeast Queensland PTW crash distribution .................................. 279

Appendix C3 Central Queensland PTW crash distribution...................................... 280

Appendix C4 North Queensland PTW crash distribution ........................................ 281

Appendix C5 Moped crashes in Brisbane Statistical Local Areas (SLAs) .............. 282

Appendix C6 Moped crashes in Gold Coast Statistical Local Areas ....................... 283

Appendix C7 Moped crashes in Sunshine Coast Statistical Local Areas ................ 284

Appendix C8 Moped crashes in Fraser/Coral Coast Statistical Local Areas ........... 285

Appendix C9 Moped crashes in Mackay/Whitsunday Statistical Local Areas ........ 286

Appendix C10 Moped crashes in Townsville Statistical Local Areas ....................... 287

Appendix C11 Moped crashes in Cairns Statistical Local Areas ............................... 288

Appendix D1 Study 3a recruitment flyer ................................................................. 289

Appendix D2 Study 3a information sheet and consent form .................................... 290

Appendix D3 Study 3a guiding questions for focus groups ..................................... 292

Appendix D4 Study 3b Queensland Scooter and Moped Rider Survey

Questionnaire ..................................................................................... 295


Appendix A1: Moped rider licensing in European countries 2003

Source: SWOV report R-2004-10 (Schoon, 2004).


Appendix B1: Brisbane City Council designated motorcycle parking areas

Source: Brisbane City Council (2010)

http://www.brisbane.qld.gov.au/2010%20Library/2009%20PDF%20and%20Docs/3.Traffic%20and%

20Transport/3.1%20Parking/2009_motorcycle_parking_map.pdf




Appendix C1: Queensland PTW crash distribution, July 2003 – June 2008


Appendix C2: Southeast Queensland PTW crash distribution


Appendix C3: Central Queensland PTW crash distribution


Appendix C4: North Queensland PTW crash distribution


Appendix C5: Moped crashes in Brisbane Statistical Local Areas (SLAs)


Appendix C6: Moped crashes in Gold Coast Statistical Local Areas


Appendix C7: Moped crashes in Sunshine Coast Statistical Local Areas


Appendix C8: Moped crashes in Fraser/Coral Coast Statistical Local Areas


Appendix C9: Moped crashes in Mackay/Whitsunday Statistical Local Areas


Appendix C10: Moped crashes in Townsville Statistical Local Areas


Appendix C11: Moped crashes in Cairns Statistical Local Areas


Appendix D1: Study 3a recruitment flyer


Appendix D2: Study 3a information sheet and consent form


Appendix D3: Guiding questions for focus groups

How long have you been riding mopeds/scooters?

Are you the owner of the moped/scooter you mostly ride?

o What make/model/type is it?

What are the reasons people ride scooters/mopeds?

What is the main reason you ride?

o Transport or recreation, in what proportions?

o Expand: Work, education, shopping, other?

Where do you mostly ride?

o To/from?

o What types of roadways ridden (i.e. backstreets, arterial, motorway,

racetrack, other)?

o What types of roadways preferred?

o Where else do you ride?

Can you tell me what people like about riding mopeds/scooters?

What do people dislike about riding mopeds/scooters?

Do you prefer riding alone, with one or two others, or in larger groups?

o Why?

Is riding with other moped/scooter riders competitive?

o If so, to what extent?

What kinds of things do you consider risky in terms of riding?

Do you deliberately take risks when riding?

o If so, how often?

o If so, what kind of risks?

Overall, how safe do feel when riding?

o Breakdown by road, traffic and environmental conditions as necessary

What hazards do you come across while riding?

o Other road users?

Cars

Trucks

Cyclists

Pedestrians

Motorcyclists

Other scooterists

Other

o Environmental

Poor road surface

Poor delineation

Contaminated surface

Roadside furniture

Weather conditions

Other


In what situations do you feel safest?

o Why?

In what situations do feel most unsafe?

o Why?

What kinds of things do you hear from other riders about safety?

o To what extent do you agree with what they say?

From where do you get information about safety issues relating to

scooter/moped riding?

Do you get information about other road and traffic issues related to

scooter/moped riding?

o If so, from what sources?

Do/did any of your family or relatives ride mopeds/scooters or motorcycles?

o If so, which?

Do you own a car?

Do you regularly use any of the following other transport modes?

o Car as driver

o Car as passenger

o Public transport

o Pedal cycling

o Walking

If you did not ride a moped/scooter,

o Would you still make the regular trips you presently make using your

moped/scooter?

o If you would, which different transport mode/s would you most likely

use instead?

Do weather conditions on the day/night affect your decision to ride?

What do you think about rider training programs?

o Have you taken any?

What do you think about rider education programs?

o Have you taken any?

What do other riders you talk to say about rider training and education?

Do you think training and/or education should be compulsory to gain a

motorcycle licence?

Should moped riders require a motorcycle licence?

Do scooterists feel any affinity with motorcyclists generally?

Do you think moped/scooter riding is a phase people ‘move out of’?

o Do some see it as a stepping stone to ‘motorcycling’?

What circumstances make people stop riding mopeds/scooters?

Do you ride any other type of motorcycle, other than moped/scooter?

o Do you have any desire to?

How important is vehicle performance (power, braking, handling etc) to

scooterists generally?


How much does the average rider know about after market performance

modifications?

o Do many people modify a moped/scooter to change its performance?

o If so, how/what/ for what purpose?

How do riders perceive the attitude of government toward moped/scooter

riding?

o Is this attitude seen to be reflected in government policy?

How do riders perceive the attitude of police toward moped/scooter riders?

How do you perceive the attitude of the following motorists toward

moped/scooter riders?

o Motorcyclists?

o Car drivers?

o Bus drivers?

o Truck drivers?

Do you think your perceptions are similar to those of other riders?

What do you think about scooterists’ use of protective clothing?

o Is fashion more or less important than protective value to the average

rider?

o How do riders decide which items are best in terms of protective

value?

Does use of protective clothing depend on the weather?

o If yes, to what extent?

How beneficial do you consider is protective clothing in the event of a crash

Is there anything else you want to say about scooter safety and related

issues?


Appendix D4: Queensland Scooter and Moped Rider Survey Questionnaire