Factors affecting the intention to use a personal carbon trading system

Southern Cross UniversityePublications@SCU

Theses

2014

Factors affecting the intention to use a personalcarbon trading systemAlexander David HendrySouthern Cross University

ePublications@SCU is an electronic repository administered by Southern Cross University Library. Its goal is to capture and preserve the intellectualoutput of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around theworld. For further information please contact [email protected].

Publication detailsHendry, A 2014, 'Factors affecting the intention to use a personal carbon trading system', MBus thesis, Southern Cross University,Lismore NSW.Copyright A Hendry 2014

http://epubs.scu.edu.au

http://epubs.scu.edu.au/theses

mailto:[email protected]

1

Factors Affecting the Intention to

Use a Personal Carbon Trading

System

Master of Business

Alexander Hendry

BIT (Information Systems)

Southern Cross Business School

Southern Cross University

Gold Coast

20 May 2014

2

Certification

I certify that the work presented in this thesis is, to the best of my knowledge and belief,

original, except as acknowledged in the text, and that the material has not been submitted,

either in whole or in part, for a degree at this or any other university.

I acknowledge that I have read and understood the University's rules, requirements,

procedures and policy relating to my higher degree research award and to my thesis. I

certify that I have complied with the rules, requirements, procedures and policy of the

University (as they may be from time to time).

Alex Hendry

20 May 2014

3

Abstract

The aims of the Norfolk Island Carbon Health Evaluation (NICHE) project are:

To assess whether personal carbon allowances are effective in reducing an

individual's carbon footprint and what impact this has on health behaviours

associated with obesity such as active transport and diet.

To assess the political acceptability of introducing personal carbon trading (PCT) into

a population as a tool for reducing emission reductions

To explore the relationship between positive health behaviours and pro-

environmental behaviours.

This thesis covers a sub section of the NICHE project and aims to understand how an

individual’s attitudes and behaviours associated with their health and well-being, the

environment, carbon emissions and climate change affect their personal carbon trading

system (PCTS) usage intentions.

After reviewing all of the available literature a conceptual model was proposed that

hypothesizes that PCTS usage intentions are influenced by:

Self-health evaluation

Attitudes and behaviours towards health

Attitudes and behaviours towards the environment

Attitudes and behaviours towards carbon emissions and climate change

Data was collected via a survey that was hand delivered to every house on Norfolk Island.

The first stage of the statistical analysis examined the distribution of respondents and

analyzed the responses to each section of the survey that corresponded with the constructs

of the conceptual model to see if they meet with researchers expectations.

The second stage of the statistical analysis examined the relationships amongst key

variables from the study. The initial examination of the relationships between variables was

undertaken using correlations analysis. Following the correlation analysis, Exploratory

Factor Analysis (EFA) was run on the blocks of variables expected to have an underlying

4

factor structure as per the design of the survey and the directionality of the relationships

was examined using linear regression analysis.

The statistical analysis resulted in some slightly different structures and the addition of a

further construct (‘Optimism’) that takes into account the perceived impact that technology

could have in relation to improving health and environmental change. The constructs

discovered during the statistical analysis were:

'Self Health Evaluation'

'Health Consciousness'

'Environmental Action'

‘Optimism’

'Environmental Consciousness'

The constructs that were found to be significant predictors of the dependent variable

'Usage Intentions Towards a PCTS' depended on the gender and carbon footprint of the

respondent.

While there are a few limitations that need to be addressed in future research the NICHE

project has resulted in the identification of validated predictive model to measure usage

intentions towards a PCTS.

5

Acknowledgements

I would like to thank Dr Bruce Armstrong, Dr Bill Smart and Dr Garry Egger of Southern Cross

University for all of their input, guidance and support throughout the project. In particular I

would like to thank Bruce for all of the time, effort and advice he gave me while I wrote this

thesis. His help was invaluable and I could not have written this thesis without him.

I would also like to thank Garry Egger and Gary Webb for the all the work they put into the

NICHE project in the initial stages to get it up and running and Tania Christian and PJ Wilson

for all of the work, passion and local knowledge they have put into running the Norfolk

Island end of the NICHE project

Lastly I would like to Thank Mum, Dad and Fiefield for all of their support over the last

couple of years.

6

Contents

1 Introduction ..................................................................................................................... 17

1.1 Background to the Research ..................................................................................... 17

1.2 Originality of the Research ........................................................................................ 18

1.3 Literature Review and the Conceptual Model .......................................................... 19

1.4 Methodology and the Research Questions ............................................................... 22

1.5 Preview of the Research Findings ............................................................................. 24

1.6 Organization of the Thesis ........................................................................................ 25

2 Literature Review ............................................................................................................. 28

2.1 Introduction............................................................................................................... 28

2.2 Health and Obesity .................................................................................................... 28

2.3 Carbon Emissions and Climate Change ..................................................................... 32

2.4 Links Between Obesity and an Individual's Carbon Footprint .................................. 37

2.5 PCTS ........................................................................................................................... 43

2.6 Information System Success ..................................................................................... 49

2.6.1 Theory of Reasoned Action (1975, 1980) .......................................................... 51

2.6.2 Technology Acceptance Models ........................................................................ 52

2.6.3 User Satisfaction Models ................................................................................... 59

2.7 Summary ................................................................................................................... 62

3 The NICHE PCTS................................................................................................................ 64

3.1 Introduction............................................................................................................... 64

3.2 NICHE POS ................................................................................................................. 65

3.3 NICHE Server ............................................................................................................. 69

3.4 NICHE Database ......................................................................................................... 70

3.5 NICHE Administration Website ................................................................................. 70

3.6 NICHE Participants Website ...................................................................................... 71

7

3.7 Conclusion ................................................................................................................. 74

4 Methodology .................................................................................................................... 75

4.1 Introduction............................................................................................................... 75

4.2 Survey Design and Construction ............................................................................... 75

4.2.1 The Broader NICHE Project ................................................................................ 75

4.2.2 Examination of PCTS Usage Intentions .............................................................. 77

4.3 Focus Groups - Pilot Study ........................................................................................ 82

4.4 Survey Administration ............................................................................................... 83

4.5 Sample Selection ....................................................................................................... 87

4.6 Limitations of the Study and the Sample Frame ....................................................... 88

4.7 Summary ................................................................................................................... 89

5 Descriptive Analysis ......................................................................................................... 90

5.1 Introduction............................................................................................................... 90

5.2 Description of Respondents ...................................................................................... 90

5.3 Self Health Evaluation ............................................................................................... 93

5.4 Attitudes and Behaviours Towards Health ............................................................... 97

5.5 Attitudes and Behaviours Towards the Environment ............................................. 101

5.6 Attitudes and Behaviours Towards Carbon Emissions and Climate Change .......... 104

5.7 PCTS Usage Intentions ............................................................................................. 111

5.8 PCTS Usage Intentions Compared with Self-Health Evaluation .............................. 113

5.9 PCTS Usage Intentions Compared with Attitudes and Behaviours Towards Health

114

5.10 PCTS Usage Intentions Compared with Attitudes and Behaviours Towards the

Environment ....................................................................................................................... 116

5.11 PCTS Usage Intentions Compared with Attitudes and Behaviours Towards Carbon

Emissions and Climate Change ........................................................................................... 119

8

6 Data Analysis .................................................................................................................. 124

6.1 Introduction............................................................................................................. 124

6.2 Correlation Analysis................................................................................................. 124

6.2.1 Correlation Analysis - Self Health Evaluation................................................... 126

6.2.2 Correlation Analysis - Attitudes Towards Health, the Environment, Carbon

Emissions and Climate Change ...................................................................................... 128

6.2.3 Correlation Analysis - Behaviours Towards Consumption and the Environment

130

6.2.4 Correlation Analysis - PCTS Usage Intentions .................................................. 131

6.2.5 Correlation Analysis - PCTS vs. Self Health Evaluation .................................... 133

6.2.6 Correlation Analysis - PCT vs. Behaviours Towards Consumption and the

Environment................................................................................................................... 134

6.2.7 Correlation Analysis - PCT vs. Attitudes Towards Health, the Environment,

Carbon Emissions and Climate Change .......................................................................... 136

6.3 Exploratory Factor Analysis ..................................................................................... 139

6.3.1 EFA - Attitudes Towards Health, the Environment, Carbon Emissions and

Climate Change .............................................................................................................. 142

6.3.2 EFA - Behaviours Towards Consumption and the Environment...................... 148

6.3.3 EFA - PCTS Usage Intentions ............................................................................ 149

6.4 Correlations Between the factors and BMI ............................................................. 152

6.5 Linear Regression Analysis ...................................................................................... 154

6.5.1 Linear Regression Analysis – All Responses ..................................................... 155

6.5.2 Linear Regression Analysis - Gender ................................................................ 159

6.5.3 Linear Regression Analysis – Carbon Footprint ............................................... 165

6.5.4 Linear Regression Analysis – Males and Females by Carbon Footprint .......... 174

6.6 Conclusion ............................................................................................................... 188

7 Conclusion ...................................................................................................................... 192

9

7.1 Introduction............................................................................................................. 192

7.2 Results of the statistical analysis ............................................................................. 192

7.3 Limitations of the study .......................................................................................... 198

7.4 Implications and relevance for future research ...................................................... 200

7.5 Future NICHE research ............................................................................................ 201

7.6 Conclusion ............................................................................................................... 201

10

Figures and Tables

Figure 1-1 NICHE Conceptual Model - Hypothesized Influences on PCTS Usage Intentions .. 22

Figure 2-1 World Obesity Rates for Males and Females (International Association for the

Study of Obesity, 2012) ............................................................................................................ 31

Figure 2-2 Radiative forcing caused by various greenhouse gases (United States

Environmental Protection Agency, 2010, p. 18) ...................................................................... 34

Figure 2-3 Concentrations of carbon dioxide in the atmosphere from hundreds of thousands

of years ago through 2009 (United States Environmental Protection Agency, 2010, p. 17) .. 35

Figure 2-4 The Link Between Obesity and Greenhouse Gases (Delpeuch et al. 2009) ........... 42

Figure 2-5 Theory of Reasoned Action (F.D. Davis, R.P. Bagozzi, & P.R. Warshaw, 1989, p.

984) .......................................................................................................................................... 51

Figure 2-6 Technology Acceptance Model (Fred D Davis, Richard P Bagozzi, & Paul R

Warshaw, 1989, p. 984) ........................................................................................................... 53

Figure 2-7 Technology Acceptance Model 2 (Viswanath. Venkatesh & Davis, 2000, p. 188) . 56

Figure 2-8 Unified Theory of Acceptance and Use of Technology (UTAT) (V. Venkatesh,

Morris, Davis, & Davis, 2003, p. 447) ....................................................................................... 59

Figure 2-9 Delone and Mclean Information System Success Model (DeLone & McLean, 1992,

p. 87) ........................................................................................................................................ 61

Figure 2-10 Updated Delone and Mclean Information System Success Model (DeLone &

McLean, 2003, p. 24)................................................................................................................ 62

Figure 3-1 Overview of N-PCTS ................................................................................................ 65

Figure 3-2 - NICHE Card ............................................................................................................ 66

Figure 3-3 NICHE Tablet at Participating Outlet ...................................................................... 67

Figure 3-4 NICHE POS ............................................................................................................... 68

Figure 3-5 NICHE POS ............................................................................................................... 68

Figure 3-6 NICHE Admin Website ............................................................................................ 71

Figure 3-7 NICHE Participants website .................................................................................... 73

Figure 4-1 Norfolk Island Census Districts ............................................................................... 85

Figure 5-1 Distribution of Survey Respondents by Gender and Age ....................................... 91

Figure 5-2 Age Distribution of Survey Respondents vs. 2011 Census ..................................... 92

Figure 5-3 Residential Status of Survey Respondents vs. 2011 Census ................................... 92

11

Table 5-4 Response Distributions - Self Health Evaluation ...................................................... 93

Figure 5-5 Response Distributions – Self Health Evaluation and Age ..................................... 94

Figure 5-6 Response Distributions – Self Health Evaluation and Gender................................ 95

Table 5-7 Response Distributions - Body Weight .................................................................... 95

Figure 5-8 Response Distributions – Self Health Evaluation and Body Weight ....................... 96

Table 5-9 Response Distributions – Diet .................................................................................. 97

Figure 5-10 Response Distributions – Diet and Age ................................................................ 98

Figure 5-11 Response Distributions – Diet and Gender .......................................................... 98

Table 5-12 Response Distributions – Exercise ......................................................................... 99

Figure 5-13 Response Distributions – Exercise and Gender .................................................. 100

Figure 5-14 Response Distributions – Diet and Exercise ....................................................... 101

Table 5-15 Response Distributions – Environmentally Friendly Products ............................. 102

Table 5-16 Response Distributions – Turn Lights Off ............................................................ 102

Table 5-17 Response Distributions – Reduce Waste / Recycle ............................................. 102

Figure 5-18 Response Distributions – Environmentally Friendly Products and Gender ....... 103

Figure 5-19 Response Distributions – Turn Lights Off and Gender ....................................... 104

Figure 5-20 Response Distributions – Reduce Waste / Recycle and Gender ........................ 104

Table 5-21 Response Distributions – Thoughts About Climate Change ................................ 105

Figure 5-22 Response Distributions – Thoughts About Climate Change and Age ................. 106

Figure 5-23 Response Distributions – Thoughts About Climate Change and Gender ........... 107

Table 5-24 Response Distributions – Important to Have a Low Carbon Footprint ............... 108

Table 5-25 Response Distributions – Households Can Reduce GHG Emissions .................... 108

Table 5-26 Response Distributions – Worried About Climate Change .................................. 108

Figure 5-27 Response Distributions – Thoughts About Climate Change and Important to

Have a Low Carbon Footprint ................................................................................................ 109

Figure 5-28 Response Distributions – Thoughts About Climate Change and Households Can

Reduce GHG Emissions .......................................................................................................... 110

Figure 5-29 Response Distributions – Thoughts About Climate Change and Worried About

Climate Change ...................................................................................................................... 111

Table 5-30 Response Distributions – Measure Carbon Footprint is Important .................... 111

Figure 5-31 Response Distributions – Measure Carbon Footprint is Important and Gender

................................................................................................................................................ 112

12

Figure 5-32 Response Distributions – Measure Carbon Footprint is Important and Carbon

Footprint Size ......................................................................................................................... 113

Figure 5-33 Response Distributions – Measure Carbon Footprint is Important and Self Health

Evaluation .............................................................................................................................. 114

Figure 5-34 Response Distributions – Measure Carbon Footprint is Important and Diet..... 115

Figure 5-35 Response Distributions – Measure Carbon Footprint is Important and Exercise

................................................................................................................................................ 115

Figure 5-36 Response Distributions – Measure Carbon Footprint is Important and

Environmentally Friendly Products ........................................................................................ 117

Figure 5-37 Response Distributions – Measure Carbon Footprint is Important and Turn Off

Lights ...................................................................................................................................... 118

Figure 5-38 Response Distributions – Measure Carbon Footprint is Important and Reduce

Waste / Recycle...................................................................................................................... 119

Figure 5-39 Response Distributions – Measure Carbon Footprint is Important and Thoughts

About Climate Change ........................................................................................................... 120

Figure 5-40 Response Distributions – Measure Carbon Footprint is Important and Important

to Have a Low Carbon Footprint ............................................................................................ 121

Figure 5-41 Response Distributions – Measure Carbon Footprint is Important and

Households Can Reduce GHG Emissions ............................................................................... 122

Figure 5-42 Response Distributions – Measure Carbon Footprint is Important and Worried

About Climate Change ........................................................................................................... 123

Table 6-1 BMI Bands .............................................................................................................. 126

Table 6-2 Self Health Evaluation Correlation Analysis Results .............................................. 127

Table 6-3 Attitudes Towards Health, the Environment, Carbon Emissions and Climate

Change Correlation Analysis Results ...................................................................................... 129

Table 6-4 Behaviours Towards Consumption and the Environment Correlation Analysis

Results .................................................................................................................................... 130

Table 6-5 Usage Intentions Towards a PCTS Correlation Analysis Results ............................ 132

Table 6-6 PCTS vs. Self Health Evaluation Correlation Analysis Results ................................ 133

Table 6-7 PCT vs. Behaviours Towards Consumption and the Environment Correlation

Analysis Results ...................................................................................................................... 135

13

Table 6-8 PCT vs. Attitudes Towards Health, the Environment, Carbon Emissions and Climate

Change Correlation Analysis Results ...................................................................................... 137

Table 6-9 Pattern Matrixa- 1 – Attitudes Towards Health, the Environment, Carbon

Emissions and Climate Change .............................................................................................. 143

Table 6-10 Pattern Matrixa- 2 – Attitudes Towards Health, the Environment, Carbon


Table 6-11 Total Variance Explained – Attitudes Towards Health, the Environment, Carbon


Table 6-12 Factor Correlation Matrix – Attitudes Towards Health, the Environment, Carbon


Table 6-13 Pattern Matrixa – Behaviours Towards Consumption and the Environment ...... 149

Table 6-14 Total Variance Explained – Behaviours Towards Consumption and the

Environment........................................................................................................................... 149

Table 6-15 Pattern Matrixa – Usage Intentions Towards a PCTS ........................................... 151

Table 6-16 Total Variance Explained – Usage Intentions Towards a PCTS ............................ 152

Table 6-17 Correlations Between the Factors and BMI ......................................................... 153

Table 1-18 Regression Modeling – Complete Dataset........................................................... 156

Figure 6-19 Variance In Usage Intentions Towards a PCTS - Complete Dataset ................... 158

Table 6-20 Regression Modeling – Gender ............................................................................ 160

Figure 6-21 Variance In Usage Intentions Towards a PCTS - Males ...................................... 162

Figure 6-22 Variance In Usage Intentions Towards a PCTS - Females ................................... 164

Table 6-23 - Scale Descriptors – Comparative Carbon Footprint .......................................... 166

Table 6-24 - Regression Modeling – Carbon Footprint .......................................................... 168

Figure 6-25 Variance In Usage Intentions Towards a PCTS - Better Than Average Carbon

Footprint ................................................................................................................................ 170

Figure 6-26 Variance In Usage Intentions Towards a PCTS - About Average Carbon Footprint

................................................................................................................................................ 172

Figure 6-27 Variance In Usage Intentions Towards a PCTS - Males Better Than Average

Carbon Footprint .................................................................................................................... 176

Figure 6-28 Variance In Usage Intentions Towards a PCTS - Males About Average Carbon

Footprint ................................................................................................................................ 178

14

Figure 6-29 Variance In Usage Intentions Towards a PCTS - Females Better Than Average

Carbon Footprint .................................................................................................................... 180

Figure 6-30 Variance In Usage Intentions Towards a PCTS - Females About Average Carbon

Footprint ................................................................................................................................ 182

Table 6-31 Regression Modeling – Gender and Carbon Footprint ........................................ 187

Table 6-32 Significant Predictors of Usage Intentions Towards a PCTS ................................ 191

15

Abbreviations

API Application Programming Interface

BMI Body Mass Index

BMR Basal Metabolic Rate

CFA Confirmatory Factor Analysis

CPRS Carbon Pollution Reduction Scheme

C&S Cap and Share

CSIRO Commonwealth Scientific and Industrial Research Organisation

DEFRA Department for Environment, Food & Rural Affairs

DTQ Domestic Tradable Quotas

EC2 Amazon Elastic Cloud

EFA Exploratory Factor Analysis

EU European Union

EUETS European Union Emission Trading System

GEP General Entry Permit – Norfolk Island visa renewal needed every 5 years

GHG Greenhouse Gas

Gt Gigatonnes

IPCC Intergovernmental Panel on Climate Change

HFCS High Fructose Corn Syrup

IS Information System

KMO Kaiser-Meyer-Olkin

MSA Measure of Sampling Adequacy

N-PCTS NICHE Personal Carbon Trading System

NHANES National Health and Nutrition Examination Survey

NICHE Norfolk Island Carbon Health Evaluation

PAF Principal Axis Factoring

PCA Personal Carbon Allowances

PCA Principal Component Analysis

PCT Personal Carbon Trading

POS Point Of Sale

PCTS Personal Carbon Trading System

16

RDS Relational Database Service

RF Radiative Forcing

SEM Structural Equation Modeling

TAM Technology Acceptance Model

TAM2 Technology Acceptance Model 2

TEQ Tradable Energy Quotas

TPB Theory of Planned Behaviour

TRA Theory of Reasoned Action

TEP Temporary Entry Permit – Norfolk Island visa renewal needed every 12

months

USDA United States Department of Agriculture

USEPA United States Environmental Protection Agency

UTAT Unified Theory of Acceptance and Use of Technology

WHO World Health Organization

W/m² Watts per square metre

17

1 Introduction

1.1 Background to the Research

The Norfolk Island Carbon Health Evaluation (NICHE) project is a multi-disciplinary study

that is exploring the link between health, obesity and an individual’s carbon footprint (see

http://www.norfolkislandcarbonhealthevaluation.com). The study is being conducted on

Norfolk Island by researchers from the Southern Cross Business School and the School of

Health and Human Sciences.

The primary objective of the NICHE project is to assess whether personal carbon allowances

are effective in reducing an individual's carbon footprint and what impact this has on health

behaviours associated with obesity such as active transport and diet.

The secondary objectives of the NICHE project are to assess the political acceptability of

introducing PCT (Personal Carbon Trading) into a population as a tool for reducing carbon

emissions and to explore the relationship between positive health behaviours and pro-

environmental behaviours.

This thesis covers a sub section of the NICHE project and is primarily concerned with

influences on PCTS (Personal Carbon Trading System) usage intentions and aims to

understand how an individual’s attitudes and behaviours associated with their health and

well-being, the environment, carbon emissions and climate change affect their usage

intentions towards a PCTS.

An initial survey of households on Norfolk Island and health evaluations of a sample of

residents were conducted in mid 2012. At the start of March 2013 a PCTS that was

programmed by the author of this thesis was rolled out to monitor the carbon footprint of

selected residents of Norfolk Island. A follow up survey and health evaluations will be

conducted approximately 12 months after the introduction of the PCTS.

18

1.2 Originality of the Research

The link between health and obesity in the developed world has been widely researched by

a variety of organizations such as the World Health Organization (World Health

Organization, 2000), the American Medical Association (Hedley et al., 2004) and the

National Bureau of Economic Research (Bleich, Cutler, Murray, & Adams, 2007).

A number of other organizations including the United States Environmental Protection

Agency (United States Environmental Protection Agency, 2010), the National Academy of

Sciences (Solomon, Plattner, Knutti, & Friedlingstein, 2009) and the Intergovernmental

Panel on Climate Change (Field et al., 2012) have conducted research looking at the

environment, carbon emissions and climate change.

It has been proposed that there is a link between obesity and climate change (see Chapter 2

– Literature Review). However after reviewing all of the available literature it does not

appear that this theory has been tested. The researchers believe that the NICHE project is

the first study of its kind to explore the relationship between health, obesity and an

individual’s carbon footprint.

The literature surrounding the most detailed and well researched proposed PCT systems

was reviewed as part of the process of developing the conceptual model (see Figure 1-1)

and writing this thesis. Considerable research has been undertaken studying how PCT could

work and what areas of energy expenditure it would cover. In addition an increasing

number of studies have looked at the political acceptability of PCT (see Chapter 2 –

Literature Review). However an extensive review of the literature has not found any

research into what factors influence PCTS usage intentions. Researchers believe that the

NICHE project and this thesis in particular is the first study of its kind to explore how an

individual's attitudes and behaviours in a range of areas influence their PCTS usage

intentions.

Norfolk Island is a relatively closed environment that makes it an ideal location to undertake

the study. The Norfolk Island Government is supporting the study and providing

infrastructure and subsidizing travel for the research team. This support makes it possible

19

to track resource usage by residents and also to obtain comprehensive data related to the

use of products on the island, both imported and locally produced.

1.3 Literature Review and the Conceptual Model

The broader NICHE project is examining links between an individual's health and their

carbon footprint. However the focus of the research for this thesis is PCTS usage intentions.

To understand where this research fits into the boarder NICHE project and to understand

what factors affect PCTS usage intentions, examination of literature relevant to the research

being undertaken has focused on the following areas:

Information systems success

PCT

Health and obesity

Carbon emissions and climate change

Links between obesity and an individual’s carbon footprint

The literature on carbon emissions and climate change and health and obesity was reviewed

to provide a background on the broader project. This literature overlaps with the literature

that is relevant to this research. The literature on information systems success, PCT and the

links between obesity and an individual’s carbon footprint was of particular interest and

provided the most insight when designing the conceptual model (see Figure 1-1) that forms

the basis of the research of this thesis.

The main PCT schemes that were examined include:

Cap and Share (C&S) – Proposed by the Foundation for the Economics of

Sustainability (FEASTA) in 2008

Tradable Energy Quotas (TEQ) - Originally known as Domestic Tradable Quotas (DTQ)

TEQ’s were first proposed by David Fleming in June 1996

Personal carbon allowances (PCA)– Proposed by Mayer Hillman and Tina Fawcett in

2004

Household Carbon Trading – Proposed by Debbie Niemeier and others in 2008

20

These four were identified as having the most detail and being the most developed of all the

proposed PCTS models and gave researchers a clear idea of how a PCTS should be

implemented, what areas of energy expenditure it should cover and how it would affect an

individual's everyday life. This in turn helped determine what factors might affect PCTS

usage intentions.

A number of researchers have begun to speculate that the growing obesity epidemic and

climate change may be related (see Chapter 2 – Literature Review). An increase in the

consumption of meat and highly processed energy dense food products and a decrease in

active transport (walking, cycling etc) offset by an increase in fossil fuel powered transport

has led to the obesity epidemic and is a major cause of carbon emissions leading to global

warming. The literature that was reviewed in this area was instrumental in determining

what factors affect PCTS usage intentions.

As a PCTS falls under the definition of an information system the literature concerning

information system success was invaluable in the development of the conceptual model and

deciding how to test it. There are a number of models that measure information system

success that have been proposed over the last thirty years. An extensive review of the

literature surrounding these models was undertaken to ascertain what factors affect usage

intentions towards information systems and how best to measure these usage intentions.

Information success models can be separated into two groups, user satisfaction models and

technology acceptance models.

From the user satisfaction group the models that were reviewed include:

DeLone and McLean Information System Success Model (1992)

DeLone and McLean Information System Success Model Revisited (2002, 2003)

From the technology acceptance group the models that were reviewed include:

Unified Theory of Acceptance and Use of Technology (UTAUT - 2003)

The Technology Acceptance Model (TAM - 1989)

The Technology Acceptance Model 2 (TAM2 - 2000)

21

TAM2 is a key model in the evaluation of the effectiveness and usefulness of information

systems and after the review it was decided that TAM2 offered the most insight. It has been

used to evaluate acceptance with a variety of systems in a variety of areas across all sections

of education and industry. While the constructs from the conceptual model used for the

research described in this thesis do not follow the relationship paths between the variables

in TAM2, insight into information system success and in particular the effect that attitudes

have on information system success were developed from TAM2.

All the information success models referred to above used surveys to capture measures of

information system satisfaction and success. The majority of these surveys used 7 point

Likert scales to measure responses (discussed in greater detail in Chapter 4 – Methodology).

Similar statistical techniques based on these studies will be used to test the conceptual

model’s validity.

A full review of all the of literature is reported in Chapter 2 – Literature Review.

After reviewing the all of the literature the conceptual model hypothesizes that an

individual’s PCTS usage intentions are influenced by:

Self-health evaluation – Information on an individual’s self-evaluation of their own

health and weight

Attitudes and behaviours towards health – Individuals who have positive attitudes

and behaviours about health and weight management and who exercise regularly

are more likely to have a positive PCTS usage intentions

Attitudes and behaviours towards the environment - Individuals who have positive

attitudes and behaviours towards the environment are more likely to have a positive

PCTS usage intentions

Attitudes and behaviours towards carbon emissions and climate change - Individuals

who have positive attitudes and behaviours towards their carbon footprint,

reduction of their carbon emissions and are concerned about climate change are

more likely to have a positive PCTS usage intentions

The conceptual model constructs are shown in Figure 1-1 below.

22

1.4 Methodology and the Research Questions

The NICHE project is a multi-disciplinary study and this was reflected in the construction of

the survey (see Appendix A - 'NICHE Survey'). The NICHE survey contained a range of

questions covering the information systems, health, environment and the carbon emissions

components of the NICHE study and included questions specific to the aims of this study and

questions relevant to the other components of the broader study.

Some of the questions contained in the survey were aimed at the individual respondent and

others were aimed at their household.

The survey contained the following sections:

General information (questions A1 – A13) - general information about the

respondents and their household, their health and their beliefs about their own and

their households carbon foot print



Attitudes and behaviours

towards carbon emissions and

climate change


towards health


towards the environment

Figure 1-1 NICHE Conceptual Model - Hypothesized Influences on PCTS Usage Intentions

23

Attitudes (questions B1 – B13) - the respondents and their households attitudes and

towards health, the environment, carbon emissions and climate change

Behaviours (questions B14 – B19) - the respondents and their households behaviours

towards consumption and climate change

Physical activity (questions C1 – C15) - the respondents and their households

physical activity

Nutrition (questions D1 – D31) - the respondents diet and nutrition

Personal carbon trading (questions E1 – E10) - the respondents attitudes and beliefs

about a personal carbon trading systems

Demographic data (questions F1 – F28) - general demographic information about the

respondent including age, gender, income, education, height, weight, health, home

ownership and motor vehicle ownership

Where specific answers were not sought from the survey respondent 5 and 7 point Likert

scales were used to gather responses. General comment fields were also used to gain

additional information from the respondent.

Some of the questions used in the NICHE survey were derived in part or based survey items

from the following sources:

Environmental behaviours – Department of Environment 2012

Health behaviours - WHO GPAQ, Cancer Council Food Frequency

Attitudes towards climate change – CSIRO 2011

Personal carbon trading – Stuart and Capstick 2009, TAM2 surveys

Demographic and socioeconomic status – NI census 2011

Anthropometrics – WHO STEP wise 2008

Other questions were developed entirely by NICHE researchers.

The following questions were used from the NICHE survey as the basis for this thesis:

Questions B1 - B13 - The questions in this section of the survey were included as measures

of the attitudes and behaviours towards health, attitudes and behaviours towards the

environment and attitudes and behaviours towards carbon emissions and climate change

components of the conceptual model.

24

Questions B14 - B19 - The questions in this section of the survey were included as measures

of the attitudes and behaviours towards the environment component of the conceptual

model.

Questions A9, A10, A12 and C1 - The questions in this section of the survey were included as

measures of the self health evaluation component of the conceptual model.

Questions E1 - E10 - As the focus of this thesis is PCTS usage intentions this section of the

survey included questions specifically related to users usage intentions towards a PCTS. The

ten questions in this section were all adapted from scales used by researchers examining

user information system satisfaction and the technology acceptance models. The questions

in this section are intended to measure the information needed for the dependent variable

PCTS usage intentions of the ‘information systems’ component of the study.

The survey was open to any permanent resident of Norfolk Island or any long term

nonresident over the age of 18. Five NGO’s were employed to hand deliver a survey to

every household on Norfolk Island. Completed surveys could be returned by mail, dropped

off at the local post office or supermarket or picked up by NICHE staff.

In total over 400 responses were received which represents over 60% of the occupied

households on Norfolk Island and over 20% of the total population or almost 30% of the

population over 18 years of age.

The full methodology of the NICHE project is reported in Chapter 4 - Methodology.

1.5 Preview of the Research Findings

The final stages of the statistical analysis involved testing of linear regression models. The

first model tested was for the entire data set. The remaining models that were tested were

moderated by gender, carbon footprint, and carbon footprint by gender. In all 12 regression

models were tested, although 3 were inadmissible due to the small number of degrees of

freedom and small number of observations. The notable outcomes of key interest to the

researchers examining PCTS are summarised below.

25

The models moderated by gender showed that 'Health Consciousness', ‘Optimism’

and 'Environmental Consciousness' were significant predictors of 'Usage Intentions

Towards a PCTS' for both males and females and that 'Self Health Evaluation' was

also significant for females but not for males.

The models moderated by carbon footprint showed that 'Self Health Evaluation', and

'Environmental Consciousness' were significant across all models but that

'Environmental Action' was also significant in the model for those with a better than

average carbon footprint and that 'Health Consciousness' and ‘Optimism’ were

significant in the model for those with an average carbon footprint.

The models moderated by gender and carbon footprint showed that:

o 'Environmental Action' was the only significant predictor for males with a

better than average carbon footprint

o 'Health Consciousness' and ‘Optimism’ were the significant predictors for

males with an average carbon footprint

o 'Health Consciousness' and 'Environmental Consciousness' were the

significant predictors for females with a better than average carbon footprint

o 'Health Consciousness' and 'Self Health Evaluation' were the significant

predictors for females with an average carbon footprint

1.6 Organization of the Thesis

Chapter 1 - Introduction

Chapter 1 provides an overview and an introduction to the research. The originality of the

research, research methodology and a overview of the main findings is covered and the

conceptual model that forms the basis of this thesis is introduced.

Chapter 2 - Literature Review

Chapter 2 examines and summarizes the literature that is relevant to the broader NICHE

project and the subsection of the NICHE project that is covered by this thesis. Literature

covering health, obesity, carbon emissions, climate change and the links between them was

reviewed to understand the aims of the broader NICHE project and this projects place

26

within the larger project. This is followed by a review of the literature covering information

systems success and PCT that is more relevant to this project.

Chapter 3 - NICHE PCTS

Chapter 3 provides an overview of the hardware and the software components that make

up the PCTS that is used to track the carbon footprint of selected participants in the NICHE

project.

Chapter 4 - Methodology

Chapter 4 describes the research methodology for the NICHE project. The NICHE survey

contained questions that are relevant to the broader study and the research undertaken for

this thesis . This chapter describes the surveys sample selection, its administration and its

limitations. The construction of the survey and research questions that are relevant to the

broader study are covered with attention given to the components of the survey that are

relevant to this project.

Chapter 5 - Descriptive Analysis

Chapter 5 presents the descriptive analysis of the data that was returned from the NICHE

survey. The survey of households on Norfolk Island included questions specific to an

individual’s PCTS usage intentions and to each of the key components of the conceptual

model. In this chapter the responses to these questions are analyzed to see if they meet

with researchers expectations.

Chapter 6 - Data Analysis

In this chapter the data that was collected in the NICHE survey is analyzed using a number of

statistical procedures. Correlation analysis was used to evaluate the strength of the

relationships between blocks of variables contained in the survey. Exploratory Factor

Analysis (EFA) was then run on the blocks of variables expected to have an underlying factor

structure as per the design of the survey. Finally linear regression analysis was run on the

factors that were extracted from the EFA to estimate the relationships among them.

27

Chapter 7 - Conclusion

Chapter 7 summarizes the factors that were found to influence the dependent variable

'usage intentions towards a PCTS'. The differences between the results of the statistical

analysis and the conceptual model are discussed and the reasons that NICHE researchers

believe are behind the differences are examined. Finally the limitations of the study are

examined and recommendations for further research are made.

28

2 Literature Review

2.1 Introduction

This chapter summarizes the research and background literature relevant to the NICHE

project that is the basis for this thesis. The examination of literature relevant to the

research focuses on the following five key areas:

Health and obesity

Carbon emissions and climate change

Links between obesity and an individual’s carbon foot print

Personal Carbon Trading Systems (PCTS)

Information Systems Success

The information examined in this chapter looks at the existing literature in these five areas

in order to provide the reader with a background to the research for the broader NICHE

project and this sub project in particular.

The broader NICHE project is examining links between an individual's actual health, their

perception of their health, their weight and the relationship this has with their carbon

footprint. Literature relevant to health, obesity, carbon emissions and climate change and

the links between them has been reviewed here to provide an overview of the research of

the broader NICHE project. This literature overlaps with and is also relevant to the sub

section of the NICHE project covered by research investigated in this thesis.

However as the main focus of the research for this thesis is an individual’s PCTS usage

intentions the literature relevant to PCTS and information systems success is covered in

greater detail in this chapter.

2.2 Health and Obesity

The broader NICHE project aims to explore the relationship between health, obesity and an

individual’s carbon footprint. To understand the research being conducted by the broader

29

NICHE project and where the focus of the research of this thesis fits into the broader project

literature on health, specifically in relation to obesity is reviewed.

This literature is also relevant to and provides part of the basis of the conceptual model (see

Figure 1-1) that underpins the research undertaken for this thesis. In part the conceptual

model hypothesizes that an individual’s PCTS usage intentions is influenced by:

their health – measured through a self-health evaluation

their attitudes towards their health

activities and behaviours that influence their health.

The literature that follows in this section of the thesis was used during the development of

some of the constructs in the conceptual model. It was also used in the development of the

questions in the NICHE survey (see Appendix A - 'NICHE Survey') designed to collect data

about each of the conceptual models constructs.

Obesity is "a condition of abnormal or excessive fat accumulation in adipose tissue, to the

extent that health may be impaired” (World Health Organization, 2000, p. 6). The link

between health and obesity in the developed world has been investigated by a variety of

organizations such as the World Health Organization (World Health Organization, 2000), the

American Medical Association (Hedley et al., 2004) and the National Bureau of Economic

Research (Bleich et al., 2007). As a result obesity is widely used as an indicator of an

individuals’ and a society’s health.

Obesity in the developed world is fast reaching epidemic proportions (Caballero, 2007).

Thirty percent of children under the age of 18 in the developed world are now classified as

overweight or obese and approximately two thirds of men and over half of women over the

age of 18 are classified as overweight or obese (G. Egger & Swinburn, 2011, p. 11). In 2009

– 2010 the Centre for Disease Control and Prevention in the US reported that 35.7% of all

adults in the United States were obese (Ogden, Carroll, Kit, & Flegal, 2012, p. 1). This

problem is not just confined to the western world, as the wealthy Middle Eastern countries

of Kuwait and Qatar report the highest incidence of obesity in the world (Al-Nesf et al.,

2006; Bener et al., 2004; International Association for the Study of Obesity, 2013)

30

As countries in the developing world become more industrialized, wealthier, and have

greater access to western technology and western diets or heavily modified indigenous diets

this pattern of obesity is being repeated. Mexico, China and Thailand have shown dramatic

increases in obesity in recent decades (Caballero, 2007). In 1980 less than 5% of the

Chinese population was overweight or obese, this figure has now risen to 22%. (G. Egger &

Swinburn, 2011, p. 13). Similar rates of obesity are being observed in other countries and

regions undergoing rapid development and the associated increase in consumer wealth with

obesity rates in some developing countries now rivaling or exceeding the developed world

(Prentice, 2006). India and Brazil have seen massive increases over the last two decades

and Mexico now has some of the highest obesity rates in the world with Mexican women

coming third and Mexican men coming sixth in world obesity rankings, well in excess of

many developed countries (International Association for the Study of Obesity, 2012). While

obesity in many African and Asian countries is still relatively uncommon, obesity is becoming

increasingly prevalent in urban areas where the population’s indigenous diet is slowly being

replaced with more processed food than in rural areas where the population has greater

access to traditional food sources. In fact in many developing countries obesity coexists

with malnutrition as the indigenous diet is often replaced with a high calorie low nutrient

diet (Hawkes, 2006, p. 2; Stuckler, McKee, Ebrahim, & Basu, 2012, p. 1; World Health

Organization, 2000, p. 16). Figure 2-1 below from the International Association for the

Study of Obesity depict the percentage of the population that is obese in the countries that

are most affected by the obesity epedemic.

31

Figure 2-1 World Obesity Rates for Males and Females (International Association for the Study of Obesity,

2012)

32

The rise in the prevalence of obesity has been linked to the increase in diabetes (Hossain,

Kawar, & El Nahas, 2007). In the developed world where obesity rates are high at least one

in three and possibly as many as one in two citizens will develop Type 2 diabetes at some

point in their lives (G. Egger & Swinburn, 2011, p. 11). However it is the developing world

that is now at the forefront of the diabetes epidemic with the largest increases now

occurring in the Asian and Western Pacific regions particularly in India and China (Hossain

et al., 2007).

Other long term effects of obesity include increased risk of many cancers including cancer of

the breast, endometrium, esophagus, kidney, pancreas, gall bladder, thyroid, ovary, cervix,

and prostate, as well as multiple myeloma and Hodgkin’s lymphoma (Kushi et al., 2006, p.

258). “For the great majority of Americans who do not use tobacco, weight control, dietary

choices, and levels of physical activity are the most important modifiable determinants of

cancer risk. Evidence suggests that one-third of the more than 500,000 cancer deaths that

occur in the United States each year can be attributed to diet and physical activity habits,

including overweight and obesity (sic)”(Kushi et al., 2006, p. 254)

The linkage between obesity and illnesses highlighted in this section of the thesis is

underlying basis of the NICHE project. The tenet of the overall research is that there is a link

between obesity and a high carbon footprint (Norfolkislandcarbonhealthevaluation.com,

2013). That is, people who consume high levels of meat and processed foods and an

inadequate supply of whole plant based foods combined with a decrease in active transport

are more likely to suffer from obesity and other related illnesses. And it is this diet and the

increase in fossil fuel powered transport that leads to the higher carbon footprint.

2.3 Carbon Emissions and Climate Change

"Climate change refers to any significant change in measures of climate (such as

temperature, precipitation, or wind) lasting for an extended period (decades or longer)"

(United States Environmental Protection Agency, 2010, p. 1)

Current research into carbon emissions and climate change is fundamental to the broader

NICHE project. Aspects of the research surrounding carbon emissions and climate change

33

are also crucial to the specific focus of the research described in this thesis. Therefore a

solid understanding of carbon emissions and their link with climate change required.

The Literature described in this section provided the basis for the development of the

conceptual model constructs (see Figure 1-1) listed below:



Examination of the literature also guided the researchers in development of the survey

instrument (see Appendix A - 'NICHE Survey') that was used to gather data to test these

constructs.

One of the main roles of the atmosphere is regulating the earth’s temperature. When

energy from the sun hits the earth some of it is absorbed by the earth and the rest is

radiated back into space as heat. Naturally occurring greenhouse gases in the atmosphere

absorb some of the radiated energy and re-emit it, keeping it in the lower atmosphere

making life on earth possible. This is known as the greenhouse effect (United States

Environmental Protection Agency, 2010, p. 9). The blanketing effect caused by these

greenhouse gases are the reason the surface of the earth is warm. In recent times humans

have altered the chemical composition of the atmosphere through the release of

greenhouse gases leading to global warming (Le Treut et al., 2007, p. 97).

Radiative Forcing (RF) is used to measure how the greenhouse gases affect the amount of

energy that is absorbed in the atmosphere (United States Environmental Protection Agency,

2010, p. 4). The Intergovernmental Panel on Climate Change (IPCC) defines RF as the

change in the downward irradiance minus the upward irradiance measured in Watts per

square metre (W/m²) (Intergovernmental Panel on Climate Change, 2007, p. 86). Positive RF

leads to global warming, negative RF leads to global cooling. The total RF for all greenhouse

gas (GHG) emissions in 2011 relative to 1750 was 2.29 W/m² with the most rapid increases

seen since 1970 (Intergovernmental Panel on Climate Change, 2013, p. 11).

While other greenhouse gases such as methane also contribute to global warming carbon

dioxide is seen as the major contributor (see Figure 2-2 below) and is the most important of

the greenhouse gases that contribute towards the greenhouse effect (Le Treut et al., 2007,

34

p. 97). Carbon dioxide does not break down in the atmosphere like other greenhouse gases

that are produced by human activities. Methane, the greenhouse gas that accounts for the

second highest RF increase has an average lifespan of 12 years in the atmosphere. In

contrast “carbon dioxide’s lifetime is poorly defined because the gas is not destroyed over

time, but instead moves between different parts of the ocean–atmosphere–land system.

Some of the excess carbon dioxide will be absorbed quickly (for example, by the ocean

surface), but some will remain in the atmosphere for thousands of years” (United States

Environmental Protection Agency, 2010, p. 18).

Figure 2-2 Radiative forcing caused by various greenhouse gases (United States Environmental Protection

Agency, 2010, p. 18)

Since pre-industrial times the atmospheric concentrations of carbon dioxide has increased

by almost 40% (Intergovernmental Panel on Climate Change, 2013, p. 9). “Before the

industrial era began around 1780, carbon dioxide concentrations measured approximately

270–290 ppm. Concentrations have risen steadily since then, reaching 387 ppm in 2009—a

35

38 percent increase. Almost all of this increase is due to human activities” (United States

Environmental Protection Agency, 2010, p. 15). Figure 2-3 below depicts this rapid increase.

Ice cores samples have shown that carbon dioxide levels are now substantially higher than

at any time in the last 800,000 years (Intergovernmental Panel on Climate Change, 2013, p.

9). Of the 2.29 W/m² RF increase for GHG between 1750 and 2011 reported by the IPCC,

1.68 W/m² was caused by carbon dioxide alone (Intergovernmental Panel on Climate

Change, 2013, p. 11).

Figure 2-3 Concentrations of carbon dioxide in the atmosphere from hundreds of thousands of years ago

through 2009 (United States Environmental Protection Agency, 2010, p. 17)

There are three main sources of carbon dioxide emissions. The burning of fossil fuels is the

primary source of carbon dioxide emissions. Changes in land use practice resulting in

deforestation and cement production are the secondary sources. "From 1750 to 2011, CO2

emissions from fossil fuel combustion and cement production have released 375 Gigatonnes

(GtC) to the atmosphere, while deforestation and other land use change are estimated to

have released 180 GtC. This results in cumulative anthropogenic emissions of 555 GtC"

(Intergovernmental Panel on Climate Change, 2013, p. 9).

36

The end result of this increase in carbon dioxide in the atmosphere and its associated

positive RF increase is climate change. "Each of the last three decades has been successively

warmer at the Earth’s surface than any preceding decade since 1850" (Intergovernmental

Panel on Climate Change, 2013, p. 3). In addition to global warming other changes include

ocean warming and acidification, the loss in mass of ice sheets, glaciers and arctic sea ice,

rises in sea levels and changes in weather patterns (Field et al., 2012; Intergovernmental

Panel on Climate Change, 2013; Le Treut et al., 2007; Solomon et al., 2009; United States

Environmental Protection Agency, 2010)

There are still 'climate change skeptics' or ‘climate change deniers’ in the general public who

dispute the evidence presented above and also dispute that any change at all is taking place

(Anderegg, Prall, Harold, & Schneider, 2010; Doran & Zimmerman, 2009; Gallup, 2014;

Legras, 2013; Leviston, Leitch, Greenhill, Leonard, & Walker, 2011; Pew Research Center,

2009; Poortinga, Spence, Whitmarsh, Capstick, & Pidgeon, 2011; Whitmarsh, 2011). There

are also many people who believe in climate change but dispute various aspects, causes,

and/or the severity and seriousness it poses (Gallup, 2014; Leviston et al., 2011; Pew

Research Center, 2009; Poortinga et al., 2011; Whitmarsh, 2011).

The reasons for climate change skepticism are varied. Whitmarsh et al. (Whitmarsh, 2011,

p. 690) found that skepticism was "strongly determined by individuals’ environmental and

political values (and indirectly by age, gender, location and lifestyle) rather than by

education or knowledge" however "findings show denial of climate change is less common

than the perception that the issue has been exaggerated". Poortinga et al. (Poortinga et al.,

2011) found that "climate skepticism appeared particularly common among older individuals

from lower socio-economic backgrounds who are politically conservative and hold traditional

values". Polls conducted by the Pew Research centre and Gallup consistently show that in

the United States a large gap exists between supporters of the Democrat and Republican

parties on perceptions of climate change (Dunlap & McCright, 2008; Gallup, 2014; Pew

Research Center, 2009).

While the vast majority of climate scientists agree with the tenants of climate change as

outlined by the IPCC there are also a small group of climate scientists who contest these

conclusions (Anderegg et al., 2010; Doran & Zimmerman, 2009). Research by Anderegg et

37

al. found that 2% of the top 50 climate researchers, 3% of the top 100 and 2.5% of the top

200 climate researchers contest the conclusions of climate change by the IPCC (Anderegg et

al., 2010, p. 12107). A similar study conducted by Dorian et al. found that 97.4% of actively

publishing climatologists believe in climate change (Doran & Zimmerman, 2009, p. 23).

The researchers conducting the NICHE study believe that climate change is real and that

human activity is the cause. It is this belief that underpins the study and its efforts to

present evidence of the link between climate change as assessed by an individual’s carbon

footprint and their health.

2.4 Links Between Obesity and an Individual's Carbon Footprint

Researching the link between obesity and an individual’s carbon footprint is the primary

objective of the NICHE project. This link has been proposed by a number of researchers

however an extensive search of the literature has not shown any research where it has been

tested. Based on the literature review, the researchers believe that the NICHE project is the

first study of its kind to explore the relationship between obesity and an individual’s carbon

footprint. This relationship is also fundamental to the focus of this thesis as an individual’s

attitudes and behaviours towards the components of this relationship underlie the

relationships and constructs of the conceptual model (see Figure 1-1).

An increase in calorie consumption and a decrease in calorie expenditure are largely to

blame for the obesity epidemic (G. Egger, 2007, p. 185; G. Egger & Swinburn, 2011; J. O. Hill,

Wyatt, Reed, & Peters, 2003; Pi-Sunyer, 2003, p. 859). Egger and Swinburne (G. Egger &

Swinburn, 2011) identified the following points as the key issues leading to increases in

calorie consumption:

Increased consumption of meat and meat based food products

Increase in consumption of highly-processed, energy-dense foods

Decreased consumption of whole plant based foods

The actions and activities leading to decreases in calorie expenditure were identified by

Egger and Swinburne as:

38

Increased use of fossil fuel powered transport

Decrease in active transport

The IPCC estimates that agriculture accounts for 10% - 12% of the worlds GHG emissions (P.

Smith et al., 2007, p. 499). In Finland it is estimated that the food chain accounts for 14% of

GHG emissions (Virtanen et al., 2011, p. 1849). And in the United States it is estimated that

food accounts for more than 15% of GHG emissions (Kim & Neff, 2009, p. 186).

To a large extent meat in industrialized countries is produced using feed grains. This is very

inefficient and on average "10 g of vegetable protein are needed to generate 1 g of animal

protein" (Reijnders & Soret, 2003, p. 665). The protein efficiency conversion is the most

inefficient for beef (6%), followed by pork (9%) and then chicken (18%) (Reijnders & Soret,

2003, p. 665).

In addition to being a highly inefficient protein source, meat is the largest contributor

towards climate change of any food source and it production produces large amounts of

carbon dioxide, methane and nitrous oxide (Carlsson-Kanyama & González, 2009; P. Smith

et al., 2007; Virtanen et al., 2011; Wallén, Brandt, & Wennersten, 2004). Methane is second

only to carbon dioxide when it comes to the overall contributions to the increase in RF that

leads to climate change (Carlsson-Kanyama & González, 2009, p. 1704) and meat accounts

for approximately one third of global methane emissions (P. Smith et al., 2007, p. 510).

Of all the varieties of meat beef has the largest GHG footprint (Carlsson-Kanyama &

González, 2009, p. 1706). Up to 30 kg of GHG emissions are produced for each kg of beef.

11 kg are produced for each kg of cheese and 9.3 kg are produced for each kg of pork. In

comparison fresh fruit and vegetables range from 0.42 kg for domestically grown carrots up

to 1.2kg for imported oranges (Carlsson-Kanyama & González, 2009, p. 1707). Fishing does

not require the same resource input however as most fish consumed in western countries

are predominantly caught using trawlers that require large amounts of fossil fuel. "The input

of fossil fuels for catching fish may be up to roughly 14 times larger than for the production

of vegetable protein" (Reijnders & Soret, 2003, p. 667).

In addition to the GHG emissions that meat production directly produces the IPCC states

that the world demand for meat also contributes indirectly to GHG emissions due to the

39

competition between land use for bio fuel and feed stocks. Due to the production of feed

stocks there is less land available for bio fuel production thereby increasing fossil fuel

consumption (P. Smith et al., 2007, p. 503).

An excess of animal based food products has been shown to increase the likelihood of

obesity. (Spencer, Appleby, Davey, & Key, 2003; Wang & Beydoun, 2009). In their 2003

study of 37,875 healthy individuals Spencer et al. found that on average vegans had a

substantially lower body mass index (BMI) than meat eaters and that it was the excess

consumption of meat as opposed to other lifestyle factors that was associated with the

increase in BMI (Spencer et al., 2003). In their 2009 study analyzing the associations

between meat consumption and obese individuals Wang et al. (Wang & Beydoun, 2009, p.

621) found "consistent positive associations between meat consumption and BMI, waist

circumference, obesity and central obesity" and those who had "high meat consumption (in

the top quintile) were 27% more likely to be obese, and 33% more likely to have central

obesity compared to those with low meat consumption (bottom quintile)".

Processed food and drink that contain sugars requires the second highest energy use of any

food product behind meat (Wallén et al., 2004, p. 527). In the United States this processing

accounts for 14% of energy use in the food system. Only home refrigeration and preparation

of food (31%) and agricultural production (21%) account for higher energy use in the food

system. Additionally, packaging of food accounts for a further 7% of the energy used in the

food system (H. Hill, 2008, p. 3).

Excess consumption of highly processed energy dense food sources have also been shown

to increase the likelihood of obesity (G. Egger & Swinburn, 2011; Hawkes, 2006; Stuckler et

al., 2012). Studies have shown that obesity in America is partly the result of increased food

consumption due to technological advances in the preparation and transport of ready to eat

processed foods (Cutler, Glaeser, & Shapiro, 2003). An example of this is the potato. Prior

to World War 2, French fries were rare and potatoes were normally boiled, baked or

mashed. Today, mass produced processed frozen French fries account for most of Americas

potato consumption and are Americas favorite vegetable (Cutler et al., 2003, p. 2). High

fructose corn syrup (HFCS), a processed caloric sweetener manufactured from corn is used

as an additive in most processed food in the United states and its uptake in the US diet since

40

its development in the 1960's mirrors the increase in obesity rates (Bray, Nielsen, & Popkin,

2004). Data from the United States Department of Agriculture (USDA) between 1994 - 1996

found the average US resident aged 2 or above consumes 318 calories a day from caloric

sweeteners added to processed food (Bray et al., 2004, p. 540). Data from the National

Health and Nutrition Examination Survey (NHANES) from 1999-2006 found that this figure

had risen to 359 calories a day (Welsh et al., 2010, p. 1492).

There is a growing obesity problem in the developing world (see Section 2.2 - 'Health and

Obesity'). In these areas "intensive production of beef, poultry, and pork is increasingly

common, leading to increases in manure with consequent increases in GHG emissions. This is

particularly true in the developing regions of South and East Asia, and Latin America" (P.

Smith et al., 2007, p. 503). From 1967 - 1997 demand for meat in developing countries grew

5% a year, rising from 11 to 24 kg/capita/yr (P. Smith et al., 2007, p. 502). This demand is

forecast to rise a further 57% by 2020 (P. Smith et al., 2007, p. 502). The growing demand

for meat will induce further changes in land use (e.g. from forestland to grassland) and will

increase carbon dioxide, methane and nitrous oxide emissions as well as increasing demand

for animal feeds (P. Smith et al., 2007, p. 503). The consumption of highly-processed,

energy-dense food and added sweeteners is also increasing in the developing world and this

is implicated in the rapid rises in obesity and diet related chronic disease (G. Egger &

Swinburn, 2011; Hawkes, 2006; Stuckler et al., 2012).

In addition to the GHG emissions that are produced by the transport of meat products and

mass produced highly-processed, energy-dense foods the obesity epidemic has an

additional relationship with GHG emissions in the transport sector. Reliance on Fossil fuel

powered transport for individuals and its associated decrease in active forms of transport

are seen as a major contributor towards the obesity epidemic (Edwards & Roberts, 2009;

Higgins, 2005; Higgins & Higgins, 2005; Woodcock et al., 2009). Fossil fuel powered

transport is also a major contributor towards carbon emissions and climate change (Kahn

Ribeiro et al., 2007). "In 2004 transport was responsible for 23% of world energy-related

greenhouse gas (GHG) emissions with about three quarters coming from road vehicles. Over

the past decade, transport’s GHG emissions have increased at a faster rate than any other

energy using sector" (Kahn Ribeiro et al., 2007, p. 325). The obesity epidemic itself has now

become a major contributor towards the increase in fossil fuel transport due to the

41

increased energy needs required to move the greater mass of an obese individuals and the

decreased likelihood of using active transport such as walking or cycling among obese

individuals (Delpeuch, Maire, & Emmanuel, 2009; Edwards & Roberts, 2009, p. 1138).

An additional impact on the system is that an obese population requires additional food due

to the increase in Basal metabolic rate (BMR) in obese individuals. Edwards et al. (Edwards

& Roberts, 2009, p. 1137) found that “compared with a normal population distribution of

BMI, a population with 40% obesity requires 19% more food energy for its total energy

expenditure. Greenhouse gas emissions from food production and car travel due to increases

in adiposity in a population of 1 billion are estimated to be between 0.4 Giga tonnes (GT)

and 1.0 GT of carbon dioxide equivalents per year.

Delpeuchet al. (2009) have developed a model that outlines the link between Obesity and

greenhouse gases (see Figure 2-4 ). The model has been included here as it provides a

graphical description of the relationship between an individual, food production, the

environment and greenhouse gas emissions and demonstrates how the relationship

changes as the individuals body weight increases.

42

The literature has shown that:

Animal based food products and highly processed energy dense food contributes

towards obesity

Animal based food products and highly processed energy dense food is associated

with increased carbon and other GHG emissions and climate change

Increased use of fossil fuel powered transport and a decrease in active transport

contributes towards obesity

Increased use of fossil fuel powered transport and a decrease in active transport is

associated with increased carbon and other GHG emissions and climate change

Increases in food

production +

intake

Increases in

fuel input Increases in

fertilizer input

Waste volume

increases

Increases in

ruminant

digestion

Increases in

livestock

numbers

Body weight

increases

Vehicle weight

increases

Decrease in physical activity

Less walking / cycling

At work – sedentary

jobs

At home – labour saving

Household appliances

Watching TV

CH4

CH4

N2

O

CO2

CO2

Figure 2-4 The Link Between Obesity and Greenhouse Gases (Delpeuch et al. 2009)

43

In recent years a number of researchers have put all of these factors together and have

begun speculating that there is a direct link between an individual’s carbon footprint,

greenhouse gases, climate change and the rising obesity epidemic (Delpeuch et al., 2009;

Edwards & Roberts, 2009; G. Egger, 2007; G Egger, 2008; G. Egger & Swinburn, 2011;

Faergeman, 2007; Friel et al., 2009). Garry Egger, one of the principal researchers in the

NICHE project, proposed Personal Carbon Trading (PCT) as an option for dealing with both

climate change and obesity (G. Egger, 2007; G. Egger & Swinburn, 2011). However while

there is an emerging interest in the links between carbon emissions and obesity there is

little research examining it in detail. Exploring this link is one of the primary goals of the

NICHE project and researchers believe that the NICHE project is the first study of its kind to

explore the relationship between obesity and an individual’s carbon footprint.

2.5 PCTS

PCT can be defined as "a downstream trading mechanism normally understood to involve an

initial allocation of carbon permits to individuals based on carbon reduction targets, with

individuals able to buy and sell permits according to their desired carbon consumption and

prevailing permit prices" (Bristow, Wardman, Zanni, & Chintakayala, 2010, p. 1824)

Understanding PCT and carbon allowances is fundamental to the NICHE project. The

primary objective of the NICHE project is to assess whether personal carbon allowances are

effective in reducing an individual's carbon footprint and what impact this has on health

behaviours associated with obesity such as active transport and diet. One of the secondary

objectives of the NICHE project is to assess the political acceptability of introducing PCT into

a population as a tool for reducing emission reductions. PCT is also highly significant to the

research conducted in this thesis. The literature covered in this section was used to develop

parts of the NICHE survey (see Section 4.2 - 'Survey Design and Construction') and provided

the basis for the development of the dependent variable of the conceptual model (see

Figure 1.1) that underlies this thesis.

An element of discussions about ways to deal with climate change has been Personal

Carbon Trading Systems (PCTS). It is felt that the implementation of these systems would

move some of the responsibility for greenhouse gas emissions to households and heighten

44

their awareness of their emission levels. The following four schemes are the ones identified

from the literature that have the most detail and are seemingly the most well-developed of

the proposed PCTS identified in the review of literature:

Cap and Share (C&S) – Proposed by the Foundation for the Economics of

Sustainability (FEASTA) in 2008 “independent committee sets a national carbon cap.

All adults periodically receive certificates entitling them to an equal share of national

emissions. Certificates are sold by individuals via banks or post offices to companies

that import or extract fossil fuels. These suppliers require surrendering certificates

equal to emissions from the use of the fossil fuels that they introduce into the

economy. The price of emissions flows through the economy.”(Fawcett & Parag,

2010, p. 330)

Tradable Energy Quotas (TEQ) - TEQ’s were first proposed by David Fleming in June

1996. They are also known as Domestic Tradable Quotas (DTQ). “TEQs aim to tackle

climate change and peak oil. A TEQ budget sets a limit on annual carbon emissions

over the next 20 years, which then rolls forward week by week. 40% of the

allowances are distributed free to individuals on an equal per capita basis. Personal

emissions allocations cover household energy use and personal travel, but not air

travel. The remaining 60% are sold by tender to all other energy users. All fuels have

a carbon rating and purchasers must surrender carbon units to cover related

emissions. Transactions are carried out electronically and all carbon units are

tradable.“(Fawcett & Parag, 2010, p. 330)

Personal carbon allowances (PCA)– Proposed by Mayer Hillman and Tina Fawcett in

2004 “a national cap is set for emissions from household energy use and personal

travel, including air travel. Allowances are allocated periodically on an equal per

capita basis to individuals for free to cover these emissions. For every purchase of

electricity, gas, transport fuels and services, allowances are surrendered.

Transactions are carried out electronically and allowances are tradable in the

personal carbon market.” (Fawcett & Parag, 2010, p. 330)

Household Carbon Trading – Proposed by Debbie Niemeier and others in 2008. “A

yearly carbon emissions cap is set for residential energy use based on emissions

reduction targets. Allowances are allocated to each household on an equal per

45

household allocation basis via utility service providers who place the allowances in

each user’s account. These are deducted periodically by the utility according to

energy use, and additional allowances must be purchased if the account is in deficit.

The carbon allowances are fully tradable. At the end of a compliance period, the

state collects the permits from the utilities and determines compliance with the

cap.”(Fawcett & Parag, 2010, p. 331)

The idea for PCT was originally proposed in the UK. TEQ’s and PCA’s were the first main

schemes proposed for individual carbon trading and they were both developed in the UK

(Roberts & Thumim, 2006, p. 3).

In 2006 the UK Department for Environment, Food & Rural Affairs (DEFRA) commissioned

the Centre for Sustainable Energy to undertake a study "to provide some initial analysis of

the ideas and issues involved in the concept of individual carbon trading" (Roberts &

Thumim, 2006, p. 3). In their report to DEFRA entitled 'A Rough Guide to Individual Carbon

Trading' Roberts and Thumim found that all of the proposed schemes advocate setting an

emissions cap in the form of a permit that is a tradable commodity however they identified

the following design choices that differentiate the implementations of downstream carbon

trading (Roberts & Thumim, 2006, pp. 11-12):

"Who participates in the system: individuals, organizations, or both?"

"What proportion of the emissions cap is allocated to individuals versus

organizations, and on what basis?"

"Do children receive an allowance, and if so what size?"

"Are permits issued free of charge, and if not, how are they distributed?"

"Which fuels or activities are included in the scheme?"

In addition to the differences between the schemes Roberts and Thumim identify political

acceptability as one of the major factors delaying the implementation of PCT. Roberts and

Thumim (Roberts & Thumim, 2006, p. 3) found that “a range of interests largely focused on

the operational minutiae of specific schemes and on examining the minor theological

differences between them. Yet the differences between the schemes appear to be less

important at this stage than the largely untested assumptions shared by them all about

public responses and political feasibility”.

46

Since Robert and Thumin published their report in 2006 there have been a growing number

of studies that have looked directly or indirectly at the political acceptability of PCT (Bird,

Jones, & Lockwood, 2008; Bristow et al., 2010; Capstick & Lewis, 2009; Energy Saving Trust,

2007; Jagers, Löfgren, & Stripple, 2009; Keay-Bright, Fawcett, Howell, & Parag, 2008; Owen,

Edgar, Prince, & Doble, 2008; Parag & Eyre, 2010; Wallace, 2009). As with other PCT

research, with a few exceptions these studies have largely focused on the UK.

In 2008 a pre-feasibility study into TEQ's was undertaken by DEFRA. The study found that

"personal carbon trading has potential to engage individuals in taking action to combat

climate change, but is essentially ahead of its time and expected costs for implementation

are high" (Department for Environment Food and Rural Affairs, 2008).

While most of the research into PCT has been undertaken in the UK, particularly in the

earlier years, some contributions have been made from other European and American

researchers (Fawcett & Parag, 2010, p. 332). In 2008 C&S (see above for definition) was

developed for the Irish economy and has been examined by the Irish government. In the

same year the proposed Household Carbon Trading (see above for definition) was

developed in California and has been examined against California's emission targets

(Fawcett & Parag, 2010, pp. 330 - 331).

In 2009 Capstick and Lewis ran a detailed PCT computer based simulation on 65 participants

for the UK Energy Research Centre. The results were published in their 2009 paper

“Personal Carbon Allowances: A Pilot Simulation and Questionnaire”. Their study involved

two components. The first “entails a computer-based simulation of PCA in which

participants complete a simple carbon footprint calculator and then make a series of

decisions in light of a personal carbon allowance allocated to them” (Capstick & Lewis, 2009,

p. 3). The second “entails a comparative questionnaire, in which participants are asked to

indicate willingness to reduce emissions-relevant behaviors” (Capstick & Lewis, 2009, p. 3).

The questions used by Capstick and Lewis in the simulation and questionnaire were used in

part to develop the PCT questions in the survey that was administered to participants in the

NICHE project (see Section 4.2 - 'Survey Design and Construction').

47

In Australia there has been far less research into the political acceptability and the

implementation of a PCT than in the UK. However in recent years several research projects

have taken place. In 2008 the Carbon Pollution Reduction Scheme (CPRS) was proposed by

the Australian government. In November of that year Sonia Akter and Jeff Bennet

conducted a web based survey “in which over 600 households from the state of New South

Wales were asked for their willingness to bear extra household expenditure to support the

Carbon Pollution Reduction Scheme” (Akter & Bennett, 2011, p. 417). In 2011 they

published the results of their study in their paper “Household perceptions of climate change

and preferences for mitigation action: the case of the Carbon Pollution Reduction Scheme in

Australia”. Their study resulted in four key findings as follows (Akter & Bennett, 2011, p.

417):

1. “Respondents’ willingness to pay for climate change mitigation is significantly

influenced by their beliefs of future temperature rise”

2. “Perceptions of policy failure have a significant negative impact on respondents’

support for the proposed mitigation measure”

3. “Respondent preferences for the proposed policy are influenced by the possibility of

reaching a global agreement on emissions reduction”

4. “Respondents’ willingness to take action against climate change, both at the national

and household level, is found to be influenced by their level of mass-media exposure”

These findings are highly pertinent to this thesis and the broader NICHE project as one of

the objectives of the broader project is to assess the political acceptability of introducing

PCT into a population as a tool for reducing carbon emissions. Other research conducted

into the CPRS was focused on specific industries and are outside the scope of this thesis.

In 2011 the Commonwealth Scientific and Industrial Research Organisation (CSIRO)

examined "Australians’ views of climate change, their beliefs about the role of human

activities in producing climate change, and their support for various policy responses to

climate change" (Leviston et al., 2011). The study had the following aims (Leviston et al.,

2011, p. 1):

"the extent of community views in Australia that climate change is happening, and

whether it is attributable to human activity or is a natural variation"

48

"whether community views of climate change have altered in recent years"

"differences between sectors of the community (e.g. rural vs urban) in views about

climate change"

"comparability of views in Australia with views in the UK and the USA"

"community support for mitigation actions and community willingness to pay for

actions, policies, and other responses"

"community support for specific policy options (e.g. an emissions trading scheme, a

carbon price, etc)"

The paper identified and analysed a total of 22 studies with the following criteria (Leviston

et al., 2011, p. 2):

"time frame (January 2008 to March 2011)"

"location (Australia, New Zealand, UK, or USA)"

"methodology (a quantitative survey that attempted to sample the whole population

or at least a broad segment of the population; sufficient information to enable

analysis of the methodology was available to the review team)"

"relevance of questions asked (survey questions at least probed perceptions of

climate change and, where possible, perceptions of climate-change related policies)"

The relevant information from each study was extracted and the results were published in

the 2011 report "Australians views of climate change" (Leviston et al., 2011). The study

concluded that:

"Most Australians believe the climate is changing, but fewer believe that it is

attributable to human activity"

"Belief in climate change and its anthropogenic drivers has waned in recent years,

matching trends in other Western countries"

"Responses to questions about climate change vary systematically with question

wording and response format, but these differences do not negate the overall

conclusions above"

49

"Beliefs about climate change are strongly related to political beliefs / voting

behaviours and gender, but relationships to location, age or income have not clearly

emerged"

"Most Australians believe that Australia should take action on climate change

without waiting for global consensus"

"There is no clear consensus of what policy actions Australians prefer such as setting

a carbon price, an emissions trading scheme, or a carbon pollution reduction scheme"

The questions of interest that were identified in the "Australians views of climate change"

were of particular interest to NICHE researchers and were a major influence when designing

the PCT section of the NICHE survey (see Section 4.2 - 'Survey Design and Construction')

While there is merit in the objectives of a PCTS there also needs to be the impetus for its

implementation. This impetus needs to be political as well as social. The beneficial impacts

that a PCTS could have in terms of household carbon emissions is not in dispute. However

the intricacies of implementation and political acceptance are presently the main

impediments. The NICHE project is implementing as PCTS on Norfolk Island with the aim of

assessing its impact on individuals and their households. While the system does not have a

‘trading’ component, it has the features that allow tracking of household consumption and

carbon emission for key household items. The data from system operation won’t be

available for analysis until mid 2014. However, it is this data that the researchers believe

will provide evidence related to the effectiveness of PCTS and identification of PCTS features

that developers need to consider in the implementation of PCTS across broader

populations.

2.6 Information System Success

An information system (IS) is "a set of interrelated computer components that collects,

processes, stores and provides as output the information needed to complete business tasks"

(Satzinger, Jackson, & Burd, 2012). A PCTS falls under this definition. How to judge the

success of an information system after its implementation is a contentious issue and there

are a number of models that measure information system success that have been proposed

50

over the last thirty years (Davis, 1989; DeLone & McLean, 1992, 2002; Viswanath. Venkatesh

& Davis, 2000; V. Venkatesh et al., 2003). In order to develop and test and validate our

conceptual model (see Figure 1-1) an extensive review of the IS success modeling literature

was undertaken to ascertain what factors affect usage intentions towards information

systems and how best to measure these usage intentions in the context of the NICHE

project.

While these models have not specifically been developed for or used to measure the

success of a PCTS we need to understand these factors and how best to measure them to

develop, test, and validate the conceptual model. The focus of the research in this thesis is

an individual’s PCTS usage intentions. While not exactly the same thing as IS success, it is

very closely related as usage intentions towards an information system is a component of

success. Therefore while we are not able to simply pick one of the competing IS success

theories and use it as is, we can pick the most relevant relationships and constructs from

each theory and adapt them to create our own model.

Information success models can be separated into two groups, technology acceptance

models and user satisfaction models.

From the technology acceptance group, the models that were reviewed for this thesis

include:




From the user satisfaction group, the models that were reviewed for this thesis include:

DeLone and McLean Information System Success Model (1992)

DeLone and McLean Information System Success Model Revisited (2002, 2003)

Although it is not an information system success model the Theory of Reasoned Action (TRA

-1975; 1980) was also reviewed as most IS success models are derived in part from the TRA

(F.D. Davis et al., 1989, p. 983).

51

2.6.1 Theory of Reasoned Action (1975, 1980)

The Theory of Reasoned Action (TRA) is an intention model that has been successful across a

range of areas predicting and explaining behaviour and is appropriate to use as a theoretical

base to explain information systems usage (F.D. Davis et al., 1989, p. 983).

The TRA was proposed by Martin Fishbein and IcekAjzen (1967; 1975; 1980). The aim of the

TRA is to explain volitional behaviour. It proposes that volitional behaviour is influenced by

an individual’s behavioural intentions, which in turn is the result of the individual’s attitude

towards performing the behaviour and the subjective norms associated with the behaviour

(Hale, Householder, & Greene, 2002, pp. 259-260). The TRA model is shown in Figure 2-5

below.

Katherine Miller defines each of the three components of the theory as follows (Miller,

2005, p. 127):

Attitudes: “the sum of beliefs about a particular behavior weighted by evaluations of

these beliefs”

Subjective norms: “looks at the influence of people in your social environment on

your behavioral intentions; the beliefs of these people, weighted by the importance

you attribute to each of their opinions, will influence your behavioral intention”

Normative beliefs and motivation to comply

Subjective Norm

Beliefs and evaluations

Attitude toward act or behavior

Behavior Behavioral Intention

Figure 2-5 Theory of Reasoned Action (F.D. Davis, R.P. Bagozzi, & P.R. Warshaw, 1989, p. 984)

52

Behavioural intention: “a function of both attitudes toward a behavior and subjective

norms toward that behavior”

The TRA provides a useful model for explaining and predicting the behaviour of an individual

however it has some limitations and has received some criticism due to the fact that

attitudes and subjective norms are not weighted equally in predicting behaviour. "Indeed,

depending on the individual and the situation, these factors might have very different effects

on behavioral intention; thus a weight is associated with each of these factors in the

predictive formula of the theory. For example, you might be the kind of person who cares

little for what others think. If this is the case, the subjective norms would carry little weight

in predicting your behavior" (Miller, 2005, p. 127). This led to the revision and extension of

the TRA by Ajzen into the Theory of Planned Behaviour (TPB).

The NICHE project is predicated by the researchers’ belief that behaviours related to health

and the environment can be influenced if people are aware of the impacts. The PCTS being

rolled out on the Island aims to raise user’s awareness of their household’s carbon

emissions and more broadly the influence that a ‘high carbon emission’ lifestyle can have on

individual’s health. Thus an understanding of human behaviours underlies the research.

2.6.2 Technology Acceptance Models

After reviewing the literature on the TAM, TAM2 and UTAUT, the technology acceptance

models were determined to be more relevant to the research than the user satisfaction

models. This is because the area of study for this thesis involves acceptance of a PCTS. As a

result the technology acceptance models, in particular TAM2 are covered in greater detail

than the user satisfaction models.


The Technology Acceptance Model (TAM) was developed by Fred Davis (1989) and is one of

the most influential adaptations of the TRA model to the field of information systems.

While TAM is now dated and has been superseded by TAM2 it is still valid to this research as

a starting point for understanding IS success and the studies conducted by Davis and others

53

were a valuable resource for user acceptance survey items in the questionnaire (see

Appendix A - 'NICHE Survey')

TAM proposes that how and when a user is motivated to use a new technology is influenced

by two major factors perceived usefulness and perceived ease of use (Miller, 2005, p. 127).

The TAM model constructs are show in Figure 2-6 below.

TAM defines perceived usefulness as "the degree to which a person believes that using a

particular system would enhance his or her job performance" (Davis, 1989, p. 320). In an

employment setting this behaviour could result in or at least be perceived to result in better

pay, promotions and/or other rewards. A system that an individual perceives to be useful is

one that will provide a “positive use-performance relationship”(Davis, 1989, p. 320).

TAM defines perceived ease of use as "the degree to which a person believes that using a

particular system would be free from effort" (Davis, 1989, p. 320). If everything else is equal

TAM theorizes that a system that is easier to use is more likely to result in actual system use.

It proposes that perceived usefulness is influenced by perceived ease of use, as the easier an

information system is to use the more useful it is likely to be.

External variables

Perceived usefulness

Perceived ease of use

Attitude toward using

Behavioral intention to

use

Actual system use

Figure 2-6 Technology Acceptance Model (Fred D Davis, Richard P Bagozzi, & Paul R Warshaw, 1989, p. 984)

54

Davis conducted two studies using psychometric scales to measure perceived usefulness and

perceived ease of use to validate the Model. In his first study (Davis, 1989, p. 326), he

evaluated acceptance of an email system at IBM. The second study (Davis, 1989, p. 330)

evaluated user acceptance of two graphics-based programs at Boston University.

Davis’s results testing TAM have been replicated a number of times since its inception. For

example, Adams, Nelson and Todd (1992) tested the validity of TAM's major constructs by

measuring MBA students’ acceptance and use of computer applications that were core to

their studies. As well, Hendrickson, Massey and Cronan (1993) tested the reliability of

perceived usefulness and perceived ease of use evaluating undergraduate student’s

acceptance of a spread sheet application. In their study they were able to validate the

reliability of TAM’s measurement scales. These, and other studies (e.g. Subramanian 1994)

replicated Davis’s results thereby supporting its validity as a measure of technology

acceptance.


In 2000 Davis and Viswanath Venkatesh proposed a major adaptation of the TAM. Named

TAM2, it is the most relevant IS success model reviewed and hence a comprehensive review

is included in this section. The constructs of the TAM2 model were influential in designing

our conceptual model. The research instruments and scales Davis and Venkatesh used to

field test TAM2 were used in the development of a number of the survey items in the

questionnaire (see Appendix A - 'NICHE Survey').

TAM has consistently been found to explain a significant proportion of “the variance

(typically about 40%) in usage intentions and behavior” (Viswanath. Venkatesh & Davis,

2000, p. 186) and has been validated by a number of researchers (Adams, Nelson, & Todd,

1992; Hendrickson, Massey, & Cronan, 1993; Segars & Grover, 1993; Subramanian, 1994)

however it is unable to identify the reasons why a system is perceived to be useful or easy

to use. Davis and Venkatesh (2000) identified that this was particularly true for perceived

usefulness and that this was the fundamental driver of usage intentions. “Since perceived

usefulness is such a fundamental driver of usage intentions, it is important to understand the

determinants of this construct and how their influence changes over time with increasing

experience using the system. Perceived ease of use, TAM’s other direct determinant of

55

intention, has exhibited a less consistent effect on intention across the studies”(Viswanath.

Venkatesh & Davis, 2000, p. 187).

This led Davis and Venkatesh to review the empirical studies and research that has been

conducted since its inception to determine the factors that lead to perceived usefulness and

as a result intention to use and user acceptance. “A better understanding of the

determinants of perceived usefulness would enable us to design organizational interventions

that would increase user acceptance and usage of new systems. Therefore, the goal of the

present research is to extend TAM to include additional key determinants of TAM’s perceived

usefulness and usage intention constructs, and to understand how the effects of these

determinants change with increasing user experience over time with the target system”

(Viswanath. Venkatesh & Davis, 2000, p. 187).

In 2000, Davis and Venkatesh proposed a new model, referred to as TAM2. It posits that

perceived usefulness and perceived ease of use are the two major factors that determine

intention to use and introduces new variables to explain perceived usefulness and usage

intentions in terms of social influence and cognitive instrumental processes. The new

variables and relationships TAM2 defines are:

Social influence processes

Subjective norm and its effect on image, perceived usefulness and intention to use

Voluntariness and the moderating effect it has on subjective norms intention to use

Experience, its effect on perceived usefulness and its moderating effect on subjective

norms intention to use

Image and its effect on perceived usefulness

Cognitive instrumental processes

Job relevance and its effect on perceived usefulness

Output quality and its effect on perceived usefulness

Result demonstrability and its effect on perceived usefulness

The TAM 2 model constructs are show in Figure 2-7 below.

56

The hypotheses related to TAM2 are reproduced below so that the reader can appreciate

the relationships among the model’s constructs. It also highlights relevancy to the aims of

the current research in establishing the importance of setting measures for acceptance of

PCTS as part of the NICHE project.

“HYPOTHESIS 1a. Subjective norm will have a positive direct effect on intention to use when

system use is perceived to be mandatory.”(Viswanath. Venkatesh & Davis, 2000, p. 188)

“HYPOTHESIS 1b. Subjective norm will have no significant direct effect on intention to use

when system use is perceived to be voluntary.”(Viswanath. Venkatesh & Davis, 2000, p. 188)

“HYPOTHESIS 1c. Voluntariness will moderate the effect of subjective norm on intention to

use.”(Viswanath. Venkatesh & Davis, 2000, p. 188)

“HYPOTHESIS 2. Subjective norm will have a positive direct effect on perceived usefulness.”

(Viswanath. Venkatesh & Davis, 2000, p. 189).

Subjective Norm

Image

Job Relevance

Output Quality

Experience Voluntariness

Result Demonstrability

Perceived Usefulness

Perceived Ease of Use

Intention to Use Usage Behavior

Technology Acceptance Model

Figure 2-7 Technology Acceptance Model 2 (Viswanath. Venkatesh & Davis, 2000, p. 188)

57

“HYPOTHESIS 3a. Subjective norm will have a positive effect on image.” (Viswanath.


“HYPOTHESIS 3b. Image will have a positive effect on perceived usefulness.”(Viswanath.


“HYPOTHESIS 4a. The positive direct effect of subjective norm on intention for mandatory

systems will attenuate with increased experience” (Viswanath. Venkatesh & Davis, 2000, p.

190)

“HYPOTHESIS 4b. The positive direct effect of subjective norm on perceived usefulness will

attenuate with increased experience for both mandatory and voluntary systems”

(Viswanath. Venkatesh & Davis, 2000, p. 190)

“HYPOTHESIS 5. Job relevance will have a positive effect on perceived usefulness”


“HYPOTHESIS 6. Output quality will have a positive effect on perceived usefulness”


“HYPOTHESIS 7. Result demonstrability will have a positive effect on perceived usefulness”


“HYPOTHESIS 8. Perceived ease of use will have a positive effect on perceived

usefulness”(Viswanath. Venkatesh & Davis, 2000, p. 192)


The unified theory of acceptance and use of technology (UTAUT) was developed by

Viswanath Venkatesh, Michael Morris, Gordon Davis and Fred Davis (2003). The UTAUT

attempts to “(1) review user acceptance literature and discuss eight prominent models, (2)

empirically compare the eight models and their extensions, (3) formulate a unified model

that integrates elements across the eight models, and (4) empirically validate the unified

model” (V. Venkatesh et al., 2003, p. 425). The eight models that the UTAUT attempts to

integrate are:

Theory of Reasoned Action (Ajzen and Fishbein, 1967; 1975; 1980)

58

Technology Acceptance Model (Davis, 1989) and the extended Technology

Acceptance Model (Venkatesh and Davis, 2000)

Motivational Model (Vallerand, 1997)

Theory of Planned Behaviour (Ajzen, 1991)

Combined TAM and TPB (Taylor and Todd, 1995)

Model of PC Utilization (Thompson et al., 1991)

Innovation Diffusion Theory (Rogers, 1995) adapted for information systems by

Moore and Benbasat (1991)

Social Cognitive Theory (Bandura, 1986) adapted for information systems by

Compeau and Higgins (1995)

After reviewing the competing models seven constructs that appear in one or more models

were identified that appeared to be “significant direct determinants of intention or usage”

(V. Venkatesh et al., 2003, p. 446). Of these seven constructs only four were theorized to

“play a significant role as direct determinants of user acceptance and usage behavior” (V.

Venkatesh et al., 2003, p. 447). These constructs are:

Performance expectancy: “the degree to which an individual believes that using the

system will help him or her to attain gains in job performance” (V. Venkatesh et al.,

2003, p. 447)

Effort expectancy: “the degree of ease associated with the use of the system” (V.

Venkatesh et al., 2003, p. 450).

Social influence: “the degree to which an individual perceives that important others

believe he or she should use the new system” (V. Venkatesh et al., 2003, p. 451).

Facilitating conditions: “the degree to which an individual believes that an

organizational and technical infrastructure exists to support use of the system” (V.

Venkatesh et al., 2003, p. 453).

The other three constructs identified were theorized “not to be direct determinants of

intention”(V. Venkatesh et al., 2003, p. 447). These constructs were attitude toward using

technology, self efficacy and anxiety.

59

Within each of UTAUT’s four constructs there are four key moderators that Venkatesh et al.

have theorized will mediate the impact of each construct. These are gender, age,

experience, and voluntariness of use.

The UTAUT model constructs are show in Figure 2-8 below.

While UTAUT is the most comprehensive of the models, the constructs are drawn from

existing models and the influences amongst its variables have been explored in individual

studies. Nonetheless, it is one of the recognized models and needs to be examined in any

context dealing with IS success.

2.6.3 User Satisfaction Models

User satisfaction models were also reviewed as part of the background information that was

thought useful in the development of the conceptual model. There are constructs from this

Performance expectancy

Effort expectancy

Social influence

Facilitating conditions

Gender Age Experience Voluntariness

of use

Behavioural intention

Usage

behaviour

Figure 2-8 Unified Theory of Acceptance and Use of Technology (UTAT) (V. Venkatesh, Morris, Davis, & Davis,

2003, p. 447)

60

set of models that were thought to influence attitudes towards information systems. After

reviewing the UIS and its adaptations it was decided that the constructs from TAM2 were

most closely aligned with the directions of the current study. However this group of models

provide invaluable insight into IS success and the case studies validating the models were

useful when designing the survey. The statistical techniques from these studies are also of

particular relevance and informed the data analysis undertaken for the NICHE project.

The DeLone and McLean Information System Success Model (1992)

Throughout the late 1970’s, 1980’s and in to the 1990’s a large number of studies were

conducted to try and identify the key factors that led to information systems success.

However almost all of the studies produced different measures of information systems

success.

In 1992 William DeLone and Ephraim McLean reviewed 100 conceptual and empirical

studies in an attempt to “bring some awareness and structure to the ‘dependent variable’ —

IS success — in IS research” (DeLone & McLean, 2003, p. 10). After reviewing the studies

they concluded there was not just one success measure but many. “However, on more

careful examination, these many measures fall into six major categories—SYSTEM QUALITY,

INFORMATION QUALITY, INFORMATION USE, USER SATISFACTION, INDIVIDUAL IMPACT and

ORGANIZATIONAL IMPACT. Moreover, these categories or components are interrelated and

interdependent, forming an I/S success model. By studying the interactions along these

components of the model, as well as the components themselves, a clearer picture emerges

as to what constitutes information systems success” (DeLone & McLean, 1992, p. 88)

The DeLone and McLean model constructs are show in Figure 2-9 below.

61

The DeLone and McLean Information System Success Model Revisited (2002, 2003)

In 2002 DeLone and McLean revisited their model resulting in an updated version. The main

changes were:

Service Quality being added to System Quality and Information Quality as

measurements of information system success

Use being split into Intention to Use and Use

Individual Impact and Organizational Impact being replaced with Net Benefits

The updated DeLone and McLean model constructs are show in Figure 2-10 below.

Individual Impact

Organizational Impact

Use

User Satisfaction

Information Quality

Figure 2-9 Delone and Mclean Information System Success Model (DeLone & McLean, 1992, p. 87)

System Quality

62

2.7 Summary

In this chapter research relevant to the broader NICHE project and this specific sub project

was examined.

The chapter started with an overview of the literature on health and more specifically

obesity in order to understand the research being conducted by the broader NICHE project

and where the focus of the research of this thesis fits into the broader project. The

literature identified the relationship between health and the obesity epidemic that started

in the developed world and is now spreading in to the developing world.

Current research into carbon emissions and climate change was covered next. Aspects of

the research surrounding carbon emissions and climate change is fundamental to the

broader NICHE project and is also relevant to the specific focus of the research described in

this thesis. The literature described in this section of the chapter provided the some of the

basis for the development of the conceptual model (see Figure 1-1).

The next section of the chapter reviewed the literature on the causes of obesity and the

causes of carbon and GHG emissions and what links exist between them. This link has been

proposed by a number of researchers however an extensive search of the literature has not

Information quality

System quality

Service quality

Net benefits

Intention to

use

User satisfaction

Use

Figure 2-10 Updated Delone and Mclean Information System Success Model (DeLone & McLean, 2003, p. 24)

63

shown any research where it has been tested. Based on the literature review, the

researchers believe that the NICHE project is the first study of its kind to explore the

relationship between obesity and an individual’s carbon footprint. This relationship is also

fundamental to the focus of this thesis as an individual’s attitudes and behaviours towards

the components of this relationship underlie the relationships and constructs of the

conceptual model.

A review of the literature covering PCT was covered in detail. Understanding PCT and

carbon allowances is fundamental to the NICHE project. The primary objective of the NICHE

project is to assess whether personal carbon allowances are effective in reducing an

individual's carbon footprint and what impact this has on health behaviours associated with

obesity such as active transport and diet. One of the secondary objectives of the NICHE

project are to assess the political acceptability of introducing PCT (Personal Carbon Trading)

into a population as a tool for reducing emission reductions. PCT is also highly relevant to

this thesis. The literature covered in this section was used to design parts of the NICHE

survey (see Section 4.2 - 'Survey Design and Construction') and provided the basis for the

development of the dependent variable of the conceptual model that is the basis of this

thesis.

Finally the literature on information success was covered. Models from the user satisfaction

and technology acceptance groups were reviewed and the technology acceptance models

were determined to be more relevant to the research. The literature in this section was

used to understand how best to develop, test, and validate the conceptual model and to

develop the NICHE survey and the scales used to measure the responses to its containing

variables.

The next chapter describes the NICHE PCTS that was programmed by the author of this

thesis to track the carbon footprint of the participants of the NICHE project.

64

3 The NICHE PCTS

3.1 Introduction

The NICHE project is unique and the project team believes it is the first real trial attempting

to identify links between obesity and an individual’s carbon footprint. This meant the

project required systems for monitoring carbon consumption. The system used for the

project was developed specifically for the project. The NICHE Personal Carbon Trading

System (N-PCTS) is used to track the carbon emissions on selected products and services for

all of the individuals who are participating in the NICHE project. This chapter describes the

N-PCTS system’s major components and features.

It was constructed as a web based system with the following components:

A web enabled Point Of Sale (POS) system at each sales outlet

A central database

An system administration website

A user’s/participant’s website

All of the sales outlets participating in the NICHE project have been fitted with a custom POS

terminal that records selected purchases made by individuals participating in the NICHE

project. The POS terminals synchronize the data that they collect with the central

database on the web server. Consumption data for electricity and gas is provided by the

utility companies on Norfolk Island and is entered into the database via a custom web

service. Software running on the server calculates the carbon emissions produced by these

products and services. Users can view their carbon emission and personal data by logging

into the Participants website. The system is managed by NICHE staff through the

Administrators website.

An overview of the N-PCTS can be seen in Figure 3-1 below.

65

3.2 NICHE POS

The POS terminals use a custom application installed on a 10 inch android tablet to record

selected purchases made by individuals participating in the NICHE project. All participants

were assigned a NICHE card and key ring (see Figure 3.2) during the registration process.

The card and key ring contain a unique identification number that is used to identify the

participant. When the participant makes a purchase their card is scanned using the camera

on the rear of the tablet and their unique identification number is registered against their

purchases.

POS 1

POS 2

POS 3

Server

Database

Administration

website

Participant’s

website

Utilities

Figure 3-1 Overview of N-PCTS

66

Figure 3-2 - NICHE Card

A number of Android tablets were tested before settling on the Asus Transformer TF300T

(https://www.asus.com/Tablets_Mobile/ASUS_Transformer_Pad_TF300T/). The Asus

Transformer TF300T was chosen as it has an 8MP auto-focus camera on the rear of the

tablet which was the most suitable camera of any tablet tested for using as a barcode

scanner. Other tablets tested include the Samsung Galaxy Tab 10.1, the Motorola Xoom

and the Acer Iconia. The tablets are mounted on the counter at each sales outlet using RAM

mounts (http://www.rammount.com) as shown in Figure 3-3 below.

67

Figure 3-3 NICHE Tablet at Participating Outlet

The custom Android POS application (see Figures 3-4 and 3-5 below) was programmed using

the Android Java API (http://developer.android.com/index.html). The application uses a

self-contained embeddable Sqlite database (http://www.sqlite.org) to store all of the sales

data, along with data unique to each sales outlet. The application periodically synchronises

the Sqlite database with the central database via the server. Sales data recorded by the

application is uploaded from each POS terminal and new products and user data is

downloaded to the POS terminals from the server.

68

Figure 3-4 NICHE POS

Figure 3-5 NICHE POS

69

3.3 NICHE Server

The NICHE server links the NICHE POS terminals, the database, the administration website

and the participant’s website together to create the N-PCTS.

It has three main components:

The N-PCTS software

The web server and servlet container (software) that runs the N-PCTS software

The physical server (hardware) that hosts the web server software

The N-PCTS software was written in Oracle Java version 7 (http://java.com/en/). It runs on

an Apache Tomcat version 7 server (http://tomcat.apache.org). The NICHE Tomcat server is

hosted on an (http://aws.amazon.com/ec2/) Amazon Elastic Compute Cloud (EC2) instance.

The N-PCTS software performs the following functions:

Syncs the POS databases with the central database

Allows NICHE staff to upload data and assign it to individuals or groups for services

like electricity and household gas that are not captured by the POS terminals

Calculates the carbon emissions produced by participants based upon their

purchases or services

Produces a monthly statement for each of the sales outlets listing all the purchases

that NICHE participants have made with them and calculates the refund that they

are owed for the discounts given on selected products

Produces a quarterly statement for each of the participating individuals in the NICHE

project. The statement lists all the purchases they made and services they have used

individually or as part of their group and the carbon emissions associated with each

product or service. The carbon emissions are charted to show the individual how

their carbon emissions for that quarter ranked against previous quarters and against

other individuals and groups who are participating in the project

Provides an object relational persistence architecture framework layer for all the N-

PCTS components that need access to the database. This framework uses Eclipse

Link (http://www.eclipse.org/eclipselink/) for its underlying architecture

70

3.4 NICHE Database

All of the data associated with the N-PCTS is stored in a MySQL database

(http://www.mysql.com). The database runs on an Amazon Relational Database Service

(RDS) on Amazons cloud (http://aws.amazon.com/rds/).

3.5 NICHE Administration Website

The NICHE administration website (see Figure 3-6) allows project staff to administer all

aspects of the N-PCTS. Administration of the NICHE system falls into the following four

areas.

Participants

All individuals who register to participate in the NICHE project must have their personal

details recorded in the NICHE database with their unique NICHE number. They are assigned

a username and password they use to logon to the NICHE participant’s website. NICHE staff

can add new participants to the N-PCTS using the administration website and update the

details of existing participants.

Groups

Individuals participating on the NICHE project are allocated to a group which is defined in

the N-PCTS system as a collection of individuals who live at the same residence. Purchases

that are tracked by the N-PCTS fall into two categories, individual purchases or group

purchases. Group purchases include items such as electricity or gas that are for the

household. These purchases are recorded against the group so that the carbon emissions

produced by the purchase can be split among the individuals in the household (NICHE

Group). Using the NICHE administration system staff can assign individuals to a group or

update groups if members leave a group or join a new group. Individuals participating in the

NICHE project can only be a member of one group at any one time.

71

Products

The N-PCTS tracks participant’s purchases of selected products. Some of these products are

offered at a discount to participants as an incentive to take part in the project. Before a

product can be tracked it must be entered into the NICHE database along with any discount

attached to it. NICHE staff manages the products that are tracked by the N-PCTS using the

administration website.

Suppliers

Sales outlets that are participating in the NICHE project and are fitted with a NICHE POS

system are recorded in the database. Supplier details can be added or updated using the

administration website.

Figure 3-6 NICHE Admin Website

3.6 NICHE Participants Website

The NICHE participant’s website (see Figure 3-7) allows individuals participating in the NICHE

project to view the products that they have purchased and the services that they have used

and track the associated carbon emissions values. The carbon emission value is shown for

each purchase as well as a total calculated from the sum of the individual carbon emission

72

values of their purchases. Participants can view their household’s (group’s) carbon emission

values as well as their individual carbon emission values. The carbon emissions are charted

to show the individual how their carbon emissions for that quarter ranked against previous

quarters and against other individuals and groups who are participating in the project.

73

Figure 3-7 NICHE Participants website

74

3.7 Conclusion

In this chapter the N-PCTS that is used to track the carbon emissions on selected products

and services for all of the individuals who are participating in the NICHE project was

reviewed. The POS, database, administration website and the user’s/participant’s website

that were programmed specifically for the NICHE project were examined. The next chapter

describes the methodology used for the NICHE project.

75

4 Methodology

4.1 Introduction

This section describes the methodology used for the study. The study reported in this thesis

is part of a larger study that has a number of components and projected outcomes. The

outcomes being reported in this thesis are one component of that study and focus on the

usage intentions of residents of Norfolk Island to a Personal Carbon Trading System (PCTS)

and the relationships between Personal Carbon Trading (PCT) and attitudes to their own

health and environmental concerns. The data that is being examined and reported on for

this component of the study was collected via survey that included questions specific to the

aims of this study and questions collecting data relevant to the other components of the

broader study.

This chapter commences with a discussion of the design and construction of the NICHE

survey. It details the pilot study and the administration of the survey and provides a brief

comparison of the NICHE process to the guidelines and administrative processes used for

the 2011 Norfolk Island census. The sample selection is detailed. A comparison of

respondents characteristics with demographic characteristics from the 2011 census is

provided in the next chapter. Finally the limitations of the survey and its administration are

discussed.

4.2 Survey Design and Construction

This section of this chapter details the design and construction of the NICHE survey

4.2.1 The Broader NICHE Project

The NICHE project is a multi-disciplinary study involving information systems (the focus of

this thesis), health, environmental concerns, global warming, carbon emissions and carbon

trading. The multi-disciplinary nature of the study was reflected in the design and

76

construction of the survey (see Appendix A - 'NICHE Survey'). The survey contained a range

of questions covering the information systems, health, environment and the carbon

emissions components of the NICHE program.

To cover all of the objectives of the NICHE project some of the questions were specifically

aimed at the individual respondent while others were aimed at their household. Most of

the questions related to health, obesity, physical activity, PCTS usage intentions and some

behaviours towards the environment and climate change are only relevant for an individual

and as such had an 'individual' focus. Other questions related to behaviours towards the

environment and climate change are only relevant for a household and as such had a

'household' focus. This is discussed further in Section 4.5 - 'sample selection' of this

chapter.

The survey was constructed in sections which aligned with the researcher’s expectations

that there were higher order factors that were expected to emerge from analysis of the

data.

The survey contained the following sections:

General information (questions A1 – A13) - general information about the

respondents and their household, their health and their beliefs about their own and

their households carbon foot print

Attitudes (questions B1 – B13) - the respondent and their households attitudes and

towards health, the environment, carbon emissions and climate change

Behaviours (questions B14 – B19) - the respondent and their households behaviours

towards consumption and climate change

Physical activity (questions C1 – C15) - the respondent and their households physical

activity

Nutrition (questions D1 – D31) - the respondent diet and nutrition

Personal carbon trading (questions E1 – E10) - the respondents usage intentions

towards and beliefs about a PCTS

Demographic data (questions F1 – F28) - the respondent and their households

demographic data

77

Some of the questions used in the NICHE survey were derived in part or based on survey

items from existing surveys and publications. Additional questions were developed by the

NICHE research team to cover aspects of the research for which there were no existing

studies. The unique nature of the study meant that a number of questions had to be

developed. The questions and scale descriptors developed by the research team specifically

for this sub project and the publications and surveys that influenced them are discussed in

the next section.

4.2.2 Examination of PCTS Usage Intentions

As this thesis is concerned with PCTS usage intentions, a subset of the questions from the

survey were of relevance to this thesis. This section of the chapter details the questions in

the survey that are the key focus of the thesis and the scale descriptors that were developed

to measure the responses to the questions. The full questionnaire has been included in

Appendix A - 'NICHE Survey'.

Measurement Scale Descriptors

The user satisfaction and technology acceptance group of models that were reviewed in the

literature review section of this thesis used surveys to capture measures of information

system satisfaction and success. These studies, survey questions and measurement scale

descriptors that the models authors and other researchers have used to empirically test the

models were used as reference points when designing parts of the survey for the NICHE

project.

The survey used for the research conducted for the original version of TAM (Davis, 1989)

contained questions with 7 point Likert scales. 7 point scales were also used in subsequent

versions of the model including TAM2 (Venkatesh and Davis, 2000), UTAT (Venkatesh,

Morris, Davis and Davis 2003) and the DeLone and McLean Information System Success

Model (DeLone and McLean 1992). The number of responses that were expected from the

survey administered to households on Norfolk Island meant that the use of a 7 point scale

would provide a greater level of granularity when analyzing responses than the 5 point

scales used in other parts of the study. Therefore 7 point scales were used for all survey

78

items seeking attitudinal and behavioural measures relating to the environment, carbon

emissions and PCTS. The scale descriptors for attitudinal scales ranged from 1 - "strongly

agree" to 7 – "strongly disagree" with midpoint of 4 - "neutral". The scale descriptors for

behavioural scales ranged from 1 - "never" to 7 – "always" with midpoint of 4 -

"sometimes".

However, it should be noted that questions relating to the respondents self-health

evaluation used 5 point Likert scales with dedicated scale descriptors, rather than the 7

point descriptors previously discussed. This was done to match the data collection

requirements of the broader NICHE project.

Survey Questions

The following questions were used from the NICHE survey as the basis for this thesis.

Attitudes Towards Health, the Environment, Carbon Emissions and Climate Change

The questions in this section of the survey were used to assess Norfolk Islanders attitudes

towards health, the environment, carbon emissions and climate change. They were

included as measures of the attitudes and behaviours towards health, attitudes and

behaviours towards the environment and attitudes and behaviours towards carbon

emissions and climate change components of the conceptual model introduced in Figure 1-

1. They were developed by NICHE researchers but some questions were influenced by the

following surveys and publications:

Baseline Survey of Australian attitudes to climate change (Leviston & Walker, 2011)

Australians’ views of climate change (Leviston et al., 2011)

Who cares about the environment in 2009? (Department of Environment Climate

Change and Water NSW, 2010)

The WHO STEPwise approach to chronic disease risk factor surveillance (STEPS)

(World Health Organization, 2008)

This section included the following questions:

B1. I buy environmentally friendly products as much as I can.

79

B2. Technology will solve future environmental problems

B3. Being overweight can have serious health effects

B4. Obesity will be solved in the future by medical advances

B5. It is important for me to have a low carbon footprint

B6. A financial incentive would encourage me to reduce my environmental impact

B7. Collectively, households can reduce the impacts of greenhouse gas emissions

B8. I always try to eat healthy food

B9. I am confident I could maintain a healthy body weight if I wanted to

B10. I would consider purchasing an electric car or bike if the price was right

B11. Walking or cycling instead of using the car can help reduce a person’s weight

B12. I am unlikely to ever be obese

B13. I am worried about climate change

All of these questions used scale descriptors ranging from 1 - "strongly agree" to 7 –

"strongly disagree" with midpoint of 4 - "neutral".

Behaviours Towards Consumption and the Environment

The questions in this section of the survey were used to assess Norfolk Islanders behaviours

towards consumption and the environment. They were included as measures of the

attitudes and behaviours towards the environment component of the conceptual model.

The questions in this section were developed by NICHE researchers but some questions

were influenced by the paper Who cares about the environment in 2009? (Department of

Environment Climate Change and Water NSW, 2010). This section included the following

questions:

B14. I turn the tap off when cleaning my teeth

B15. I turn lights off when not in use

B16. I sort my rubbish

B17. I look to buy second hand over brand new

B18. I consciously try to reduce waste and recycle

B19.I buy local produce, even if imported is cheaper

80

All of these questions used scale descriptors ranging from 1 - "never" to 7 – "always" with

midpoint of 4 - "sometimes".

Self Health Evaluation

The conceptual model includes a self health evaluation component. The NICHE survey did

not include a dedicated self health evaluation section however the broader NICHE study has

a large health related component and there were a number of questions in different

sections that provided a basis to measure the self reported health of the respondent. The

following questions were used for the self health evaluation component of the study and

are intended to measure the self health evaluation component of the conceptual model:

A9. Do you generally consider your health to be………

A10. How would you best describe yourself?

A12. Compared to others on the island of similar age and gender do you consider

your body weight to be…..

C1. How often do you engage in leisure time physical activity for the sole purpose of

improving or maintaining your health?

All of these questions had dedicated scale descriptors. For more information the full NICHE

survey is included in Appendix A - 'NICHE Survey'.

The self health evaluation questions were developed by NICHE researchers but some

questions were influenced by the following surveys and publications:

Development of the World Health Organization Global Physical Activity

Questionnaire (GPAQ) (Armstrong & Bull, 2006)

The WHO STEPwise approach to chronic disease risk factor surveillance (STEPS)

(World Health Organization, 2008)

PCTS Usage Intentions

To assess PCTS usage intentions a section of the survey included questions specifically

related to users’ acceptance of and satisfaction of a PCTS. The 10 questions in this section

were influenced in part by the questions used to test the user satisfaction models and

technology acceptance models (see Section 2.6 - 'Information System Success'). While the

81

questions in these surveys were not directly suitable for the NICHE study they provided an

overview of how to best measure user satisfaction and the success of information systems

in general and were reviewed in order to develop suitable scales to test PCTS usage

intentions which falls under the definition of an information system. The questions in this

section were also influenced by Personal carbon allowances: a pilot simulation and

questionnaire (Capstick & Lewis, 2009). The questions in this section were included as

measures of the dependent variable PCTS usage intentions of the conceptual model and

included:

E1. Being able to measure my carbon footprint is important to me

E2. Most people would accept a PCT system as a tool for improving the environment

E3. A PCT system would encourage me to reduce my carbon footprint

E4. A PCT system would encourage me to walk or cycle more and drive less

E5. People who reduce their carbon footprint should be rewarded in some way

E6. People with a greater carbon footprint should have to pay for it in some way

E7. A PCT system would encourage me to eat more healthy, locally grown produce

E8. A PCT system would be useful for me to help monitor my environmental impact

E9. Comparing my carbon usage to the average would influence my consumption

habits

E10. There is a strong link between a person’s carbon footprint and their health

All of these questions used scale descriptors ranging from 1 - "strongly agree" to 7 –

"strongly disagree" with midpoint of 4 - "neutral".

Demographics

Questions A1 and A2 detailing the respondent’s gender and age were used to categorize

responses in the descriptive analysis in Chapter 5. Other demographic data collected was

used to compare the sample from the NICHE survey to the 2011 Census to make sure that

the survey sample was consistent with the population of the Island (see Chapter 5). These

questions were influenced in part by Norfolk Island 2011 Census of Population and Housing

(Taylor & McNiel, 2011).

82

4.3 Focus Groups - Pilot Study

After the design and construction of the NICHE survey focus groups were convened to

evaluate the baseline survey and assist in development of the final survey to be

administered to the Island’s residents. The focus groups contained 20 male and female

residents of Norfolk Island ranging in ages from 12 to 86. While the NICHE survey was only

open to Islanders over the age of 18 younger individuals were asked to take part in the focus

groups to make sure that the survey questions were easily understood. 18 members of the

focus groups were asked to take the survey home and return it when they had completed it.

The other 2 members completed the survey with the NICHE project officers as an interview

as this was a planned option for residents who were unable to complete the survey by

themselves. Upon completing the survey all members of the focus groups were asked:

How long the survey took?

Did they understand all of the questions?

Were there any questions they were uncomfortable with or they felt were

offensive?

Were there any spelling or grammatical errors?

Did they have any further suggestions / feedback?

Further goals of the focus groups were to:

Inform participants about the study

Come up with ideas to increase participation in the baseline survey and project

Brainstorm lists of items for the survey involving common local foods and activities

Feedback and comments from the focus groups that informed changes to survey items and

the survey administration and promotion were:

Minor changes in the wording, spelling and grammar used in the survey.

The survey being conducted anonymously owning to the sensitive nature of some of

the self health and demographic questions.

Sell the benefits of the NICHE project e.g. healthier environment, healthier people

and a self-sustainable island.

83

Be positive – not doom and gloom about environment

Be mindful of the language in the promotion and survey - don’t alienate people by

talking about obesity too much.

Don’t come across as climate change fear-mongers.

Education is important e.g. Hint of the week in the local paper and/or information

about products

Promote participation as being for the greater good of the community and island –

some residents might perceive that only people already interested would

participate.

Consider literacy problems with the survey

Do survey drives at the Food-Land supermarket

In addition to the focus groups, the survey was also examined by the 2011 Norfolk Island

census field supervisor and the NICHE project field officers on Norfolk Island.

4.4 Survey Administration

After considering alternatives researchers decided to use a paper-based survey. This

decision was made for a number of reasons. Internet speeds and access on the Island is not

reliable with some areas having no access at all, so it was not feasible to conduct the survey

online. The census was conducted on Norfolk Island in 2011 was also paper-based. By using

paper-based surveys the NICHE survey was able to use the same guidelines and

administration practices as the census. Finally, rather than collecting data via differing

media it was decided to use the paper-based version in consultation with the Norfolk Island

Government representatives, whom preferred this method.

The survey and an accompanying cover letter detailing the NICHE project were delivered by

hand to Island residents in an envelope that had instructions for filling out the survey and

contact information for the two Norfolk Island based NICHE project officers printed on the

front (see Appendix C - 'NICHE Cover Letter'). Ethics approval for the survey was obtained

from the Human Research Ethics Committee at Southern Cross University in January 2012.

The ethics approval number allocated by the university was ECN-12-012.

84

Two local assistants were recruited to assist the two Norfolk Island project officers who

administered the survey. The two project officers were responsible for survey distribution

and collection. In the lead up to the survey the project coordinator and project supervisor

conducted training sessions for the four members of the team.

Section 8 of the Census and Statistics Act 1961 defines six census districts (see Figure 4-1).

The districts do not have any administrative status but provide a geographical framework

for surveys on the island. Each member of the team was assigned one and a half districts.

The survey was hand delivered to every house on the island that was identified in the

census as having resident occupants. A large number of houses on the island are for tourist

use and information was only being sought from Norfolk Island residents. Maps of the

island and the census districts were produced for the 2011 census collectors and these were

supplied to the NICHE project officers. The census maps were compared with aerial maps of

the island that were provided by the Norfolk Island administration to avoid any omissions

when the NICHE surveys were delivered. The four team members used their personal

vehicles to deliver the surveys and were reimbursed for expenses.

85

Figure 4-1 Norfolk Island Census Districts

To encourage residents of the island to complete their surveys 6 local clubs were selected (1

for each census district) and $10,000 was budgeted for donations to the clubs as an

incentive. On Wednesday the 22nd of February 2012 a lottery was held on Radio Norfolk

(the local radio station) to assign a club to each census district. The results were as follows:

District 1 - Rotary Club of Norfolk Island

86

District 2 - Lions Club of Norfolk Island

District 3 - Norfolk Island Women's Hospital

District 4 - The Youth Centre

District 5 - St John Ambulance - Norfolk Island Division

District 6 - The Sunshine Club of Norfolk Island

The $10,000 was then divided by 6 to for a total of $1666.66 for each district. This figure

was then divided by the number of surveys delivered in each district to assign a donation

amount per survey for each district. This amount was donated to the club assigned to the

district for each returned survey. The money that was left over for unreturned surveys was

then awarded as a prize to the club for the district that had the highest proportion of

returned surveys (district 5).

Prior to the survey the NICHE project had received island wide publicity through a series of

community meetings, the local media and information stalls at local events such as the

annual show day.

A 6 week media campaign was conducted in the lead up to the survey to further publicize

the NICHE project and create awareness among the local population about the aims and

goals of the project and explain why their help was needed. The media campaign included

the following activities:

Advertisements and articles were published in the Norfolk islander (Norfolk Island

local newspaper) and in Norfolk Online (Norfolk Island online news site).

Interviews with the local NICHE project officers and advertisements were aired on

Radio Norfolk (Norfolk Islands local radio station)

Posters were displayed in prominent positions around the island

An email campaign was undertaken using a mailing list that had been assembled

from previous publicity drives

A promotion was run on the NICHE website and the NICHE Facebook page

The service clubs that were nominated to receive donations also conducted their own

promotions in their designated census district.

87

Residents of the island originally had until Thursday the 22nd of March 2012 to return their

completed survey however this was extended until Monday the 26th of March 2012 to

allow residents to have a full week to complete the survey. A second media campaign was

conducted during this week targeting any households that had been omitted during the

delivery of the survey. During this media campaign the donations to the local clubs were

publicized to ensure maximum participation among the islands residents. However due to

the anonymous nature of the returned surveys it was not possible to follow up households

who had not completed their survey.

The completed surveys were dropped off at the supermarket or the post office in town.

Completed surveys could also be picked up by NICHE staff if required. Each survey was

given an ID number when it was received. These numbers were used only for coding

purposes when the survey data was entered. There was no way of identifying the

respondents from the survey information and there was no link between the survey number

and the respondent.

After the completed surveys were collected the data was converted into electronic format.

To facilitate this a data entry program was created so that the results of the survey could be

input and exported to a Microsoft Excel file. The survey ID number was attached to the

electronic copy for quality control purposes. The data entry was completed by the NICHE

project officers and the assistants on the island. 1 in 10 paper surveys were checked against

their electronic counterparts as part of a quality control process. On completion of the data

entry all of the survey forms were archived.

4.5 Sample Selection

One resident from each household was requested to fill out the survey. The survey was

open to any permanent resident of Norfolk Island or any long term nonresident holding a

GEP (General Entry Permit - visa renewal needed every 5 years) or a TEP (Temporary Entry

Permit - visa renewal needed every 12 months) over the age of 18.

Because of the mix of 'individual' and 'household' focused questions contained in the survey

and given only one occupant of each household was required to fill out the survey, it was

88

recognized that the attitudinal data reflected the attitudes of the respondent and not

necessarily the views of other members of the household. It is also recognized that while

this fits the needs of the project it reduced the potential number of responses and the

representativeness of individual responses.

The aims and objectives of the NICHE project are population wide and it was important to

researchers to have a sample selection that represented this as opposed to having a sample

of a sub section of the population. To try to randomize the sample of residents from each

household who filled out the survey it was requested that of all the individuals living in the

household who were over 18 years of age, the person whose birthday was closest to the 1st

of any month fill out the survey. If that person was not available then it was requested that

the person whose birthday was next closest to the 1st of any month fill it out.

The size of the total resident population identified in the 2011 census was 1795, with 1368

residents being identified as over 19 years of age. A total of 423 households responded to

the survey out of the 805 households that were identified in the census as occupied by

residents of Norfolk Island. This sample represents over 20% of the total population or

almost 30% of the population over 19 years of age.

4.6 Limitations of the Study and the Sample Frame

Norfolk Island is a closed system which makes it ideal for this study. However this suitability

also means that while the results of the study are comprehensive the extrapolation of

results to the general population needs to be undertaken cautiously. The population of the

Island is reasonably representative of other developed locations yet it needs to be

recognized that the geographic and demographic characteristics can reasonably be

expected to affect attitudes to climate change and health. While this may not be a

limitation of the study as such, and more an issue of generalizability, care must be taken

when applying the results of the study to other population groups. For example, as all

resources have to be shipped to the island electricity and fuel are more expensive and a

larger proportion of people have solar power and solar hot water than in other developed

locations which will ultimately affect an individual’s carbon foot print.

89

Another limitation of the study is that the survey was administered to gather both

household data and individual data. Given the survey was completed by a member of the

household the attitudinal items reflect that person’s views and not necessarily the views of

all members of the household. However, given the number of responses and the range of

characteristics, then the survey could be held to be representative of the views of the

broader population of Norfolk Island.

It also needs to be recognized that this survey has been administered to gather data to

develop a baseline on attitudes and household characteristics as a precursor to the roll-out

of a PCTS on the Island. A second survey will be conducted 12 months after system roll-out

that will aim to investigate if an understanding of personal carbon footprints influences

personal behaviours in relation to health. The results being reported here are the baseline

parameters that will be re-evaluated in a secondary study that will be conducted 2014.

4.7 Summary

This section described the methodology used for the study. The design and construction of

the NICHE survey was examined. This was followed by a discussion of the pilot study that

was run prior to the commencement of the main survey. The survey administration is

examined in detail and this is compared with the guidelines and administrative processes

used for the 2011 Norfolk Island census. Finally the sample selection and the limitations of

the study and sample frame were covered.

The following chapters detail the descriptive analysis of the demographic characteristics of

the sample and responses to survey items followed by the data analysis.

90

5 Descriptive Analysis

5.1 Introduction

The conceptual model (see Figure 1-1) that forms the basis of this research hypothesizes

that PCTS usage intentions are influenced by the following factors:


Attitudes and behaviours towards health



The survey of households on Norfolk Island that was conducted in mid-2012 included

questions specific to an individual’s PCTS usage intentions and to each of these key

components of the conceptual model.

In this chapter the distribution of respondents is discussed and the responses to these

questions are analyzed to see if they meet with researchers expectations. The responses

are then examined against key demographic variables including age and gender.

Finally the questions specific to each of these key components of the conceptual model are

examined against the questions specific to an PCTS usage intentions.

5.2 Description of Respondents

The 2011 Norfolk Island census identified 1795 individuals who reside on Norfolk Island.

Residents include either permanent residents of Norfolk Island or long term non residents

holding either a GEP (General Entry Permit - visa renewal needed every 5 years) or a TEP

(Temporary Entry Permit - visa renewal needed every 12 months). 1368 of the 1795

residents were over the age of 19. In total 805 households were identified as being

occupied by residents (730 houses, 5 flats, 4 tourist accommodation, 6 other).

Of the 805 households that were identified in the 2011 a total of 423 households responded

to the NICHE survey. This represents 52.5% of households on the island, 23.6% of the total

91

population or 30.9% of the population over 19 years of age. It is recognized that the views

of non-responding households may differ significantly from responding households. This is

addressed in more detail in section 7.3 – ‘Limitations of the Study’.

Responses to the survey indicated that 40.2% of respondents were male and 59.8 % were

female. Figure 5-1 below shows the breakdown of survey respondents by gender and age.

Figure 5-1 Distribution of Survey Respondents by Gender and Age

The average age of the resident population of Norfolk Island as reported in the 2011 census

was 46 years (45 years for men and 47 years for women). The average age of the

respondents for the NICHE survey was 56 years (58 years for men, 54 for women). This

difference was expected as the 2011 census average age was calculated on all age groups

where as the NICHE survey specified that only adults were eligible to complete the survey.

However, as can be seen in Figure 5-2 the age distribution of survey respondents was very

similar to the age distribution of Norfolk Island residents in the 2011 census.

0%

2%

4%

6%

8%

10%

12%

14%

16%

<=25 26-35 36-45 46-55 56-65 66-75 76-85 >85

Males

Females

92

Figure 5-2 Age Distribution of Survey Respondents vs. 2011 Census

The residential status of respondents was also very similar (Figure 5-3), when comparing the

proportion of NICHE respondents who are residents of Norfolk Island, those who are on a

GEP and those who are on a TEP to the 2011 census data.

Figure 5-3 Residential Status of Survey Respondents vs. 2011 Census

0%

5%

10%

15%

20%

25%

20-29 30-39 40-49 50-59 60-69 70+

Census data

NICHE survey

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Resident GEP TEP

Census data

NICHE survey

93

The average household responding to the survey had 2.19 members and 2.22 vehicles. In

the 2011 census the average Norfolk Island household had 2.23 members and 2.24 vehicles.

95.0% of NICHE survey respondents use gas for cooking and 97.6% have a water tank. In the

2011 census 95.9% of residents use gas for cooking and 97.0% had a water tank.

The brief description of the demographic characteristics of respondents, and the

comparison with 2011 census information shown in this section highlights that responses to

the project’s survey are representative of the broader Norfolk Island population.

5.3 Self Health Evaluation

The NICHE survey included the question A9 "do you generally consider your health to be...."

that was specific to the self health evaluation component of the conceptual model. The

question had a 5 point scale enabling respondents to choose from the following:

1. Poor

2. Fair

3. Good

4. Very good

5. Excellent

Analysis of the responses indicates that the majority of Norfolk Islanders considered

themselves to be healthy by evaluating their health as "Good" or "Very Good". Table 5-4

below shows the distribution of the 409 valid responses to each of the scale items related to

this question.

Poor Fair Good Very good Excellent Total

8 45 183 128 45 409

Table 5-4 Response Distributions - Self Health Evaluation

94

The responses were compared with key demographic items from the survey to provide

more detailed categorization of respondents’ views of their health. This also presents a

baseline for future implementation of work associated with PCTS and personal health.

The data summarized in Figure 5-5 shows that overall residents of the island in the younger

age groups were more likely to think their health was "excellent" or "very good" and the

older age groups had a higher percentage who thought that their health was "good" or

"fair". 25% of the under 25's age group consider their health to be "excellent". This figure

falls for successive age groups and is offset by a rise in the percentage of residents who

consider their health to be "fair" or "poor”.

Figure 5-5 Response Distributions – Self Health Evaluation and Age

However gender was not a differentiating characteristic related to how Norfolk Islanders

viewed their health (see Figure 5-6). There were 4.6% more females than males that

considered their health to be "very good" while 4.3% more males than females considered

their health to be "fair". In all other categories the difference between males and females

was 1% or less.

0%

10%

20%

30%

40%

50%

60%

<=25 26-35 36-45 46-55 56-65 66-75 76-85 >85

Poor

Fair

Good

Very good

Excellent

95

Figure 5-6 Response Distributions – Self Health Evaluation and Gender

Question A10 "how would you best describe yourself" was included in the survey to

ascertain how residents of Norfolk Island view their body weight. The following scale items

were included with this question:

1. Very underweight

2. A bit underweight

3. Ideal weight

4. A bit overweight

5. Very overweight

Analysis of the responses indicates that the majority of Norfolk Islanders considered

themselves to be an "ideal weight" or "a bit overweight". Table 5.7 below shows the

distribution of the 402 valid responses to each of the scale items related to this question.

Very underweight

A bit underweight

Ideal weight A bit overweight

Very overweight

Total

1 26 156 204 15 402

Table 5-7 Response Distributions - Body Weight

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Poor Fair Good Very good Excellent

All

Males

Females

96

The response distribution to this question and the question asking respondents to evaluate

their health were cross tabulated. The results are summarized in Figure 5-8 and show that

75% of the survey respondents who rated their health as "poor" thought they were a "bit

overweight" or "very overweight". As respondents perception of their health improves from

"fair" to "excellent" they were less likely to evaluate themselves as being a "bit overweight",

"very overweight" or "underweight" and were the more likely to rate themselves as an

"ideal weight".

Figure 5-8 Response Distributions – Self Health Evaluation and Body Weight

The information on health and obesity gives a solid baseline for interpreting PCTS usage

intentions. Given the research believes there is a strong link between health and attitude

towards the environment, the data developed around these questions enables us to

ascertain whether or not the introduction of a PCTS does affect behaviours towards health,

particularly those who have a lower self-evaluation of their own health.

0%

10%

20%

30%

40%

50%

60%

70%

80%


Very underweight

A bit underweight

Ideal weight

A bit overweight

Very overweight

97

5.4 Attitudes and Behaviours Towards Health

The following two questions were included in the survey to measure attitudes and

behaviours towards health:

B8 "I always try to eat healthy food"

C1 "How often do you engage in leisure time physical activity for the sole purpose of

improving or maintaining your health?"

A seven point Likert scale was used for question B8 "I always try to eat healthy food" with

scale descriptors ranging from 1 - "strongly agree" to 7 – "strongly disagree" with midpoint

of 4 - "neutral".

Of the 408 valid responses shown in Table 5.9 below, 24.5% "strongly agree" and 58.9%

"agree" that they "always try to eat healthy food", 12.5% were "neutral" and only 4.1%

"disagree" or "strongly disagree" that they "always try to eat healthy food".

1 (Strongly

agree) 2 3

4 (Neutral)

5 6 7

(Strongly disagree)

Total

98 111 130 51 12 4 2 408

Table 5-9 Response Distributions – Diet

The responses were the grouped by the respondent's age and gender. As Figure 5-10 shows

older respondents were more likely to "strongly agree" that they "always try to eat healthy

food". Younger respondents were more likely to "agree" or have "neutral" feelings when

asked whether they "always try to eat healthy food".

98

Figure 5-10 Response Distributions – Diet and Age

Gender had a definite impact on Norfolk Islanders agreement that they "always try to eat

healthy food". The data contained in Figure 5.11 shows that females were more likely to

"strongly agree" that they "always try to eat healthy food" and males were more likely to

"agree", be "neutral" or "disagree".

Figure 5-11 Response Distributions – Diet and Gender

0%

10%

20%

30%

40%

50%

60%

70%

<=25 26-35 36-45 46-55 56-65 66-75 76-85 >85

1 (Strongly agree)

2

3

4 (Neutral)

5

6

7 (Strongly disagree)

0%

5%

10%

15%

20%

25%

30%

35%

40%

1 (Stronglyagree)

2 3 4 (Neutral) 5 6 7 (Stronglydisagree)

All

Males

Females

99

Question C1 "how often do you engage in leisure time physical activity for the sole purpose

of improving or maintaining your health" had the following possible options:

1. Daily

2. 3-5 times/week

3. 1-3 times/week

4. < once a week

5. Never

Table 5-12 below shows the distribution of the 402 valid responses to each of the scale

items related to this question. Analysis of the valid responses to question C1 "how often do

you engage in leisure time physical activity for the sole purpose of improving or maintaining

your health" revealed that almost three quarters of Norfolk Island residents "engage in

leisure time physical activity" at least "1 – 3 times" a week. Of these 26.9% "engage in

leisure time physical activity "3 – 5 times" a week and 19.7% "daily".

Daily 3-5

times/week 1-3

times/week < Once a

week Never Total

79 108 104 51 60 402

Table 5-12 Response Distributions – Exercise

The responses to this question were compared with the residents age and gender. There

was no clear relationship between exercise patterns and age however gender did play a

part. The histogram in Figure 5-13 shows that a higher percentage of males engage in

exercise in all frequency categories except “3 – 5 times per week” where 14.4% more

females than males engaged in exercise for health reasons.

100

Figure 5-13 Response Distributions – Exercise and Gender

Respondents who answered above the scale mid-point to the question “how often do you

engage in leisure time physical activity for the sole purpose of improving or maintaining your

health” were also more likely to answer above the scale mid-point to the question "I always

try to eat healthy food".

A summary of these response distributions is shown in Figure 5-14. Of the 95 residents

who "engage in leisure time physical activity" "daily", 40.5% "strongly agree" that they

"always try to eat healthy food". Only 3.8% had a "neutral" stance and none "strongly

disagreed". As the exercise frequency of the respondents fall the percentage who "strongly

agree" that they "always try to eat healthy food" falls and they are more likely to have a

"neutral" stance.

0%

5%

10%

15%

20%

25%

30%

35%

Daily 3-5 times 1-3 times < Once Never

All

Males

Females

101

Figure 5-14 Response Distributions – Diet and Exercise

5.5 Attitudes and Behaviours Towards the Environment

The NICHE survey included the following questions to ascertain Norfolk Island residents

attitudes and behaviours towards the environment:

B1 "I buy environmentally friendly products as much as I can"

B15 "I turn lights off when not in use"

B18 "I consciously try to reduce waste and recycle"

Scale descriptors for question B1 "I buy environmentally friendly products as much as I can"

ranged from 1 - "strongly agree" to 7 – "strongly disagree" with midpoint of 4 - "neutral".

The distribution of valid responses is shown in Table 5-15 below. Scale descriptors for the

other questions ranged from 1 - "never" to 7 – "always" with midpoint of 4 - "sometimes".

The distribution of valid responses for question B15 "I turn lights off when not in use" are

shown in Table 5-16 and Table 5-17 shows the distribution of the valid responses to each of

the scale items for B18 "I consciously try to reduce waste and recycle".

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Daily 3-5 times 1-3 times < Once Never

1 (Strongly agree)

2

3

4 (Neutral)

5

6


102

1 (Strongly

agree) 2 3

4 (Neutral)

5 6 7

(Strongly disagree)

Total

78 93 104 102 15 5 5 402

Table 5-15 Response Distributions – Environmentally Friendly Products

1 (Never) 2 3 4

(Sometimes) 5 6

7 (Always)

Total

6 3 8 19 35 97 243 411

Table 5-16 Response Distributions – Turn Lights Off

1 (Never) 2 3 4

(Sometimes) 5 6

7 (Always)

Total

9 9 18 85 80 104 100 405

Table 5-17 Response Distributions – Reduce Waste / Recycle

Analysis of the responses to these questions showed that the majority of Norfolk Islander

share positive attitudes and behaviours towards the environment. 68.4% "agreed" or

"strongly agreed" that they "buy environmentally friendly products as much as I can" and a

further 25.4% were "neutral". 59.1% of respondents "always" "turn lights off when not in

use". Only 4.1% responded below the scale midpoint ("rarely" or "never"). 24.7% of

respondents "always" "consciously try to reduce waste and recycle" with most of the

remainder indicating they "sometimes" or "nearly always""consciously try to reduce waste

and recycle".

There were no discernible differences in environmental behaviours or attitudes between

age groups. However, 23.8% of females indicated that they "strongly agree" that they "buy

environmentally friendly products as much as I can" compared to only 13.2% of males.

103

Males indicated they were more likely to "agree" than females. The data is summarized in

Figure 5-18 below.

Figure 5-18 Response Distributions – Environmentally Friendly Products and Gender

Figure 5-19 below shows that females were more likely than males to "always" "turn lights

off when not in use" with 63.8% indicating they "always" do so compared with 53.4% of

males. Males were more likely to choose a scale descriptor between "always" and

"sometimes". Females were also far more likely to "always" or "nearly always" "consciously

try to reduce waste and recycle" (see Figure 5-20). Males were more likely to "sometimes"

"consciously try to reduce waste and recycle".

0%

5%

10%

15%

20%

25%

30%

35%

1(Strongly

agree)

2 3 4(Neutral)

5 6 7(Stronglydisagree)

All

Males

Females

104

Figure 5-19 Response Distributions – Turn Lights Off and Gender

Figure 5-20 Response Distributions – Reduce Waste / Recycle and Gender

5.6 Attitudes and Behaviours Towards Carbon Emissions and Climate Change

The survey included question F23 "what best describes your thoughts about climate

change". This question was included as one of the measures of attitudes and behaviours

towards carbon emissions and climate change on the conceptual model. They were asked

0%

10%

20%

30%

40%

50%

60%

70%

All

Males

Females

0%

5%

10%

15%

20%

25%

30%

All

Males

Females

105

to identify their thoughts about climate change by choosing one of the four options listed

below.

1. I don’t think climate change is happening

2. I have no idea whether climate change is happening or not

3. I think climate change is happening but it’s a natural fluctuation in earth

temperatures

4. I think climate change is happening and I think humans are largely causing it

Table 5-21 below shows the distribution of the 407 valid responses to each of the scale

items related to this question. The responses to this question show that the majority of

Norfolk Islanders believe climate change is happening. Of the responses to this question,

93.1% "think climate change is happening", 1.7% "don’t think climate change is happening"

and 5.2% "have no idea whether climate change is happening or not". However of those

who "think climate change is happening", 65.7% thought that "humans are largely causing

it". The remaining 34.3% thought that "climate change is being caused by a natural

fluctuation in the earth’s temperatures".

1 2 3 4 Total

7 21 130 249 407

Table 5-21 Response Distributions – Thoughts About Climate Change

The responses were then examined against the demographic variables age and gender.

Figure 5-22 below shows the distribution of responses to the question on "thoughts about

climate change" in age bands and Figure 5.23 shows the distribution by gender.

Attitude towards climate change does seem to be in part determined by the age of the

survey respondent. The data summarized in Figure 5-22 shows that with the exception of

residents over 85, older residents on the island were more likely to believe that climate

change was not happening or that it is a "natural fluctuation in earth temperatures". 75% of

106

the under 25’s age group believed that human behaviour is causing climate change

compared with 12.5% who believe it is a "natural fluctuation in earth temperatures". With

each successive age group the percentage of respondents who believe that "humans are

largely causing" climate change falls reaching a low of 41.1% in the 66 – 75 age groups and

the percentage of respondents who believe that climate change is a "natural fluctuation in

earth temperatures" rises to a high of 51.8% in the same age group

Figure 5-22 Response Distributions – Thoughts About Climate Change and Age

Figure 5-23 shows that women (66.5%) were more likely than men (53.7%) to believe that

climate change was caused by human behaviour. Males (36.4%) were more likely to believe

that climate change is a "natural fluctuation in earth temperatures" than females (28.9%).

0%

10%

20%

30%

40%

50%

60%

70%

80%

<=25 26-35 36-45 46-55 56-65 66-75 76-85 >85

I don’t think climate change is happening

I have no idea whether climatechange is happening or not

I think climate change is happening but it’s a natural fluctuation in earth temperatures

I think climate change ishappening and I think humans arelargely causing it

107

Figure 5-23 Response Distributions – Thoughts About Climate Change and Gender

The responses to question F23 "what best describes your thoughts about climate change"

were compared to the other questions that were specific to the attitudes and behaviours

towards carbon emissions and climate change component of the conceptual model

including:

B5 "it is important for me to have a low carbon footprint"

B7 "collectively, households can reduce the impacts of greenhouse gas emissions"

B13 "I am worried about climate change"

Scale descriptors for these questions ranged from 1 - "strongly agree" to 7 – "strongly

disagree" with midpoint of 4 - "neutral". The distribution of valid responses for question B5

"it is important for me to have a low carbon footprint" are shown in Table 5-24 below. Table

5-25 shows the distribution of the valid responses to each of the scale items for question B7

"collectively, households can reduce the impacts of greenhouse gas emissions" and Table 5-

26 contains the valid responses for question B13 "I am worried about climate change".

0%

10%

20%

30%

40%

50%

60%

70%

1 2 3 4

All

Male

Female

108

1 (Strongly

agree) 2 3

4 (Neutral)

5 6 7

(Strongly disagree)

Total

116 91 119 68 5 3 3 405

Table 5-24 Response Distributions – Important to Have a Low Carbon Footprint

1 (Strongly

agree) 2 3

4 (Neutral)

5 6 7

(Strongly disagree)

Total

136 125 87 36 8 6 6 404

Table 5-25 Response Distributions – Households Can Reduce GHG Emissions

1 (Strongly

agree) 2 3

4 (Neutral)

5 6 7

(Strongly disagree)

Total

106 104 84 75 13 12 13 407

Table 5-26 Response Distributions – Worried About Climate Change

The data summarized in Figure 5-27 below shows that most NICHE respondents "agreed" or

"strongly agreed "that "it is important for me to have a low carbon footprint". Only 15.5%

were "neutral" and only 2.8% "disagreed" or "strongly disagreed" that it was "important for

me to have a low carbon footprint". Of those who thought that climate change is a "natural

fluctuation in earth temperatures" 23.6% were "neutral" and 70.7% "agreed" or "strongly

agreed" that "it is important for me to have a low carbon footprint". Of those who thought

that change is happening and that humans are largely causing it, 87.7% "agreed" or

"strongly agreed" that it was "important for me to have a low carbon footprint". Even those

who did not believe that climate change was happening were all "neutral" or "agreed "that

it was "important for me to have a low carbon footprint".

109

Figure 5-27 Response Distributions – Thoughts About Climate Change and Important to Have a Low Carbon

Footprint

Figure 5-28 below shows that while most Norfolk Islanders believe that "households can

reduce the impacts of greenhouse gas emissions", those who believe climate change is

caused by human behaviour were more likely to "agree" or "strongly agree" that

"households can reduce the impacts of greenhouse gas emissions" than those who believe

climate change is a "natural fluctuation in earth temperatures". Of those who believe

climate change is caused by human behaviour, 42.1% "strongly agreed" that "households

can reduce the impacts of greenhouse gas emissions", 4.5 % were "neutral" and 3.3%

disagreed or strongly disagreed. Of the respondents who thought that climate change is a

"natural fluctuation in earth temperatures" 20.3% "strongly agreed", 15.4% were "neutral"

and 8.9% "disagreed" or "strongly disagreed" that "households can reduce the impacts of

greenhouse gas emissions".

0%

10%

20%

30%

40%

50%

60%

1 2 3 4

1 (strongly agree)

2

3

4 (Neutral)

5

6


110

Figure 5-28 Response Distributions – Thoughts About Climate Change and Households Can Reduce GHG

Emissions

Figure 5-29 below shows that Norfolk Islanders who believe climate change is caused by

human behaviour are more likely to be "worried about climate change" than those who

believe climate change is a "natural fluctuation in earth temperatures". Of the 244

respondents who believe climate change is caused by human behaviour 88.5% "agreed" or

"strongly agreed" that they were "worried about climate change". Of the 124 who believe

climate change is a "natural fluctuation in earth temperatures" exactly 50% "agreed" or

"strongly agreed" that they were "worried about climate change".

0%

10%

20%

30%

40%

50%

60%

1 2 3 4

1 (Strongly agree)

2

3

4 (Neutral)

5

6


111

Figure 5-29 Response Distributions – Thoughts About Climate Change and Worried About Climate Change

5.7 PCTS Usage Intentions

Respondents were asked question E1 "being able to measure my carbon footprint is

important to me". This question was included as one of the measures of the construct PCTS

usage intentions of the conceptual model.

Scale descriptors for the question ranged from 1 - "strongly agree" to 7 – "strongly

disagree" with midpoint of 4 - "neutral". Table 5-30 below shows the distribution of the 409

valid responses to each of the scale items related to this question.

1 (Strongly

agree) 2 3

4 (Neutral)

5 6 7

(Strongly disagree)

Total

54 68 111 158 5 8 9 413

Table 5-30 Response Distributions – Measure Carbon Footprint is Important

0%

10%

20%

30%

40%

50%

60%

70%

1 2 3 4

1 (strongly agree)

2

3

4 (Neutral)

5

6


112

Overall 38.3% of respondents chose "neutral", 13.1% "strongly agreed" and 43.3% "agreed".

Only 5.33% "disagreed" or "strongly disagreed".

Age did not appear to be an influencing characteristic. However, the data summarized in

Figure 5-31 shows that greater percentages of females were more likely to "strongly agree"

(15.4% - males 10.4%) or "agree" (47.3% - males 38.4%) that "being able to measure my

carbon footprint is important".

Figure 5-31 Response Distributions – Measure Carbon Footprint is Important and Gender

Response distributions to the questions E1 “being able to measure my carbon footprint is

important to me” and A6 "compared to others on Norfolk Island, do you think your carbon

footprint is/would be" are shown in Figure 5-32 below. It shows that residents of the Island

who thought their "carbon footprint is/would be" "well below average" were more likely to

"strongly agree" that "being able to measure my carbon footprint is important to me". None

of this particular group "disagreed" or "strongly disagreed" that "being able to measure my

carbon footprint is important to me". Residents who thought their "carbon footprint

is/would be" "well above average" were more likely to "strongly disagree" that "being able

to measure my carbon footprint is important to me".

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

1 (Stronglyagree)


All

Males

Females

113

Figure 5-32 Response Distributions – Measure Carbon Footprint is Important and Carbon Footprint Size

5.8 PCTS Usage Intentions Compared with Self-Health Evaluation

The response distributions to the questions E1 "being able to measure my carbon footprint

is important to me" and A9 "do you generally consider your health to be" from the self-

health evaluation and PCTS usage intentions constructs are examined in this section.

The data summarized in Figure 5-33 shows that Norfolk Islanders who "generally consider

your (their) health" to be "excellent" were more likely to "strongly agree" that "being able to

measure my carbon footprint is important". As self evaluated health decreases from

"excellent" to "fair" the percentage of respondents who "strongly agree" that "being able to

measure my carbon footprint is important" decreases and the percentage who are "neutral"

increases. Of the 43 respondents who thought their health was "excellent", 27.9% strongly

believed that "being able to measure my carbon footprint is important to me" while the

same number were "neutral". Of the 44 respondents who thought their health was "fair"

11.4% believed that "being able to measure my carbon footprint is important to me" while

52.3% were "neutral".

0%

10%

20%

30%

40%

50%

60%

70%

80%

1 (Stronglyagree)

2 3 4(Neutral)

5 6 7 (Stronglydisagree)

Well below average

Below average

About average

Above average

Well above average

114

Figure 5-33 Response Distributions – Measure Carbon Footprint is Important and Self Health Evaluation

5.9 PCTS Usage Intentions Compared with Attitudes and Behaviours Towards Health

The response distributions to the questions the questions E1 "being able to measure my

carbon footprint is important to me", B8 "I always try to eat healthy food" and C1 "how

often do you engage in leisure time physical activity for the sole purpose of improving or

maintaining your health" from the attitudes and behaviours towards health and PCTS usage

intentions constructs are examined in this section.

Examination of the response distributions appears to suggest that the more positive an

individual's attitudes and behaviours are towards their health, the more likely they are to

have positive PCTS usage intentions.

Figure 5-34 below shows that Norfolk Island residents who "strongly agree" that "being able

to measure my carbon footprint is important to me" were more likely to "strongly agree"

that they "always try to eat healthy food" than those who "agree" or were "neutral".

Residents who "agree" that "being able to measure my carbon footprint is important to me"

were most likely to "agree" that they "always try to eat healthy food".

0%

10%

20%

30%

40%

50%

60%


1 (Strongly agree)

2

3

4 (Neutral)

5

6


115

Figure 5-34 Response Distributions – Measure Carbon Footprint is Important and Diet

The data summarized in Figure 5-35 below shows that residents who "strongly agree" that

"being able to measure my carbon footprint is important to me" were more likely to exercise

daily for the "sole purpose of improving or maintaining" their health.

Figure 5-35 Response Distributions – Measure Carbon Footprint is Important and Exercise

0%

10%

20%

30%

40%

50%

60%

E1) 1(Strongly

agree)

E1) 2 E1) 3 E1) 4(Neutral)

E1) 5 E1) 6 E1) 7(Stronglydisagree)

B8) 1 (Strongly agree)

B8) 2

B8) 3

B8) 4 (Neutral)

B8) 5

B8) 6

B8) 7 (Strongly disagree)

0%

10%

20%

30%

40%

50%

60%

70%

1 (Stronglyagree)


Daily

3-5 times

1-3 times

< Once

Never

116

5.10 PCTS Usage Intentions Compared with Attitudes and Behaviours Towards the

Environment

To determine if there is any relationship between Norfolk Islanders PCTS usage intentions

and their attitudes and behaviours towards the environment the response distributions to

the questions E1 "being able to measure my carbon footprint is important to me", B1 "I buy

environmentally friendly products as much as I can", B15 "I turn lights off when not in use"

and B18 "I consciously try to reduce waste and recycle" were examined.

Analysis of the responses indicates that there is a strong link between "being able to

measure my carbon footprint" and buying "environmentally friendly products as much as I

can" (see Figure 5-36 below). 50.9% of residents who "strongly agreed" that "being able to

measure my carbon footprint is important to me" also "strongly agreed" that they "buy

environmentally friendly products as much as I can" and another 28.3% selected option 2

indicating agreement / strong agreement. Respondents who "agreed" that "being able to

measure my carbon footprint is important to me" were most likely to "agree" that they "buy

environmentally friendly products as much as I can". And those who were "neutral" when

asked whether "being able to measure my carbon footprint is important to me" were most

likely to answer "neutral" when asked if they "buy environmentally friendly products as

much as I can".

117

Figure 5-36 Response Distributions – Measure Carbon Footprint is Important and Environmentally Friendly

Products

Figure 5-37 below shows the response distributions to the questions "being able to measure

my carbon footprint is important to me" and "I turn lights off when not in use". While the

majority of Norfolk Island residents who responded to the survey "always" "turn lights off

when not in use", residents who "strongly agreed " that "being able to measure my carbon

footprint is important to me" were more likely to "always" "turn lights off when not in use"

than those who "agreed" or were "neutral".

0%

10%

20%

30%

40%

50%

60%

70%

E1) 1(Strongly

agree)

E1) 2 E1) 3 E1) 4(Neutral)



B1) 2

B1) 3

B1) 4 (Neutral)

B1) 5

B1) 6


118

Figure 5-37 Response Distributions – Measure Carbon Footprint is Important and Turn Off Lights

Figure 5.-38 below shows Norfolk Islanders who "strongly agree" that "being able to

measure my carbon footprint is important to me" are more likely to "always" "consciously

try to reduce waste and recycle" and as their agreement falls so does the amount that they

"always" "consciously try to reduce waste and recycle". 42.3% of survey respondents who

"strongly agree" that "being able to measure my carbon footprint is important to me"

indicated that they "always" "consciously try to reduce waste and recycle" and 32.69%

selected option 2 indicating they nearly always "consciously try to reduce waste and

recycle". Of the respondents who were "neutral" when asked whether "being able to

measure my carbon footprint is important to me" only 18.2% indicated that they "always"

"consciously try to reduce waste and recycle".

0%

10%

20%

30%

40%

50%

60%

70%

80%

1 (Stronglyagree)


1 (Never)

2

3

4 (Sometimes)

5

6

7 (Always)

119

Figure 5-38 Response Distributions – Measure Carbon Footprint is Important and Reduce Waste / Recycle

5.11 PCTS Usage Intentions Compared with Attitudes and Behaviours Towards Carbon

Emissions and Climate Change

To further examine responses to questions measuring aspects of the conceptual model

constructs PCTS usage intentions and attitudes and behaviours towards carbon emissions

and climate change responses to the questions E1 "being able to measure my carbon

footprint is important to me" and F23 "what best describes your thoughts about climate

change".

The responses are summarized in Figure 5-39 below. 80.8% of islanders who "strongly

agree" that "being able to measure my carbon footprint is important to me" selected "I think

climate change is happening and I think humans are largely causing it" and only 13.5%

"think climate change is happening but it’s a natural fluctuation in earth temperatures". As

the respondents agreement that "being able to measure my carbon footprint is important to

me" decreases so too does the percentage of respondents who think "think humans are

largely causing climate change" while the percentage who think climate change is "a natural

fluctuation in earth temperatures" rises. Of those who were "neutral" when asked whether

"being able to measure my carbon footprint is important to me" 48.7% think "humans are

largely causing it" and 43.2% think "it’s a natural fluctuation in earth temperatures". Of

0%

10%

20%

30%

40%

50%

60%

1 (Stronglyagree)


1 (Never)

2

3

4 (Sometimes)

5

6

7 (Always)

120

those who "strongly disagree" when asked whether "being able to measure my carbon

footprint is important to me" all 100% think "it’s a natural fluctuation in earth

temperatures".

Figure 5-39 Response Distributions – Measure Carbon Footprint is Important and Thoughts About Climate

Change

Response distributions to the question"E1 being able to measure my carbon footprint is

important to me" were also cross-tabulated with the following questions:

B5 "it is important for me to have a low carbon footprint"

B7 "collectively, households can reduce the impacts of greenhouse gas emissions"

B13 "I am worried about climate change"

Figure 5-40 below shows that 79.3% of respondents who "strongly agreed" that "being able

to measure my carbon footprint is important to me" also "strongly agreed" that "it is

important for me to have a low carbon footprint". Respondents who "agree" that "being

able to measure my carbon footprint is important to me" were most likely to "agree" that "it

is important for me to have a low carbon footprint". As the importance of being able to

0%

20%

40%

60%

80%

100%

120%

1 (Stronglyagree)


1

2

3

4

121

measure their carbon footprint decreases from "agree" to "strongly disagree" so does the

importance of having a low carbon footprint.

Figure 5-40 Response Distributions – Measure Carbon Footprint is Important and Important to Have a Low

Carbon Footprint

As Figure 5-41 below shows respondents who chose who "strongly agreed" that "being able

to measure my carbon footprint is important to me" were most likely to "strongly agree"

that "collectively, households can reduce the impacts of greenhouse gas emissions". 75.5%

of respondents who "strongly agreed" that "being able to measure my carbon footprint is

important to me" also "strongly agreed" that "households can reduce the impacts of

greenhouse gas emissions". At the other end of the scale respondents who "strongly

disagreed" that "being able to measure my carbon footprint is important to me" were most

likely to "strongly disagree" that "collectively, households can reduce the impacts of

greenhouse gas emissions".

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

E1) 1(Strongly

agree)

E1) 2 E1) 3 E1) 4(Neutral)



B5) 2

B5) 3

B5) 4 (Neutral)

B5) 5

B5) 6


122

Figure 5-41 Response Distributions – Measure Carbon Footprint is Important and Households Can Reduce

GHG Emissions

The data summarized in Figure 5-42 below shows that 62.3% of Norfolk Islanders who

"strongly agreed" that "being able to measure my carbon footprint is important to me" also

"strongly agreed" that "I am worried about climate change". Conversely 66.7% of Norfolk

Islanders who "strongly disagreed" that "being able to measure my carbon footprint is

important to me" also "strongly disagreed" that "I am worried about climate change".

0%

10%

20%

30%

40%

50%

60%

70%

80%

E1) 1(Strongly

agree)

E1) 2 E1) 3 E1) 4(Neutral)



B7) 2

B7) 3

B7) 4 (Neutral)

B7) 5

B7) 6


123

Figure 5-42 Response Distributions – Measure Carbon Footprint is Important and Worried About Climate

Change

5.12 Conclusion

This chapter described the distribution of respondents of the NICHE survey and provided a

comparison with the demographic data that was collected in the 2011 Norfolk Island census.

The responses to the questions that are specific to each of the components of the

conceptual model were analyzed and then compared against key demographic variables

including age and gender. These responses were then examined against the questions

specific to an PCTS usage intentions.

The next chapter describes the relationships amongst key variables from the study. The

conceptual model is examined using correlations analysis, factor analysis and linear

regression analysis.

0%

10%

20%

30%

40%

50%

60%

70%

E1) 1(Strongly

agree)

E1) 2 E1) 3 E1) 4(Neutral)



B13) 2

B13) 3

B13) 4 (Neutral)

B13) 5

B13) 6


124

6 Data Analysis

6.1 Introduction

This chapter describes the examination of the relationships amongst key variables from the

study. The initial examination of the relationships between variables was undertaken using

correlation analysis with the directionality of the relationships examined using linear


Following the correlation analysis, Exploratory Factor Analysis (EFA) was run on the blocks of

variables expected to have an underlying factor structure as per the design of the survey.

The EFA examining the block of variables that were used to form the dependent variable

resulted in a single factor, 'Usage Intentions Towards a PCTS’. The EFA run on the other

blocks of variables resulted in five factors that made up the independent variables.

Correlation analysis was also run on the factors that were derived from the EFA prior to the


The blocks of variables entered into the linear regression model were based on the variables

identified underlying each of the factors. The variables comprising each of the factors that

were extracted from the EFA were entered as blocks to predict the total variance in 'Usage

Intentions Towards a PCTS' and identify which variables in each block were significant. The

regression model was then examined by moderating inputs based upon survey respondents

gender and their self assessment of the size of their carbon footprint.

6.2 Correlation Analysis

Correlation analysis measures how variables are related (Lee Rodgers & Nicewander, 1988;

SPSS Inc, 2009) and was used in the first stage of the analysis of the relationships amongst

the variables from the survey. It was used to determine if significant relationships were

present among the blocks of variables that were discussed in Section 4.2 - 'Survey Design

and Construction'.

125

All of the correlations reported in this chapter are Pearson product-moment correlation

coefficients. Pearson's correlation coefficient is a measure of the degree of linear

dependence between two variables (Lee Rodgers & Nicewander, 1988; SPSS Inc, 2009). The

SPSS guide book states:

"For quantitative, normally distributed variables, choose the Pearson correlation coefficient.

If your data are not normally distributed or have ordered categories, choose Kendall’s tau-b

or Spearman, which measure the association between rank orders" (SPSS Inc, 2009, p. 333)

Pearson's correlation coefficient is an integer that can "range in value from –1 (a perfect

negative relationship) and +1 (a perfect positive relationship). A value of 0 indicates no linear

relationship" (SPSS Inc, 2009, p. 333).

All correlations in this chapter also use a two tailed test of significance as the direction of

association was not known in advance. "If the direction of association is known in advance,

select One-tailed. Otherwise, select Two-tailed" (SPSS Inc, 2009, p. 333).

The option to flag selected correlations was ‘checked’ in the SPSS analysis software to flag

the correlation coefficients that were significant at the 0.05 level (identified by a single

asterisk) and those that were significant at the 0.01 level (identified by two asterisks).

The correlations between items in the different blocks of variables have been examined to

determine what intra-item correlations were suitable for consideration in the regression

analysis.

The correlations are reported on a section-by-section basis that matches the general

sections from the survey. Correlations among each of the items in each section as well as

correlations between the items from each section are discussed. The correlations in this

section address the variables comprising the constructs shown in the conceptual model (see

Figure 1.1).

126

6.2.1 Correlation Analysis - Self Health Evaluation

Questions were included in the survey to gather data related to respondents’ health. The

responses are the individual respondent’s view of their level of health but may not be a true

reflection of their actual health. In this section the correlations among the items included as

indicators of general health are examined. These include:

Body Mass Index (BMI)

General overall health

Comparable health

Exercise frequency

Body weight

Body Mass Index (BMI) is commonly used to measure obesity and is calculated by dividing

an individual's weight in kilograms by their height in meters squared. BMI's based upon the

self-reported height and weight were calculated for all NICHE survey participants and were

categorized into bands on the following basis as shown in Table 6-1 below.

BMI Descriptor Band N %

40 + morbidly obese (category III obese) 6 6 1.8

35 - 39.9 category II obese 5 13 4.0

30 - 34.9 category 1 obese 4 54 16.5

25 - 29.9 overweight 3 119 36.28

18.5 - 24.9 ideal 2 125 38.1

< 18.5 underweight 1 6 1.8

Table 6-1 BMI Bands

127

A10 A12 C1 BMI band

A9 -.235**

-.123* -.220

** -.234

**

A10 .437**

.139**

.523**

A12 .031 .354**

C1 .094

Table 6-2 Self Health Evaluation Correlation Analysis Results

The correlations shown in Table 6-2 above are for the self-reported BMI and questions A9,

A10, A12 and C1 that represent respondent's attitudes to their weight and general health.

The correlation matrix shows that four of the six variables had relationships that were

significant at the 99% confidence interval, one at the 95% confidence interval and one

relationship was not significant. The variance explained by the relationships was not

particularly high but the presence of significant relationships between these items was

expected. The negative correlations reflect the scale descriptors and are also consistent

with expectations. The relationships among these variables are important as they are

measures of respondent’s attitudes to their own health which are measures of the

hypothesized construct Self-health Evaluation from the conceptual model. Questions A10

and A12 were both about perceptions of obesity and this explains the higher level of

variance between these two items (r = 0.437, p < 0.001). Essentially the statistical

relationships evident from the correlations among these items reflect that people with a

positive view of their bodyweight also consider themselves to be in sound health.

With the exception of question C1 the correlations between BMI band and the self-health

evaluation questions were significant at the 99% confidence interval. The correlations

among these items does indicate that individual’s assessment of their own health and

relative obesity does match their BMI, thus providing some validation that these items do

reflect the actual health and obesity of respondents. This is important in taking these items

128

forward in further analysis and seeking to identify the relationships between obesity, health

and climate change/carbon footprint.

6.2.2 Correlation Analysis - Attitudes Towards Health, the Environment, Carbon


The survey included questions to gather data about the respondents’ attitudes towards

health, the environment, carbon emissions and climate change. In this section the

correlations between the variables measuring environmental and health related attitudes,

attitudes towards carbon emissions and attitudes towards climate change are examined.

While some of the questions in this section of the survey have a ‘household’ focus only one

member of the household was required to fill out the survey and it was recognized that the

data reflected the attitudes of that respondent and not necessarily the views of other

members of their household.

129

B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13

B1 -.005 .184**

-.041 .484**

-.001 .415**

.402**

.340**

.132**

.148**

.169**

.342**

B2 .072 .317**

-.023 .166**

.120* .022 .019 .056 -.031 -.004 .054

B3 .021 .237**

.092 .213**

.237**

.297**

.200**

.405**

.213**

.184**

B4

-.075 .192**

.021 -.047 -.047 .055 -.019 -.102* .026

B5

.180**

.482**

.263**

.326**

.212**

.310**

.154**

.490**

B6 .288**

.050 .179**

.315**

.114* .083 .211

**

B7

.336**

.405**

.277**

.380**

.161**

.448**

B8

.574**

.168**

.343**

.332**

.270**

B9

.318**

.410**

.443**

.271**

B10 .229**

.224**

.253**

B11

.315**

.339**

B12

.230**

Table 6-3 Attitudes Towards Health, the Environment, Carbon Emissions and Climate Change Correlation

Analysis Results

The correlations shown in Table 6-3 above are for questions B1 to B13 that were included in

the survey to measure respondent's attitudes towards health, the environment, carbon

emissions and climate change (see Appendix A - 'NICHE Survey').

There were a large number of significant relationships among the 13 items at the 99%

confidence interval (p < 0.001). This supports their inclusion under an ‘attitudes' construct

in the survey. The questions were arranged in sections to reflect the structures it was

thought would be evident in the data. These 13 items were included to measure attributes

130

of the hypothesized constructs Attitudes and Behaviours Towards Health, Attitudes and

Behaviours Towards the Environment and Attitudes and Behaviours Towards Carbon

Emissions and Climate Change from the conceptual model. That there were relationships

among the individual items warrants them being examined using EFA (see Section 6.3 -

'Exploratory Factor Analysis').

6.2.3 Correlation Analysis - Behaviours Towards Consumption and the Environment

Questions were included in the survey to gather data about the respondents’ behaviours

that were related to consumption that impact carbon outputs. In this section the results of

the correlation analysis between the items measuring these behaviours are examined. As

stated in the previous section it was recognized that the data in this section of the survey

reflected the behaviours of the respondent and not necessarily the behaviours of other

members of their household.

B15 B16 B17 B18 B19

B14 .557**

.412**

.233**

.413**

.270**

B15 .515**

.145**

.457**

.386**

B16 .203**

.580**

.291**

B17

.409**

.276**

B18

.451**

Table 6-4 Behaviours Towards Consumption and the Environment Correlation Analysis Results

All of the items in this set of questions were significantly related (p < 0.001) therefore no

specific commentary about any of the correlations is going to provide any further support or

understanding about the nature of the relationships. That they are all significantly

131

correlated supports their inclusion under a ‘behaviours' construct in the survey which were

used to measure the construct on the conceptual model called Attitudes and Behaviours

Towards the Environment.

It was anticipated that there would be significant relationships among these items because

an individual who acts in an environmentally friendly way would be expected to act that

way across a range of environmental behaviours.

6.2.4 Correlation Analysis - PCTS Usage Intentions

In this section the correlations between items measuring PCTS usage intentions are

examined. These items were included as intended measures of the hypothesized

dependent variable PCTS Usage Intentions from the conceptual model. Once again it was

recognized that the data in this section of the survey reflected the intentions of the

respondent and not necessarily the intentions of other members of their household.

132

E2 E3 E4 E5 E6 E7 E8 E9 E10

E1 .420**

.665**

.566**

.298**

.336**

.526**

.669**

.605**

.468**

E2 .626**

.445**

.235**

.267**

.451**

.484**

.509**

.347**

E3 .705**

.383**

.417**

.614**

.719**

.675**

.512**

E4

.342**

.377**

.657**

.668**

.657**

.486**

E5

.379**

.320**

.385**

.376**

.274**

E6 .348**

.448**

.390**

.416**

E7

.677**

.706**

.544**

E8

.786**

.551**

E9

.569**

Table 6-5 Usage Intentions Towards a PCTS Correlation Analysis Results

All the items in this block of questions were positively and significantly correlated (p <

0.001). It is worth noting that some of the variance shared between the items is quite high

although no correlation coefficient in the matrix was high enough for any of the items to be

considered covariate.

It was expected that all the correlation coefficients would be positive with reasonable levels

of variance. That the items are all correlated and the high level of variance explained by

some of the relationships confirms that the items are seen as measuring similar things and

that they are all potentially related to a higher order latent construct.

The identification of the relationships among these items is key to the current project as

these questions were included in the survey as observed measures of the dependent

construct PCTS Usage Intentions from the conceptual model.

133

6.2.5 Correlation Analysis - PCTS vs. Self Health Evaluation

In the following section the results of the correlation analysis between the variables

measuring PCTS usage intentions and the variables included in the survey as indicators of

general health are examined.

A9 A10 A12 C1

E1 -.136**

-.007 .004 .216**

E2 -.035 .043 -.059 .097

E3 -.081 .026 -.031 .178**

E4 -.115* -.039 -.070 .153

**

E5 .028 .046 .008 .010

E6 -.039 .066 .026 .111*

E7 .022 -.051 -.061 .162**

E8 -.112* -.044 -.062 .169

**

E9 -.083 -.055 -.079 .163**

E10 -.081 .043 .010 .190**

Table 6-6 PCTS vs. Self Health Evaluation Correlation Analysis Results

The results of the correlation analysis among the variables measuring PCTS Usage Intentions

(E1 – E10) and the variables measuring Self Health Evaluation (A9, A10, A12, C1) from the

conceptual model are shown in Table 6-6 above.

134

There were only a small number of significant relationships between these items. In each

case the level of variance explained by the relationship was quite low – the highest between

questions C1 and E1 was only 4.7% (r = 0.216, p < 0.001). The absence of correlations

among these items was somewhat surprising, but as this is a baseline survey, correlations

among these items will be re-examined in the second stage of the study after users have

engaged with the PCTS for 12 months. It would be expected that more relationships

explaining higher levels of variance will be evident after users have had the opportunity to

use a PCTS and see the impact of their carbon footprint and link this to their own health.

6.2.6 Correlation Analysis - PCT vs. Behaviours Towards Consumption and the

Environment

In the following section the correlations between the variables measuring PCTS usage

intentions and the variables included in the survey to gather data about the respondents’

behaviours are examined.

135

B14 B15 B16 B17 B18 B19

E1 -.059 -.101* -.079 -.152

** -.164

** -.154

**

E2 .004 .002 -.023 -.039 -.031 -.058

E3 -.042 -.058 -.058 -.062 -.115* -.101

*

E4 -.065 -.061 -.056 -.087 -.051 -.076

E5 .035 .021 .024 -.090 -.033 .024

E6 .002 -.004 -.006 -.047 -.052 -.044

E7 -.042 -.021 -.011 -.044 -.004 -.032

E8 -.115* -.091 -.084 -.121

* -.144

** -.102

*

E9 -.044 -.038 .002 -.094 -.044 -.039

E10 -.060 -.059 -.052 -.128* -.117

* -.078

Table 6-7 PCT vs. Behaviours Towards Consumption and the Environment Correlation Analysis Results

Table 6-7 shows the results of the correlation analysis between the block of variables

measuring PCTS Usage Intentions (E1 – E10) and respondent's Attitudes and Behaviours

towards the environment based on the questions about consumption and environmentally

friendly behaviours (B14 – B19).

The results that were significant were expected as conceptually individuals that have

positive environmental behaviours could be expected to have positive attitudes towards

PCTS. However it was thought that there would be more significant relationships and that

the relationships that did exist would explain higher levels of variance. As can be seen in

Section 6.2.7 - ' Correlation Analysis - PCT vs. Attitudes Towards Health, the Environment,

Carbon Emissions and Climate Change' there are significant relationships between

individuals attitudes towards the environment and PCTS usage intentions. However the

136

same is not evident in relation to behaviours towards consumption and the environment

and PCTS usage intentions. The lack of a relationship and researchers hypothesized reasons

why it does not exist is discussed in further detail in the next chapter in Section 7.2 - 'Results

of the Statistical Analysis'.

6.2.7 Correlation Analysis - PCT vs. Attitudes Towards Health, the Environment, Carbon


In this section the results of the correlation analysis between the variables measuring PCTS

usage intentions and the variables included in the survey to gather data about the

respondents attitudes towards health, the environment, carbon emissions and climate

change are examined.

137

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 B13

E1 .377**

.084 .128* .047 .522

** .168

** .386

** .259

** .216

** .286

** .163

** .049 .487

**

E2 .183**

.102* .092 .067 .238

** .153

** .194

** .163

** .155

** .142

** .082 .036 .288

**

E3 .332**

.014 .116* .043 .416

** .209

** .348

** .148

** .205

** .339

** .144

** .096 .411

**

E4 .240**

.035 .167**

.039 .369**

.187**

.291**

.114* .116

* .300

** .215

** .069 .311

**

E5 .089 -.016 .031 .087 .171**

.353**

.262**

.037 .110* .241

** .117

* .109

* .235

**

E6 .141**

-.043 .106* .032 .249

** .199

** .219

** .121

* .097 .251

** .131

** .054 .298

**

E7 .226**

.080 .096 .069 .315**

.234**

.215**

.081 .124* .252

** .115

* .065 .334

**

E8 .331**

.108* .137

** .051 .426

** .218

** .362

** .193

** .177

** .331

** .183

** .112

* .397

**

E9 .255**

.099 .104* .080 .367

** .247

** .297

** .141

** .136

** .304

** .186

** .058 .412

**

E10 .281**

.012 .210**

.030 .415**

.204**

.316**

.250**

.277**

.262**

.248**

.109* .389

**

Table 6-8 PCT vs. Attitudes Towards Health, the Environment, Carbon Emissions and Climate Change

Correlation Analysis Results

The large number of positive correlations shown in Table 6.8 above shows that people who

are concerned about the environment and carbon emissions do have a positive view of the

role a PCTS can have in addressing environmental concerns. Correlations between some of

the items were expected and in fact would have been noticeable if they had not existed.

These were:

Question B5 and E1 - E10 as B5 is related to the respondents carbon footprint

Question B6 and E1 - E10 as B6 is related to financial incentives for individuals to

reduce their environmental impact

Question B7 and E1 - E10 as B7 is related to the reducing the impacts of green house

gas emissions

138

Question B13 and E1 - E10 as B13 is related to the respondents fears of climate

change

The reason relationships among these particular items were expected is because

conceptually individuals who have positive attitudes towards the environment, carbon

emissions and climate change would also have positive attitudes towards PCT. Conversely

individuals who have negative attitudes towards the environment, carbon emissions and

climate change could also be expected to have negative attitudes towards PCT.

The EFA run on questions B1 to B13 (see Section 6.3.1 - 'EFA - Attitudes Towards Health, the

Environment, Carbon Emissions and Climate Change') resulted in the definition of 3 factors –

'Health Consciousness', 'Environmental Consciousness' and 'Optimism'. A full description of

the EFA on this block of questions is contained in the next section of this chapter and is not

described any further in this section. A brief discussion of the correlations among the items

comprising each of the factors and the items measuring PCTS Usage Intentions follows. The

correlations are addressed separately under headings in the following sections that form the

names given to each factor derived from the EFA.

Health Consciousness

Question B12 ‘I am unlikely to ever be obese’ had no significant relationships with any of the

variables measuring Usage Intentions Towards a PCTS and question B3 ‘Being overweight

can have serious health effects’ was significantly related to just three of the 10 variables

measuring PCTS Usage Intentions and the level of variance explained by the relationships

was in each case very low. There were a higher number of significant relationships among

the remaining three variables measuring ‘Health Consciousness’ and the 10 variables

measuring PCTS Usage Intentions but once again the variances explained by the

relationships were low although significant at p < 0.001.

Optimism

The results from the correlation analysis among the variables measuring ‘Optimism’ (B2, B4

and B6) and PCTS Usage Intentions showed a number of interesting relationships. Question

B2 ‘Technology will solve future environmental problems' and question B4 ‘Obesity will be

solved in the future by medical advances’ had no significant relationships with any of the

139

variables measuring PCTS Usage Intentions. However question B6 ‘A financial incentive

would encourage me to reduce my environmental impact’ had significant relationships at the

99% confidence interval with all of the PCTS Usage Intentions variables. This relationship

was expected to be significant as a financial incentive for reducing carbon emissions is the

basis for PCTS.

Environmental Consciousness

The correlation analysis between the variables measuring ‘Environmental Consciousness’

and those measuring PCTS Usage Intentions showed a high number of significant

relationships. With one exception every correlation was significant at the 99% confidence

interval. The researchers expected there would be significant, positive correlations among

these variables. Conceptually, an individual who has a positive attitude towards PCT would

also be expected to have strong feelings about the environment and carbon emissions/

climate change. That relationships exist between the variables means there is a case for

further analysis to look at higher order factors and to examine the nature of the

relationships among the variables.

6.3 Exploratory Factor Analysis

The following sections discuss the factor analysis undertaken on the data set to determine if

there were underlying structures evident as per the relationships already determined from

the correlation analysis. Exploratory Factor Analysis (EFA) is a statistical method that is used

to identify the underlying relationships between measured variables (Norris & Lecavalier,

2010). SPSS states that EFA:

" attempts to identify underlying variables, or factors, that explain the pattern of

correlations within a set of observed variables. Factor analysis is often used in data reduction

to identify a small number of factors that explain most of the variance that is observed in a

much larger number of manifest variables." (SPSS Inc, 2009, p. 395)

EFA was run on the blocks of variables expected to have an underlying factor structure as

per the design of the survey. The blocks of variables were discussed in detail in 4.2 - Survey

Design and Construction.

140

Suitability of the Data for Factor Analysis

Before undertaking factor analysis the data must be tested to assess its suitability for factor

analysis. There are a number of different tests that can be undertaken to assess the

suitability of data for factor analysis.

Tobias and Carlson (Tobias & Carlson, 1969) and Stewart (Stewart, 1981) recommend that

Bartlett's test of sphericity be applied prior to factor analysis. Bartlett's test of sphericity

checks that "the hypothesis tested is that the correlation matrix came from a population of

variables that are independent. Rejection of the hypothesis is an indication that the data are

appropriate for factor analysis" (Stewart, 1981, p. 57)

The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (MSA) is also recommended.

Stewart states that "a final test of the appropriateness of a matrix for factoring is a Kaiser-

Meyer-Olkin measure of sampling adequacy, MSA" (Stewart, 1981, p. 57). The MSA

"provides a measure of the extent to which the variables belong together and are thus

appropriate for factor analysis" (Stewart, 1981, p. 57).

Bartlett’s test of sphericity and the KMO MSA were both used to assess the suitability of the

data set from the NICHE survey for factor analysis.

Methods Used for Factor Analysis

EFA was used instead of confirmatory factor analysis (CFA) due to the exploratory nature of

the research and that the scales were largely developed by researchers specifically for the

project, albeit informed by current research where applicable. "CFA is appropriate when

evaluating scales that have empirical support or where there are clear expectations about

which scales will load on which factors. EFA is appropriate when there is less certainty, such

as when a new scale is developed or substantial revisions are made to an existing scale"

(Smart, 2009, p. 138)

Principal Axis Factoring (PAF) was used for factor extraction as opposed to Principal

Component Analysis (PCA) as PAF accounts for co-variation and PCA results in orthogonal

rotation (uncorrelated factors). Simply put "EFA’s purpose is to identify latent constructs

underlying a set of manifest variables. These latent constructs are hypothesized to account

141

for the patterns of correlations between the manifest variables" (Norris & Lecavalier, 2010,

p. 9). Whereas "PCA is intended to reduce data while preserving as much information from

the original data set as possible. The components extracted in a PCA are not latent

constructs, but are determinate variables that represent a combination of common, specific,

and error variance" (Norris & Lecavalier, 2010, p. 9)

Oblique (direct oblimin) rotation was chosen over orthogonal rotation as it obtains a simple

structure and allows for factors to be either correlated or uncorrelated. Orthogonal

rotations require the factors to be uncorrelated. Norris and Lecavalier state that "since this

is rarely the case in social and behavioral research, methodologists often recommend using

oblique rotations. Oblique rotations allow factors to be correlated, which often reflects

reality more accurately and allows for more clinically useful results. Additionally, if factors

are unrelated, results obtained with oblique rotations will be almost identical to those

produced by orthogonal rotations, making orthogonal rotation essentially obsolete" (Norris

& Lecavalier, 2010, p. 8).

The Kaiser-Guttman criterion (Gorsuch, 1974) was used for the selection of factors. This

criteria identifies factors for selection as those having Eigen-values greater than one, called

the 'root one criterion'.

Normally all items in a survey would be examined simultaneously. However the NICHE

survey was constructed in a number of sections which aligned with the researcher’s

expectations that there were higher order factors overlaying groups of questions measuring

key aspects of respondent’s health and understanding of environmental issues. Because of

this and due to the number of questions contained in the survey it was deemed appropriate

to run factor analysis on each block of relevant constructs.

The groups of variables examined in the EFA are as follows:

Attitudes towards health, the environment, carbon emissions and climate change

Behaviours towards consumption and the environment


142

6.3.1 EFA - Attitudes Towards Health, the Environment, Carbon Emissions and Climate

Change

This section describes the EFA run on the block of variables measuring attitudes towards

health and the environment (questions B1 to B13).

The appropriateness of this group of items for factor analysis was established via Bartlett’s

test of sphericity and the KMO MSA. Both of these measures showed the data suitable for

factor analysis (x2 [78] = 1116.98, p < .001 and MSA = 0.80).

143

1 2 3 4

B9 .734 .060 -.051 .194

B8 .585 .029 -.070 .428

B12 .583 -.084 .081 .010

B11 .504 -.118 -.132 -.225

B3 .395 .015 -.065 -.075

B10 .359 .181 -.127 -.167

B4 -.054 .616 .093 .003

B2 -.021 .520 -.007 .058

B6 .092 .329 -.186 -.267

B5 -.027 -.117 -.781 -.005

B7 .089 .137 -.667 -.031

B1 .002 -.033 -.655 .467

B13 .072 .011 -.576 -.117

Extraction Method: Principal Axis Factoring.

Rotation Method: Oblimin with Kaiser Normalization.

Rotation converged in 18 iterations.

Table 6-9 Pattern Matrixa- 1 – Attitudes Towards Health, the Environment, Carbon Emissions and Climate

Change

144

Pattern Matrix 1 (Table 6-9) shows four factors were extracted (Eigen-values > 1.0) from the

13 scales with three clear underlying factors evident explaining over 60% of the variance

among the items.

The following six scales that are predominantly linked to obesity loaded on the first factor:




B10. I would consider purchasing an electric car or bike if the price was right



The obvious exception among the items was question B10 ‘I would consider purchasing an

electric car or bike if the price was right’. This question was correlated with 10 of the other

12 items in this block of variables. However, while significant at the 99% confidence

interval, the correlation coefficients were not very high therefore the variance explained by

question B10’s relationship with the other variables in this group was not very high. The

highest correlation that this question had with any of the other questions in this section was

0.318 with question B9 ‘I am confident I could maintain a healthy body weight if I wanted to’

indicating that only a little over 10% of the variance was explained by the relationship

between these 2 items. Correlations with the other questions in this group of variables

showed much lower levels of variance. As a result the researchers decided to eliminate this

variable from the analysis and re-run the EFA.

After removing variable B10, Bartlett’s test of sphericity and the KMO MSA were run again

to confirm the suitability of the revised data set for factor analysis. On both counts, the

matrix was deemed to be factorable with x2 [66] = 1085.74, p < 0.001 and MSA = 0.78.

145

1 2 3 4

B9 .713 .063 -.077 .172

B8 .635 .025 -.057 .392

B11 .560 -.073 -.131 -.350

B12 .559 -.063 .058 -.010

B3 .424 .047 -.068 -.132

B4 -.038 .640 .097 .009

B2 .000 .534 -.006 .061

B6 .039 .310 -.219 -.165

B5 -.017 -.110 -.786 -.003

B7 .083 .141 -.681 -.014

B1 .036 -.077 -.614 .425

B13 .059 .030 -.588 -.107



Rotation converged in 25 iterations.

Table 6-10 Pattern Matrixa- 2 – Attitudes Towards Health, the Environment, Carbon Emissions and Climate

Change

As can be seen in Pattern Matrix 2 (Table 6-10), removing variable B10 results in the same

factors minus variable B10 from the first factor. However, the variance explained by the

structures underlying these data items increased from 59.2% in the first solution to 62.2% in

146

the second solution. Given that the variance explained among the items by these structures

has increased, the decision to remove variable B10 from the analysis was justified.

The survey items that loaded in the revised first factor are:






As these items measure an individual’s attitude towards obesity and health this factor has

been labeled ‘Health Consciousness’.

The survey items that loaded on the second factor are:


B4. Obesity will be solved in the future by medical advances

B6. A financial incentive would encourage me to reduce my environmental

impact

As these items measure an individual’s attitude towards the perceived impact that

technology could have in relation to improving health and environmental change this factor

has been labeled ‘Optimism’.

The survey items that loaded on the third factor are:

B1. I buy environmentally friendly products as much as I can.


B7. Collectively, households can reduce the impacts of greenhouse gas emissions


As these items measure an individual’s attitude towards their carbon foot print and climate

change this factor has been labeled ‘Environmental Consciousness’.

147

Factor Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings

a

Total % of Variance Cumulative % Total % of Variance Cumulative % Total

1 3.699 30.822 30.822 3.215 26.792 26.792 2.552

2 1.556 12.970 43.792 .913 7.607 34.399 .886

3 1.207 10.061 53.853 .680 5.665 40.064 2.686

4 1.007 8.394 62.248 .499 4.158 44.222 .557

5 .868 7.235 69.483

6 .735 6.126 75.609

7 .646 5.380 80.989

8 .600 5.001 85.989

9 .498 4.149 90.138

10 .426 3.547 93.685

11 .406 3.383 97.068

12 .352 2.932 100.000


When factors are correlated, sums of squared loadings cannot be added to obtain a total variance.

Table 6-11 Total Variance Explained – Attitudes Towards Health, the Environment, Carbon Emissions and

Climate Change

The factor correlation matrix in Table 6-12 below shows a single significant relationship

between factors 1 and 3 ‘Health Consciousness’ and ‘Environmental Consciousness’. Factor

2 ‘Optimism’ was not correlated with the other factors but has emerged as a construct to

take forward in the analysis. This suggests that although 'Optimism' may be a construct

worth examining further, it is not part of a higher order variable overlaying this set of items

and it will be used in the regression analysis as a separate structure.

148

Factor 2 3 4

1 .003 -.530 -.014

2 -.138 -.123

3 .007



Table 6-12 Factor Correlation Matrix – Attitudes Towards Health, the Environment, Carbon Emissions and

Climate Change

6.3.2 EFA - Behaviours Towards Consumption and the Environment

This section describes the EFA of the following block of variables measuring behaviours

towards consumption and the environment:







The appropriateness of this group of items for factor analysis was established via Bartlett’s

test of sphericity and the KMO MSA. This showed the data for this group of items suitable

for factor analysis (x2 [15] = 603.392, p < 0.001 and MSA = 0.76).

As shown in Table 6-13 below factor analysis resulted in a single factor. As questions B13 -

B19 measure behaviours towards consumption and the environment this factor has been

labeled ‘Environmental Action’. This factor explained nearly 47.5% of the total variance

among these items as shown in Table 6-14.

149

Factor

1

B18 .763

B15 .674

B16 .672

B14 .602

B19 .512

B17 .398


a. 1 factors extracted. 6 iterations required.

Table 6-13 Pattern Matrixa – Behaviours Towards Consumption and the Environment

Total Variance Explained

Factor Initial Eigenvalues Extraction Sums of Squared Loadings

Total % of Variance Cumulative % Total % of Variance Cumulative %

1 2.848 47.474 47.474 2.271 37.845 37.845

2 .964 16.075 63.549

3 .735 12.251 75.799

4 .670 11.159 86.958

5 .426 7.096 94.054

6 .357 5.946 100.000


Table 6-14 Total Variance Explained – Behaviours Towards Consumption and the Environment

6.3.3 EFA - PCTS Usage Intentions

This section describes the EFA of the following block of variables measuring PCTS usage

intentions.

150

E1. Being able to measure my carbon footprint is important to me

E2. Most people would accept a PCT system as a tool for improving the environment

E3. A PCT system would encourage me to reduce my carbon footprint

E4. A PCT system would encourage me to walk or cycle more and drive less

E5. People who reduce their carbon footprint should be rewarded in some way

E6. People with a greater carbon footprint should have to pay for it in some way

E7. A PCT system would encourage me to eat more healthy, locally grown produce

E8. A PCT system would be useful for me to help monitor my environmental impact

E9. Comparing my carbon usage to the average would influence my consumption

habits

E10. There is a strong link between a person’s carbon footprint and their health

Bartlett’s test of sphericity and the KMO MSA showed the data was suitable for factor

analysis (x2 [45] = 2254.859, p < 0.001 and MSA = 0.926).

As shown in Tables 6-15 and 6-16 below the EFA resulted in a single factor accounting for

over 56% of the total variance, with no other factor explaining more than 10%. This factor

has been labeled 'Usage Intentions Towards a PCTS'. Given the correlations among these

items, this was not unexpected as they all relate clearly to ways respondents would use a

PCTS and how information from a PCTS would support personal decision making about their

health and environmental related behaviours.

151

Factor

1

E1 .729

E2 .598

E3 .851

E4 .785

E5 .468

E6 .501

E7 .789

E8 .865

E9 .863

E10 .644


a. 1 factors extracted. 4 iterations required.

Table 6-15 Pattern Matrixa – Usage Intentions Towards a PCTS

152

Total Variance Explained

Factor Initial Eigenvalues Extraction Sums of Squared Loadings

Total % of Variance Cumulative % Total % of Variance Cumulative %

1 5.643 56.428 56.428 5.230 52.300 52.300

2 .924 9.241 65.670

3 .683 6.826 72.496

4 .641 6.411 78.906

5 .538 5.384 84.290

6 .488 4.884 89.174

7 .376 3.755 92.929

8 .290 2.899 95.829

9 .231 2.306 98.135

10 .187 1.865 100.000


Table 6-16 Total Variance Explained – Usage Intentions Towards a PCTS

6.4 Correlations Between the factors and BMI

Correlation analysis was run between the factors that were derived from the EFA and

respondents’ Body Mass Index (BMI) to see what relationships may exist between these

variables prior to undertaking regression analysis.

The result of the correlation analysis between factors derived from the EFA and BMI are

shown in Table 6-17 below. All of the relationships between the factors measuring

'Environmental Consciousness', 'Health Consciousness', 'Usage Intentions Towards a PCTS'

and self-reported BMI were significant at the 99% confidence interval. In some cases the

level of variance explained by the relationships was quite low. The relationships between

'Usage Intentions Towards a PCTS' and 'Environmental Consciousness' (r = 0.553, p < 0.001)

and 'Usage Intentions Towards a PCTS' and 'Health Consciousness' (r = 0.447, p < 0.001)

explained highest levels of variance in the matrix.

153

The factors labeled ‘Optimism’ and 'Environmental Action' had non-significant relationships

with some of the other variables. ‘Optimism’ had only one significant relationship with

'Usage Intentions Towards a PCTS' and the variance explained by the relationship was low

(r = 0.198, p < 0.001). 'Environmental Action' had two significant relationships but the level

of variance explained by each relationship was also low. The absence of a significant

relationship between 'Usage Intentions Towards a PCTS' and 'Environmental Action' was

unexpected. It would seem reasonable to expect that these two factors would have a

significant relationship given that individuals that display positive actions towards the

environment would also be reasonably expected to have positive usage intentions towards

a PCTS. However, among this group of respondents, this is not the case.

Factors Optimism Environmental Consciousness

Usage Intentions Towards a PCTS

Environmental Action

BMI

Health consciousness

.052 .477**

.247**

.137**

.396**

Optimism

.124

* .198

** .012 .080

Environmental consciousness

.553**

.206**

.227**

Usage Intentions Towards a PCTS

.110* .162**

Environmental Action

-.026

Table 6-17 Correlations Between the Factors and BMI

154

The correlations explained above have shown there are key relationships existing among the

factors. The next section explores these relationships using regression modeling and tests

the conceptual model.

6.5 Linear Regression Analysis

Regression analysis is a statistical procedure for estimating the relationships among

variables. There are a number of kinds of regression analysis. In this thesis linear regression

analysis was used. It models the relationship between a dependent variable and

independent variables. SPSS states that "linear Regression estimates the coefficients of the

linear equation, involving one or more independent variables, that best predict the value of

the dependent variable" (SPSS Inc, 2009, p. 343). The dependent and independent variables

must be quantitative and are added to a single regression model (SPSS Inc, 2009).

In this stage of the statistical analysis, linear regression analysis was run on the 'Self Health

Evaluation' group of variables and all of the factors that were extracted from the EFA. The

'Self Health Evaluation' group of variables were not suitable for EFA however due to the

significant relationships that exist between the variables in the group and the expectation

that there was a relationship between this group of variables and 'Usage Intentions Towards

a PCTS' the 'Self Health Evaluation' group of variables were entered into the regression

model as a single block. The variables comprising each of the factors that were extracted

from the EFA were also entered as single blocks of variables.

The EFA examining questions E1 - E10 resulted in a single factor, 'Usage Intentions Towards

a PCTS' . The regression analyses that follow use the 'Usage Intentions Towards a PCTS'

factor as the dependent variable.

The blocks of independent variables were entered in the following order:

'Self Health Evaluation' (variables A9 - A12, C1)

'Health Consciousness' (variables B3, B8, B9, B11, B12)

'Environmental Action' (variables B14 - B19)

155

‘Optimism’(variables B2, B4, B6)

'Environmental Consciousness' (variables B1, B5, B7, B13)

The regression analysis proceeded in four stages as follows:

Analysis of the complete data set with all responses

Analysis moderated by gender

Analysis moderated by carbon footprint

Analysis moderated by gender and carbon footprint

6.5.1 Linear Regression Analysis – All Responses

In the first stage of the regression analysis the complete data set was used. The model

summary for this is shown in Table 6-18 below. The significant relationships are shown in

the shaded rows of the table.

156

Model Summary

Model R R Square Adjusted

R Square

Std. Error of

the

Estimate

Change Statistics

R Square

Change

F Change df1 df2 Sig. F

Change

1 .221a .049 .037 .91936 .049 3.993 4 312 .004

2 .337b .113 .088 .89470 .065 4.487 5 307 .001

3 .364c .133 .090 .89370 .019 1.115 6 301 .353

4 .444d .197 .148 .86439 .064 7.920 3 298 .000

5 .580e .337 .287 .79095 .140 15.476 4 294 .000

a. Predictors: (Constant), A10, C1, A9, A12

b. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9

c. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18

d. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18, B4, B6, B2

e. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18, B4, B6, B2,

B13, B1, B5, B7

Table 1-18 Regression Modeling – Complete Dataset

The model summary shows that 33.4% of the total variance in 'Usage Intentions Towards a

PCTS' is predicted by the 5 blocks of variables.

The 'Self Health Evaluation', 'Health Consciousness', and ‘Optimism’ blocks of variables were

significant at the 95% confidence interval with 4.9%, 6.5%, and 6.4% of the total variance in

'Usage Intentions Towards a PCTS' explained by each block respectively. The

'Environmental Consciousness' block of variables was also significant and it explained the

highest level of variance of any of the 5 blocks, accounting for 14% of the total variance in

'Usage Intentions Towards a PCTS' . The 'Environmental Action' block of variables was not

significant and accounted for only 1.9% of the total variance.

157

Examination of the coefficient matrix (see Appendix B.1.4 - 'Coefficients') and Figure 6.19

below shows that 5 of the 22 variables were significant in explaining the variance in 'Usage

Intentions Towards a PCTS'.

None of the variables in the 'Environmental Action' block were significant, however variable

B17 "I look to buy second hand over brand new" was close (p = 0.053). This block of

variables explained only 1.9% of the overall variance in 'Usage Intentions Towards a PCTS'.

Of the four blocks of variables that were significant in Table 6-18 the following individual

variables from each were significant at the 95% confidence interval or greater:

'Self Health Evaluation' - C1 "How often do you engage in leisure time physical

activity for the sole purpose of improving or maintaining your health" (p = 0.002)

'Health Consciousness' - B9 "I am confident I could maintain a healthy body weight if

I wanted to" (p = 0. 022)

‘Optimism’ - B6 "A financial incentive would encourage me to reduce my

environmental impact" (p < 0.001)

'Environmental Consciousness' - B1 "I buy environmentally friendly products as much

as I can" (p = 0.002) and B13 "I am worried about climate change" (p < 0.001)

It was expected that the 'Environmental Action' block of variables would explain a significant

level of variance in 'Usage Intentions Towards a PCTS' because it was surmised that

individuals who display positive environmental actions could be expected to have positive

usage intentions towards a PCTS and vice versa. Therefore, that the relationship between

‘Environmental Action' and 'Usage Intentions Towards a PCTS' was not significant is of

particular note. However, given the focus of the broader project is the impact of an

understanding of an individual’s carbon footprint on their health, then it expected that a

significant relationship between these variables may become evident from the next stage of

the project.

158

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.163*

β=0.180**

14%**

6.4%**

1.9%

6.5%**

4.9%**

β=0.242**

Environmental

Consciousness

B7

B5

B1

B13

β=0.221**

β=0.198**

β=-0.117*

* 95% confidence interval, ** 99% confidence interval

Figure 6-19 Variance In Usage Intentions Towards a PCTS - Complete Dataset

159

6.5.2 Linear Regression Analysis - Gender

To see what effect gender would have on the level of variance explained in the dependent

variable the regression models were examined by moderating inputs by the variable A1

"What is your gender?" The model summaries from the two analyses have been combined

into Table 6-20 below to more easily enable comparisons. The significant relationships have

been highlighted.

160

Model Summary - MALES


R Square

Std. Error

of the

Estimate

Change Statistics

Male R Square

Change

F Change df1 df2 Sig. F Change

1 .196a .038 .008 .98160 .038 1.269 4 127 .286

2 .347b .120 .055 .95800 .082 2.267 5 122 .052

3 .406c .165 .057 .95722 .045 1.033 6 116 .407

4 .512d .262 .144 .91178 .097 4.950 3 113 .003

5 .619e .383 .258 .84877 .121 5.350 4 109 .001

Model Summary - FEMALES

Model R R

Square

Adjusted

R Square

Std. Error

of the

Estimate

Change Statistics

Female R Square

Change


1 .294a .087 .066 .83501 .087 4.122 4 174 .003

2 .406b .165 .120 .81011 .078 3.172 5 169 .009

3 .442c .195 .121 .80984 .030 1.019 6 163 .415

4 .490d .240 .155 .79406 .045 3.181 3 160 .026

5 .597e .357 .266 .74002 .116 7.056 4 156 .000






B13, B1, B7, B5

Table 6-20 Regression Modeling – Gender

161

Males

The summary in Table 6-20 above shows that among male respondents 38.3% of the total

variance in 'Usage Intentions Towards a PCTS' was predicted by the 5 blocks of variables

entered to the model.

The coefficient matrix (see Appendix B.2.4 - 'Coefficients') and Figure 6.21 below shows that

in the 'Self Health Evaluation' block of variables, variable C1 "How often do you engage in

leisure time physical activity for the sole purpose of improving or maintaining your health"

was significant (p = 0.033). No variables in the 'Environmental Action' block were significant

although variable B17 "I look to buy second hand over brand new" was close (p = 0.063).

However neither of these two blocks of variables explained a significant amount of variance

in 'Usage Intentions Towards a PCTS'.

The 'Health Consciousness' block of variables was close to the 95% confidence interval (sig F

= 0.052) and explained 8.2% of the total variance in 'Usage Intentions Towards a PCTS’.

‘Optimism’ and 'Environmental Consciousness' were significant at the 95% confidence

interval and explained 9.7% and 12.1% of the variance respectively. Of the blocks of

variables that were significant the following individual variables were identified in the

coefficient matrix as being significant at the 95% confidence interval or greater:


I wanted to" (p = 0.007)

‘Optimism’ - B2 "Technology will solve future environmental problems" (p = 0.037)

and B6 "A financial incentive would encourage me to reduce my environmental

impact" (p = 0.003)


as I can" (p = 0.041) and B13 "I am worried about climate change" (p = 0.028)

162

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.299**

β=0.193*

β=0.190*

12.1%**

9.7%**

4.5%

8.2%*

3.8%

β=0.269**

Environmental

Consciousness

B7

B5

B1

B13

β=0.237*

β=0.216*


Figure 6-21 Variance In Usage Intentions Towards a PCTS - Males

163

Females

In the multiple regression model for women, 35.6% of the total variance in 'Usage Intentions

Towards a PCTS' was predicted by the 5 blocks of variables, with 4 of the 5 being significant.

'Environmental Action' was the only block that was not significant.

Of the 4 blocks that were significant ‘Optimism’ explained 4.5%, 'Health Consciousness'

explained 7.8%, 'Self Health Evaluation' explained 8.7% and 'Environmental Consciousness'

explained 11.6% of the variance in 'Usage Intentions Towards a PCTS' .

The coefficient matrix (see Appendix B.3.4 - 'Coefficients') and Figure 6.22 below shows that

none of the individual variables in the 'Health Consciousness' block were significant. Of the

other blocks of variables that were significant the following variables were identified in the

coefficient matrix as being significant at the 95% confidence interval or greater:


environmental impact" (p = 0.020)

'Self Health Evaluation' - A9 "Do you generally consider your health to be" (p = 0.020)

and C1 "How often do you engage in leisure time physical activity for the sole

purpose of improving or maintaining your health" (p = 0.044)


as I can" (p = 0.019) and B13 "I am worried about climate change" (p = 0.036)

164

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.152*

β=-0.180*

11.6%**

4.5%*

3%

7.8%**

8.7%**

β=0.190*

Environmental

Consciousness

B7

B5

B1

B13

β=0.180*

β=0.214*


Figure 6-22 Variance In Usage Intentions Towards a PCTS - Females

165

Differences Reflected in Responses Between Males and Females

A higher percentage of the total variance in 'Usage Intentions Towards a PCTS' was

explained by the responses from males (38.3%) than females (35.6%).

The regression analyses in the previous section and this section show that the

'Environmental Action' block of variables was not significant in any of the models (entire

data set, males or females).

The 'Self Health Evaluation' block of variables explained 8.7% of the total variance in 'Usage

Intentions Towards a PCTS' in the model for females however was not significant in the

model for males.

The responses from males to the questions in the ‘Optimism’ block of variables explained

more than twice the amount of variance in 'Usage Intentions Towards a PCTS' than the

responses to these items from females (9.7% for males compared to 4.5% for females).

The variance explained by the 'Health Consciousness' block of variables (males 8.2% :

females 7.8%) and the 'Environmental Consciousness' block of variables (males 12.1% :

females 11.6%) was similar for both genders.

'Environmental Consciousness' explained the highest level of variance in 'Usage Intentions

Towards a PCTS' in all 3 models that were tested so far (entire data set, males, females). In

all three models the individual variables that were significant at the 95% confidence interval

were B1 "I buy environmentally friendly products as much as I can" and B13 "I am worried

about climate change".

6.5.3 Linear Regression Analysis – Carbon Footprint

In the next stage, the regression models were examined by moderating inputs by the

variable A6 "Compared to others on NI, do you think your carbon footprint is…?" The scale

responses to this question are listed in Table 6-23 below. Responses were entered to the

regression model based on being above the scale mid-point, below the scale mid-point or at

the scale mid-point. This was done to see if differences were evident in the model based on

166

respondent’s evaluation of the size of their carbon footprint, that is, worse than average,

about average or better than average.

A6. Compared to others on Norfolk Island, do you think your carbon

footprint is/would be….

Well below average 1

Below average 2

About average 3

Above average 4

Well above average 5

Table 6-23 - Scale Descriptors – Comparative Carbon Footprint

The regression outputs from the 3 analyses have been combined into Table 6-24 and the

significant relationships have been highlighted.

167

Model Summary A6 > 3 – carbon footprint worse than average

Model R R

Square

Adjusted R

Square

Std. Error of

the

Estimate

Change Statistics

A6 > 3 R Square

Change


Change

1 .359a .129 .016 1.20041 .129 1.146 4 31 .354

2 .536b .287 .041 1.18556 .158 1.156 5 26 .357

3 .694c .481 .092 1.15308 .194 1.248 6 20 .325

4 .766d .586 .149 1.11679 .105 1.440 3 17 .266

5 .903e .816 .505 .85190 .230 4.054 4 13 .024

Model Summary A6 < 3 – carbon footprint better than average

Model R R

Square

Adjusted R

Square

Std. Error of

the

Estimate

Change Statistics

A6 < 3 R Square

Change


Change

1 .338a .114 .077 .90227 .114 3.085 4 96 .019

2 .426b .182 .101 .89048 .068 1.512 5 91 .194

3 .557c .310 .188 .84617 .128 2.630 6 85 .022

4 .573d .328 .180 .85019 .018 .732 3 82 .536

5 .693e .480 .334 .76670 .152 5.708 4 78 .000

Model Summary A6 = 3 – carbon footprint about average

Model R R

Square

Adjusted R

Square

Std. Error of

the

Estimate

Change Statistics

A6 = 3 R Square

Change


Change

1 .240a .058 .035 .85503 .058 2.546 4 166 .041

2 .433b .187 .142 .80644 .129 5.121 5 161 .000

3 .465c .216 .140 .80707 .029 .958 6 155 .456

168

4 .526d .276 .191 .78304 .060 4.219 3 152 .007

5 .577e .332 .233 .76225 .056 3.102 4 148 .017

a. Predictors: (Constant), C1, A10, A9, A12

b. Predictors: (Constant), C1, A10, A9, A12, B11, B8, B3, B9, B12

c. Predictors: (Constant), C1, A10, A9, A12, B11, B8, B3, B9, B12, B15, B17, B19, B14, B16, B18

d. Predictors: (Constant), C1, A10, A9, A12, B11, B8, B3, B9, B12, B15, B17, B19, B14, B16, B18, B4, B6, B2

e. Predictors: (Constant), C1, A10, A9, A12, B11, B8, B3, B9, B12, B15, B17, B19, B14, B16, B18, B4, B6, B2, B1,

B13, B5, B7

Table 6-24 - Regression Modeling – Carbon Footprint

Worse Than Average Carbon Footprint

The first model examining responses by those who rated their carbon footprint as worse

than average (question A6 > 3) was discounted for a number of reasons. While the output

shows that the model explains 81.6% of the total variance in 'Usage Intentions Towards a

PCTS' the first 4 blocks of variables are not significant and the ‘standard error of the

estimate’ is greater than 1 in each instance. The 'Environmental Consciousness' block is

significant and standard error of the estimate is less than 1 however the degrees of freedom

(df2 = 13) is quite low and shows there were only a small number of observations in the

data set that met the model criteria.

Better Than Average Carbon Footprint

The model summary in Table 6-24 above shows that among respondents who thought that

their carbon footprint was better than average, that 3 of the 5 independent variable blocks

were significant in predicting 48% of the total variance in 'Usage Intentions Towards a PCTS'.

‘Optimism’ and 'Health Consciousness' were not significant in this model.

The 'Self Health Evaluation', 'Environmental Action' and 'Environmental Consciousness'

explained 11.4%, 12.8% and 15.2% respectively of the total variance in 'Usage Intentions

Towards a PCTS'. The coefficient matrix (see Appendix B.5.4 - 'Coefficients') and Figure 6-25

169

below shows that the following individual variables in each block were significant at the 95%

confidence interval or greater:

'Self Health Evaluation' - C1 "How often do you engage in leisure time physical

activity for the sole purpose of improving or maintaining your health" (p = 0.002)

'Environmental Action' - B17 "I look to buy second hand over brand new" (p = 0.006),

B18 "I consciously try to reduce waste and recycle" (p = 0.022) and B19 "I buy local

produce, even if imported is cheaper" (p = 0.045)


as I can" (p =0.017) and B13 "I am worried about climate change" (p = 0.018)

170

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.328**

β=-0.232*

β=-0.292**

15.2%**

1.8%

12.8%*

6.8%

11.4%*

Environmental

Consciousness

B7

B5

B1

B13

β=0.267*

β=0.309*


Figure 6-25 Variance In Usage Intentions Towards a PCTS - Better Than Average Carbon Footprint

171

About Average Carbon Footprint

The model in Table 6-24 of the data set of the respondents who thought that their carbon

footprint was about average (question A6 = 3) predicted 33.2% of the total variance in

'Usage Intentions Towards a PCTS'.

The 'Environmental Action' block of variables was not significant. The variances predicted by

the 3 of the 4 other blocks of variables were all quite small - 'Self Health Evaluation' 5.8% ,

‘Optimism’ (6%) and 'Environmental Consciousness' (5.6%). 'Health Consciousness'

explained 12.9% of the total variance in the model.

None of the individual variables (see Appendix B.6.4 - 'Coefficients' and Figure 6-26 below)

in 'Environmental Consciousness' accounted for significant variance in the model despite the

block itself being significant. The following variables in the remaining 3 blocks were

significant:

'Self Health Evaluation' - A9 "Do you generally consider your health to be" (p = 0. 021)


I wanted to" (p = 0. 019)


environmental impact" (p = 0.002)

172

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.214*

β=-0.190*

5.6%*

6%**

2.9%

12.9%**

5.8%*

β=0.255**

Environmental

Consciousness

B7

B5

B1

B13


Figure 6-26 Variance In Usage Intentions Towards a PCTS - About Average Carbon Footprint

173

Discussion of Differences in the Models Moderated on Carbon Footprint


explained among respondents who thought that their carbon footprint was better than

average (48%) than among the respondents who thought that their carbon footprint was

about average (33.2%).

The 'Self Health Evaluation' block of variables explained almost twice the amount of

variance in 'Usage Intentions Towards a PCTS' among respondents who thought that their

carbon footprint was better than average (11.4%) compared with those who thought their

carbon footprint was about average (5.8%).

The 'Health Consciousness' block of variables explained the highest level of variance of any

of the 5 blocks in 'Usage Intentions Towards a PCTS' (12.9%) among respondents who

thought that their carbon footprint was about average. However, it did not explain a

significant amount of variance in the model for respondents who thought that their carbon

footprint was better than average.

For those who thought that their carbon footprint was about average the 'Environmental

Consciousness' block of variables explained only 5.8% of the total variance in 'Usage

Intentions Towards a PCTS' and there were no variables in the block that were individually

significant. The amount of variance explained in this particular model by this block was far

less than when no moderating inputs were used (14%) and when the input was moderated

by gender (males 12.1% and females 11.6%). However the level of variance explained by

the 'Environmental Consciousness' block was 15.2% among respondents who thought their

carbon footprint was better than average and explained the greatest level of variance in

'Usage Intentions Towards a PCTS' of any of the blocks of variables in this model. The

individual variables, B1 "I buy environmentally friendly products as much as I can" (p =0.017)

and B13 "I am worried about climate change" (p = 0.018) were significant in this model as

the models described in the previous sections of this chapter.

‘Optimism’ was only significant in the model moderated by respondents with a carbon

footprint that was about average. That it was not significant in the model for respondents

with a better than average carbon footprint seems to indicate that people that are trying to

174

assist in reducing carbon emissions through having a better than average carbon footprint

are less optimistic of the likelihood of technology being able to solve issues associated with

global warming.

These results were not unexpected although statistical confirmation sets a sound platform

for progressing to the next stages of the project. The reasoning behind the findings being

expected is that individuals who thought that their carbon footprint was better than

average are more likely to display positive environmental attitudes and behaviours and

could be expected that to have positive usage intentions towards a PCTS. Individuals who

did not think that their carbon footprint was better than average could be expected to have

more positive attitudes towards the perceived impact that technology could have in relation

to improving health and environmental change.

6.5.4 Linear Regression Analysis – Males and Females by Carbon Footprint

Table 6-31 at the end of this section shows the regression models moderated by gender and

carbon footprint. These were examined to see if both gender and the size of an individual’s

carbon footprint were related to usage intentions towards a PCTS given that there were

differences evident in the models in the previous section based on gender. The table shows

all models combined into one table for ease of comparison. The significant relationships

have been highlighted.

Neither of the models for male and female respondents with a carbon footprint that was

worse than average were considered admissible due to the small number of degrees-of-

freedom for each block of variables. The remaining four models were admissible with

differing significant relationships evident among some of the variables in each model.

Males - Better than Average Carbon Footprint

The model summary in Table 6-31 at the end of this section shows that among male

respondents with a better than average carbon footprint (question A6 < 3), 68.1% of the

total variance in 'Usage Intentions Towards a PCTS' was predicted by the 5 blocks of

variables entered into the model.

175

Examination of the coefficient matrix (see Appendix B.8.4 - 'Coefficients') and Figure 6-27

below shows that 6 of the 22 variables predicted significant variance individually. The 'Self

Health Evaluation' block of variables had one significant variable (C1 "How often do you

engage in leisure time physical activity for the sole purpose of improving or maintaining your

health", p = 0.022) as did the Optimism’ block (B4 "Obesity will be solved in the future by

medical advances", p = 0.029) however neither of the blocks were significant. The 'Health

Consciousness' block of variables also had one significant variable (B12 "I am unlikely to ever

be obese", p = 0.014) however the block was not significant (Sig. F Change 0.074).

The 'Environmental Action' block of variables was the only block that was significant at the

95% confidence interval in this model, accounting for 22.3% of the total variance and had

the following variables that were individually significant:

B17 "I look to buy second hand over brand new" (p = 0.010)

B18 "I consciously try to reduce waste and recycle" (p = 0.025)

B19 "I buy local produce, even if imported is cheaper" (p = 0.047)

176

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=-0.428*

β=0.365*

β=0.333*

β=-0.362*

β=0.510*

β=-0.395**

6.1%

7.1%

22.3%*

19.5%

13.1%

Environmental

Consciousness

B7

B5

B1

B13


Figure 6-27 Variance In Usage Intentions Towards a PCTS - Males Better Than Average Carbon Footprint

177

Males - About Average Carbon Footprint

The model summary in Table 6-31 at the end of this section shows that among male

respondents with an average carbon footprint (question A6 = 3), 53.1% of the total variance

in 'Usage Intentions Towards a PCTS' was predicted by the 5 blocks of variables entered into

the model.

The 'Health Consciousness' and ‘Optimism’ blocks of variables were the only blocks that

were significant at the 95% confidence interval with 19%, and 9.6% of the variance in 'Usage

Intentions Towards a PCTS' explained by each block respectively.

The coefficient matrix (see Appendix B.9.4 - 'Coefficients') and Figure 6-28 below shows that

variable B6 "A financial incentive would encourage me to reduce my environmental impact"

(p = 0.013) in the ‘Optimism’ block was the only variable that was individually significant at

the 95% confidence interval in this model. The only other significant block, 'Health

Consciousness' had no variables significant individually although variable B9 "I am confident

I could maintain a healthy body weight if I wanted to" was close to the 95% confidence

interval (p = 0.077).

178

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

7.3%

9.6%*

9.3%

19%*

7.9%

β=0.321*

Environmental

Consciousness

B7

B5

B1

B13


Figure 6-28 Variance In Usage Intentions Towards a PCTS - Males About Average Carbon Footprint

179

Females - Better Than Average Carbon Footprint

Among female respondents with a better than average carbon footprint (question A6 < 3),

58.3% of the total variance in 'Usage Intentions Towards a PCTS' was predicted by the 5

blocks of variables entered into the model (Table 6-31).

'Health Consciousness' and 'Environmental Consciousness' were the only significant blocks of

variables at the 95% confidence interval and explained 20.8% and 16.9% of the variance

respectively in 'Usage Intentions Towards a PCTS'.

The coefficient matrix (see Appendix B.11.4 - 'Coefficients') and Figure 6-29 below showed

that variable B11 "Walking or cycling instead of using the car can help reduce a person’s

weight" (p = 0.020) in the 'Health Consciousness' block was significant at the 95% confidence

interval. The other significant block, 'Environmental Consciousness' had no individually

significant variables.

180

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.409*

16.9%*

3.5%

7.5%

20.8%*

9.6%

Environmental

Consciousness

B7

B5

B1

B13


Figure 6-29 Variance In Usage Intentions Towards a PCTS - Females Better Than Average Carbon Footprint

181

Females - About Average Carbon Footprint

Table 6-31 shows that among female respondents with an average carbon footprint

(question A6 = 3), 37.7% of the total variance in 'Usage Intentions Towards a PCTS' was

predicted by the 5 blocks of variables entered to the model.

The 'Self Health Evaluation' and 'Health Consciousness' blocks were significant at the 95%

confidence interval and explained 11.6% and 10.8% of the total variance respectively. The

‘Optimism’ block was not significant however it was very close (Sig F change = 0.057).

'Environmental Action' and 'Environmental Consciousness' were not significant.

The coefficient matrix (see Appendix B.12.4 - 'Coefficients') and Figure 6-30 below shows

that 2 of the 22 variables predicted significant variances individually. Variable A9 "Do you

generally consider your health to be" in the 'Self Health Evaluation' (p = 0.008), B9 "I am

confident I could maintain a healthy body weight if I wanted to" (p = 0.022) were significant

and B12 "I am unlikely to ever be obese" was very close to being significant at p = 0.054.

182

Self Health

Evaluation

Optimism

Environmental

Action

Health

Consciousness

A9

A12

A10

B12

B11

B9

B8

B3

B19

B18

B17

B16

B15

B14

B6

B4

B2

C1

Usage Intentions

Towards a PCTS

β=0.269*

β=-0.284**

3.8%

6.1%*

5.4%

10.8%*

11.6%*

Environmental

Consciousness

B7

B5

B1

B13

β=0.231*

* 95% confidence interval, ** 99% confidence

intervalC

Figure 6-30 Variance In Usage Intentions Towards a PCTS - Females About Average Carbon Footprint

183

Differences Reflected in Responses Between Males and Females by Carbon Footprint


explained among male and female respondents who thought that their carbon footprint was

better than average than among male and female respondents who thought that their

carbon footprint was about average. In both cases the total variance explained by male

respondents was greater than female respondents. Among those who thought that their

carbon footprint was better than average 68.1% of the variance in the model was explained

by male respondents compared with 58.3% for females. Among those who thought that

their carbon footprint was about average 53.1% of the variance in the model was explained

by male respondents compared with 37.7% for females.

The regression model for males with a better than average carbon footprint was the only

model where the 'Environmental Action' block of variables was significant at the 95%

confidence interval when the input was moderated by gender and carbon footprint. Among

males with a better than average carbon footprint the 'Environmental Action' block of

variables was the only block that was significant, accounting for 22.3% of the total variance

in 'Usage Intentions Towards a PCTS'.

The regression model for males with a better than average carbon footprint was also the

only model where the 'Health Consciousness' block was not significant at the 95%

confidence interval when the input was moderated by gender and carbon footprint. In all of

the other models, when the input was moderated by gender and carbon footprint, the

'Health Consciousness' block of variables was significant. Among males with an average

carbon footprint there were no individually significant variables in the 'Health

Consciousness' block although variable B9 "I am confident I could maintain a healthy body

weight if I wanted to" was close (p = 0.077). Variable B9 was also the only individually

significant variable in the 'Health Consciousness' block among females with an average

carbon footprint. However for females with a better than average carbon footprint variable

B11 "Walking or cycling instead of using the car can help reduce a person’s weight" was

significant (p = 0.020).

Among males with an average carbon footprint, in addition to the 'Health Consciousness'

block of variables, the ‘Optimism’ block was significant at the 95% confidence interval, with

184

19%, and 9.6% of the variance in 'Usage Intentions Towards a PCTS' explained by each block

respectively.

Among females with a better than average carbon footprint, in addition to the 'Health

Consciousness' block of variables, the 'Environmental Consciousness' block was significant at

the 95% confidence interval, with 9.6% and 3.5% of the variance in 'Usage Intentions

Towards a PCTS' explained by each block respectively.

Among females with an average carbon footprint, in addition to the 'Health Consciousness'

block of variables, the 'Self Health Evaluation' block was significant at the 95% confidence

interval, with 10.8% and 11.6% of the variance in 'Usage Intentions Towards a PCTS'

explained by each block respectively.

185

Model Summarya – MALES worse than average carbon footprint Model Summary

a – FEMALES worse than average carbon footprint

Model R R

Square

Adjuste

d R

Square

Std. Error

of the

Estimate

Change Statistics Model R R

Square

Adjuste

d R

Square

Std. Error

of the

Estimate

Change Statistics

A6 > 3 R Square

Change

F

Change

df1 df2 Sig. F

Change

A6 > 3 R Square

Change

F

Change

df1 df2 Sig. F

Change

1 .543b .295 .060 1.28041 .295 1.253 4 12 .341 1 .672

a .451 .268 .93412 .451 2.467 4 12 .101

2 .927c .859 .678 .74937 .564 5.607 5 7 .021 2 .853

b .728 .379 .86057 .277 1.428 5 7 .322

3 .998d .996 .931 .34643 .137 5.292 6 1 .321 3 .999

c .999 .980 .15321 .270 36.639 6 1 .126

4 1.000e 1.000 . . .004 . 1 0 . 4 1.000

d 1.000 . . .001 . 1 0 .

Model Summarya – MALES better than average carbon footprint Model Summary

a – FEMALES better than average carbon footprint

Model R R

Square

Adjuste

d R

Square

Std. Error

of the

Estimate


Square

Adjuste

d R

Square

Std. Error

of the

Estimate

Change Statistics

A6 < 3 R Square

Change

F

Change

df1 df2 Sig. F

Change

A6 < 3 R Square

Change

F

Change

df1 df2 Sig. F

Change

1 .362b .131 .050 .90787 .131 1.620 4 43 .187 1 .309

a .096 .019 .89211 .096 1.242 4 47 .306

2 .571c .326 .166 .85050 .195 2.199 5 38 .074 2 .551

b .304 .155 .82783 .208 2.517 5 42 .044

3 .741d .549 .338 .75816 .223 2.637 6 32 .034 3 .616

c .379 .120 .84473 .075 .723 6 36 .634

4 .787e .620 .384 .73121 .071 1.801 3 29 .169 4 .644

d .414 .095 .85683 .035 .663 3 33 .580

5 .825f .681 .400 .72143 .061 1.198 4 25 .336 5 .764

e .583 .267 .77121 .169 2.934 4 29 .038

186

a. A1 = Male a. A1 = Female

b. Predictors: (Constant), C1, A12, A9, A10 b. Predictors: (Constant), C1, A12, A9, A10

c. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11 c. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11

d. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14,

B18

d. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14, B18

e. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14,

B18, B6, B4, B2


B18, B6, B4, B2

f. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14,

B18, B6, B4, B2, B1, B5, B13, B7

f. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14, B18,

B6, B4, B2, B1, B5, B13, B7

Model Summarya – MALES about average carbon footprint Model Summary

a – FEMALES about average carbon footprint

Model R R

Square

Adjuste

d R

Square

Std. Error

of the

Estimate


Square

Adjuste

d R

Square

Std. Error

of the

Estimate

Change Statistics

A6 = 3 R Square

Change

F

Change

df1 df2 Sig. F

Change

A6 = 3 R Square

Change

F

Change

df1 df2 Sig. F

Change

1 .281b .079 .018 .92421 .079 1.288 4 60 .285 1 .341

a .116 .080 .77512 .116 3.246 4 99 .015

2 .518c .269 .149 .86013 .190 2.855 5 55 .023 2 .473

b .224 .150 .74533 .108 2.614 5 94 .029

3 .602d .362 .167 .85116 .093 1.194 6 49 .325 3 .527

c .278 .155 .74307 .054 1.096 6 88 .371

4 .677e .458 .246 .80967 .096 2.717 3 46 .055 4 .582

d .339 .198 .72357 .061 2.602 3 85 .057

5 .729f .531 .285 .78825 .073 1.633 4 42 .184 5 .613

e .376 .207 .71978 .038 1.224 4 81 .307

187

a. A1 = Male a. A1 = Female

b. Predictors: (Constant), C1, A12, A9, A10 b. Predictors: (Constant), C1, A12, A9, A10

c. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11 c. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11

d. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14,

B18



B18, B6, B4, B2


B18, B6, B4, B2

f. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14,

B18, B6, B4, B2, B1, B5, B13, B7

f. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14, B18,

B6, B4, B2, B1, B5, B13, B7

Table 6-31 Regression Modeling – Gender and Carbon Footprint

6.6 Conclusion

The analysis described in this chapter has resulted in the identification of key relationships

among the research constructs that provide interesting insights into the motivations that

drive usage intentions towards a PCTS. The analysis has resulted in identification of

differences in usage intentions towards a PCTS based on gender and an individual's carbon

footprint. Table 6-32 below highlights these differences and shows which constructs are

significant predictors of 'Usage Intentions Towards a PCTS' based on gender and carbon

footprint.

The differences emerging from the analysis are as follows.

The responses from females on Norfolk Island showed that 'Self Health Evaluation' was a

significant predictor of 'Usage Intentions Towards a PCTS' but was not statistically significant

for male respondents on the Island. This seems to indicate that among females usage

intentions towards a PCTS were influenced in part by their view of their health, body weight

and level of exercise however this is not the case for males.

The responses among Norfolk Islanders with an average carbon footprint showed that

'Health Consciousness' and ‘Optimism’ were significant predictors of 'Usage Intentions

Towards a PCTS' whereas 'Environmental Action' was not. However among Norfolk

Islanders with a better than average carbon footprint the opposite is true. 'Health

Consciousness' and ‘Optimism’ were not significant predictors of 'Usage Intentions Towards

a PCTS' whereas 'Environmental Action' was. The level of variance explained by

'Environmental Consciousness' is also almost three times higher among this group than

among those with an average carbon footprint. This seems to indicate that among Norfolk

Islanders with a better than average carbon footprint attitudes towards their carbon foot

print and climate change and behaviours towards consumption and the environment

influence their usage intentions towards PCTS. Whereas among Norfolk Islanders with an

average carbon footprint the perceived impact that technology could have in relation to

189

improving health and environmental change and their attitudes towards obesity and health

influence their usage intentions towards PCTS.

Among male Norfolk Islanders with a better than average carbon footprint 'Environmental

Action' is the only significant predictor of 'Usage Intentions Towards a PCTS' . The significant

predictors for male Norfolk Islanders with an average carbon footprint are 'Health

Consciousness' and ‘Optimism’. For females with a better than average carbon footprint the

significant predictors are 'Health Consciousness' and 'Environmental Consciousness'. Among

females with an average carbon footprint the significant predictors are 'Health

Consciousness' and 'Self Health Evaluation' with ‘Optimism’ being an almost significant

predictor. This indicates that:

Usage intentions towards a PCTS among males with a better than average carbon

footprint are influenced by their behaviours towards consumption and the

environment however this is not the case with any other group

Usage intentions towards a PCTS among females with a better than average carbon

footprint are influenced in part by their attitudes towards their carbon foot print and

climate change however this is not the case with any other group

Usage intentions towards a PCTS among females with an average carbon footprint

are influenced in part by their view of their health, body weight and level of exercise

however this is not the case with any other group

With the exception of males with a better than average carbon footprint, Norfolk

Islanders usage intentions towards a PCTS are influenced in part by their attitudes

towards obesity and health

Usage intentions towards a PCTS among males and females with an average carbon

footprint are influenced in part by the perceived impact that technology could have

in relation to improving health and environmental change where as this is not the

case for males and females with a better than average carbon footprint

190

It must be noted that no constructs were significant across all the models. 'Environmental

Consciousness' was the only construct that was significant for the entire dataset and when

input was moderated by gender OR carbon footprint however when gender AND carbon

footprint was used as a moderating input 'Environmental Consciousness' was only significant

for females with a better than average carbon footprint. No other construct was significant

for the entire dataset and when input was moderated by gender or carbon footprint.

The non-significant relationships emerging from the analysis were unexpected but

nonetheless important in developing a profile of key indicators related to implementation of

PCTS. What was most surprising for NICHE researchers was the non-significance of

'Environmental Action' in all of the regression models with the exception of male

respondents who thought that their carbon footprint was better than average. It had been

expected that 'Environmental Action' would explain a significant levels of variance in 'Usage

Intentions Towards a PCTS' for all groups as it could be expected that individuals who have

positive environmental actions would also have positive usage intentions towards PCTS. A

possible reason for this block of variables not being significant may be due to the positive

environmental actions of the vast majority of Norfolk Islanders being for financial reasons,

regardless of their usage intentions towards a PCTS. Another reason may be the ‘value-

action gap’ which can occur when an individuals' values and beliefs do not correlate with

their actions. A more in-depth discussion of this and the reasons NICHE researchers believe

are behind the other significant and non-significant relationships emerging from the analysis

is undertaken in Section 7.2 'Results of the Statistical Analysis' of the next chapter. Also

discussed in the next chapter are the unique characteristics of the Norfolk Island sample and

their environment which in some cases could mean they are atypical and their responses

may not necessarily reflect the attitudes of respondents from other locations. However

given that this is the first stage of the broader NICHE project the statistical analysis has

identified important information to predict usage intentions towards a PCTS and aid

implementation of PCTS in other locations.

191

'Self Health Evaluation'

'Health Consciousness'

'Environmental Action'

‘Optimism’ 'Environmental Consciousness'

Complete dataset

Males

Females

Better than average carbon

footprint

About average carbon footprint

Males better than average carbon

footprint

Males about average carbon

footprint

Females better than average

carbon footprint

Females about average carbon

footprint

Table 6-32 Significant Predictors of Usage Intentions Towards a PCTS

192

7 Conclusion

7.1 Introduction

This chapter summarizes the key findings from the statistical analysis of the baseline survey

undertaken for the NICHE project. The key outcomes of the regression modelling and the

differences that were found among the models are discussed. The regression models are

then compared to the conceptual model. The author also proposes possible reasons to

account for the differences that occur among the models.

This is followed by a summary of the limitations of the study and the ways these limitations

could be addressed in future research. Finally the implications and relevance of these

findings and an examination of the focus of future research within the NICHE project is

discussed.

7.2 Results of the statistical analysis

Five factors were identified from the analysis as the independent variables to be entered to

the regression models. While similar to the constructs in the conceptual model that was

introduced in chapter 1, these factors were subsequently given labels to appropriately

reflect the items that loaded on each factor identified in the EFA. The factors were:

'Self Health Evaluation' (variables A9 - A12, C1)

'Health Consciousness' (variables B3, B8, B9, B11, B12)

'Environmental Action' (variables B14 - B19)

‘Optimism’(variables B2, B4, B6)

'Environmental Consciousness' (variables B1, B5, B7, B13)

193

Linear regression analysis was run on the entire dataset. Input was then moderated by

gender, carbon footprint and carbon footprint by gender. The main findings and the

differences that were found between the regression models are as follows.

The regression model run on the whole dataset showed that 'Self Health Evaluation',

'Health Consciousness', ‘Optimism’ and 'Environmental Consciousness' were

significant predictors of 'Usage Intentions Towards a PCTS'.

These four variables were also significant in the regression model moderated by

gender for females. However in the model for males 'Self Health Evaluation' was not

significant.

The models moderated by carbon footprint showed that:

o 'Self Health Evaluation', and 'Environmental Consciousness' were significant

across all models;

o 'Environmental Action' was also significant in the model for those with a

better than average carbon footprint; and

o 'Health Consciousness' and ‘Optimism’ were also significant in the model for

those with an average carbon footprint.

The models moderated by gender and carbon footprint showed that:

o 'Environmental Action' was the only significant predictor for males with a

better than average carbon footprint

o 'Health Consciousness' and ‘Optimism’ were the significant predictors for

males with an average carbon footprint

o 'Health Consciousness' and 'Environmental Consciousness' were the

significant predictors for females with a better than average carbon footprint

o 'Health Consciousness' and 'Self Health Evaluation' were the significant

predictors for females with an average carbon footprint

Somewhat surprising was the non-significance of 'Environmental Action' in the model for the

whole dataset and the models moderated by gender and 3 of the 4 models moderated by

194

carbon footprint and gender. It was thought that 'Environmental Action' would be significant

as individuals who have positive environmental actions could reasonably be expected to

have positive usage intentions towards PCTS and vice versa, hence its inclusion in the

conceptual model. Being found not to be significant has led to it being surmised that the

positive environmental actions of the vast majority of Norfolk Islanders results from financial

reasons. Due to the geographic location of Norfolk Island all resources have to be shipped to

the island and electricity, fuel and imported food sources are far more expensive than on the

mainland. Because of this extra financial cost residents are far more likely to have solar

electricity and solar hot water systems and engage in positive environmental actions and

behaviours regardless of their usage intentions towards PCTS. While some residents do

engage in positive environmental actions for environmental reasons, it is thought that others

do it out of necessity. However this has not been tested in the current research. An

effective method of testing this supposition would be to test the willingness of respondents

to adopt even more demanding environmental behaviours for non-financial reasons.

Another possible reason for the non-significance of 'Environmental Action' may be the

‘value-action gap’ which can occur when an individuals’ values and beliefs do not correlate

with their actions. This gap is especially prevalent in environmental policy where

environmental beliefs do not always translate into behaviours and actions (Blake, 1999).

While this theory was not tested it is interesting to contrast ‘Environmental Action’ with

'Environmental Consciousness'. The latter variable measures an individual’s attitudes and

beliefs towards their carbon foot print and climate change as opposed to their behaviours.

In the regression model for the entire dataset, the models moderated by gender and the

model for those with a better than average carbon footprint, 'Environmental Consciousness'

explained the highest level of variance of all of the constructs.

While ‘Environmental Action’ was found to be non-significant, it will still be investigated in

the next stages of the project to assess users’ usage intentions towards a PCTS after having

used the PCTS on the Island for 12 months. While it was not significant in the baseline

survey reported in this thesis, it is thought that an individual’s understanding of their carbon

195

footprint will inform their environmental behaviours as well as their attitudes to the

environment and climate change.

The regression models show that ‘Health Consciousness’ was significant for both males and

females and that ‘Self Health Evaluation’ was also significant for females. With the

exception of females with a better than average carbon footprint, 'Self Health Evaluation'

was significant for females in all of the regression models moderated by gender and by

carbon footprint and gender, but was not significant in any of the models for males. This is

in contrast to 'Health Consciousness' that was significant for all of the models for females

and in all of the models for males with the exception of males with a better than average

carbon footprint.

The literature shows that males use health services less often than females, their health is

poorer and their mortality rates are higher (Doherty & Kartalova-O'Doherty, 2010;

Jayasinghe, Harris, Taggart, Christl, & Black, 2013; Jeffries & Grogan, 2012; Mansfield, Addis,

& Mahalik, 2003; J. A. Smith, Braunack-Mayer, & Wittert, 2006; Tabenkin, Goodwin,

Zyzanski, Stange, & Medalie, 2004). The exact reasons for this disparity in the use of health

services is not clear however it is thought that due to masculine stereotypes and cultural

reasons males are less interested or concerned about their health and less likely to seek help

than females for health related problems (Jeffries & Grogan, 2012; Mansfield et al., 2003; J.

A. Smith et al., 2006). It is not clear from the baseline study why ‘Self Health Evaluation’

was significant for females and not significant for males. However it appears that the

literature is indicating that males are less likely to seek help related to their health and may

in fact be less concerned with their own health. While it is recognised that males may

endeavor to remain healthy and understand the risks associated with obesity, it appears

their actions are not necessarily influenced by their evaluation of their own health to the

same extent as females. However, these are suppositions only and areas that other

researchers could explore in greater detail.

While 'Self Health Evaluation' was not significant in the models for males it will be further

investigated in the next stages of the NICHE project. After users have used the PCTS for 12

196

months this relationship will be revisited to see if males have a better understanding of the

relationship between their own health and how it relates to their carbon footprint.

When input was moderated by carbon footprint the regression models show that

'Environmental Action' was significant for those with a better than average carbon footprint

and ‘Optimism’ was significant for those with an average carbon footprint. It was thought

that both of these constructs would be significant because individuals who have a better

than average carbon footprint could be expected to believe that environmental actions are

needed to lower carbon emissions and bring about environmental change. Whereas

individuals with an average carbon footprint could be expected to believe that future

technology could reduce the impact of carbon emissions and bring about environmental

change. However it is not clear why 'Health Consciousness' was significant in the model for

those with an average carbon footprint but not for those with a better than average carbon

footprint.

In the next stages of the NICHE project the relationship between 'Health Consciousness' and

'Usage Intentions Towards a PCTS' will be further investigated for those with a better than

average carbon footprint. It is clear from the model that these users understand the

relationship between their environmental actions and their carbon footprint and that this

influences their usage intentions towards a PCTS. However, this is not the case for 'Health

Consciousness'. Once individuals with a better than average carbon footprint have had the

chance to use a PCTS for 12 months this relationship will be revisited to see if they are more

fully informed about the relationship between their health related behaviours and their

carbon footprint.

The significance of ‘Optimism’ for males with an average carbon footprint was surmised as

occurring for the same reasons listed in the previous paragraphs. Although this factor was

not significant for females with an average carbon footprint, it was very close (p = 0.57)

which means it should not be completely discounted. It was also expected that the

‘environmental’ constructs would be significant for males and females with a better than

average carbon footprint as individuals with a better than average carbon footprint could be

197

expected to have positive environmental beliefs. The analysis showed 'Environmental

Action' to be significant for males with a better than average carbon footprint and

'Environmental Consciousness' for females with a better than average carbon footprint.

However the reasons why 'actions' are significant for males and 'consciousness' is significant

for females were not clear from the data collected in the baseline survey. It was thought

that both of these constructs would be significant for males and females with a better than

average carbon footprint. It was also expected that 'Health Consciousness' would be

significant for all models moderated by carbon footprint and gender as the belief that there

is a link between positive health behaviours and pro-environmental behaviours is a

fundamental tenant of the NICHE project. It was surmised that males lack of interest about

their health and help seeking behaviour that was discussed in the paragraph above may

explain the non-significance of 'Health Consciousness' for males with a better than average

carbon footprint. However, it is not clear from the baseline survey why 'Health

Consciousness' was not significant for males with a better than average carbon footprint but

was significant for males with an average carbon footprint.

After users have used the PCTS for 12 months these relationships will be revisited. At this

time users will have a better understanding of their carbon footprint and how it relates to

their behaviours than they did when the baseline survey was conducted and the

relationships between the constructs and 'Usage Intentions Towards a PCTS' may change.

NICHE researchers recognize that the survey that is reported on as the basis of this thesis

was a baseline survey to gather data on individual and household attitudes prior to the roll-

out of a PCTS on the Island and that the significant predictors of 'Usage Intentions Towards a

PCTS' may change once the users have used the NICHE PCTS and more fully understand how

their actions and behaviours affect their carbon footprint. As part of the next stages of the

NICHE project an interim survey is planned for mid to late 2014 after the PCTS has been in

operation for over a year. The relationships reported in this thesis between the constructs

and the dependent variable will be revisited at that time to see if using a PCTS changes what

constructs are significant predictors of usage intentions towards a PCTS.

198

7.3 Limitations of the study

The limitations of the study and the sample frame were covered in chapter 4. However, in

light of the findings from the statistical analysis some aspects of the limitations are discussed

further in this section.

The fact that Norfolk Island is a closed system makes it ideal for this study but it also

presents some limitations. While the population is reasonably representative of other

developed locations its geographic isolation can reasonably be expected to affect attitudes

and behaviours towards the environment, climate change and health. As discussed in the

previous section most food products and all gas, diesel and electricity have to be shipped to

the island. Electricity and fuel are far more expensive than they are in main land Australia

and a larger proportion of households have solar power and solar hot water. Given the

high level of imported goods which makes them more expensive than they would be at the

point of origin, it is reasonable to expect that residents on the Island would be more

conscious of their consumption. This would be expected to be true of general goods as well

as energy based resources, fuel and natural gas. Therefore the relationships between

‘Environmental Action’ and PCTS evident in the models may be driven by necessity, as well

as, or even more so, than concerns about the environment.

A limitation of the study is the extent to which non-respondents could differ to respondents.

In general survey work it can be held that people do not respond to written surveys due to

low literacy levels, disinterest in the subject matter related to the research, and/or other

concerns such as privacy and confidentiality. Dillman’s formula (Dillman, 2007; Needham,

Vaske, & Vaske, 2008, p. 180) was used to assess the generalizability of the sample to the

broader population on the island. To ensure generalizability at the 95% confidence interval a

sample size of 260 would be required given there are 805 households on the island. The

current study received responses from 423 of the 805 households which represents a

response rate of 52.5%. While extrapolation of the results to broader populations should be

undertaken cautiously, the response rate for this study indicates that it is representative of

199

the broader population.Another limitation of the study is that the survey was administered

to gather both household data and individual data. Since only one member of each

household was required to fill out the survey it has to be recognized that the attitudes and

beliefs reflect that person’s views rather than the views held by all members of the

household. Attempts were made to try and randomize the sample of residents from each

household who filled out the survey. The age distribution of survey respondents was similar

to the NICHE 2011 census (see chapter 5) however there were some differences, particularly

in the 20 - 29 year age band. A higher percentage of females (59.8%) filled out the survey for

their household than males (40.2%). However, the due to the high number of responses (n =

423) the survey could be held to be representative of the views of the broader population of

Norfolk Island.

The unique nature of the NICHE study also makes it difficult to identify other research of this

nature for comparative purposes. The lack of comparative research means that the

conceptual model and the scales developed to measure the constructs were tested for the

first time in this study. ‘Optimism’ was identified from the study and is a new construct that

was not identified from any related literature. This suggests that there may be other

constructs that could affect usage intentions towards PCTS that were not considered by the

NICHE researchers.

A further limitation of the research was the anonymity of the NICHE survey. This did not

allow researchers to follow up individual respondents or households with further questions.

For example, it would have been interesting to follow up with respondents about the factor

'Environmental Action' and its non-significance in some of the models. This was an

unexpected outcome of the research and may be a reflection of the individual survey items

or could truly reflect the attitudes of those respondents.

It also needs to be recognized that this was a baseline survey to gather data on individual

and household attitudes prior to the roll-out of a PCTS on the Island. A follow up survey and

health evaluations will be conducted in mid to late 2014 when the PCTS has been in

200

operation for over 12 months and the baseline parameters that are reported here will be re-

evaluated and compared with the acceptance of the PCTS.

7.4 Implications and relevance for future research

A number of studies have looked at the political acceptability of PCT in different countries.

There has also been considerable research studying carbon emissions and climate change

and the link between them. Much of this research has looked at individuals beliefs

surrounding the causes of climate change. Considerable research has also looked at health

and obesity and the causes behind and links between them. However an extensive review of

the literature was unable to identify research into how all of these factors fit together and

influence an individual's usage intention towards PCTS. It is felt that the NICHE project is the

first study of its kind to explore how an individual's attitudes and behaviours in these areas

influence their usage intention towards a PCTS. This research is highly relevant to any future

research in these areas and gives future researchers a solid foundation to further explore the

relationship between these attitudes and behaviours and their effect on PCTS.

Capping emissions through the use of PCT is one of the main proposals put forward to

reduce carbon emissions. The use of PCT has been proposed in many countries and a similar

‘cap and trade’ system, the European Union Emission Trading System (EUETS) has been

implemented in the European Union (EU). There have been a number of studies that have

looked at the level of support for PCT however there has been no research into what

attitudes and behaviours affect these levels of support. The research contained in this thesis

into what attitudes and behaviours affect an individual's usage intention towards a PCTS is

highly relevant for researchers seeking to understand how PCTS will be accepted by a

population as a whole.

As this appears to be the first research of its kind to measure usage intentions towards PCTS

it will provide researchers with a validated model as a basis for future research. The scales

201

that were developed to measure the constructs for the conceptual model will also provide a

starting point for any future research of this nature.

7.5 Future NICHE research

At the start of March 2013 a PCTS (See chapter 3) that was developed by the author was

rolled out to monitor the carbon footprint of selected residents of Norfolk Island. The NICHE

PCTS tracks electricity and gas usage and selected purchases made by NICHE participants

and calculates the carbon footprint of participants based upon these purchases and services.

The NICHE PCTS trial is scheduled to run until mid 2014. A follow up survey and health

evaluations will be conducted at the end of the PCTS trial. Future NICHE research and the

follow up survey will focus on the following areas:

Examine the acceptance of a PCTS as opposed to usage intention towards a PCTS

among the NICHE projects participants

Determine what impact the PCTS trial has had on an individual's attitudes and

behaviours towards consumption, the environment and carbon emissions

Determine what impact the PCTS trial has had on an individual's attitudes and

behaviours towards health, their health evaluation and their weight

Determine what impact the PCTS trial has had on which constructs are significant

predictors of 'Usage Intentions Towards a PCTS' among different segments of the

population

The refinement and further testing of the scales that were developed specifically for

this project to measure the constructs contained in the conceptual model

7.6 Conclusion

This thesis has made a unique contribution to the literature surrounding PCT. The

identification of attitudes and behaviours that affect an individual's usage intention towards

202

a PCTS and the results of the statistical analysis contained in this thesis will allow future

researchers to more fully understand usage intentions towards and acceptance of PCT as a

means to reduce or at least understand their own and their household’s carbon footprint.

The scales that were developed will also provide a starting point for other researchers

looking to conduct research in this domain.

The development of the conceptual model that was used as the basis of the research for this

thesis and the constructs that were identified from the statistical analysis serve as a baseline

for comparison for further research once the NICHE PCTS trial has concluded. The

refinements made to the constructs and the addition of the construct ‘Optimism’ provide

NICHE researchers with an important insight into the design of the follow up survey and

areas to focus their future research in the NICHE project.

203

References

Adams, D. A., Nelson, R. R., & Todd, P. A. (1992). Perceived usefulness, ease of use, and

usage of information technology: a replication. MIS quarterly, 227-247.

Akter, S., & Bennett, J. (2011). Household perceptions of climate change and preferences for

mitigation action: the case of the Carbon Pollution Reduction Scheme in Australia.

Climatic change, 109(3-4), 417-436.

Al-Nesf, Y., Kamel, M., El-Shazly, M., Makboul, G., Sadek, A., El-Sayed, A., & El-Fararji, A.

(2006). Kuwait STEPS 2006: Kuwait Minsitry of Health.

Anderegg, W. R., Prall, J. W., Harold, J., & Schneider, S. H. (2010). Expert credibility in climate

change. Proceedings of the national academy of sciences, 107(27), 12107-12109.

Armstrong, T., & Bull, F. (2006). Development of the World Health Organization Global

Physical Activity Questionnaire (GPAQ).

Bener, A., Al-Suwaidi, J., Al-Jaber, K., Al-Marri, S., Dagash, M., & Elbagi, I. (2004). The

prevalence of hypertension and its associated risk factors in a newly developed

country. Saudi Medical Journal, 25(7), 918 - 922.

Bird, J., Jones, N., & Lockwood, M. (2008). Political acceptability of personal carbon trading:

Findings from primary research. London, IPPR.

Blake, J. (1999). Overcoming the ‘value‐action gap’in environmental policy: Tensions

between national policy and local experience. Local environment, 4(3), 257-278.

Bleich, S., Cutler, D., Murray, C., & Adams, A. (2007). Why is the developed world obese? :

National Bureau of Economic Research.

Bray, G. A., Nielsen, S. J., & Popkin, B. M. (2004). Consumption of high-fructose corn syrup in

beverages may play a role in the epidemic of obesity. The American Journal of Clinical

Nutrition, 79(4), 537-543.

Bristow, A. L., Wardman, M., Zanni, A. M., & Chintakayala, P. K. (2010). Public acceptability

of personal carbon trading and carbon tax. Ecological Economics, 69(9), 1824-1837.

Caballero, B. (2007). The global epidemic of obesity: an overview. Epidemiologic reviews,

29(1), 1-5.

204

Capstick, S., & Lewis, A. (2009). Personal carbon allowances: a pilot simulation and

questionnaire. UK Energy Research Centre, London.

Carlsson-Kanyama, A., & González, A. D. (2009). Potential contributions of food consumption

patterns to climate change. The American Journal of Clinical Nutrition, 89(5), 1704S-

1709S.

Cutler, D., Glaeser, E., & Shapiro, J. (2003). Why have Americans become more obese? :

National Bureau of Economic Research.

Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of

information technology. MIS quarterly, 319-340.

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer

technology: a comparison of two theoretical models. Management Science, 982-

1003.

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer

technology: a comparison of two theoretical models. Management Science, 35(8),

982-1003.

DeLone, W. H., & McLean, E. R. (1992). Information Systems Success: The Quest for the

Dependent Variable. Information systems research, 3(1), 60-95.

DeLone, W. H., & McLean, E. R. (2002). Information systems success revisited.

DeLone, W. H., & McLean, E. R. (2003). The DeLone and McLean Model of Information

Systems Success: A Ten-Year Update. Journal of Management Information Systems,

19(4), 9-30.

Delpeuch, F., Maire, B., & Emmanuel, M. (2009). Globesity: a planet out of control?

Department for Environment Food and Rural Affairs. (2008). Retrieved 17 August, 2013,

from

http://webarchive.nationalarchives.gov.uk/20130123162956/http:/www.defra.gov.u

k/news/2008/080508c.htm/

Department of Environment Climate Change and Water NSW. (2010). Who cares about the

environment in 2009?

http://webarchive.nationalarchives.gov.uk/20130123162956/http:/www.defra.gov.uk/news/2008/080508c.htm/

http://webarchive.nationalarchives.gov.uk/20130123162956/http:/www.defra.gov.uk/news/2008/080508c.htm/

205

Dillman, D. A. (2007). Mail and Internet surveys: The tailored design method--2007 Update

with new Internet, visual, and mixed-mode guide: John Wiley & Sons.

Doherty, D. T., & Kartalova-O'Doherty, Y. (2010). Gender and self-reported mental health

problems: Predictors of help seeking from a general practitioner. British Journal of

Health Psychology, 15(1), 213-228. doi: 10.1348/135910709X457423

Doran, P. T., & Zimmerman, M. K. (2009). Examining the scientific consensus on climate

change. Eos, Transactions American Geophysical Union, 90(3), 22.

Dunlap, R. E., & McCright, A. M. (2008). A Widening Gap: Republican and Democratic Views

on Climate Change. Environment, 50(5), 26-35.

Edwards, P., & Roberts, I. (2009). Population adiposity and climate change. International

Journal of Epidemiology, 38(4), 1137-1140.

Egger, G. (2007). Personal carbon trading: a potential" stealth intervention" for obesity

reduction? Medical Journal of Australia, 187(3), 185.

Egger, G. (2008). Dousing our inflammatory environment (s): is personal carbon trading an

option for reducing obesity–and climate change? obesity reviews, 9(5), 456-463.

Egger, G., & Swinburn, B. (2011). Planet Obesity: How we're eating ourselves and the planet

to death: Allen and Unwin.

Energy Saving Trust. (2007). Green barometer: Measuring environmental attitudes. Energy

Saving Trust, London.

Faergeman, O. (2007). Climate change and preventive medicine. European Journal of

Cardiovascular Prevention & Rehabilitation, 14(6), 726-729.

Fawcett, T., & Parag, Y. (2010). An introduction to personal carbon trading.

Field, C. B., Barros, V., Stocker, T. F., Qin, D., Dokken, D., Ebi, K., . . . Allen, S. (2012).

Managing the risks of extreme events and disasters to advance climate change

adaptation. A Special Report of Working Groups I and II of the Intergovernmental

Panel on Climate Change Cambridge University Press, Cambridge, UK, and New York,

NY, USA.

206

Friel, S., Dangour, A. D., Garnett, T., Lock, K., Chalabi, Z., Roberts, I., . . . McMichael, A. J.

(2009). Public health benefits of strategies to reduce greenhouse-gas emissions: food

and agriculture. The Lancet, 374(9706), 2016-2025.

Gallup. (2014). Environment. Retrieved 6 January, 2014, from

http://www.gallup.com/poll/1615/Environment.aspx

Gorsuch, R. L. (1974). Factor Analysis. Philadelphia: W.B. Saunders Company.

Hale, J. L., Householder, B. J., & Greene, K. L. (2002). The theory of reasoned action. The

persuasion handbook: Developments in theory and practice, 259-286.

Hawkes, C. (2006). Uneven dietary development: linking the policies and processes of

globalization with the nutrition transition, obesity and diet-related chronic diseases.

Globalization and health, 2(1), 4.

Hedley, A. A., Ogden, C. L., Johnson, C. L., Carroll, M. D., Curtin, L. R., & Flegal, K. M. (2004).

Prevalence of overweight and obesity among US children, adolescents, and adults,

1999-2002. JAMA: the journal of the American Medical Association, 291(23), 2847-

2850.

Hendrickson, A. R., Massey, P. D., & Cronan, T. P. (1993). On the test-retest reliability of

perceived usefulness and perceived ease of use scales. MIS quarterly, 227-230.

Higgins, P. A. (2005). Exercise-based transportation reduces oil dependence, carbon

emissions and obesity. Environmental Conservation, 32(3), 197-202.

Higgins, P. A., & Higgins, M. (2005). A healthy reduction in oil consumption and carbon

emissions. Energy Policy, 33(1), 1-4.

Hill, H. (2008). Food miles: background and marketing: ATTRA.

Hill, J. O., Wyatt, H. R., Reed, G. W., & Peters, J. C. (2003). Obesity and the environment:

where do we go from here? Science, 299(5608), 853-855.

Hossain, P., Kawar, B., & El Nahas, M. (2007). Obesity and diabetes in the developing world—

a growing challenge. New England Journal of Medicine, 356(3), 213-215.

Intergovernmental Panel on Climate Change. (2007). IPCC AR4 SYR Appendix Glossary.

Retrieved 22 August, 2013, from http://www.ipcc.ch/pdf/assessment-

report/ar4/syr/ar4_syr_appendix.pdf

http://www.gallup.com/poll/1615/Environment.aspx

http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr_appendix.pdf

http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr_appendix.pdf

207

Intergovernmental Panel on Climate Change. (2013). Climate Change 2013.

International Association for the Study of Obesity. (2012). Retrieved 22 July, 2013, from

http://www.iaso.org/resources/obesity-data-portal/resources/charts/

International Association for the Study of Obesity. (2013). Retrieved 15 July 2013, from

http://www.iaso.org/resources/world-map-obesity/

Jagers, S. C., Löfgren, Å., & Stripple, J. (2009). Attitudes to Personal Carbon Allowances: The

effect of trust in politicians, perceived fairness and ideology. rapport nr.: Working

Papers in Economics 360.

Jayasinghe, U. W., Harris, M. F., Taggart, J., Christl, B., & Black, D. A. (2013). Gender

differences in health-related quality of life of Australian chronically-ill adults: patient

and physician characteristics do matter. Health And Quality Of Life Outcomes, 11,

102-102. doi: 10.1186/1477-7525-11-102

Jeffries, M., & Grogan, S. (2012). ‘Oh, I’m just, you know, a little bit weak because I’m going

to the doctor's’: Young men's talk of self-referral to primary healthcare services.

Psychology & Health, 27(8), 898-915. doi: 10.1080/08870446.2011.631542

Kahn Ribeiro, S., Kobayashi, S., Beuthe, M., Gasca, J., Greene, D., Lee, D. S., . . . Zhou, P. J.

(2007). Transport and its infrastructur. Climate Change 2007: Mitigation. Contribution

of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel

on Climate Change.

Keay-Bright, S., Fawcett, T., Howell, R., & Parag, Y. (2008). Personal Carbon Trading (PCT):

Bringing together the research community. Paper presented at the Workshop Report.

Kim, B., & Neff, R. (2009). Measurement and communication of greenhouse gas emissions

from US food consumption via carbon calculators. Ecological Economics, 69(1), 186-

196.

Kushi, L. H., Byers, T., Doyle, C., Bandera, E. V., McCullough, M., Gansler, T., . . . Thun, M. J.

(2006). American Cancer Society Guidelines on Nutrition and Physical Activity for

cancer prevention: reducing the risk of cancer with healthy food choices and physical

activity. CA: a cancer journal for clinicians, 56(5), 254-281.

http://www.iaso.org/resources/obesity-data-portal/resources/charts/

http://www.iaso.org/resources/world-map-obesity/

208

Le Treut, H., Somerville, R., Cubasch, U., Ding, Y., Mauritzen, C., Mokssit, A., . . . Prather, M.

(2007). Historical Overview of Climate Change Science. Climate Change 2007: The

Physical Science Basis. Contribution of Working Group I to the Fourth Assessment

Report of the Intergovernmental Panel on Climate Change.

Lee Rodgers, J., & Nicewander, W. A. (1988). Thirteen ways to look at the correlation

coefficient. The American Statistician, 42(1), 59-66.

Legras, B. (2013). Climate Change Sceptics. European Review, 21(S1), S85-S93.

Leviston, Z., Leitch, A., Greenhill, M., Leonard, R., & Walker, I. (2011). Australians’ views of

climate change. Canberra: CSIRO Report. Retrieved January, 15, 2013.

Leviston, Z., & Walker, I. (2011). Baseline Survey of Australian attitudes to climate change:

PRELIMINARY REPORT: CSIRO.

Mansfield, A. K., Addis, M. E., & Mahalik, J. R. (2003). " Why Won't He Go to the Doctor?":

The Psychology of Men's Help Seeking. International Journal of Men's Health, 2(2),

93-109.

Miller, K. (2005). Communications theories: perspectives, processes, and contexts: McGraw-

Hill.

Needham, M., Vaske, J., & Vaske, J. (2008). Survey implementation, sampling, and weighting

data. Survey Research and Analysis: Applications in Parks, Recreation, and Human

Dimensions’.(Ed. JJ Vaske)(Venture Publishing: State College, PA).

Norris, M., & Lecavalier, L. (2010). Evaluating the use of exploratory factor analysis in

developmental disability psychological research. Journal of autism and

developmental disorders, 40(1), 8-20.

Ogden, C. L., Carroll, M. D., Kit, B. K., & Flegal, K. M. (2012). Prevalence of obesity in the

United States, 2009-2010: US Department of Health and Human Services, Centers for

Disease Control and Prevention, National Center for Health Statistics.

Owen, L., Edgar, L., Prince, S., & Doble, C. (2008). Personal carbon trading: public

acceptability: a report to the Department for Environment, Food and Rural Affairs.

Opinion Leader & Enviros Consulting, London.

209

Parag, Y., & Eyre, N. (2010). Barriers to personal carbon trading in the policy arena. Climate

Policy, 10(4), 353-368.

Pew Research Center. (2009). Fewer Americans see solid evidence of global warming:

Modest support for cap and trade policy. Retrieved 2013, 18 December, from

http://pewresearch.org/pubs/1386/cap-and-trade-global-warming-opinion

Pi-Sunyer, X. (2003). A clinical view of the obesity problem. Science, 299(5608), 859-860.

Poortinga, W., Spence, A., Whitmarsh, L., Capstick, S., & Pidgeon, N. F. (2011). Uncertain

climate: An investigation into public scepticism about anthropogenic climate change.

Global Environmental Change, 21(3), 1015-1024.

Prentice, A. M. (2006). The emerging epidemic of obesity in developing countries.

International Journal of Epidemiology, 35(1), 93-99.

Reijnders, L., & Soret, S. (2003). Quantification of the environmental impact of different

dietary protein choices. The American Journal of Clinical Nutrition, 78(3), 664S-668S.

Roberts, S., & Thumim, J. (2006). A rough guide to individual carbon trading. The ideas, the

issues and the next steps, Centre for Sustainable Energy, UK.

Satzinger, J., Jackson, R., & Burd, S. (2012). Systems Analysis and Design in a Changing World.

Segars, A. H., & Grover, V. (1993). Re-examining perceived ease of use and usefulness. MIS

quarterly, 17(4), 517-525.

Smart, W. (2009). Information system success: evaluation of a carbon accounting and

sequestration system. Southern Cross University.

Smith, J. A., Braunack-Mayer, A., & Wittert, G. (2006). What do we know about men's help-

seeking and health service use? Medical Journal of Australia, 184(2), 81.

Smith, P., Martino, D., Cai, Z., Gwary, D., Janzen, H., Kumar, P., . . . Sirotenko, O. (2007).

Agriculture. Climate Change 2007: Mitigation. Contribution of Working Group III to

the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.

Solomon, S., Plattner, G. K., Knutti, R., & Friedlingstein, P. (2009). Irreversible climate change

due to carbon dioxide emissions. Proceedings of the national academy of sciences,

106(6), 1704-1709.

http://pewresearch.org/pubs/1386/cap-and-trade-global-warming-opinion

210

Spencer, E., Appleby, P., Davey, G., & Key, T. (2003). Diet and body mass index in 38 000

EPIC-Oxford meat-eaters, fish-eaters, vegetarians and vegans. International Journal

of Obesity, 27(6), 728-734.

SPSS Inc. (2009). SPSS Statistics Base 17.0 User’s Guide.

Stewart, D. W. (1981). The Application and Misapplication of Factor Analysis in Marketing

Research. Journal of Marketing Research, 18(1), 51-62.

Stuckler, D., McKee, M., Ebrahim, S., & Basu, S. (2012). Manufacturing epidemics: the role of

global producers in increased consumption of unhealthy commodities including

processed foods, alcohol, and tobacco. PLoS Medicine, 9(6), e1001235.

Subramanian, G. H. (1994). A Replication of Perceived Usefulness and Perceived Ease of Use

Measurement*. Decision Sciences, 25(5‐6), 863-874.

Tabenkin, H., Goodwin, M. A., Zyzanski, S. J., Stange, K. C., & Medalie, J. H. (2004). Gender

differences in time spent during direct observation of doctor-patient encounters.

Journal of Women's Health, 13(3), 341-349.

Taylor, B., & McNiel, P. (2011). Norfolk Island 2011 Census of Population and Housing.

Tobias, S., & Carlson, J. (1969). Brief Report: Bartlett's Test of Sphericity and Chance Findings

in Factor Analysis. Multivariate Behaviourial Research, 4(3), 3.

United States Environmental Protection Agency. (2010). Climate Change Indicators in the

United States.

Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the technology acceptance

model: Four longitudinal field studies. Management Science, 46(2), 186-204.

Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of

information technology: Toward a unified view. MIS quarterly, 425-478.

Virtanen, Y., Kurppa, S., Saarinen, M., Katajajuuri, J.-M., Usva, K., Mäenpää, I., . . . Nissinen,

A. (2011). Carbon footprint of food–approaches from national input–output statistics

and a LCA of a food portion. Journal of Cleaner Production, 19(16), 1849-1856.

Wallace, A. A. (2009). Reducing carbon emissions by households: the effects of footprinting

and personal allowances. Institute of Energy and Sustainable Development, De

Montfort University, Leicester.

211

Wallén, A., Brandt, N., & Wennersten, R. (2004). Does the Swedish consumer's choice of

food influence greenhouse gas emissions? Environmental Science & Policy, 7(6), 525-

535.

Wang, Y., & Beydoun, M. A. (2009). Meat consumption is associated with obesity and central

obesity among US adults. International Journal of Obesity, 33(6), 621-628.

Welsh, J. A., Sharma, A., Abramson, J. L., Vaccarino, V., Gillespie, C., & Vos, M. B. (2010).

Caloric sweetener consumption and dyslipidemia among US adults. JAMA: the journal

of the American Medical Association, 303(15), 1490-1497.

Whitmarsh, L. (2011). Scepticism and uncertainty about climate change: dimensions,

determinants and change over time. Global Environmental Change, 21(2), 690-700.

Woodcock, J., Edwards, P., Tonne, C., Armstrong, B. G., Ashiru, O., Banister, D., . . . Cohen, A.

(2009). Public health benefits of strategies to reduce greenhouse-gas emissions:

urban land transport. The Lancet, 374(9705), 1930-1943.

World Health Organization. (2000). Obesity: preventing and managing the global epidemic.

World Health Organization technical report series(894).

World Health Organization. (2008). The WHO STEPwise approach to chronic disease risk

factor surveillance (STEPS).

212

Appendix A - NICHE Survey

213

NICHE Household Questionnaire

The questionnaire is divided into 5 sections. Please answer the questions by circling the appropriate number in

the box or ticking in the space provided.

A. General Information

These first questions are designed to find out a bit about you and your feelings

A1. What is you gender? Male 1 Female 2

A2. What was your year of birth? 19 ___

A3. What is the highest level of education you have

completed?

Less than primary 1

Primary school 2

Secondary School 3

High school 4

College/University 5

A4. In the past 12 months did you catch or produce

any of the following for your own consumption?

(circle each applicable number)

If so, roughly what percentage of your annual

household consumption of this food did this

make up?

(write in the percentage)

Yes %

Fruit 1 _____%

Vegetables 2 _____%

Eggs 3 _____%

Meat 4 _____%

Fish/seafood 5 _____%

A5. Which of the following items do you use at your primary Water tank 1

214

Residence?

(circle as many numbers as apply).

Solar hot water 2

Gas hot water 3

Gas cooking 4

Gas heater 5

Wood fireplace 6 7

A person’s ‘carbon footprint’ is the amount of greenhouse gas emissions, expressed as carbon dioxide

equivalents (CO2-e) involved with a person’s activities over a year.

A6. Compared to others on Norfolk Island, do you think your carbon

footprint is/would be….


Below average 2

About average 3

Above average 4


A7. Which of these activities do you think contributes the most

to your carbon footprint? (rank from 1 to 6)

Car Usage

Electricity

Heating/cooling

food /drink

Waste disposal

Plane flights

A8. Which of these activities do you think contributes the most

to the world’s carbon footprint?

Car Usage

Electricity

Heating/cooling

food /drink

Waste disposal

215

Plane flights

And now some questions about your health

A9. Do you generally consider your health to be……… Poor 1

Fair 2

Good 3

Very good 4

Excellent 5

A10. How would you best describe yourself? Very underweight 1

A bit underweight 2

Ideal weight 3

A bit overweight 4

Very overweight 5

A11. What is your height and weight Height ______m

_____cm

Weight ______kg

A12. Compared to others on the island of similar age and gender

do you consider your body weight to be…..


Below average 2

About average 3

Above average 4


A13. What do you think would be your most healthy body

weight?

____kg

216

B. Attitudes

Your views about health and the environment are important. Please state to what extent you agree or

disagree with the following statements (circle a number)

Strongly Neutral Strongly

Agree Disagree

B1. I buy environmentally friendly products as much as I

can.

1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7

B3. Being overweight can have serious health effects 1_____2____3_____4____5_____6____7

B4. Obesity will be solved in the future by medical

advances

1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7

B6. A financial incentive would encourage me to reduce my

environmental impact

1_____2____3_____4____5_____6____7

B7. Collectively, households can reduce the impacts of

greenhouse gas emissions

1_____2____3_____4____5_____6____7

B8. I always try to eat healthy food 1_____2____3_____4____5_____6____7

B9. I am confident I could maintain a healthy body weight if

I wanted to

1_____2____3_____4____5_____6____7

B10. I would consider purchasing an electric car or bike if

the price was right

1_____2____3_____4____5_____6____7

B11. Walking or cycling instead of using the car can help

reduce a person’s weight

1_____2____3_____4____5_____6____7


217

1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7

….. and some questions about your behaviours Never Sometimes Always


1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7


1_____2____3_____4____5_____6____7

B20. I use the toy library for my kids/grandkids

(leave blank if not applicable)

1_____2____3_____4____5_____6____7

C. Physical activity

These questions give us an idea of your activity levels

C1. How often do you engage in leisure time physical activity Daily 1

218

for the sole purpose of improving or maintaining your health? 3-5 times/week 2

1-3 times/week 3

< once a week 4

Never 5

C2. Do you ever walk or use a bicycle for at least 10 minutes

continuously to get to and from places?

(If no go to question C5)

Yes walk 1

Yes cycle 2

No 3

C3. In a typical week, how many days do you walk or bicycle for at least

10 minutes to get to and from places?

Days a week _____

C4. How much time (in hours &/or minutes) would you spend walking

or cycling to and from places on a typical day?

_____ hrs ______mins

C5. Are you employed outside your home?

(if no go to question C8 )

Yes 1

No – work at home 2

No – don’t work 3

C6. Approximately how far is it to your usual place of work? ___________km or

__________

metres

varies 3

C7. How do you usually get to and from work? Drive car 1

Motorbike 2

Car pool 3

Cycle 4

Walk 5

219

C8. Do you have school age children?

(If no, go to question C13)

Yes 1

C9. If you have children in infants school how do they normally get to

and from school? (most common method)

(If no children of this age, leave blank)

Driven by car 1

Walk 2

Cycle 3

Car pool 4

C10. If you have children in primary school to year 8 how do they

normally get to and from school? (most common method)


Driven by car 1

Walk 2

Cycle 3

Car pool 4

C11. If you have children in years 9 to 12 how do they normally get to

and from school? (most common method)


Driven by car 1

Walk 2

Cycle 3

Motor-bike 4

Drive themselves 5

Car pool 6

C12. Approximately how many km (or metres) is it to school? ________ km or

_______ Metres

C13. If you had to travel from Foodlands to the ATM, how

would you usually do it?

Car 1

Walk 2

Cycle 3

Get a lift 4

220

C14. If you had to travel from Emily Bay to Kingston Pier, how

would you usually do it?

Car 1

Walk 2

Cycle 3

Get a lift 4

C15. Before the recent petrol shortage, did you ever car pool if going to

a meeting or event with others?

Never 1

Rarely 2

Sometimes 3

Often 4

D. Nutrition

… and now some questions about what you eat and drink

Thinking about your eating habits

on most days, how often would you consume each of

the following?

Rarely

or

Never

1-3

times a

month

1-3

times

a

week

4-6

times

a

week

Usually

Daily

D1. Pastries, croissants, muffins, doughnuts, cake or

sweet biscuits

D2. Imported soft drink, or fruit juice

D3. Local soft drink or fruit juice

D4. Full fat milk or cream (more than 1/2 cup)

D5. Low fat milk or cream (more than 1/2 cup)

D6. Ice cream

221

D7. Chocolate, chocolate biscuits or sweet snacks

D8. Canned fruit/dried fruit (2 or more pieces)

D9. Chips, crisps or corn chips

D10. Rice crackers, nuts, popcorn, seeds

D11. Yoghurt, sorbet, frozen fruit

D12. Canned or frozen vegetables

D13. Fresh local fruit and vegetables/salad packs

D14. Imported meat

D15. Local meat

D16. High sugared breakfast cereals (eg. Nutri-grain;

Coco-pops etc)

D17. Porridge/Muesli/Weetbix

D18. White bread

D19. Grain/wholemeal bread

D20. Packaged meals/instant noodles

D21. Home made noodle dishes

D22. Oils (safflower; sunflower)

D23. Oils (olive/canola/peanut)

D24. Local fish/crustaceans (eg. hihi)

D25. Imported fish

D26. Tahitian fish

D27. Pilahai (banana bake)

222

D28. Poi (banana/pumpkin and arrowroot)

D29. Mudda

D30. Green plun fritters

D31. Coconut bread/coconut pie

E. Personal carbon trading

Circle the number on the following statements that best represents your view

Strongly Neutral Strongly

Agree Disagree

E1. Being able to measure my carbon footprint is

important to me

1_____2____3_____4____5_____6____7

E2. Most people would accept a PCT system as a tool

for improving the environment

1_____2____3_____4____5_____6____7

E3. A PCT system would encourage me to reduce my

carbon footprint

1_____2____3_____4____5_____6____7

E4. A PCT system would encourage me to walk or

cycle more and drive less

1_____2____3_____4____5_____6____7

E5. People who reduce their carbon footprint should

be rewarded in some way

1_____2____3_____4____5_____6____7

E6. People with a greater carbon footprint should have to

pay for it in some way

1_____2____3_____4____5_____6____7

E7. A PCT system would encourage me to eat more

1_____2____3_____4____5_____6____7

223

healthy, locally grown produce

E8. A PCT system would be useful for me to help

monitor my environmental impact

1_____2____3_____4____5_____6____7

E9. Comparing my carbon usage to the average

would influence my consumption habits

1_____2____3_____4____5_____6____7

E10. There is a strong link between a person’s

carbon footprint and their health

1_____2____3_____4____5_____6____7

F. Demographic data

And finally, some questions about yourself

F1. Do you rent your primary residence? Yes ____ 1

No _____ 2

F2. Which best describes you current residential status on

Norfolk Island?

TEP ____ 1

GEP _____2

Resident ______3

Temporary visitor ____4

F3. How many people live in your house?

How many Adults (>18yrs)?

How many Children (<18 yrs)?

Total _______

Number ______

Number ______

F4. How many motor vehicles are there in the household (if any)?

Cars/vans Number ______

Motorbikes Number ______

224

Trucks/Utes Number ______

(If zero, go to question F7)

F6. What percentage of those vehicle expenses do you normally claim as a business expense?

_____%

F8. How many months would it take you to use a full gas bottle

in your house? Number ____

F9. How many small (BBQ) gas bottles do you use yearly? Number ____

F10. Roughly, what is your electricity cost per quarter in your

primary residence?

$0 - $300 1

$301- $600 2

$601 - $900 3

$901 - $1200 4

$1200+ 5

F11. Do you own a business that generates income out of this

residence?

Yes 1

F12. If yes, about what proportion of your electricity costs would

be related to business? _______%

F13. Do you have solar hot water in your primary residence? (if no, go

to question F15)

Yes 1

No 2

225

F14. If yes, how long have you had it? 0-6 months 1

6-12 months 2

1 – 2 yrs 3

2 years + 4

F15. Do you have solar panels or wind turbines that generate electricity

at your primary residence?

(if no, skip to question F19)

Yes 1

F16. If yes, how long have you had these? 0-6 months 1

6-12 months 2

1 – 2 yrs 3

2 years + 4

F17. What is your main reason for using solar electricity/wind turbines

(rank from 1, most important to 4, least important)

Govt Rebate ____

Reduce power costs____

Environment ___

Reliability ____

F 18. How many kilowatts of electricity can you generate? (if unsure

leave blank)

Size _______kW

F19. Roughly, what would be the average weekly expenditure for your

entire household on each of the following fuels?

Petrol $ ________

Diesel $ ________

Bio-fuel $________

F20. During the past 12 months how many flights have you

taken off Norfolk Island to the following locations?

(if none answer 0)

Number

Australia _______

NZ _______

226

Other ________

F21. In the last 12 months have you made a purchase of any of

the following major items? And so roughly how much did you spend on

these

Amount

Car $________

Renovations $________

TV/video etc $________

Other__________

$________

No ______

F22. Roughly, what is your total weekly household income from all

sources?

<$500 1

$501- 1000 2

$1001-1500 3

$1501-2000 4

>$2000 5

F23. What best describes your thoughts about climate change? (please circle a number)

I don’t think climate change is happening 1

I have no idea whether climate change is happening or not 2

I think climate change is happening but it’s a natural fluctuation in earth temperatures 3

I think climate change is happening and I think humans are largely causing it 4

The following questions relate to the recent fuel shortage on Norfolk Island.

F24. Did the recent petrol shortage on Norfolk Island result in you doing more of any of the following:

227

Walking Yes / No

Cycling Yes / No

getting lifts with others Yes / No

car pooling Yes / No

not doing some things you might otherwise do Yes / No

other (please write in)____________________________________________________________

F25. Did the recent petrol shortage increase your interest in any of the following: (please circle if appropriate)

electric/solar powered bicycles

electric/solar powered cars

ordinary bicycle

better walking shoes

other (please write in)_______________________________________________________________

F26. Based on your experience of living on Norfolk Island what would the best way of reducing the Island’s

carbon footprint? (write in - optional)

___________________________________________________________________________________________

_________

228

F27. Based on your experience of living on Norfolk Island what would the best way of improving the Island’s

health status? (write in - optional)

___________________________________________________________________________________________

_________

___________________________________________________________________________________________

_________

F28. Are there any other comments you would like to add?

___________________________________________________________________________________________

_________

229

Appendix B - Regression Analysis

230

B.1 Regression Analysis – All Responses

Table B.1.1 Variables Entered/Removeda

Model Variables Entered Variables Removed Method

1 A10, C1, A9, A12b . Enter

2 B11, B8, B3, B12, B9b . Enter

3 B17, B15, B19, B14,

B16, B18b

. Enter

4 B4, B6, B2b . Enter

5 B13, B1, B5, B7b . Enter

a. Dependent Variable: REGR factor score questions E1 to E10-personal carbon

trading system

b. All requested variables entered.

Table B.1.2 Model Summary

Mod

el

R R

Square

Adjusted

R Square

Std. Error of

the Estimate

Change Statistics

R Square

Change


Change

1 .221a .049 .037 .91936 .049 3.993 4 312 .004

2 .337b .113 .088 .89470 .065 4.487 5 307 .001

3 .364c .133 .090 .89370 .019 1.115 6 301 .353

4 .444d .197 .148 .86439 .064 7.920 3 298 .000

5 .580e .337 .287 .79095 .140 15.476 4 294 .000






B13, B1, B5, B7

Table B.1.3 ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression 13.501 4 3.375 3.993 .004

b

Residual 263.711 312 .845

231

Total 277.212 316

2

Regression 31.461 9 3.496 4.367 .000c

Residual 245.751 307 .800

Total 277.212 316

3

Regression 36.805 15 2.454 3.072 .000d

Residual 240.407 301 .799

Total 277.212 316

4

Regression 54.558 18 3.031 4.057 .000e

Residual 222.654 298 .747

Total 277.212 316

5

Regression 93.285 22 4.240 6.778 .000f

Residual 183.927 294 .626

Total 277.212 316

a. Dependent Variable: REGR factor score questions E1 to E10-personal carbon trading system

b. Predictors: (Constant), A10, C1, A9, A12

c. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9

d. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18

e. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18, B4, B6, B2

f. Predictors: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18, B4, B6, B2,

B13, B1, B5, B7

Table B.1.4 Coefficientsa

Model Unstandardized Coefficients Standardized

Coefficients

t Sig.

B Std. Error Beta

1

(Constant) .130 .431 .301 .764

C1 .133 .042 .180 3.148 .002

A12 -.075 .082 -.055 -.906 .365

A9 -.092 .065 -.082 -1.419 .157

A10 .008 .087 .006 .096 .924

2

(Constant) -.444 .444 -1.001 .318

C1 .121 .042 .165 2.918 .004

A12 -.101 .081 -.075 -1.241 .216

A9 -.032 .065 -.029 -.486 .628

A10 -.031 .091 -.022 -.343 .732

B11 .079 .057 .089 1.387 .166

B8 .045 .055 .054 .823 .411

232

B3 .032 .046 .043 .707 .480

B12 -.003 .036 -.006 -.093 .926

B9 .128 .056 .163 2.301 .022

3

(Constant) -.207 .579 -.357 .721

C1 .117 .042 .160 2.785 .006

A12 -.087 .082 -.064 -1.052 .294

A9 -.035 .066 -.031 -.531 .596

A10 -.015 .092 -.011 -.159 .874

B11 .076 .058 .085 1.316 .189

B8 .041 .056 .050 .730 .466

B3 .038 .046 .051 .814 .416

B12 -.008 .036 -.015 -.220 .826

B9 .126 .057 .160 2.223 .027

B17 -.066 .034 -.117 -1.946 .053

B15 .028 .061 .034 .465 .643

B19 -.054 .041 -.084 -1.324 .186

B16 -.011 .048 -.018 -.237 .813

B14 -.010 .046 -.015 -.227 .820

B18 .037 .053 .056 .711 .478

4

(Constant) -.838 .589 -1.421 .156

C1 .135 .041 .184 3.296 .001

A12 -.076 .080 -.056 -.954 .341

A9 -.058 .064 -.052 -.912 .362

A10 -.010 .089 -.007 -.107 .915

B11 .069 .056 .077 1.225 .222

B8 .055 .054 .067 1.013 .312

B3 .023 .045 .031 .510 .610

B12 -.018 .035 -.034 -.521 .603

B9 .079 .056 .100 1.419 .157

B17 -.046 .033 -.080 -1.367 .173

B15 .066 .060 .080 1.106 .270

B19 -.062 .040 -.097 -1.547 .123

B16 -.011 .047 -.018 -.244 .807

B14 -.034 .045 -.050 -.749 .454

B18 .041 .051 .061 .799 .425

B4 -.019 .035 -.032 -.550 .583

B6 .142 .034 .242 4.226 .000

B2 .058 .034 .097 1.688 .092

5 (Constant) -1.189 .555 -2.140 .033

C1 .101 .038 .138 2.676 .008

233

A12 -.087 .073 -.065 -1.196 .233

A9 -.102 .059 -.092 -1.729 .085

A10 .001 .082 .001 .015 .988

B11 -.004 .055 -.004 -.068 .946

B8 -.016 .051 -.019 -.313 .755

B3 .008 .042 .011 .194 .846

B12 -.022 .032 -.042 -.694 .488

B9 .006 .053 .008 .117 .907

B17 -.029 .031 -.052 -.956 .340

B15 .049 .056 .060 .886 .376

B19 -.030 .037 -.047 -.813 .417

B16 -.030 .043 -.047 -.701 .484

B14 -.028 .041 -.042 -.685 .494

B18 .064 .047 .096 1.345 .180

B4 -.008 .033 -.013 -.251 .802

B6 .115 .032 .197 3.564 .000

B2 .033 .031 .055 1.040 .299

B13 .141 .039 .221 3.590 .000

B1 .148 .048 .198 3.100 .002

B5 .065 .054 .083 1.205 .229

B7 .064 .054 .084 1.197 .232


Table B.1.5 Excluded Variablesa

Model Beta In t Sig. Partial Correlation Collinearity

Statistics

Tolerance

1

B11 .181b 3.300 .001 .184 .982

B8 .174b 3.104 .002 .173 .940

B3 .137b 2.461 .014 .138 .968

B12 .100b 1.556 .121 .088 .739

B9 .240b 4.209 .000 .232 .888

B17 -.125b -2.267 .024 -.128 .985

B15 -.026b -.461 .645 -.026 .988

B19 -.127b -2.293 .022 -.129 .983

234

B16 -.018b -.322 .748 -.018 .969

B14 -.072b -1.298 .195 -.073 .998

B18 -.079b -1.410 .160 -.080 .972

B4 .052b .935 .350 .053 .994

B6 .299b 5.633 .000 .304 .986

B2 .119b 2.168 .031 .122 .998

B13 .410b 8.025 .000 .414 .970

B1 .353b 6.763 .000 .358 .980

B5 .388b 7.535 .000 .393 .977

B7 .395b 7.768 .000 .403 .991

2

B17 -.117c -2.169 .031 -.123 .979

B15 -.010c -.174 .862 -.010 .974

B19 -.086c -1.558 .120 -.089 .942

B16 -.023c -.425 .671 -.024 .951

B14 -.033c -.591 .555 -.034 .950

B18 -.028c -.496 .620 -.028 .904

B4 .061c 1.127 .261 .064 .978

B6 .258c 4.785 .000 .264 .925

B2 .112c 2.074 .039 .118 .973

B13 .377c 6.753 .000 .360 .810

B1 .322c 5.613 .000 .306 .796

B5 .341c 5.915 .000 .320 .782

B7 .365c 5.979 .000 .323 .696

3

B4 .057d 1.035 .301 .060 .950

B6 .253d 4.566 .000 .255 .879

B2 .130d 2.356 .019 .135 .934

B13 .379d 6.704 .000 .361 .785

B1 .319d 5.346 .000 .295 .742

B5 .330d 5.626 .000 .309 .761

B7 .355d 5.769 .000 .316 .689

4

B13 .341e 6.063 .000 .332 .760

B1 .332e 5.762 .000 .317 .733

B5 .301e 5.185 .000 .288 .737

B7 .295e 4.710 .000 .264 .640


b. Predictors in the Model: (Constant), A10, C1, A9, A12

c. Predictors in the Model: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9

d. Predictors in the Model: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18

235

e. Predictors in the Model: (Constant), A10, C1, A9, A12, B11, B8, B3, B12, B9, B17, B15, B19, B14, B16, B18,

B4, B6, B2

B.2 Regression Analysis – Males

Table B.2.1 Variables Entered/Removeda,b

Model Variables

Entered

Variables

Removed

Method

1 C1, A12, A9,

A10c

. Enter

2 B3, B8, B11,

B12, B9c

. Enter

3 B17, B15, B19,

B16, B14, B18c

. Enter

4 B2, B6, B4c . Enter

5 B1, B13, B5,

B7c

. Enter

a. Dependent Variable: REGR factor score questions E1

to E10-personal carbon trading system

b. Models are based only on cases for which Gender =

Male

c. All requested variables entered.


Mod

el

R R

Square

Adjusted R

Square

Std. Error

of the

Estimate

Change Statistics

Gender =

Male

(Selected)

R Square

Change

F

Change

df1 df2 Sig. F

Change

1 .196a .038 .008 .98160 .038 1.269 4 127 .286

2 .347b .120 .055 .95800 .082 2.267 5 122 .052

3 .406c .165 .057 .95722 .045 1.033 6 116 .407

4 .512d .262 .144 .91178 .097 4.950 3 113 .003

5 .619e .383 .258 .84877 .121 5.350 4 109 .001




236


e. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9, B17, B15, B19, B16, B14, B18, B2, B6, B4,

B1, B13, B5, B7

Table B.2.3 ANOVAa,b


1

Regression 4.890 4 1.223 1.269 .286c

Residual 122.368 127 .964

Total 127.259 131

2

Regression 15.291 9 1.699 1.851 .066d

Residual 111.968 122 .918

Total 127.259 131

3

Regression 20.971 15 1.398 1.526 .107e

Residual 106.288 116 .916

Total 127.259 131

4

Regression 33.317 18 1.851 2.226 .006f

Residual 93.942 113 .831

Total 127.259 131

5

Regression 48.735 22 2.215 3.075 .000g

Residual 78.524 109 .720

Total 127.259 131


b. Selecting only cases for which Gender = Male

c. Predictors: (Constant), C1, A12, A9, A10

d. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9

e. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9, B17, B15, B19, B16, B14, B18

f. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9, B17, B15, B19, B16, B14, B18, B2,

B6, B4

g. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9, B17, B15, B19, B16, B14, B18, B2,

B6, B4, B1, B13, B5, B7

Table B.2.4 Coefficientsa,b


Coefficients

t Sig.

B Std. Error Beta

1 (Constant) -.069 .653 -.105 .916

237

A9 .043 .103 .037 .413 .680

A10 -.103 .136 -.072 -.755 .452

A12 .041 .121 .032 .337 .737

C1 .142 .066 .193 2.153 .033

2

(Constant) -.854 .688 -1.240 .217

A9 .117 .107 .103 1.099 .274

A10 -.093 .148 -.065 -.628 .531

A12 .063 .123 .049 .513 .609

C1 .147 .067 .200 2.183 .031

B3 .059 .086 .065 .681 .497

B8 -.107 .096 -.117 -1.115 .267

B9 .272 .099 .299 2.733 .007

B11 .069 .092 .075 .744 .458

B12 -.061 .062 -.104 -.981 .329

3

(Constant) -.863 .984 -.877 .382

A9 .144 .108 .127 1.329 .186

A10 -.086 .149 -.060 -.574 .567

A12 .077 .128 .060 .600 .550

C1 .178 .069 .242 2.572 .011

B3 .024 .090 .026 .267 .790

B8 -.114 .097 -.126 -1.175 .242

B9 .292 .102 .320 2.852 .005

B11 .080 .094 .087 .844 .400

B12 -.065 .063 -.110 -1.030 .305

B14 -.076 .071 -.109 -1.076 .284

B15 -.008 .098 -.009 -.086 .932

B16 .032 .075 .046 .426 .671

B17 -.107 .057 -.172 -1.878 .063

B18 .118 .084 .159 1.399 .164

B19 -.017 .068 -.024 -.252 .801

4

(Constant) -1.334 .965 -1.383 .170

A9 .085 .105 .075 .812 .419

A10 -.122 .144 -.086 -.853 .395

A12 .062 .123 .049 .505 .614

C1 .185 .066 .252 2.790 .006

B3 -.014 .086 -.015 -.162 .872

B8 -.052 .095 -.057 -.545 .587

B9 .216 .100 .238 2.164 .033

B11 .067 .091 .073 .734 .464

B12 -.047 .061 -.080 -.778 .438

238

B14 -.127 .069 -.183 -1.840 .068

B15 .045 .095 .045 .469 .640

B16 .005 .075 .007 .062 .951

B17 -.087 .056 -.139 -1.549 .124

B18 .123 .082 .166 1.507 .135

B19 -.034 .066 -.048 -.510 .611

B2 .123 .058 .190 2.110 .037

B4 -.010 .060 -.015 -.168 .867

B6 .175 .058 .269 3.004 .003

5

(Constant) -1.320 .967 -1.366 .175

A9 -.042 .104 -.037 -.404 .687

A10 -.158 .139 -.111 -1.129 .261

A12 .035 .115 .027 .301 .764

C1 .140 .063 .191 2.208 .029

B3 -.039 .082 -.044 -.479 .633

B8 -.054 .090 -.059 -.598 .551

B9 .113 .098 .124 1.147 .254

B11 -.034 .089 -.037 -.376 .708

B12 -.065 .057 -.111 -1.137 .258

B14 -.094 .066 -.136 -1.425 .157

B15 .031 .091 .032 .344 .732

B16 .002 .071 .003 .030 .976

B17 -.087 .052 -.139 -1.661 .100

B18 .152 .079 .205 1.925 .057

B19 -.003 .064 -.005 -.051 .960

B2 .087 .055 .134 1.573 .119

B4 -.013 .058 -.020 -.224 .823

B6 .123 .060 .189 2.043 .044

B1 .174 .084 .216 2.067 .041

B5 .069 .097 .079 .713 .477

B7 .031 .094 .039 .327 .744

B13 .147 .066 .237 2.222 .028


b. Selecting only cases for which Gender = Male

239


Model Beta In t Sig. Partial

Correlation

Collinearity

Statistics

Tolerance

1

B3 .121b 1.360 .176 .120 .955

B8 .056b .613 .541 .055 .908

B9 .259b 2.900 .004 .250 .896

B11 .134b 1.513 .133 .134 .949

B12 -.017b -.165 .870 -.015 .731

B14 -.092b -1.031 .304 -.091 .952

B15 -.001b -.007 .994 -.001 .984

B16 .036b .392 .696 .035 .909

B17 -.154b -1.781 .077 -.157 .992

B18 .028b .305 .761 .027 .938

B19 -.057b -.650 .517 -.058 .983

B2 .181b 2.090 .039 .183 .982

B4 .067b .772 .442 .069 .993

B6 .321b 3.862 .000 .325 .987

B1 .356b 4.283 .000 .356 .962

B5 .389b 4.688 .000 .385 .945

B7 .393b 4.836 .000 .396 .975

B13 .407b 4.841 .000 .396 .910

2

B14 -.077c -.852 .396 -.077 .893

B15 .021c .248 .805 .022 .965

B16 .053c .589 .557 .053 .887

B17 -.142c -1.666 .098 -.150 .982

B18 .080c .878 .382 .080 .879

B19 -.005c -.055 .956 -.005 .892

B2 .160c 1.863 .065 .167 .959

B4 .062c .716 .475 .065 .963

B6 .273c 3.221 .002 .281 .930

B1 .321c 3.427 .001 .297 .755

B5 .339c 3.780 .000 .325 .807

B7 .358c 3.766 .000 .324 .720

B13 .400c 4.238 .000 .360 .713

3 B2 .205d 2.319 .022 .211 .884

240

B4 .090d .993 .323 .092 .883

B6 .281d 3.153 .002 .282 .842

B1 .356d 3.621 .000 .320 .675

B5 .342d 3.705 .000 .327 .761

B7 .369d 3.877 .000 .340 .710

B13 .416d 4.381 .000 .378 .689

4

B1 .330e 3.497 .001 .314 .666

B5 .266e 2.716 .008 .249 .645

B7 .274e 2.722 .008 .249 .611

B13 .350e 3.668 .000 .327 .645


b. Predictors in the Model: (Constant), C1, A12, A9, A10

c. Predictors in the Model: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9

d. Predictors in the Model: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9, B17, B15, B19,

B16, B14, B18

e. Predictors in the Model: (Constant), C1, A12, A9, A10, B3, B8, B11, B12, B9, B17, B15, B19,

B16, B14, B18, B2, B6, B4

B.3 Regression Analysis – Females


Model Variables

Entered

Variables

Removed

Method

1 C1, A12, A9,

A10c

. Enter

2 B11, B8, B3,

B12, B9c

. Enter

3 B16, B17, B19,

B14, B18, B15c

. Enter


5 B13, B1, B7, B5c . Enter

a. Dependent Variable: REGR factor score questions E1 to

E10-personal carbon trading system

b. Models are based only on cases for which Gender =

Female


241


Mod

el

R R

Square

Adjusted R

Square

Std. Error

of the

Estimate

Change Statistics

Gender =

Female

(Selected)

R Square

Change

F

Change

df1 df2 Sig. F

Change

1 .294a .087 .066 .83501 .087 4.122 4 174 .003

2 .406b .165 .120 .81011 .078 3.172 5 169 .009

3 .442c .195 .121 .80984 .030 1.019 6 163 .415

4 .490d .240 .155 .79406 .045 3.181 3 160 .026

5 .597e .357 .266 .74002 .116 7.056 4 156 .000






B13, B1, B7, B5



1

Regression 11.495 4 2.874 4.122 .003c

Residual 121.321 174 .697

Total 132.817 178

2

Regression 21.905 9 2.434 3.709 .000d

Residual 110.912 169 .656

Total 132.817 178

3

Regression 25.914 15 1.728 2.634 .001e

Residual 106.903 163 .656

Total 132.817 178

4

Regression 31.932 18 1.774 2.813 .000f

Residual 100.885 160 .631

Total 132.817 178

5

Regression 47.387 22 2.154 3.933 .000g

Residual 85.430 156 .548

Total 132.817 178


242

b. Selecting only cases for which Gender = Female





B4, B6


B4, B6, B13, B1, B7, B5



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) .227 .571 .398 .691

A9 -.190 .081 -.180 -2.344 .020

A10 .131 .109 .099 1.201 .231

A12 -.172 .112 -.123 -1.531 .128

C1 .108 .053 .152 2.027 .044

2

(Constant) -.220 .597 -.368 .713

A9 -.122 .081 -.116 -1.507 .134

A10 .052 .113 .039 .464 .643

A12 -.208 .114 -.148 -1.826 .070

C1 .096 .052 .135 1.835 .068

B3 .022 .051 .035 .424 .672

B8 .056 .071 .074 .790 .430

B9 .098 .067 .143 1.456 .147

B11 .058 .073 .069 .802 .424

B12 .039 .043 .081 .899 .370

3

(Constant) .108 .758 .143 .886

A9 -.145 .083 -.138 -1.759 .080

A10 .059 .118 .044 .498 .619

A12 -.231 .116 -.165 -1.988 .048

C1 .078 .053 .110 1.476 .142

B3 .034 .053 .054 .640 .523

B8 .050 .073 .066 .679 .498

B9 .107 .070 .156 1.521 .130

B11 .056 .074 .067 .765 .446

B12 .033 .044 .069 .754 .452

243

B14 .101 .066 .156 1.542 .125

B15 -.016 .084 -.023 -.194 .847

B16 -.032 .065 -.054 -.488 .626

B17 -.026 .043 -.050 -.604 .547

B18 .005 .068 .008 .072 .943

B19 -.081 .051 -.139 -1.600 .112

4

(Constant) -.611 .788 -.775 .439

A9 -.149 .081 -.142 -1.836 .068

A10 .081 .116 .061 .699 .485

A12 -.221 .115 -.158 -1.917 .057

C1 .097 .053 .136 1.832 .069

B3 .021 .053 .034 .405 .686

B8 .057 .073 .076 .792 .430

B9 .069 .070 .101 .991 .323

B11 .060 .073 .072 .824 .411

B12 .014 .044 .030 .326 .745

B14 .094 .065 .145 1.449 .149

B15 -.016 .083 -.023 -.189 .850

B16 -.005 .064 -.009 -.079 .937

B17 -.014 .043 -.026 -.324 .746

B18 .006 .067 .009 .084 .933

B19 -.081 .051 -.139 -1.602 .111

B2 .047 .042 .088 1.119 .265

B4 .003 .044 .006 .074 .941

B6 .099 .042 .190 2.354 .020

5

(Constant) -1.347 .759 -1.775 .078

A9 -.149 .076 -.142 -1.958 .052

A10 .125 .109 .094 1.146 .254

A12 -.198 .108 -.142 -1.836 .068

C1 .082 .049 .116 1.659 .099

B3 .004 .050 .007 .081 .935

B8 -.051 .071 -.068 -.722 .471

B9 -.011 .068 -.016 -.159 .874

B11 .042 .072 .050 .579 .563

B12 .027 .041 .056 .645 .520

B14 .071 .062 .109 1.147 .253

B15 .003 .081 .004 .033 .974

B16 -.036 .061 -.063 -.600 .549

244

B17 .008 .040 .015 .198 .843

B18 .010 .063 .017 .163 .871

B19 -.037 .048 -.063 -.769 .443

B2 .028 .040 .051 .697 .487

B4 .021 .041 .039 .513 .609

B6 .088 .040 .169 2.180 .031

B1 .147 .062 .214 2.378 .019

B5 .032 .071 .045 .449 .654

B7 .090 .069 .128 1.309 .193

B13 .116 .055 .180 2.113 .036


b. Selecting only cases for which Gender = Female



Correlation

Collinearity

Statistics

Tolerance

1

B3 .150b 2.049 .042 .154 .961

B8 .222b 2.999 .003 .222 .914

B9 .259b 3.462 .001 .255 .880

B11 .195b 2.727 .007 .203 .989

B12 .201b 2.432 .016 .182 .744

B14 .013b .181 .857 .014 .977

B15 -.034b -.460 .646 -.035 .970

B16 -.036b -.495 .621 -.038 .976

B17 -.079b -1.064 .289 -.081 .948

B18 -.114b -1.564 .120 -.118 .971

B19 -.156b -2.134 .034 -.160 .964

B2 .156b 2.157 .032 .162 .981

B4 .084b 1.149 .252 .087 .977

B6 .283b 4.056 .000 .295 .987

B1 .335b 4.840 .000 .345 .968

B5 .355b 5.222 .000 .369 .984

B7 .388b 5.788 .000 .403 .985

B13 .372b 5.508 .000 .386 .987

245

2

B14 .051c .701 .484 .054 .944

B15 -.023c -.320 .750 -.025 .944

B16 -.057c -.787 .432 -.061 .945

B17 -.076c -1.052 .294 -.081 .934

B18 -.063c -.862 .390 -.066 .919

B19 -.124c -1.721 .087 -.132 .936

B2 .133c 1.851 .066 .141 .937

B4 .104c 1.448 .149 .111 .948

B6 .227c 3.130 .002 .235 .890

B1 .304c 3.967 .000 .293 .772

B5 .291c 3.703 .000 .275 .743

B7 .359c 4.212 .000 .309 .619

B13 .313c 4.167 .000 .306 .801

3

B2 .136d 1.863 .064 .145 .917

B4 .095d 1.293 .198 .101 .912

B6 .215d 2.849 .005 .218 .827

B1 .288d 3.634 .000 .275 .733

B5 .288d 3.570 .000 .270 .709

B7 .340d 3.916 .000 .294 .601

B13 .309d 3.952 .000 .297 .743

4

B1 .316e 4.055 .000 .306 .713

B5 .291e 3.678 .000 .280 .704

B7 .303e 3.459 .001 .265 .580

B13 .278e 3.566 .000 .272 .726





B14, B18, B15


B14, B18, B15, B2, B4, B6

B.4 Regression Analysis – Worse Than Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

246

1 C1, A9, A12,

A10c

. Enter

2 B3, B11, B12,

B9, B8c

. Enter

3 B19, B17, B14,

B16, B18, B15c

. Enter


5 B1, B13, B5, B7c . Enter

a. Dependent Variable: FRS_QE_1

b. Models are based only on cases for which A6 > 3



Mode

l

R R Square Adjusted

R Square

Std. Error of

the Estimate

Change Statistics

A6 > 3 R Square

Change


Change

1 .359a .129 .016 1.20041 .129 1.146 4 31 .354

2 .536b .287 .041 1.18556 .158 1.156 5 26 .357

3 .694c .481 .092 1.15308 .194 1.248 6 20 .325

4 .766d .586 .149 1.11679 .105 1.440 3 17 .266

5 .903e .816 .505 .85190 .230 4.054 4 13 .024






B1, B13, B5, B7



1

Regression 6.604 4 1.651 1.146 .354c

Residual 44.670 31 1.441

Total 51.274 35

2

Regression 14.729 9 1.637 1.164 .357d

Residual 36.545 26 1.406

Total 51.274 35

247

3

Regression 24.682 15 1.645 1.238 .323e

Residual 26.592 20 1.330

Total 51.274 35

4

Regression 30.071 18 1.671 1.339 .276f

Residual 21.203 17 1.247

Total 51.274 35

5

Regression 41.839 22 1.902 2.620 .038g

Residual 9.435 13 .726

Total 51.274 35


b. Selecting only cases for which A6 > 3





B4, B6


B4, B6, B1, B13, B5, B7



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) -2.342 1.779 -1.317 .198

A9 .096 .291 .061 .330 .743

A10 .388 .447 .195 .868 .392

A12 .125 .388 .068 .323 .749

C1 .205 .178 .208 1.152 .258

2

(Constant) -5.795 2.675 -2.167 .040

A9 .440 .330 .281 1.334 .194

A10 .762 .544 .384 1.400 .173

A12 .147 .394 .080 .374 .712

C1 .067 .206 .068 .324 .748

B3 -.081 .195 -.090 -.415 .682

B8 .256 .232 .309 1.104 .280

B9 .048 .212 .062 .228 .822

B11 .317 .223 .293 1.424 .166

248

B12 -.012 .146 -.017 -.079 .938

3

(Constant) -9.388 3.561 -2.636 .016

A9 .497 .344 .317 1.444 .164

A10 .608 .590 .306 1.029 .316

A12 .588 .465 .318 1.266 .220

C1 -.027 .228 -.028 -.119 .906

B3 -.127 .220 -.140 -.576 .571

B8 .264 .269 .319 .980 .339

B9 -.086 .260 -.111 -.331 .744

B11 .510 .266 .470 1.914 .070

B12 .006 .169 .009 .036 .972

B14 .051 .212 .073 .240 .813

B15 .255 .309 .261 .825 .419

B16 .345 .229 .424 1.509 .147

B17 -.013 .165 -.018 -.076 .940

B18 -.275 .212 -.359 -1.296 .210

B19 .077 .176 .090 .439 .666

4

(Constant) -7.528 3.773 -1.995 .062

A9 .370 .365 .236 1.014 .325

A10 .539 .602 .272 .896 .383

A12 .108 .570 .059 .190 .851

C1 .171 .244 .174 .703 .492

B3 -.086 .217 -.095 -.398 .696

B8 .229 .264 .277 .866 .399

B9 -.107 .264 -.137 -.403 .692

B11 .267 .284 .246 .939 .361

B12 .002 .165 .003 .011 .991

B14 .064 .207 .091 .308 .762

B15 .337 .310 .344 1.086 .293

B16 .047 .274 .058 .171 .866

B17 .038 .179 .055 .213 .833

B18 -.178 .217 -.232 -.822 .423

B19 .158 .177 .185 .892 .385

B2 -.066 .161 -.091 -.408 .689

B4 -.142 .186 -.189 -.765 .455

B6 .362 .175 .559 2.072 .054

5

(Constant) -7.325 3.784 -1.936 .075

A9 .156 .337 .100 .463 .651

A10 .824 .571 .415 1.443 .173

A12 .059 .469 .032 .125 .902

249

C1 .003 .207 .003 .016 .987

B3 -.163 .181 -.180 -.899 .385

B8 .368 .263 .445 1.400 .185

B9 -.253 .213 -.325 -1.185 .257

B11 -.028 .254 -.026 -.109 .915

B12 -.014 .137 -.022 -.105 .918

B14 .017 .166 .025 .105 .918

B15 .699 .319 .715 2.187 .048

B16 .031 .216 .038 .144 .888

B17 .127 .141 .185 .900 .385

B18 -.325 .187 -.423 -1.733 .107

B19 .034 .147 .040 .232 .820

B2 -.125 .133 -.172 -.941 .364

B4 -.081 .163 -.108 -.497 .627

B6 .394 .139 .608 2.834 .014

B1 -.565 .267 -.777 -2.112 .055

B5 -.311 .249 -.358 -1.247 .235

B7 1.319 .358 1.774 3.689 .003

B13 -.349 .234 -.540 -1.491 .160





Correlation

Collinearity

Statistics

Tolerance

1

B3 .078b .436 .666 .079 .911

B8 .360b 1.709 .098 .298 .596

B9 .307b 1.752 .090 .305 .856

B11 .326b 1.894 .068 .327 .878

B12 .105b .576 .569 .105 .858

B14 .032b .189 .852 .034 .979

B15 .205b 1.146 .261 .205 .871

B16 .287b 1.429 .163 .252 .676

B17 -.042b -.248 .806 -.045 .992

250

B18 -.120b -.676 .504 -.123 .908

B19 .200b 1.163 .254 .208 .937

B2 .083b .455 .653 .083 .871

B4 -.122b -.669 .508 -.121 .858

B6 .451b 2.804 .009 .456 .887

B1 .430b 2.780 .009 .453 .967

B5 .441b 2.781 .009 .453 .919

B7 .599b 4.301 .000 .618 .925

B13 .625b 3.758 .001 .566 .714

2

B14 .267c 1.432 .164 .275 .756

B15 .356c 1.949 .063 .363 .742

B16 .418c 1.876 .072 .351 .504

B17 -.012c -.063 .950 -.013 .755

B18 -.027c -.132 .896 -.026 .679

B19 .188c 1.051 .303 .206 .851

B2 -.025c -.127 .900 -.025 .730

B4 -.134c -.723 .476 -.143 .815

B6 .347c 1.725 .097 .326 .629

B1 .394c 1.667 .108 .316 .458

B5 .351c 1.625 .117 .309 .554

B7 .812c 3.425 .002 .565 .346

B13 .622c 2.467 .021 .442 .360

3

B2 .025d .116 .909 .027 .591

B4 .031d .138 .892 .032 .546

B6 .422d 1.900 .073 .400 .464

B1 .124d .384 .706 .088 .261

B5 .091d .307 .762 .070 .313

B7 .741d 2.548 .020 .505 .240

B13 .442d 1.425 .170 .311 .256

4

B1 .085e .266 .794 .066 .251

B5 -.022e -.064 .950 -.016 .229

B7 .771e 2.640 .018 .551 .211

B13 .368e 1.122 .279 .270 .223





B16, B18, B15

251


B16, B18, B15, B2, B4, B6

Analysis – Better Than Average Carbon Footprint


Model Variables Entered Variables

Removed

Method

1 C1, A12, A9, A10c . Enter

2 B11, B3, B8, B12,

B9c

. Enter

3 B17, B16, B19,

B14, B15, B18c

. Enter


5 B7, B13, B1, B5c . Enter


b. Models are based only on cases for which A6 < 3



Mode

l

R R

Square

Adjusted R

Square

Std. Error

of the

Estimate

Change Statistics

A6 < 3 R Square

Change


Change

1 .338a .114 .077 .90227 .114 3.085 4 96 .019

2 .426b .182 .101 .89048 .068 1.512 5 91 .194

3 .557c .310 .188 .84617 .128 2.630 6 85 .022

4 .573d .328 .180 .85019 .018 .732 3 82 .536

5 .693e .480 .334 .76670 .152 5.708 4 78 .000






B7, B13, B1, B5

252



1

Regression 10.047 4 2.512 3.085 .019c

Residual 78.153 96 .814

Total 88.199 100

2

Regression 16.040 9 1.782 2.248 .026d

Residual 72.159 91 .793

Total 88.199 100

3

Regression 27.340 15 1.823 2.546 .004e

Residual 60.860 85 .716

Total 88.199 100

4

Regression 28.928 18 1.607 2.223 .008f

Residual 59.271 82 .723

Total 88.199 100

5

Regression 42.349 22 1.925 3.275 .000g

Residual 45.851 78 .588

Total 88.199 100


b. Selecting only cases for which A6 < 3




f. Predictors: (Constant), C1, A12, A9, A10, B11, B3, B8, B12, B9, B17, B16, B19, B14, B15, B18, B4, B6, B2

g. Predictors: (Constant), C1, A12, A9, A10, B11, B3, B8, B12, B9, B17, B16, B19, B14, B15, B18, B4, B6, B2,

B7, B13, B1, B5



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) -.079 .677 -.117 .907

A9 -.031 .104 -.030 -.301 .764

A10 .028 .156 .018 .178 .859

A12 -.189 .144 -.136 -1.311 .193

C1 .234 .073 .328 3.191 .002

2 (Constant) -.747 .726 -1.029 .306

253

A9 -.038 .107 -.036 -.351 .726

A10 .078 .169 .052 .461 .646

A12 -.133 .147 -.096 -.904 .368

C1 .236 .074 .331 3.207 .002

B3 .110 .076 .144 1.438 .154

B8 .072 .133 .073 .539 .591

B9 .075 .121 .086 .625 .534

B11 .083 .101 .085 .818 .416

B12 -.117 .063 -.213 -1.856 .067

3

(Constant) -1.034 .883 -1.171 .245

A9 .042 .105 .040 .395 .694

A10 .117 .167 .078 .697 .488

A12 -.100 .148 -.072 -.677 .500

C1 .259 .071 .363 3.636 .000

B3 .080 .076 .106 1.061 .292

B8 .143 .131 .146 1.095 .277

B9 .003 .117 .003 .026 .979

B11 .117 .098 .120 1.196 .235

B12 -.114 .060 -.209 -1.901 .061

B14 -.017 .080 -.028 -.210 .834

B15 .143 .103 .200 1.384 .170

B16 -.152 .085 -.260 -1.792 .077

B17 -.160 .057 -.292 -2.817 .006

B18 .223 .096 .364 2.336 .022

B19 -.148 .073 -.232 -2.034 .045

4

(Constant) -1.277 .947 -1.348 .181

A9 .056 .108 .053 .515 .608

A10 .095 .169 .063 .563 .575

A12 -.091 .149 -.066 -.612 .542

C1 .271 .073 .379 3.728 .000

B3 .064 .077 .085 .834 .407

B8 .130 .134 .133 .973 .334

B9 .002 .119 .002 .016 .987

B11 .134 .099 .136 1.345 .182

B12 -.107 .061 -.195 -1.765 .081

B14 -.049 .083 -.081 -.584 .561

B15 .144 .105 .202 1.376 .173

B16 -.137 .087 -.235 -1.583 .117

B17 -.144 .059 -.263 -2.452 .016

B18 .229 .098 .373 2.328 .022

254

B19 -.171 .076 -.268 -2.254 .027

B2 .081 .059 .140 1.388 .169

B4 -.013 .063 -.021 -.213 .832

B6 .024 .056 .043 .434 .666

5

(Constant) -1.562 .918 -1.702 .093

A9 -.002 .101 -.002 -.018 .986

A10 .080 .161 .053 .498 .620

A12 -.069 .135 -.050 -.512 .610

C1 .189 .068 .266 2.771 .007

B3 .038 .071 .050 .541 .590

B8 .014 .124 .014 .110 .912

B9 .012 .114 .013 .102 .919

B11 .052 .096 .053 .548 .586

B12 -.117 .055 -.213 -2.109 .038

B14 -.017 .077 -.028 -.219 .827

B15 .085 .098 .119 .873 .385

B16 -.141 .080 -.241 -1.771 .080

B17 -.139 .054 -.254 -2.582 .012

B18 .225 .092 .368 2.454 .016

B19 -.110 .072 -.172 -1.522 .132

B2 .067 .054 .116 1.249 .215

B4 -.015 .058 -.023 -.256 .799

B6 .045 .055 .080 .820 .414

B1 .235 .096 .309 2.433 .017

B5 .046 .104 .063 .441 .660

B7 -.054 .090 -.067 -.601 .550

B13 .163 .068 .267 2.409 .018




Model Beta In t Sig. Partial Correlation Collinearity

Statistics

Tolerance

1 B3 .152b 1.580 .117 .160 .977

255

B8 .120b 1.207 .230 .123 .923

B9 .130b 1.264 .209 .129 .870

B11 .105b 1.083 .282 .110 .985

B12 -.121b -1.109 .270 -.113 .775

B14 -.065b -.671 .504 -.069 .991

B15 .022b .226 .821 .023 .948

B16 -.054b -.553 .582 -.057 .971

B17 -.228b -2.406 .018 -.240 .980

B18 .022b .224 .823 .023 .971

B19 -.184b -1.860 .066 -.187 .915

B2 .121b 1.259 .211 .128 .991

B4 .021b .217 .829 .022 .978

B6 .094b .958 .341 .098 .953

B1 .384b 4.204 .000 .396 .943

B5 .404b 4.429 .000 .414 .929

B7 .289b 3.135 .002 .306 .992

B13 .342b 3.592 .001 .346 .908

2

B14 -.020c -.199 .843 -.021 .931

B15 .058c .589 .557 .062 .919

B16 -.038c -.387 .700 -.041 .950

B17 -.212c -2.252 .027 -.231 .968

B18 .055c .552 .582 .058 .928

B19 -.145c -1.429 .156 -.149 .862

B2 .117c 1.205 .231 .126 .954

B4 .035c .356 .723 .037 .954

B6 .099c 1.001 .319 .105 .913

B1 .393c 3.795 .000 .371 .733

B5 .412c 4.170 .000 .402 .780

B7 .258c 2.670 .009 .271 .903

B13 .363c 3.648 .000 .359 .800

3

B2 .140d 1.425 .158 .154 .830

B4 .008d .087 .931 .009 .901

B6 .051d .521 .604 .057 .840

B1 .362d 3.479 .001 .355 .662

B5 .341d 3.365 .001 .345 .706

B7 .179d 1.843 .069 .197 .832

B13 .358d 3.762 .000 .380 .778

4 B1 .408e 3.785 .000 .388 .608

256

B5 .328e 3.147 .002 .330 .680

B7 .162e 1.617 .110 .177 .804

B13 .351e 3.596 .001 .371 .752




d. Predictors in the Model: (Constant), C1, A12, A9, A10, B11, B3, B8, B12, B9, B17, B16, B19, B14, B15,

B18

e. Predictors in the Model: (Constant), C1, A12, A9, A10, B11, B3, B8, B12, B9, B17, B16, B19, B14, B15,

B18, B4, B6, B2

B.6 Regression Analysis – About Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

1 C1, A10, A9,

A12c

. Enter

2 B11, B8, B3, B9,

B12c

. Enter

3 B15, B17, B19,

B14, B16, B18c

. Enter


5 B1, B13, B5, B7c . Enter


b. Models are based only on cases for which A6 = 3




R Square

Std. Error of

the

Estimate

Change Statistics

A6 = 3 R Square

Change


Chang

e

1 .240a .058 .035 .85503 .058 2.546 4 166 .041

2 .433b .187 .142 .80644 .129 5.121 5 161 .000

3 .465c .216 .140 .80707 .029 .958 6 155 .456

257

4 .526d .276 .191 .78304 .060 4.219 3 152 .007

5 .577e .332 .233 .76225 .056 3.102 4 148 .017






B1, B13, B5, B7



1

Regression 7.446 4 1.861 2.546 .041c

Residual 121.358 166 .731

Total 128.803 170

2

Regression 24.099 9 2.678 4.117 .000d

Residual 104.704 161 .650

Total 128.803 170

3

Regression 27.843 15 1.856 2.850 .001e

Residual 100.961 155 .651

Total 128.803 170

4

Regression 35.604 18 1.978 3.226 .000f

Residual 93.199 152 .613

Total 128.803 170

5

Regression 42.812 22 1.946 3.349 .000g

Residual 85.991 148 .581

Total 128.803 170


b. Selecting only cases for which A6 = 3





B6, B2


B6, B2, B1, B13, B5, B7

258



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) 1.082 .596 1.815 .071

A9 -.202 .086 -.190 -2.337 .021

A10 -.071 .107 -.058 -.666 .507

A12 -.111 .101 -.093 -1.093 .276

C1 .057 .056 .080 1.022 .308

2

(Constant) .712 .577 1.233 .219

A9 -.120 .084 -.113 -1.428 .155

A10 -.236 .113 -.192 -2.094 .038

A12 -.190 .100 -.159 -1.910 .058

C1 .029 .053 .040 .534 .594

B3 .029 .060 .041 .481 .631

B8 .101 .065 .128 1.551 .123

B9 .164 .069 .214 2.377 .019

B11 .028 .079 .034 .356 .723

B12 .074 .045 .153 1.653 .100

3

(Constant) 1.710 .806 2.122 .035

A9 -.140 .085 -.132 -1.643 .102

A10 -.251 .116 -.204 -2.174 .031

A12 -.205 .101 -.171 -2.029 .044

C1 .022 .055 .031 .403 .687

B3 .024 .062 .034 .380 .705

B8 .101 .067 .128 1.516 .132

B9 .146 .072 .191 2.019 .045

B11 .031 .081 .037 .383 .702

B12 .075 .047 .155 1.613 .109

B14 -.049 .063 -.067 -.789 .432

B15 -.101 .078 -.119 -1.295 .197

B16 .055 .066 .082 .831 .407

B17 -.002 .045 -.003 -.036 .971

B18 .033 .071 .049 .468 .641

B19 -.064 .053 -.106 -1.216 .226

4

(Constant) .947 .817 1.159 .248

A9 -.157 .083 -.148 -1.888 .061

A10 -.211 .113 -.171 -1.871 .063

A12 -.163 .099 -.136 -1.637 .104

259

C1 .033 .054 .046 .612 .542

B3 -.020 .063 -.028 -.315 .753

B8 .098 .065 .123 1.497 .136

B9 .109 .071 .142 1.535 .127

B11 .042 .079 .051 .538 .591

B12 .041 .047 .084 .868 .387

B14 -.079 .062 -.106 -1.258 .210

B15 -.067 .076 -.080 -.881 .380

B16 .055 .065 .083 .852 .395

B17 .011 .044 .019 .239 .812

B18 .016 .070 .024 .233 .816

B19 -.037 .053 -.061 -.704 .482

B2 .019 .047 .033 .402 .688

B4 .023 .046 .042 .508 .612

B6 .151 .048 .255 3.163 .002

5

(Constant) .295 .839 .351 .726

A9 -.172 .082 -.162 -2.093 .038

A10 -.187 .111 -.151 -1.675 .096

A12 -.161 .097 -.135 -1.659 .099

C1 .039 .053 .055 .737 .462

B3 -.023 .061 -.032 -.370 .712

B8 .054 .066 .068 .826 .410

B9 .042 .076 .054 .551 .583

B11 .001 .082 .002 .015 .988

B12 .045 .048 .092 .934 .352

B14 -.070 .062 -.095 -1.141 .256

B15 -.060 .077 -.071 -.778 .438

B16 .046 .064 .070 .723 .471

B17 .014 .044 .026 .326 .745

B18 .018 .068 .026 .258 .797

B19 -.001 .053 -.001 -.010 .992

B2 .018 .046 .031 .386 .700

B4 .029 .045 .051 .632 .529

B6 .117 .053 .198 2.208 .029

B1 .087 .066 .115 1.313 .191

B5 .086 .074 .107 1.150 .252

B7 .040 .079 .053 .499 .618

260

B13 .083 .059 .127 1.407 .162





Correlation

Collinearity

Statistics

Tolerance

1

B3 .157b 2.049 .042 .157 .944

B8 .249b 3.243 .001 .245 .914

B9 .327b 4.211 .000 .311 .854

B11 .231b 3.090 .002 .234 .963

B12 .262b 2.899 .004 .220 .666

B14 -.119b -1.557 .121 -.120 .969

B15 -.130b -1.727 .086 -.133 .986

B16 -.014b -.178 .859 -.014 .961

B17 -.048b -.624 .534 -.049 .964

B18 -.142b -1.862 .064 -.143 .962

B19 -.198b -2.627 .009 -.200 .967

B2 .079b 1.046 .297 .081 .986

B4 .100b 1.328 .186 .103 .987

B6 .368b 5.246 .000 .378 .995

B1 .307b 4.253 .000 .314 .990

B5 .358b 5.031 .000 .365 .980

B7 .366b 5.201 .000 .375 .991

B13 .366b 5.216 .000 .376 .997

2

B14 -.097c -1.313 .191 -.103 .921

B15 -.121c -1.679 .095 -.132 .964

B16 -.009c -.122 .903 -.010 .923

B17 -.003c -.042 .967 -.003 .917

B18 -.025c -.322 .748 -.025 .848

B19 -.117c -1.561 .121 -.122 .883

B2 .058c .788 .432 .062 .935

B4 .120c 1.655 .100 .130 .948

B6 .282c 3.735 .000 .283 .822

B1 .216c 2.856 .005 .220 .843

261

B5 .269c 3.484 .001 .266 .794

B7 .263c 2.902 .004 .224 .589

B13 .258c 3.226 .002 .247 .746

3

B2 .084d 1.108 .270 .089 .879

B4 .112d 1.496 .137 .120 .899

B6 .270d 3.473 .001 .269 .779

B1 .207d 2.550 .012 .201 .739

B5 .251d 3.164 .002 .247 .758

B7 .268d 2.906 .004 .228 .569

B13 .237d 2.854 .005 .224 .703

4

B1 .206e 2.596 .010 .207 .728

B5 .211e 2.563 .011 .204 .680

B7 .170e 1.695 .092 .137 .465

B13 .218e 2.685 .008 .213 .695





B14, B16, B18


B14, B16, B18, B4, B6, B2

B.7 Regression Analysis – Males Worse Than Average Carbon Footprint

Table B.7.1 Variables Entered/Removeda,b,c

Model Variables

Entered

Variables

Removed

Method

1 C1, A12, A9,

A10d

. Enter

2 B9, B12, B11,

B3, B8d

. Enter

3 B19, B17, B15,

B16, B18, B14d

. Enter

4 B2e . Enter

a. A1 = Male

b. Dependent Variable: FRS_QE_1

c. Models are based only on cases for which A6 > 3

d. All requested variables entered.

262

e. Tolerance = .000 limits reached.

Table B.7.2 Model Summarya

Model R R

Square

Adjusted

R Square

Std. Error of

the Estimate

Change Statistics

A6 > 3 R Square

Change


Change

1 .543b .295 .060 1.28041 .295 1.253 4 12 .341

2 .927c .859 .678 .74937 .564 5.607 5 7 .021

3 .998d .996 .931 .34643 .137 5.292 6 1 .321

4 1.000e 1.000 . . .004 . 1 0 .

a. A1 = Male

b. Predictors: (Constant), C1, A12, A9, A10

c. Predictors: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8


e. Predictors: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8, B19, B17, B15, B16, B18, B14, B2

Table B.7.3 ANOVAa,b,c


1

Regression 8.218 4 2.054 1.253 .341d

Residual 19.673 12 1.639

Total 27.891 16

2

Regression 23.960 9 2.662 4.741 .026e

Residual 3.931 7 .562

Total 27.891 16

3

Regression 27.771 15 1.851 15.427 .198f

Residual .120 1 .120

Total 27.891 16

4

Regression 27.891 16 1.743 . .g

Residual .000 0 .

Total 27.891 16

a. A1 = Male


c. Selecting only cases for which A6 > 3

d. Predictors: (Constant), C1, A12, A9, A10

e. Predictors: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8

263

f. Predictors: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8, B19, B17, B15, B16, B18, B14

g. Predictors: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8, B19, B17, B15, B16, B18, B14, B2

Table B.7.4 Coefficientsa,b,c


Coefficients

t Sig.

B Std. Error Beta

1

(Constant) -5.523 3.206 -1.723 .111

A9 .554 .450 .334 1.231 .242

A10 .961 .814 .369 1.180 .261

A12 .492 .607 .224 .810 .434

C1 -.215 .295 -.199 -.727 .481

2

(Constant) -8.724 3.498 -2.494 .041

A9 .817 .388 .492 2.108 .073

A10 1.202 .916 .462 1.313 .231

A12 .406 .500 .184 .810 .444

C1 -.865 .392 -.801 -2.210 .063

B3 -.440 .422 -.262 -1.042 .332

B8 .425 .503 .356 .846 .426

B9 .839 .374 .574 2.242 .060

B11 .347 .326 .259 1.065 .322

B12 .432 .229 .414 1.884 .102

3

(Constant) -28.166 5.897 -4.776 .131

A9 .733 .326 .442 2.250 .266

A10 4.508 1.169 1.732 3.855 .162

A12 -.890 .473 -.405 -1.882 .311

C1 -2.660 .520 -2.461 -5.114 .123

B3 .334 .283 .199 1.183 .447

B8 3.421 .890 2.863 3.845 .162

B9 -.585 .427 -.400 -1.370 .401

B11 2.946 .740 2.199 3.983 .157

B12 -.534 .241 -.510 -2.211 .270

B14 .958 .321 1.065 2.980 .206

B15 -.364 .464 -.170 -.785 .576

B16 .549 .207 .623 2.657 .229

B17 -.014 .144 -.014 -.095 .940

B18 -.044 .201 -.039 -.217 .864

B19 .653 .150 .586 4.353 .144

264

4

(Constant) -31.235 .000 . .

A9 1.394 .000 .840 . .

A10 5.409 .000 2.078 . .

A12 -1.107 .000 -.503 . .

C1 -2.762 .000 -2.556 . .

B3 .339 .000 .202 . .

B8 3.585 .000 3.000 . .

B9 -.200 .000 -.137 . .

B11 2.637 .000 1.967 . .

B12 -.488 .000 -.467 . .

B14 .982 .000 1.092 . .

B15 -.419 .000 -.196 . .

B16 .436 .000 .495 . .

B17 -.043 .000 -.043 . .

B18 -.037 .000 -.033 . .

B19 .674 .000 .605 . .

B2 -.321 .000 -.386 . .

a. A1 = Male


c. Selecting only cases for which A6 > 3

Table B.7.5 Excluded Variablesa,b


Correlation

Collinearity

Statistics

Tolerance

1

B3 .067c .239 .815 .072 .816

B8 .694c 1.483 .166 .408 .244

B9 .507c 2.222 .048 .557 .849

B11 .482c 1.839 .093 .485 .714

B12 .623c 2.644 .023 .623 .706

B14 -.387c -1.399 .189 -.389 .713

B15 .301c 1.031 .325 .297 .687

B16 .188c .634 .539 .188 .706

B17 -.314c -1.041 .320 -.300 .644

B18 -.573c -1.560 .147 -.426 .390

265

B19 .178c .628 .543 .186 .772

B2 .296c .794 .444 .233 .435

B4 -.304c -1.022 .329 -.294 .661

B6 .511c 1.872 .088 .492 .653

B1 1.110c 3.691 .004 .744 .317

B5 .843c 4.869 .000 .826 .678

B7 .881c 4.026 .002 .772 .542

B13 .817c 3.079 .010 .680 .490

2

B14 .103d .255 .807 .103 .143

B15 .062d .239 .819 .097 .351

B16 .183d .598 .572 .237 .237

B17 -.204d -.966 .371 -.367 .456

B18 -.293d -1.212 .271 -.444 .322

B19 .290d 1.700 .140 .570 .545

B2 -.201d -.330 .753 -.133 .062

B4 -.118d -.465 .658 -.187 .352

B6 .075d .229 .826 .093 .215

B1 .190d .397 .705 .160 .100

B5 .540d 2.251 .065 .677 .221

B7 .280d .593 .575 .235 .100

B13 .215d .611 .563 .242 .179

3

B2 -.386e . . -1.000 .029

B4 2.529e . . 1.000 .001

B6 -.912e . . -1.000 .005

B1 -.313e . . -1.000 .044

B5 -.692e . . -1.000 .009

B7 -.680e . . -1.000 .009

B13 -42.564e . . -1.000 2.375E-006

4

B4 .f . . . .000

B6 .f . . . .000

B1 .f . . . .000

B5 .f . . . .000

B7 .f . . . .000

B13 .f . . . .000

a. A1 = Male


c. Predictors in the Model: (Constant), C1, A12, A9, A10

d. Predictors in the Model: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8

266


B16, B18, B14

f. Predictors in the Model: (Constant), C1, A12, A9, A10, B9, B12, B11, B3, B8, B19, B17, B15,

B16, B18, B14, B2

B.8 Regression Analysis – Males Better Than Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

1 C1, A12, A9,

A10d

. Enter

2 B11, B8, B3,

B12, B9d

. Enter

3 B15, B17, B18,

B19, B14, B16d

. Enter

4 B6, B2, B4d . Enter

5 B7, B13, B1, B5d . Enter

a. A1 = Male


c. Models are based only on cases for which A6 < 3




R Square

Std. Error of

the

Estimate

Change Statistics

A6 < 3 R Square

Change


Change

1 .362b .131 .050 .90787 .131 1.620 4 43 .187

2 .571c .326 .166 .85050 .195 2.199 5 38 .074

3 .741d .549 .338 .75816 .223 2.637 6 32 .034

4 .787e .620 .384 .73121 .071 1.801 3 29 .169

5 .825f .681 .400 .72143 .061 1.198 4 25 .336

a. A1 = Male




267

e. Predictors: (Constant), C1, A12, A9, A10, B11, B8, B3, B12, B9, B15, B17, B18, B19, B14, B16, B6, B2, B4

f. Predictors: (Constant), C1, A12, A9, A10, B11, B8, B3, B12, B9, B15, B17, B18, B19, B14, B16, B6, B2, B4,

B7, B13, B1, B5



1

Regression 5.341 4 1.335 1.620 .187d

Residual 35.442 43 .824

Total 40.783 47

2

Regression 13.295 9 1.477 2.042 .061e

Residual 27.487 38 .723

Total 40.783 47

3

Regression 22.389 15 1.493 2.597 .011f

Residual 18.394 32 .575

Total 40.783 47

4

Regression 25.277 18 1.404 2.626 .010g

Residual 15.505 29 .535

Total 40.783 47

5

Regression 27.771 22 1.262 2.425 .017h

Residual 13.012 25 .520

Total 40.783 47

a. A1 = Male


c. Selecting only cases for which A6 < 3





B2, B4

h. Predictors: (Constant), C1, A12, A9, A10, B11, B8, B3, B12, B9, B15, B17, B18, B19, B14, B16, B6,

B2, B4, B7, B13, B1, B5



Coefficients

t Sig.

B Std. Error Beta

268

1

(Constant) -.529 .892 -.593 .556

A9 .030 .147 .031 .206 .838

A10 .053 .204 .040 .258 .797

A12 -.072 .192 -.057 -.373 .711

C1 .244 .102 .365 2.380 .022

2

(Constant) -.499 .912 -.547 .588

A9 -.086 .146 -.087 -.588 .560

A10 .177 .219 .134 .809 .424

A12 .193 .197 .154 .981 .333

C1 .243 .100 .363 2.426 .020

B3 .093 .111 .123 .841 .406

B8 -.127 .179 -.131 -.708 .483

B9 .122 .155 .147 .788 .436

B11 -.154 .121 -.181 -1.267 .213

B12 -.225 .087 -.428 -2.579 .014

3

(Constant) .064 1.153 .055 .956

A9 .042 .139 .042 .300 .766

A10 .151 .211 .114 .715 .480

A12 .252 .196 .201 1.288 .207

C1 .296 .101 .442 2.913 .006

B3 .016 .118 .021 .137 .892

B8 .016 .167 .017 .097 .923

B9 -.079 .159 -.095 -.495 .624

B11 -.028 .115 -.033 -.244 .809

B12 -.244 .079 -.463 -3.081 .004

B14 .018 .102 .031 .172 .865

B15 .009 .137 .013 .067 .947

B16 -.148 .135 -.254 -1.096 .281

B17 -.230 .084 -.395 -2.748 .010

B18 .299 .127 .510 2.356 .025

B19 -.232 .112 -.362 -2.068 .047

4

(Constant) -.676 1.164 -.581 .566

A9 .023 .140 .023 .164 .871

A10 .132 .204 .100 .645 .524

A12 .312 .192 .249 1.627 .115

C1 .335 .100 .502 3.363 .002

B3 .090 .123 .119 .728 .473

B8 .069 .164 .072 .423 .675

B9 -.170 .162 -.205 -1.050 .302

269

B11 -.091 .119 -.107 -.764 .451

B12 -.228 .077 -.433 -2.951 .006

B14 .094 .110 .167 .858 .398

B15 -.046 .145 -.063 -.318 .753

B16 -.149 .131 -.257 -1.138 .264

B17 -.314 .092 -.539 -3.398 .002

B18 .235 .127 .401 1.844 .075

B19 -.199 .113 -.311 -1.760 .089

B2 -.011 .081 -.018 -.131 .897

B4 .213 .092 .333 2.305 .029

B6 .002 .079 .004 .030 .976

5

(Constant) -1.556 1.444 -1.078 .291

A9 .055 .148 .055 .369 .715

A10 .065 .216 .049 .302 .765

A12 .276 .196 .220 1.407 .172

C1 .309 .101 .463 3.047 .005

B3 .050 .128 .067 .393 .698

B8 .085 .165 .089 .518 .609

B9 -.274 .180 -.331 -1.525 .140

B11 -.139 .130 -.164 -1.072 .294

B12 -.203 .079 -.386 -2.560 .017

B14 .127 .117 .226 1.090 .286

B15 -.115 .150 -.157 -.765 .452

B16 -.093 .139 -.161 -.674 .506

B17 -.312 .093 -.536 -3.365 .002

B18 .198 .137 .338 1.446 .161

B19 -.145 .124 -.226 -1.164 .255

B2 .012 .084 .021 .146 .885

B4 .226 .100 .353 2.262 .033

B6 .020 .085 .033 .231 .820

B1 .238 .150 .283 1.591 .124

B5 .139 .155 .168 .897 .378

B7 -.061 .138 -.069 -.442 .662

B13 .062 .097 .106 .645 .525

a. A1 = Male


c. Selecting only cases for which A6 < 3

270



Correlation

Collinearity

Statistics

Tolerance

1

B3 .003c .019 .985 .003 .921

B8 -.066c -.441 .662 -.068 .911

B9 .070c .465 .645 .072 .895

B11 -.231c -1.590 .119 -.238 .924

B12 -.455c -2.928 .005 -.412 .713

B14 -.041c -.269 .789 -.041 .895

B15 -.014c -.096 .924 -.015 .950

B16 -.062c -.407 .686 -.063 .895

B17 -.412c -2.930 .005 -.412 .870

B18 .072c .470 .641 .072 .887

B19 -.242c -1.675 .101 -.250 .926

B2 .122c .841 .405 .129 .963

B4 .148c 1.016 .316 .155 .953

B6 .125c .842 .404 .129 .926

B1 .251c 1.757 .086 .262 .941

B5 .170c 1.158 .254 .176 .929

B7 .078c .526 .602 .081 .943

B13 .053c .333 .741 .051 .812

2

B14 .036d .235 .815 .039 .764

B15 .004d .027 .979 .004 .858

B16 -.004d -.029 .977 -.005 .773

B17 -.357d -2.558 .015 -.388 .796

B18 .118d .803 .427 .131 .834

B19 -.250d -1.545 .131 -.246 .652

B2 .036d .250 .804 .041 .894

B4 .157d 1.106 .276 .179 .874

B6 .063d .447 .658 .073 .901

B1 .211d 1.266 .213 .204 .627

B5 .238d 1.609 .116 .256 .781

B7 .074d .508 .614 .083 .856

B13 .145d .930 .359 .151 .728

271

3

B2 .023e .163 .872 .029 .709

B4 .331e 2.398 .023 .396 .644

B6 .035e .252 .803 .045 .773

B1 .211e 1.269 .214 .222 .500

B5 .112e .768 .448 .137 .674

B7 .065e .493 .626 .088 .819

B13 .231e 1.647 .110 .284 .681

4

B1 .290f 1.691 .102 .304 .420

B5 .196f 1.351 .188 .247 .606

B7 .084f .651 .520 .122 .795

B13 .173f 1.206 .238 .222 .627

a. A1 = Male





B19, B14, B16


B19, B14, B16, B6, B2, B4

B9 Regression Analysis – Males About Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

1 C1, A12, A9,

A10d

. Enter

2 B3, B8, B12, B9,

B11d

. Enter

3 B17, B15, B19,

B16, B14, B18d

. Enter

4 B6, B4, B2d . Enter

5 B1, B5, B13, B7d . Enter

a. A1 = Male


c. Models are based only on cases for which A6 = 3


272


Model R R

Square

Adjusted R

Square

Std. Error

of the

Estimate

Change Statistics

A6 = 3 R Square

Change


Change

1 .281b .079 .018 .92421 .079 1.288 4 60 .285

2 .518c .269 .149 .86013 .190 2.855 5 55 .023

3 .602d .362 .167 .85116 .093 1.194 6 49 .325

4 .677e .458 .246 .80967 .096 2.717 3 46 .055

5 .729f .531 .285 .78825 .073 1.633 4 42 .184

a. A1 = Male




e. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14, B18, B6, B4, B2

f. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14, B18, B6, B4, B2,

B1, B5, B13, B7



1

Regression 4.400 4 1.100 1.288 .285d

Residual 51.250 60 .854

Total 55.651 64

2

Regression 14.961 9 1.662 2.247 .032e

Residual 40.690 55 .740

Total 55.651 64

3

Regression 20.151 15 1.343 1.854 .053f

Residual 35.500 49 .724

Total 55.651 64

4

Regression 25.495 18 1.416 2.161 .018g

Residual 30.156 46 .656

Total 55.651 64

5

Regression 29.554 22 1.343 2.162 .016h

Residual 26.097 42 .621

Total 55.651 64

273

a. A1 = Male


c. Selecting only cases for which A6 = 3





B4, B2

h. Predictors: (Constant), C1, A12, A9, A10, B3, B8, B12, B9, B11, B17, B15, B19, B16, B14, B18, B6,

B4, B2, B1, B5, B13, B7



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) 1.630 .969 1.682 .098

A9 -.079 .146 -.071 -.544 .588

A10 -.342 .180 -.268 -1.897 .063

A12 -.009 .155 -.008 -.058 .954

C1 .025 .095 .034 .259 .796

2

(Constant) .305 1.026 .297 .768

A9 .124 .156 .111 .794 .431

A10 -.425 .199 -.333 -2.143 .037

A12 -.076 .158 -.067 -.479 .634

C1 .001 .092 .001 .008 .994

B3 .085 .131 .097 .650 .518

B8 .048 .127 .060 .383 .703

B9 .261 .144 .308 1.805 .077

B11 .079 .152 .093 .520 .605

B12 .043 .092 .079 .467 .642

3

(Constant) 1.627 1.700 .957 .343

A9 .076 .160 .068 .475 .637

A10 -.466 .198 -.364 -2.352 .023

A12 -.148 .167 -.132 -.888 .379

C1 .025 .093 .035 .268 .790

B3 .048 .136 .054 .351 .727

B8 .062 .132 .077 .475 .637

B9 .276 .150 .327 1.841 .072

274

B11 .089 .154 .105 .579 .565

B12 .087 .094 .158 .922 .361

B14 -.155 .101 -.201 -1.531 .132

B15 -.193 .136 -.174 -1.421 .162

B16 .053 .111 .064 .476 .636

B17 .001 .071 .002 .019 .985

B18 .182 .122 .216 1.493 .142

B19 -.022 .085 -.034 -.262 .794

4

(Constant) .670 1.759 .381 .705

A9 .007 .157 .006 .046 .964

A10 -.426 .190 -.333 -2.241 .030

A12 -.127 .161 -.113 -.789 .434

C1 .051 .090 .070 .563 .576

B3 .068 .130 .077 .518 .607

B8 .115 .127 .142 .901 .372

B9 .254 .147 .301 1.727 .091

B11 -.042 .156 -.049 -.268 .790

B12 .082 .090 .150 .913 .366

B14 -.231 .101 -.299 -2.283 .027

B15 -.086 .138 -.078 -.623 .537

B16 .034 .108 .041 .313 .755

B17 -.004 .069 -.006 -.051 .960

B18 .197 .119 .233 1.651 .106

B19 -.030 .085 -.046 -.354 .725

B2 .164 .098 .247 1.678 .100

B4 -.078 .087 -.125 -.894 .376

B6 .220 .085 .321 2.578 .013

5

(Constant) .560 1.860 .301 .765

A9 -.115 .169 -.103 -.679 .501

A10 -.393 .192 -.308 -2.048 .047

A12 -.129 .164 -.115 -.788 .435

C1 .026 .095 .036 .274 .785

B3 .050 .135 .057 .372 .712

B8 .152 .133 .188 1.145 .259

B9 .079 .165 .094 .480 .633

B11 -.121 .163 -.143 -.747 .459

B12 .059 .089 .107 .661 .512

B14 -.199 .106 -.258 -1.886 .066

B15 -.028 .137 -.025 -.202 .841

275

B16 .101 .115 .124 .882 .383

B17 -.010 .072 -.017 -.138 .891

B18 .132 .122 .157 1.088 .283

B19 -.021 .094 -.032 -.221 .826

B2 .120 .098 .180 1.225 .227

B4 -.049 .087 -.079 -.558 .580

B6 .069 .125 .101 .554 .583

B1 -.097 .151 -.123 -.643 .524

B5 .121 .155 .137 .784 .438

B7 .131 .162 .164 .813 .421

B13 .196 .111 .312 1.765 .085

a. A1 = Male


c. Selecting only cases for which A6 = 3



Correlation

Collinearity

Statistics

Tolerance

1

B3 .249c 2.004 .050 .252 .949

B8 .302c 2.290 .026 .286 .827

B9 .444c 3.469 .001 .412 .790

B11 .343c 2.781 .007 .340 .909

B12 .295c 1.947 .056 .246 .640

B14 -.181c -1.398 .167 -.179 .904

B15 -.069c -.545 .588 -.071 .983

B16 .173c 1.357 .180 .174 .935

B17 -.030c -.244 .808 -.032 .995

B18 .081c .635 .528 .082 .951

B19 -.073c -.572 .569 -.074 .963

B2 .100c .791 .432 .102 .958

B4 .044c .352 .726 .046 .985

B6 .340c 2.871 .006 .350 .976

B1 .335c 2.793 .007 .342 .960

B5 .451c 3.980 .000 .460 .956

276

B7 .448c 3.979 .000 .460 .969

B13 .541c 5.172 .000 .559 .981

2

B14 -.171d -1.326 .190 -.178 .791

B15 -.156d -1.300 .199 -.174 .914

B16 .126d 1.018 .313 .137 .869

B17 .047d .394 .695 .053 .943

B18 .189d 1.527 .133 .203 .847

B19 .018d .140 .889 .019 .861

B2 .116d .938 .352 .127 .865

B4 .091d .755 .454 .102 .925

B6 .279d 2.426 .019 .314 .923

B1 .175d 1.261 .213 .169 .687

B5 .355d 3.080 .003 .387 .866

B7 .346d 2.411 .019 .312 .594

B13 .455d 3.384 .001 .418 .618

3

B2 .157e 1.157 .253 .165 .704

B4 .044e .341 .735 .049 .807

B6 .275e 2.287 .027 .313 .828

B1 .188e 1.203 .235 .171 .528

B5 .331e 2.646 .011 .357 .743

B7 .365e 2.513 .015 .341 .557

B13 .430e 3.021 .004 .400 .552

4

B1 .153f 1.017 .315 .150 .523

B5 .221f 1.433 .159 .209 .486

B7 .258f 1.395 .170 .204 .338

B13 .340f 2.335 .024 .329 .507

a. A1 = Male





B16, B14, B18


B16, B14, B18, B6, B4, B2

277

B.10 Regression Analysis – Females Worse Than Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

1 C1, A9, A10,

A12c

. Enter

2 B12, B9, B3,

B11, B8c

. Enter

3 B17, B14, B19,

B15, B18, B16c

. Enter

4 B4d . Enter


b. Models are based only on cases for which A6 > 3


d. Tolerance = .000 limits reached.


Model R R

Square

Adjusted

R

Square

Std. Error

of the

Estimate

Change Statistics

A6 > 3 R

Square

Change

F

Change

df1 df2 Sig. F Change

1 .672a .451 .268 .93412 .451 2.467 4 12 .101

2 .853b .728 .379 .86057 .277 1.428 5 7 .322

3 .999c .999 .980 .15321 .270 36.639 6 1 .126

4 1.000d 1.000 . . .001 . 1 0 .




d. Predictors: (Constant), C1, A9, A10, A12, B12, B9, B3, B11, B8, B17, B14, B19, B15, B18, B16, B4



1 Regression 8.612 4 2.153 2.467 .101

c

Residual 10.471 12 .873

278

Total 19.083 16

2

Regression 13.899 9 1.544 2.085 .172d

Residual 5.184 7 .741

Total 19.083 16

3

Regression 19.059 15 1.271 54.128 .106e

Residual .023 1 .023

Total 19.083 16

4

Regression 19.083 16 1.193 . .f

Residual .000 0 .

Total 19.083 16






f. Predictors: (Constant), C1, A9, A10, A12, B12, B9, B3, B11, B8, B17, B14, B19, B15, B18, B16, B4



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) -.609 1.908 -.320 .755

A9 -.329 .326 -.241 -1.009 .333

A10 .308 .455 .176 .677 .511

A12 -.208 .522 -.106 -.398 .698

C1 .559 .245 .529 2.280 .042

2

(Constant) -3.175 4.161 -.763 .470

A9 -.267 .472 -.196 -.567 .589

A10 .687 .701 .392 .980 .360

A12 .035 .512 .018 .068 .948

C1 .522 .319 .493 1.638 .145

B3 .022 .184 .035 .120 .908

B8 .479 .336 .701 1.424 .197

B9 .046 .267 .084 .174 .867

B11 -.172 .362 -.199 -.475 .649

B12 -.271 .159 -.538 -1.703 .132

3 (Constant) 5.810 1.451 4.005 .156

279

A9 .006 .103 .004 .057 .964

A10 4.225 .315 2.415 13.402 .047

A12 -5.298 .467 -2.696 -11.339 .056

C1 .764 .081 .722 9.409 .067

B3 .270 .055 .434 4.959 .127

B8 .758 .098 1.111 7.738 .082

B9 .224 .054 .405 4.158 .150

B11 .418 .091 .486 4.599 .136

B12 -.495 .042 -.985 -11.792 .054

B14 .927 .083 1.546 11.145 .057

B15 .586 .071 .845 8.195 .077

B16 -2.500 .197 -3.572 -12.712 .050

B17 .931 .073 1.725 12.795 .050

B18 -.044 .069 -.072 -.647 .634

B19 -1.128 .112 -1.693 -10.041 .063

4

(Constant) 6.215 .000 . .

A9 .135 .000 .099 . .

A10 4.311 .000 2.464 . .

A12 -5.487 .000 -2.792 . .

C1 .815 .000 .771 . .

B3 .267 .000 .428 . .

B8 .745 .000 1.091 . .

B9 .235 .000 .423 . .

B11 .475 .000 .552 . .

B12 -.515 .000 -1.025 . .

B14 .928 .000 1.547 . .

B15 .561 .000 .810 . .

B16 -2.556 .000 -3.651 . .

B17 .956 .000 1.771 . .

B18 -.055 .000 -.089 . .

B19 -1.097 .000 -1.646 . .

B4 -.083 .000 -.135 . .



280



Correlation

Collinearity

Statistics

Tolerance

1

B3 -.052b -.201 .845 -.060 .728

B8 .471b 1.812 .097 .479 .567

B9 .291b 1.135 .280 .324 .680

B11 -.094b -.249 .808 -.075 .349

B12 -.168b -.665 .520 -.197 .747

B14 .107b .457 .656 .137 .898

B15 .029b .118 .908 .036 .800

B16 .063b .171 .867 .052 .363

B17 .119b .531 .606 .158 .968

B18 -.233b -.887 .394 -.258 .672

B19 .218b .769 .458 .226 .587

B2 .033b .133 .896 .040 .824

B4 .145b .509 .621 .152 .605

B6 .484b 2.513 .029 .604 .854

B1 .292b 1.337 .208 .374 .899

B5 .318b 1.058 .313 .304 .501

B7 .585b 2.462 .032 .596 .569

B13 .580b 2.347 .039 .578 .545

2

B14 .193c .756 .478 .295 .634

B15 .149c .546 .605 .218 .581

B16 -.010c -.026 .980 -.011 .331

B17 .200c .788 .461 .306 .636

B18 .126c .394 .707 .159 .428

B19 .360c 1.113 .308 .414 .359

B2 .143c .440 .675 .177 .413

B4 .178c .463 .660 .186 .297

B6 .442c 1.588 .163 .544 .412

B1 -.551c -1.356 .224 -.484 .210

B5 .070c .173 .868 .070 .277

B7 .546c .623 .556 .247 .055

B13 .357c 1.048 .335 .393 .329

3

B2 -.588d . . -1.000 .004

B4 -.135d . . -1.000 .068

B6 -2.023d . . -1.000 .000

281

B1 .227d . . 1.000 .024

B5 -.435d . . -1.000 .006

B7 .937d . . 1.000 .001

B13 -.127d . . -1.000 .077

4

B2 .e . . . .000

B6 .e . . . .000

B1 .e . . . .000

B5 .e . . . .000

B7 .e . . . .000

B13 .e . . . .000





B15, B18, B16


B15, B18, B16, B4

B11 Regression Analysis – Females Better Than Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

1 C1, A9, A10,

A12c

. Enter

2 B11, B3, B12,

B8, B9c

. Enter

3 B17, B16, B19,

B14, B18, B15c

. Enter


5 B5, B7, B13, B1c . Enter


b. Models are based only on cases for which A6 < 3


282


Model R R

Square

Adjusted R

Square

Std. Error of

the

Estimate

Change Statistics

A6 < 3 R Square

Change


1 .309a .096 .019 .89211 .096 1.242 4 47 .306

2 .551b .304 .155 .82783 .208 2.517 5 42 .044

3 .616c .379 .120 .84473 .075 .723 6 36 .634

4 .644d .414 .095 .85683 .035 .663 3 33 .580

5 .764e .583 .267 .77121 .169 2.934 4 29 .038





e. Predictors: (Constant), C1, A9, A10, A12, B11, B3, B12, B8, B9, B17, B16, B19, B14, B18, B15, B2, B6, B4, B5, B7,

B13, B1



1

Regression 3.955 4 .989 1.242 .306c

Residual 37.406 47 .796

Total 41.360 51

2

Regression 12.578 9 1.398 2.039 .058d

Residual 28.783 42 .685

Total 41.360 51

3

Regression 15.672 15 1.045 1.464 .171e

Residual 25.689 36 .714

Total 41.360 51

4

Regression 17.133 18 .952 1.296 .252f

Residual 24.227 33 .734

Total 41.360 51

5

Regression 24.112 22 1.096 1.843 .062g

Residual 17.248 29 .595

Total 41.360 51




283




B6, B4


B6, B4, B5, B7, B13, B1



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) .045 1.073 .042 .967

A9 -.051 .149 -.048 -.342 .734

A10 .089 .247 .054 .361 .720

A12 -.312 .222 -.213 -1.406 .166

C1 .199 .107 .275 1.868 .068

2

(Constant) -1.466 1.128 -1.299 .201

A9 -.081 .152 -.076 -.530 .599

A10 .181 .280 .109 .648 .521

A12 -.212 .216 -.145 -.978 .334

C1 .257 .109 .355 2.358 .023

B3 .088 .104 .122 .844 .403

B8 .110 .198 .112 .555 .582

B9 -.067 .195 -.075 -.345 .732

B11 .465 .193 .409 2.413 .020

B12 -.063 .091 -.112 -.690 .494

3

(Constant) -2.898 1.666 -1.740 .090

A9 .043 .178 .041 .243 .809

A10 .327 .373 .197 .875 .387

A12 -.239 .234 -.164 -1.021 .314

C1 .290 .118 .400 2.470 .018

B3 .099 .117 .138 .846 .403

B8 .214 .217 .219 .987 .330

B9 -.092 .213 -.103 -.433 .668

B11 .416 .213 .367 1.956 .058

B12 -.066 .094 -.118 -.705 .486

B14 -.091 .151 -.152 -.606 .548

B15 .232 .183 .350 1.266 .214

284

B16 -.171 .117 -.310 -1.459 .153

B17 -.061 .089 -.119 -.688 .496

B18 .186 .152 .306 1.227 .228

B19 -.051 .112 -.082 -.451 .655

4

(Constant) -2.998 1.806 -1.660 .106

A9 .083 .185 .078 .448 .657

A10 .218 .389 .131 .559 .580

A12 -.201 .248 -.137 -.809 .424

C1 .309 .122 .426 2.524 .017

B3 .113 .121 .157 .937 .355

B8 .132 .244 .135 .539 .594

B9 -.037 .235 -.041 -.155 .877

B11 .371 .224 .327 1.657 .107

B12 -.060 .098 -.107 -.613 .544

B14 -.129 .157 -.216 -.825 .415

B15 .227 .188 .342 1.204 .237

B16 -.134 .128 -.242 -1.047 .303

B17 -.043 .097 -.083 -.440 .663

B18 .198 .157 .326 1.263 .215

B19 -.109 .132 -.176 -.825 .415

B2 .118 .093 .215 1.272 .212

B4 -.033 .114 -.053 -.288 .775

B6 .042 .081 .083 .515 .610

5

(Constant) -2.702 1.675 -1.613 .118

A9 -.031 .174 -.029 -.177 .861

A10 .158 .380 .095 .414 .682

A12 -.116 .228 -.079 -.510 .614

C1 .210 .119 .290 1.767 .088

B3 .088 .110 .122 .802 .429

B8 -.119 .236 -.122 -.505 .617

B9 .100 .226 .112 .442 .661

B11 .289 .210 .254 1.374 .180

B12 -.064 .089 -.114 -.717 .479

B14 -.086 .147 -.143 -.582 .565

B15 .223 .177 .336 1.261 .217

B16 -.153 .124 -.277 -1.230 .228

B17 -.047 .090 -.092 -.525 .604

B18 .101 .158 .166 .642 .526

285

B19 -.029 .124 -.047 -.234 .817

B2 .070 .086 .128 .811 .424

B4 -.027 .104 -.045 -.265 .793

B6 .038 .081 .075 .464 .646

B1 .194 .171 .283 1.133 .267

B5 .023 .184 .036 .123 .903

B7 -.005 .142 -.007 -.034 .973

B13 .193 .133 .300 1.450 .158





Correlation

Collinearity

Statistics

Tolerance

1

B3 .263b 1.846 .071 .263 .903

B8 .264b 1.809 .077 .258 .859

B9 .193b 1.202 .236 .174 .739

B11 .445b 3.398 .001 .448 .914

B12 .022b .139 .890 .020 .782

B14 -.010b -.070 .945 -.010 .897

B15 .094b .625 .535 .092 .859

B16 -.020b -.140 .889 -.021 .994

B17 -.074b -.510 .613 -.075 .928

B18 .070b .480 .633 .071 .916

B19 -.068b -.446 .658 -.066 .842

B2 .214b 1.547 .129 .222 .975

B4 -.038b -.262 .794 -.039 .948

B6 .091b .639 .526 .094 .963

B1 .503b 3.912 .000 .500 .893

B5 .568b 4.712 .000 .571 .911

B7 .484b 3.913 .000 .500 .964

B13 .532b 4.121 .000 .519 .863

2 B14 .059

c .417 .679 .065 .837

B15 .159c 1.099 .278 .169 .790

286

B16 -.012c -.083 .934 -.013 .856

B17 .001c .008 .993 .001 .877

B18 .137c .980 .333 .151 .845

B19 .001c .005 .996 .001 .690

B2 .176c 1.278 .208 .196 .865

B4 .101c .699 .488 .109 .807

B6 .132c .955 .345 .147 .872

B1 .446c 3.118 .003 .438 .671

B5 .476c 3.558 .001 .486 .725

B7 .348c 2.524 .016 .367 .772

B13 .472c 3.510 .001 .481 .723

3

B2 .203d 1.332 .191 .220 .725

B4 .054d .320 .751 .054 .625

B6 .091d .579 .567 .097 .712

B1 .437d 2.896 .006 .440 .630

B5 .446d 3.034 .005 .456 .649

B7 .314d 1.944 .060 .312 .614

B13 .485d 3.392 .002 .497 .654

4

B1 .459e 2.968 .006 .465 .599

B5 .432e 2.848 .008 .450 .635

B7 .295e 1.733 .093 .293 .577

B13 .464e 2.979 .005 .466 .591





B14, B18, B15


B14, B18, B15, B2, B6, B4

B12 Regression Analysis – Females About Average Carbon Footprint


Model Variables

Entered

Variables

Removed

Method

1 C1, A12, A9,

A10c

. Enter

287

2 B11, B8, B3, B9,

B12c

. Enter

3 B14, B17, B19,

B16, B18, B15c

. Enter


5 B1, B13, B5, B7c . Enter


b. Models are based only on cases for which A6 = 3



Model R R

Square

Adjusted

R Square

Std. Error

of the

Estimate

Change Statistics

A6 = 3 R Square

Change


1 .341a .116 .080 .77512 .116 3.246 4 99 .015

2 .473b .224 .150 .74533 .108 2.614 5 94 .029

3 .527c .278 .155 .74307 .054 1.096 6 88 .371

4 .582d .339 .198 .72357 .061 2.602 3 85 .057

5 .613e .376 .207 .71978 .038 1.224 4 81 .307





e. Predictors: (Constant), C1, A12, A9, A10, B11, B8, B3, B9, B12, B14, B17, B19, B16, B18, B15, B4, B2, B6, B1,

B13, B5, B7



1

Regression 7.801 4 1.950 3.246 .015c

Residual 59.480 99 .601

Total 67.280 103

2

Regression 15.062 9 1.674 3.013 .003d

Residual 52.219 94 .556

Total 67.280 103

3 Regression 18.691 15 1.246 2.257 .010

e

Residual 48.589 88 .552

288

Total 67.280 103

4

Regression 22.779 18 1.265 2.417 .004f

Residual 44.502 85 .524

Total 67.280 103

5

Regression 25.315 22 1.151 2.221 .005g

Residual 41.965 81 .518

Total 67.280 103







B2, B6


B2, B6, B1, B13, B5, B7



Coefficients

t Sig.

B Std. Error Beta

1

(Constant) .767 .751 1.021 .310

A9 -.285 .105 -.284 -2.718 .008

A10 .114 .129 .098 .889 .376

A12 -.172 .134 -.138 -1.281 .203

C1 .057 .066 .084 .861 .391

2

(Constant) 1.055 .752 1.402 .164

A9 -.253 .103 -.253 -2.463 .016

A10 -.077 .138 -.066 -.557 .579

A12 -.307 .144 -.247 -2.129 .036

C1 .036 .064 .053 .558 .578

B3 .020 .066 .033 .298 .766

B8 .009 .082 .012 .112 .911

B9 .187 .080 .269 2.326 .022

B11 -.037 .092 -.047 -.400 .690

B12 .101 .052 .231 1.948 .054

3 (Constant) 1.610 .945 1.703 .092

289

A9 -.266 .103 -.265 -2.580 .012

A10 -.084 .144 -.072 -.582 .562

A12 -.323 .147 -.260 -2.192 .031

C1 .032 .067 .047 .481 .631

B3 .006 .070 .010 .088 .930

B8 -.031 .085 -.041 -.358 .721

B9 .177 .086 .255 2.064 .042

B11 -.020 .093 -.026 -.219 .827

B12 .093 .055 .213 1.699 .093

B14 .150 .088 .212 1.703 .092

B15 -.105 .105 -.148 -.995 .323

B16 -.026 .089 -.046 -.294 .769

B17 .041 .060 .077 .689 .493

B18 -.014 .087 -.023 -.158 .875

B19 -.098 .067 -.175 -1.472 .145

4

(Constant) .816 .965 .846 .400

A9 -.273 .100 -.272 -2.718 .008

A10 -.034 .142 -.029 -.239 .811

A12 -.256 .154 -.206 -1.666 .099

C1 .031 .066 .047 .473 .637

B3 -.063 .074 -.105 -.857 .394

B8 -.020 .084 -.027 -.237 .813

B9 .116 .087 .167 1.334 .186

B11 .027 .094 .034 .284 .777

B12 .055 .060 .126 .916 .362

B14 .134 .086 .190 1.550 .125

B15 -.140 .107 -.199 -1.315 .192

B16 .002 .088 .004 .025 .980

B17 .053 .059 .097 .890 .376

B18 -.048 .086 -.080 -.557 .579

B19 -.054 .067 -.097 -.812 .419

B2 .022 .060 .040 .358 .721

B4 .078 .054 .153 1.436 .155

B6 .101 .063 .193 1.593 .115

5

(Constant) .230 1.032 .223 .824

A9 -.259 .102 -.259 -2.551 .013

A10 -.014 .147 -.012 -.095 .924

A12 -.280 .155 -.225 -1.810 .074

C1 .027 .066 .040 .411 .682

290

B3 -.084 .076 -.141 -1.105 .273

B8 -.051 .089 -.067 -.572 .569

B9 .046 .095 .066 .483 .630

B11 .041 .100 .052 .409 .684

B12 .077 .061 .177 1.254 .213

B14 .131 .087 .186 1.507 .136

B15 -.170 .112 -.240 -1.518 .133

B16 .009 .090 .016 .100 .921

B17 .061 .060 .112 1.016 .313

B18 -.035 .086 -.059 -.410 .683

B19 -.018 .069 -.033 -.267 .790

B2 .030 .062 .056 .488 .627

B4 .098 .055 .192 1.768 .081

B6 .076 .066 .144 1.138 .258

B1 .114 .081 .159 1.417 .160

B5 .084 .094 .112 .895 .374

B7 .033 .093 .047 .351 .727

B13 -.004 .079 -.006 -.052 .959





Correlation

Collinearity

Statistics

Tolerance

1

B3 .101b 1.020 .310 .103 .903

B8 .147b 1.453 .149 .145 .860

B9 .303b 2.997 .003 .290 .810

B11 .130b 1.359 .177 .136 .969

B12 .284b 2.503 .014 .245 .660

B14 .012b .120 .905 .012 .952

B15 -.137b -1.436 .154 -.144 .973

B16 -.094b -.962 .338 -.097 .929

B17 -.060b -.602 .548 -.061 .918

B18 -.218b -2.312 .023 -.227 .961

291

B19 -.241b -2.556 .012 -.250 .950

B2 .183b 1.890 .062 .188 .923

B4 .202b 2.136 .035 .211 .962

B6 .361b 4.080 .000 .381 .984

B1 .279b 3.026 .003 .292 .972

B5 .242b 2.591 .011 .253 .966

B7 .288b 3.141 .002 .302 .978

B13 .207b 2.202 .030 .217 .971

2

B14 .036c .374 .709 .039 .924

B15 -.112c -1.179 .242 -.121 .914

B16 -.095c -.985 .327 -.102 .880

B17 .000c -.001 1.000 .000 .823

B18 -.116c -1.161 .249 -.120 .829

B19 -.169c -1.755 .083 -.179 .875

B2 .145c 1.443 .152 .148 .803

B4 .197c 2.082 .040 .211 .892

B6 .282c 2.562 .012 .257 .645

B1 .217c 2.224 .029 .225 .835

B5 .153c 1.441 .153 .148 .729

B7 .211c 1.714 .090 .175 .532

B13 .097c .888 .377 .092 .693

3

B2 .145d 1.376 .172 .146 .735

B4 .213d 2.157 .034 .225 .805

B6 .256d 2.267 .026 .236 .612

B1 .184d 1.746 .084 .184 .726

B5 .154d 1.418 .160 .150 .692

B7 .170d 1.357 .178 .144 .516

B13 .089d .793 .430 .085 .649

4

B1 .203e 1.957 .054 .209 .702

B5 .175e 1.610 .111 .173 .647

B7 .121e .940 .350 .102 .469

B13 .065e .589 .557 .064 .639





B16, B18, B15


B16, B18, B15, B4, B2, B6

292

Appendix C - NICHE Cover Letter

Factors affecting the intention to use a personal carbon trading system

Documents

Transcript of Factors affecting the intention to use a personal carbon trading system