Role of Reality TV as a Consumer Socialisation Agent of

Teenagers in a Developing Country

Submitted by:

Md. Ridhwanul Haq

MBA (Marketing), Master (Management), BBA (Marketing)

This thesis is Submitted in Fulfillment of the Requirement for the Degree

of Doctor of Philosophy of Marketing

School of Marketing

College of Business

University of Western Sydney

August, 2010

i

Table of Contents

List of Tables vii

List of Figures x List of Diagrams xi

Abbreviations xii

Certificate of Authorship xiv

Acknowledgements xv

Ethics Approval xvi

Record of Candidature xvii

Dedication xviii

Abstract xix

1 Introduction

1.1 Background to the Research 1 1.2 Research Gaps 2 1.3 Research Problem, Objectives and Hypotheses 4 1.4 Justification of Research 6 1.5 Research Methodology 7 1.6 Outline of the Thesis 8 1.7 Definitions 10 1.8 Delimitations of Scope and Key Assumptions 12 1.9 Conclusion 12

2 Literature Review

2.1 Introduction 13 2.2 Review of Parent and Immediate Disciplines 15 2.3 Development of Marketing Concept - Consumer Orientation 16 2.4 Consumer Behaviour 17 2.5 Socialisation Theories 20 2.5.1 Socialisation Theories from Multidisciplinary Perspectives 21 2.6 Consumer Socialisation Theories 24 2.7 Learning Theories 28

ii

2.8 Agents of Consumer Socialisation 32 2.8.1 Family (Parents) 33 2.8.2 Peer Group (Friends) 37 2.8.3 Schools 40 2.8.4 Other Socialisation Agents 42 2.8.4.1 Religious Institutions 42 2.8.4.2 Community 43 2.8.5 Media 43 2.9 Media Type 46 2.9.1 Television 48 2.9.2 Media Vehicle 49 2.9.3 TV Vehicle - Reality TV (RTV) 50 2.10 Conclusion 56

3 Conceptual Framework and Hypotheses

3.1 Introduction 58 3.2 Involvement with Media 58 3.2.1 Involvement with Reality TV (RTV) 60 3.3 Key Constructs of Consumer Socialisation 61 3.3.1 Cognition 62 3.3.1.1Consumption Related Cognition Development by Television (RTV Involvement) 63 3.3.2 Attitude 66 3.3.2.1 Consumption Related Attitudes Influenced by Television (RTV Involvement) 67 3.3.3 Values 71 3.3.3.1 Consumption Related Values Transmitted by Television (RTV Involvement) 72

iii

3.4 Social Structural Variables in Consumer Socialisation 76

3.4.1 Peer Group Influences on RTV Involvement of Teenagers 76 3.4.2 Parental Control on RTV Involvement of Teenagers 79 3.4.3 Gender Differences of Teenagers and RTV Involvement 83 3.4.4 Social Class Theories 86 3.4.4.1 Social Class Differences of Teenagers and their

RTV Involvement 88

3.5 Conclusion 92

4 Research Methodology

4.1 Introduction 97 4.2 Justification of the Methodology - Quantity or Quality 98 4.3 Research Procedure 101 4.3.1 Qualitative 101 4.3.1.1 Sampling for Qualitative Research 102 4.3.1.2 The Focus Group Interviews (FGI) 102 4.3.1.3 Ensuring Rigour 105 4.3.1.4 Data Management and Analysis 106 4.3.2 Need for Quantitative Research 107 4.3.3 Scale Development and Confirmatory Factor Analysis 108 4.3.4 Constructs, Domains and Scale Items 109 4.3.4.1 RTV Involvement 109 4.3.4.2 Consumption Related Cognition 110 4.3.4.3 Consumption Related Attitude 112 4.3.4.4 Consumption Related Values 113 4.3.4.5 Parental Control 114 4.3.4.6 Peer Group Influence 116 4.3.4.7 List of Constructs, Domains and Scale Items 117 4.3.5 Scale Evaluation 120 4.3.6 Questionnaire - the Research Instrument 122 4.3.7 Sample and Sample Size 123 4.3.8 Administration of Questionnaire 125 4.3.9 Data Preparation 126 4.3.10 Data Entry and Screening 126

4.3.11 Structural Equation Modelling and Hypotheses Testing 127 4.4 Ethical Considerations 127

iv

5 Refinement of Measurement Scales by Qualitative Measures

5.1 Introduction 128 5.2 RTV Involvement 129 5.3 Consumption Related Cognition 134 5.4 Consumption Related Attitude 139 5.5 Consumption Related Values 143 5.6 Social Structural Variables and RTV Involvement of Teenagers 146

5.6.1 Peer Group Influences on RTV Involvement 147 5.6.2 Parental Control on RTV Involvement 150 5.7 Social Class and Gender Differences of Teenagers and their RTV

Involvement and Consumer Socialisation 153 5.7.1 Social Class Differences and RTV Involvement of Teenagers 153 5.7.2 Gender Differences and RTV Involvement of Teenagers 155 5.8 Conclusion 157

6 Assessment and Refinement of Measurement Scales by Quantitative

Measures

6.1 Introduction 165 6.2 Sample Characteristics 165 6.3 Assessment of Measurement Scales 171 6.4 Results of Exploratory Factor Analysis 172 6.5 Measurement Model Evaluation and Specification 178 6.5.1 Deciding Goodness-of-Fit Criteria 179 6.5.2 Testing the Measurement Model Fit 181 6.6 Assessment of Measurement Scales Using CFA 182 6.6.1 CFA results of ‘Parental Control’ 184 6.6.2 CFA results of ‘Peer Group Influences’ 186 6.6.3 CFA results of ‘RTV Involvement’ 188 6.6.4 CFA results of ‘Consumption Related Cognition’ 190 6.6.5 CFA results of ‘Consumption Related Attitude’ 192 6.6.6 CFA results of ‘Consumption Related Values’ 195

v

6.7 Reliability and Validity of Constructs 197

6.7.1 Reliability 197 6.7.2 Reliability of Constructs 198 6.7.3 Item Reliability 199 6.7.4 Validity 200

7 Testing the Theoretical Model and Hypotheses

7.1 Structural Model Evaluation 202

7.1.1 Assessment of the Distribution of the Variables 202 7.1.2 Estimating the Proposed Model 204 7.1.3 Modifying the Model 205 7.1.4 Estimating the Main Model and Test Hypotheses 206 7.1.5 Testing Hypotheses of the Main Model 210 7.1.6 Interpretation/Discussion of Empirical Results 211

7.2 Conclusion 223

8 Conclusions

8.1 Overview 226 8.2 Contribution of this Research 227 8.2.1 Implication for Theory 228 8.2.2 Implication for Methodology 230 8.2.3 Implications for the Policy and Practice 230 8.3 Limitations 232 8.4 Implications for Further Research 233 8.5 Specific Conclusions from the Findings of This Research Regarding the Research Question and Hypotheses 235 8.6 Conclusion 236

vi

References 237 Appendix 1: Questionnaire 280

Appendix 2: Flyer for Volunteers 287

Appendix 3: Research Protocol 289

Appendix 4: Participant Consent Form 292

Appendix 5: Focus Group Discussion Script 294

Appendix 6: Parental Consent Form 297

Appendix 7: Permission from Schools in Dhaka 299

Appendix 8: A proposed Model for this Research 302

Appendix 9: Computer Output of the SEM Analysis of the Empirically Tested

Model (Main Model) 305

Appendix 10: Multi-Group Analysis Output of Gender Differences 308

Appendix 11: Multi-Group Analysis Output of Social Class Differences 313

Appendix 12: List of Constructs and Variables 318

Appendix 13: Questionnaire (Bengali Version) 321

Appendix 14: List of Middle and Lower Social Class Suburbs 327

vii

List of Tables

Serial No Name of the Table Page No

2.1 Parent-Children Communication Patterns 35

2.2 Classification of Media 46

2.3 Features of Various Media 47

2.4 Classification of reality TV (RTV) 50

2.5 Domain of RTV Observables and Their Dimensions 55

3.1 Comparison between Boys and Girls’ Involvement with TV

shows

84

3.2 List of Constructs, Domains and Scale Items Adapted from

Literature

93

4.1 Domain of Scale Items of RTV Involvement 110

4.2 Domain of Scale Items of Consumption Related Cognition 112

4.3 Domain of Scale Items of Consumption Related Attitude 113

4.4 Domain of Scale Items of Consumption Related Values 114

4.5 Domain of Scale Items of Parental Control on TV 115

4.6 Domain of Scale Items of Peers Group Influences 116

4.7 List of Constructs, Domains and Scale Items 117

5.1 List of Constructs, Domains and Scale Items 157

5.2 Teenagers’ Favourite RTV shows (Number within parenthesis

indicating number of times mentioned by respondents)

161

5.3 The Compressed Assembly of Data based on FGIs 162

6.1 Distribution of Respondents by Gender Differences 166

6.2 Distribution of Respondents by Age Groups 166

6.3 Distribution of Respondents by Academic Class 166

6.4 Distribution of Respondents List by Social Classes 167

6.5 A Cross-Tabulation of teenagers’ Parents Social Class and

Their Education

168

6.6 A Cross-Tabulation of teenagers’ Parents Social Class and 169

viii

Their Occupation

6.7 A Cross-Tabulation of teenagers’ Parents Social Class and

Their Annual Income

170

6.8 Factors, Variables and Factor Loading 173

6.9 Retained Factors, Variables and Factor Loading 177

6.10 Summary of Reliability, Weights, and Fit Indices Used in this

Research

179

6.11 Summary of Standardised Estimates Used for CFA Models 181

6.12 List of Variables 183

6.13 Regression Weights of Parental Control 184

6.14 Standard Regression Weights, Goodness-of-Fits Estimates and

Composite Reliability of the Parental Control

185

6.15 Regression Weights of Peer Group Influences 187

6.16 Standard Regression Weights, Goodness-of- Fits Estimates and

Composite Reliability of the Peer Group Influences

187

6.17 Regression Weights of RTV Involvement 189

6.18 Standard Regression Weights, Goodness-of-Fits Estimates and

Composite Reliability of the RTV Involvement

189

6.19 Regression Weights of Consumption Related Cognition 191

6.20 Standard Regression Weights, Goodness-of -Fits Estimates and

Composite Reliability of the Consumption Related Cognition

191

6.21 Regression Weights of Consumption Related Attitude 193

6.22 Standard Regression Weights, Goodness-of -Fits Estimates and

Composite Reliability of the Consumption Related Attitude

194

6.23 Regression Weights of Consumption Related Values 195

6.24 Standard Regression Weights, Goodness-of -Fits Estimates and

Composite Reliability of the Consumption Related Values

196

6.25 Construct Reliability of Scale Items 198

6.26 Multiple Correlation Coefficients 199

7.1 Assessment of Normality 203

ix

7.2 Fitness of Measure for the Proposed Model 204

7.3 Fitness of Measure for the Main Model 207

7.4 Comparisons of Fit Measures for Main and Proposed Model 208

7.5 Standard Estimates for the Main Model 210

7.6 Goodness-of-Fits Estimates of Male and Female 217

7.7 Cross Validation Statistics - Gender 218

7.8 Multi-Group Analysis of Gender Differences on Consumer

Socialisation

218

7.9 Goodness-of-Fits Estimates of Middle and Lower Social Class 220

7.10 Cross Validation Statistics - Social Class 221

7.11 Multi-Group Analysis of Social Class Differences on Consumer

Socialisation

222

7.12 Summary of the Research Findings 224

8.1 List of Research Objectives for this Research 226

x

List of Figures

Serial No Name of the Figure Page No

2.1 Relationships of Parent and Immediate Disciplines 15

2.2 Domains of Consumer Behaviour 18

2.3 Conceptual Model of Consumer Socialisation 26

2.4 Components of Observational Learning 31

3.1 Outcomes of Consumer Socialisation 62

3.2 Proposed Model of Consumer Socialisation of

Teenagers’ by RTV (CSRTV) 93

4.1 Qualitative Data Analysis Process 107

6.1 CFA Model of Parental Control 184

6.2 CFA Model of Peer Group Influences 186

6.3 CFA Model of RTV Involvement 188

6.4 CFA Model of Consumption Related Cognition 191

6.5 CFA Model of Consumption Related Attitude 193

6.6 CFA Model of Consumption Related Values 195

7.1 Proposed Model of CSRTV 205

7.2 Main Model of CSRTV 208

7.3 The Empirically Tested Model of CSRTV (Main Model) 223

xi

List of Diagrams

Serial No Name of the Diagrams Page no

2.1 Links among Socialisation, Communications, and Offspring

Outcomes

36

3.1 Links between Media Motives, Involvement and Media Use

and Effects

59

3.2 Links between Media Motives, Cognitive and Affective

Involvement, Media Use/Effects

59

xii

Abbreviations

AC Actual Counts

ATT Attitude

ANOVA Analysis of Variance

AMOS A SEM Software

A Coefficient Alpha

AGFI Adjusted Goodness-of-Fit Index

B Boys

CR Critical Ratio

CFA Confirmatory Factor Analysis

Cog Cognition

CSRTV Consumer Socialisation by Reality Televisions

CFI Comparative Fit Index

DF Degrees of Freedom

EC Expected Counts

E Error

EQS A SEM Software

EFA Exploratory Factor Analysis

F Factor

G Girls

GFI Goodness-of-fit Index

INV Involvement

LISERAL A SEM Software

ML Maximum Likelihood

MP3 Moving Picture Experts Group Layer-3 Audio

MNC Multi-National Companies

NFI Normed Fit Index

PI Peer Group Influences

PC Parental Control

xiii

P (%) Percentage

P Probability Level

RMSEA Root Mean Square Error of Approximation

RTV Reality Television

SEM Structural Equation Modelling

SMS Short Message Service

SPSS Statistical Package for Social System

SMC Square Multiple Co-relation

SRW (Beta) Standard Regression Weight

TLI Tucker Lewis Fit Index

TV Television

TVC Television Commercial

VAL Values

X2 (df,,p) Chi-square

X2�df Normed Chi-square

xiv

Certificate of Authorship

I hereby declare that this submission is my own work and that, to the best of my

knowledge and belief, it contains no material published or written by another person nor

material which to a substantial extent has been approved or accepted for the award of

any other degree or diploma of a university or other institution of higher learning,

except where due acknowledgement is made in the acknowledgement.

Md. Ridhwanul Haq

xv

Acknowledgements

Firstly, I would like to express my gratefulness to Almighty Allah who made this work

possible for me. I am also cordially thankful to a number of kind-hearted individuals

and organisations whose contributions either made the completion of this research

possible or helped make the final output better.

I would especially like to thank my principal supervisor Dr. Syed H. Rahman for his

excellent supervision. He took maximum interest in the research project. His guidance

and constructive criticism, quick response to the problems and encouragement during

the entire period of the study, together with careful reading and correcting of the drafts

of the report, is sincerely acknowledged. Without his close guidance it would have been

never possible to complete this research project within the time frame it has been

completed.

To my associate supervisor, Dr. Felicitas Evangelista, I express my sincere thanks. She

gave her valuable insights and feedback on all phases of this research.

I appreciate the concerned authorities of the respective schools in Dhaka, who permitted

me to conduct Focus Group Interviews (FGIs) and survey in their schools. Also this

research could not have been completed without the assistance of the interview and

survey participants, who patiently gave their time and shared their valuable experiences

and knowledge with me. They deserve big thanks.

Special thanks to other academics and administrative staff in the school of Marketing

particularly; John Stanton, Laural Jackson, Aila Khan and Kaylene bailey for giving me

support and encouragement to complete this research project. Also thanks to all my

family members, relatives, friends, students and colleagues at home and abroad who

wish me well for this research. I must thank Stephen Dicks for his comments and

editing of this thesis.

I thank my wife Sohana for her continual support. I express my deepest love and sorry

to my little angel Najat to whom I hardly gave time and care during this period.

xvi

Ethics Approval

xvii

Record of Candidature

xviii

This Ph.D thesis is dedicated to my father Md. Shamsul Haque

and mother Nurun Nahar Haque with love and gratitude.

1

Chapter 1 Introduction

1.1 Background to the Research

Consumer socialisation is a process by which young people acquire skills, knowledge

and attitude relevant to their functioning as consumers in the market place. Generally,

consumer socialisation refers to what behaviour teenage children learn from different

socialisation agents and how those behaviours are associated with purchase as well as

use of goods and services. In recent consumer socialisation research, teenagers are the

key focus of various researchers (Anderson et al. 2001; Chan & McNeal 2006a) as

growth of the teenaged market segment has got the attention of the researchers and

marketing practitioners (Blackwell et al. 2007). Particularly, the size of the teenaged

population is quite large in developing countries (1.6 billion), compared with developed

countries (240 million) (Population Reference Bureau 2006). Furthermore, the teenaged

population is the major proportion of most of the developing countries total population

(Quraishi, Bhuiya & Mohammad 2004). This teenaged population is considered to be a

lucrative market by marketers due to their purchasing patterns and consumption related

behaviour (Zolo 2004). Teenagers of developing countries are also getting the attention

of the marketing practitioners as an emerging market segment (Bhosale & Gupta 2006).

Consumer socialisation theorists have also extensively researched the effect of

socialisation agents on teenaged children (Avrahami 1995; Lueg & Finney 2007). In the

extant consumer socialisation researches media, parents, schools and peer groups are

considered to be the most common socialisation agents. Specifically, as one of the

consumer socialisation agents, teenagers’ involvement with the media and its affect on

teenagers’ consumption related behaviour has received a lot of attention from the

researchers (Butterbaugh 1999; Chan & McNeal 2006a). It has been established by the

researchers that media, particularly TV, develops teenagers’ consumption related

cognition; influences consumption related attitudes and transmits consumption related

values (Chan 2003; Moschis & Moore 1978; Schmoll et al. 2006). Teenagers in

2

developing countries are also showing interest in various media (Bhosale & Gupta

2006; Yoon & Garma 2006). These days, not only in developed countries but also in

developing countries, teenagers are getting involved with TV, particularly with reality

TV (RTV), a new generation of TV (Tedesco 2007; Yoon & Garma 2006). Reality TV

(RTV) is a form of TV programming which represents purportedly unscripted dramatic

or humorous situations, documents actual events, and features ordinary people instead

of professional actors. Particularly, different content and the format of RTV show the

influence of teenagers’ involvement which ultimately affects their consumption related

behaviour (Adams, Sutherland & Johnson 2007; Idato 2007).

However, most of the research on consumer socialisation so far has been from the

perspective of developed countries or from western cultural settings (Dotson & Hyatt

2005; John 1999; Neeley 2005). Consumer socialisation of teenagers has rarely been

examined from the perspective of developing countries (Bhosale & Gupta 2006).

Teenagers in developed countries grow up differently as a consumer, than their

developing countries' counterparts. In the context of this research, it has been addressed

that accessibility and availability of various media are comparatively higher in

teenagers of developed countries than developing countries (Mark 1998). Furthermore,

due to socio-economic structural differences, the consumer socialisation takes place

differently among the teenagers in developing countries than their counterparts in

developed countries (Choi & Ferle 2004; Erametsa 1990). This research will contribute

towards plugging this gap in the existing literature.

1.2 Research Gaps

As has already been mentioned most research on consumer socialisation is focused on

developed countries. Little attention has been given to the developing countries in this

regard. This is one of the major gaps in current consumer socialisation research. There

are also a number of other gaps observed in the current literature of consumer

socialisation, which this research will try to address. Firstly, there is no empirically

tested model of consumer socialisation of teenagers by RTV involvement in the

3

literature (Yoon & Garma 2006). Secondly, parental control, peer group influences,

gender and social class differences of teenagers that are considered as social structural

variables of consumer socialisation have been mostly researched from the perspective

of developed countries, but are rarely examined from the perspective of developing

countries (Cogdon 2007; Lueg & Finney 2007). Current literatures show that teenagers’

gender and social class influence their involvement with media which ultimately affect

their consumer socialisation. Particularly, teenagers both boys and girls in developed

countries enjoy equal opportunity and access to different media (Jung & Peterson 2007;

Moschis & Churchill 1978). On the other hand, teenagers in developing countries have

comparatively lower access and options to different media than their developed

countries counterparts (Nath 2006). Furthermore, boys have more access and

opportunity of getting involved with media than girls in developing countries (Nath

2006). Current literature also shows that in developed countries parents play moderating

roles in teenagers’ media involvement (Neeley 2005). On the other hand, teenagers’

media involvements are closely supervised by their parents in developing countries

(Cardoza 2002). However, peer group in developed countries immensely influences

teenagers’ perception towards media and this has also been researched only in

developed countries but rarely in developing countries (Byoungkwan Salmon & Hye-

Jin 2007).

Finally, applicability of the measurement scale items of the socialisation agent, RTV

involvement and the key outcome variables of consumer socialisation, e.g.,

consumption related cognition, and attitude and values of teenagers have not been

established. These measurement scale items have been mostly developed based on other

media research (Bartlett, Griffiths & Badian 2008; Schomoll et al. 2006; Chan 2003;

Yoon & Garma 2006), that has not been tested in case of Reality TV (RTV). RTV is a

generation of media which unlike traditional TV programs presents purportedly

unscripted dramatic or humorous situations, documents actual events, and features

ordinary people instead of professional actors (Lundy & Jacobson 2008). This involves

interactivity between the programs and the audience, often live. Moreover, RTV also

differ from main stream TV programming in terms of its format, reward system and

4

layout (Bown 2008; Frutkin 2008; James 2007). Though there is no known difference

between developed and developing countries regarding RTV format, reward system and

layout, there is significant difference in the nature of teenagers’ involvement in RTV.

Based on their research Cardoza (2002) and Nath (2006) concluded that developing

country teenagers’ show different levels of involvement with the media than their

developed country counterparts. This is due to the relatively lesser access of the

developing country teenagers’ to the electronic media to watch RTV as well as to

participate directly in RTV programs via SMS, internet or telephone calls during live

programs.

1.3 Research Problem, Objectives and Hypotheses

The problem addressed in this research is:

How does teenagers’ involvement with reality television (RTV) affect their

consumer socialisation?

Based on the above research problem, the following are the objectives of this research,

which links to the theory examined in chapter two:

1. To develop and test a model of the consumer socialisation of teenagers of a

developing country by RTV.

2. To determine how RTV affects the consumer socialisation of teenagers in a

developing country.

3. To determine how teenagers’ social class differences, their gender differences,

peer group and parental control influence the role of RTV in their consumer

socialisation.

5

The following hypotheses are developed on the basis of current literature which was

tested in this research:

H1: There is a positive impact of RTV involvement on consumption related cognition.

H2: There is a positive impact of RTV involvement on consumption related attitude.

H3: There is a positive impact of consumption related cognition on consumption related

attitude.

H4: There is a positive impact of RTV involvement on consumption related values.

H5: There is a positive impact of consumption related values on consumption related

attitude.

H6: There is a positive impact of peer groups influences on teenagers’ RTV

involvement.

H7: Parents through communication, mediation and instruction positively influence

teenagers’ RTV involvement.

H8: There is a difference between male and female teenagers’ involvement with RTV

and its effect on their consumer socialisation.

H9: There is a difference between middle and lower social class teenagers’ involvement

with RTV and its effect on their consumer socialisation.

6

1.4 Justification of Research

There is a lot of literature on the consumer socialisation of teenaged children (Lueg &

Finney 2007; Neeley 2005). Specifically, TV as a consumer socialisation agent of

teenagers has received a lot of attention by the researchers in the developed countries

(Butterbaugh 1999; Fletcher 2006), but has been largely overlooked from the

developing countries’ perspective. RTV in particular, has become the most popular TV

vehicle and recent phenomenon (Bellafante 2008; Scardino 2005), but has not received

much attention from academics and researchers (Yoon & Garma 2006). The countries

of the world are large in number. They are also varied in nature and complexity, and

developing countries are surely different from developed ones. Accordingly,

understanding the role of RTV in the consumer socialisation of teenagers in a

developing country, as well as how this socialisation varies in two social classes is

critical for further theory building. Findings of this research will contribute towards

that.

So far, the increasing role of TV as a consumer socialisation agent of teenagers in

developed countries has been recognised by [the] researchers (Dubow, Huesmann &

Greenwood 2007). However, even in the developed countries no empirically tested

model of consumer socialisation by TV has been developed. Moreover, in the recent

trend as a TV vehicle, RTV has received serious attention from teenagers (Peters 2007;

Powell 2003), but not much information is found in the current literature, particularly in

a developing country like Bangladesh. One of the key theoretical contributions of this

research will be to develop and test such a model.

Despite substantial interest in TV’s role in the consumer socialisation of teenagers,

particularly in the developed countries, empirical research on consumption knowledge,

skills, attitudes and values that are transmitted by TV to teenagers is quite limited. The

scales that were developed and tested in this research can enable measurement of

consumer socialisation by the media for future research. One category of applications

relates to the use of RTV involvement and influence measures on outcome variables.

7

Another set of applications of the measure was related to the study of the causal effects

of RTV involvement and its influence on a range of variables such as consumption

related cognition, attitudes and values. The scales developed in this research also will

help to examine the role of referents like parents and peers on the consumer

socialisation.

Various researchers (Carlson, Laczniak & Walsh 2001; Kamaruddin & Mokhlis 2003;

Mark 1998) have given importance to social class differences, gender differences,

parental involvement and peer group influences as social structural variables while

studying consumer socialisation. Kamaruddin and Mokhlis (2003) have recognised the

significance of social classes while studying consumer socialisation issues. However,

not much has been reported in the literature. Similarly, the gender difference of

teenagers (Wolin 2003), parental control and peer group influences on TV vehicle

selection (Cogdon 2007; Lueg & Finney 2007; Wiman 1983) play a vital role in the

consumer socialisation of teenagers. But not much has been reported on any of these

issues in the current literature from the developing countries perspective. Findings of

this research will help to plug these gaps in the comparative understanding of teenagers

of different social classes within a country.

Findings of this research will also help current and future marketing practitioners,

media strategists, sponsors and researchers who operate or may operate in the future in

a developing country, particularly if their target market segment is teenaged children.

1.5 Research Methodology

This research adopts a positivist approach. The objective of this research has been to

identify how RTV as a socialisation agent, influences the teenaged consumers’

socialisation. In addition, the study also aims to assess the effects of social class

differences, gender differences, parental control and peer group influences on the role of

RTV as a consumer socialisation agent. Both qualitative and quantitative approaches

were adopted. Qualitative research was adopted at the exploratory stage, using focus

group interviews (FGIs) of teenagers, representing urban middle and lower socio-

8

economic classes in a developing country. The purpose of these focus group interviews

(FGIs) was to refine the existing constructs and their measurement scale items, making

them relevant to a developing country.

Based on the existing literature and qualitative research findings, a quantitative survey

instrument was developed. This instrument was administered among a representative

sample of the urban middle and the lower socio-economic class teenagers of a

developing country with proportional representation of both genders. The target

population for the study was the urban teenagers (13-19 years). Using a stratified

sampling method, data was gathered from 400 respondents. The collected data was

analysed using exploratory factor analysis followed by confirmatory factor analysis.

Furthermore, the framework of the consumer socialisation by RTV was tested, using

quantitative multivariate technique, structural equation modelling (SEM).

1.6 Outline of the Thesis

This thesis is divided into 8 chapters. This chapter contains the introduction, the other

chapters are: chapter 2- literature review, chapter 3- conceptual framework and

hypotheses, chapter 4- research methodology, chapter 5- refinement of measurement

scales by qualitative measures, chapter 6- assessments and refinement of measurement

scales by quantitative measures, chapter 7- testing the theoretical model and

hypotheses, and chapter 8- conclusion.

Chapter 2 provides a theoretical foundation upon which the research is based (a review

of the relevant literature). Further, the purpose of chapter 3 is to draw and consolidate

knowledge from different related literature in an effort to develop a model of consumer

socialisation of teenagers, as well as to evaluate the affect of social structural variables

(e.g., social class differences, gender differences, parental control and peer group

influences on RTV vehicle involvement) by RTV. A range of theories were examined to

draw research problems covering issues that underpin the research questions and

hypotheses.

9

Chapter 4 describes the methodology that was used to collect and analyse the data to

address the research problem, research objectives and hypotheses. In this research, both

qualitative and quantitative methods were used, the rationale of which, along with other

aspects of the methodology, was justified. This chapter describes the sampling, data

gathering methods, research instruments, data management, and analysis and evaluation

methods in the research. This chapter also describes the quantitative data analysis

method, SEM, and the SEM software AMOS that was used for quantitative data

analysis. Chapter 4 also includes an analysis of the ethical issues concerning the

research.

As has been already mentioned, both qualitative and quantitative methods of research

were used in this study. The measurement scale items which were developed on the

basis of literature review were subjected to a qualitative study to determine its

applicability to the population of interest, e.g., teenagers of a developing country.

Chapter 5 describes the findings of the qualitative research.

Qualitative research findings were further explored through quantitative technique.

Chapter 6 reports on assessments and refinement of measurement scale items by

quantitative measures. Whereas, chapter 7 reports on test results of the theoretical

model and hypotheses.

Chapter 8 concludes and highlights the contribution of this research both in academic

and practical terms and identifies potential areas for future research.

10

1.7 Definitions

Definitions adopted by researchers are often not uniform, so the key terms used in

research reports need to be defined. The following definitions and concepts are some of

the key terms that will be used in this report.

Socialisation: “The process by which someone learns the ways of a given society or

social group so that they can function within it” Elkin and Handel (1972, p. 4).

Consumer Socialisation: “Consumer socialisation is a process by which young people

acquires skills, knowledge and attitudes relevant to their functioning as consumers in

the market place” Neal, Quester and Hawkins (2004, p.434).

Media Vehicle: “Media vehicles are specific media within each general media type,

such as specific magazines, TV shows or radio programs” (Kotler et al. 2007).

Media Content: “Media content is entertainment, violence, sex, and informational

issues which are shown in the media vehicle” (Bushman 2005).

Reality TV (RTV): “RTV is a generation of TV programming which represents

purportedly unscripted dramatic or humorous situations, documents, actual events, and

features ordinary people instead of professional actors” (Deziel 2004).

Involvement: “A person’s perceived relevance to the object based on inherent needs,

values, and interest” (Zaichkowsky 1985).

Cognition: “Cognition is the mental processes involved in gaining knowledge and

comprehension; including thinking, knowing, remembering, judging, and problem

solving” (Wagner 2008).

11

Consumption related cognition: Is the mental processes involved in gaining

knowledge and comprehension; including thinking, knowing, remembering, judging,

problem solving and skill development regarding products and brands consumption

(developed for this thesis).

Attitude: “An enduring combination of motivational, emotional, perceptual and

cognitive process with respect to some aspect of our environment, such as other people,

places, ideas, products, etc” (Neal, Quester & Hawkins 2004).

Consumption related attitude: An enduring combination of motivational, emotional,

perceptual and cognitive process with respect to imitation, fashion, inspiration and

product placement, (developed for this thesis).

Values: “Values are shared beliefs among group members as to what behaviours are

desirable or undesirable” (Pope, Brennan & Voges 2007, p. 335).

Consumption related values: Are shared beliefs among group members as to what

behaviours are desirable or undesirable towards products, prohibited product

consumption, materialism and anti-social behaviour, (developed for this thesis).

Construct: “A hypothetical variable made up of a set of component responses or

behaviours that are thought to be related” (Hair, Bush & Ortinau 2006, p.353).

Domain: “The set of identifiable and measurable components associated with an

abstract construct” (Hair, Bush & Ortinau 2006, p.354).

Model: “A model is a representation of the most important elements of a perceived real

world system” (Naert & Leflang 1978, p.9).

12

1.8 Delimitations of Scope and Key Assumptions

Like any other research, this research also has its limitations:

Firstly, this research was carried out in Bangladesh, a developing country. Due to

cultural and socio-economic differences, findings of this study cannot be held generally

applicable as representative of other developing countries. Further research in other

developing countries will be needed.

Secondly, this research involved a comparative study between middle and lower social

classes, and although there is a small urban high social class in Bangladesh the findings

of this research does not represent that class.

Thirdly, this research was conducted only among urban teenagers and as such, does not

represent rural teenagers. Future research may be conducted among that segment of the

population.

Finally, a large number of migrants of Bangladeshi origin are now living in different

parts of the world. Future research may be conducted among the children of those

migrants to see the effect of acculturation.

1.9 Conclusion

This chapter has laid the foundation for this report. In this chapter the research

objectives and the hypotheses have been introduced and the need for this research

justified. It has also briefly described and justified the methodology used in this

research. Definitions of the key terms have been presented, limitations described, and

the report has been outlined. With this foundation, the report has proceeded to give a

detailed description of the research.

13

Chapter 2 Literature Review

2.1 Introduction

Teenagers are an emerging market segment these days. This has made it crucial for the

researchers and practitioners in the field to understand the teenaged consumers’ learning

and their response process. Consumer socialisation theorists (Chan 2006; John 1999;

Lueg & Finney 2007) have extensively researched the effect of socialisation agents on

teenaged children to explore the process and outcomes of consumer socialisation.

Specifically, as a consumer socialisation agent; electronic media received serious

attention from researchers (Frideres 1973; Dotson & Hyatt 2005; Vakratsas & Ambler

1999). Also during this period teenagers as a market segment have received increasing

research attention (Bhosale & Gupta 2006; Lueg & Finney 2007). However, most of the

theories built so far, focus on developed countries or from the perspective of western

cultural settings (Dotson & Hyatt 2005; John 1999; Neeley 2005), but not much

theoretical and empirical contribution has been made on consumer socialisation in the

context of developing countries (Bhosale & Gupta 2006). Also, behavioural differences

in this regard between teenaged children from different social class backgrounds within

the same cultural context have not received much attention from researchers. Moreover,

as social structural variables, social class differences, gender differences, parental

control and peer group influences on teenagers’ media involvement have got the

attention researchers (Edward & Grantham 2009; Fujioka & Austin 2002; Keith 1986;

Kim & Shin 1996), but absence of these issues are also observed in the developing

countries consumer socialisation research.

The countries of the world are large in number (195) (About.com 2010). They are also

varied in nature and complexity, and developing countries are surely different from

developed ones. Accordingly, understanding the role of media in the consumer

socialisation of teenagers in a developing country, as well as how consumer

socialisation varies between social classes and genders, is critical for further theory

14

building. It is particularly important as the huge size of the teenaged market segment

and impressive growth of electronic media are observed in developing countries

(Bhosale & Gupta 2006; Quraishi, Bhuiya & Mohammad 2004). Teenagers are chosen

in this research not only because of the huge size of this segment in developing

countries but also because they are in the early phase in learning and formatting their

consumer behaviour. Current literatures suggest that teenagers’ learning capacity and

cognitive development, consumption related beliefs and values are significantly

influenced by media (Chernin 2005; Cram & NG 1999; Gruber & Thau 2000; Lueg and

Finney 2007). Teenagers are chosen in this research not only because of the huge size of

this segment in developing countries but also because they are in the early phase in

learning and forming their consumer behaviour. Current literatures suggest that

teenagers’ learning capacity and cognitive development, consumption related beliefs

and values are significantly influenced by media (Chernin 2005; Cram & NG 1999;

Gruber & Thau 2000; Lueg and Finney 2007). Further, theory building in this area is

also important in developing countries, where the population is relatively young. For

example, the total teenaged population in developing countries is 1,537 million,

compared to only 236 million in developed countries (Population Reference Bureau

2006). One such developing country, Bangladesh has 23% (40 million) of its total

population in the 10-19 age group, which is the biggest segment of it’s population

(Quraishi, Bhuiya & Mohammad 2004). The growth of media, particularly electronic

media is also quite impressive in Bangladesh (Nath 2006). Accordingly, the key

problem this research will address is:

How does teenagers’ involvement with reality television (RTV) affect their

consumer socialisation?

In this chapter, a review of the literature from several key areas will build a theoretical

foundation for the research. The focus of this review is to draw and consolidate

knowledge from different related literature, in an effort to synthesise a model of

consumer socialisation of teenagers by reality television (RTV). Theories from several

disciplines will be examined in order to develop hypotheses for this research and

identify relevant constructs and their domain and measurement items.

15

2.2 Review of Parent and Immediate Disciplines

Given the large number of disciplinary literatures that have bearing in some way on this

research, only the critical antecedents have been examined. This chapter introduces

some major disciplines, including parent discipline, on which this research is built.

Figure 2.1 depicts the parent and immediate disciplines underpinning this study and

their interrelationships.

Figure 2.1 Relationships of Parent and Immediate Disciplines

Development of marketing concept-consumer orientation

Socialisation and consumer socialisation theories

Consumer behaviour

Learning theories and Social learning theory

Agents of consumer socialisation

Consumer socialisation

16

2.3 Development of Marketing Concept - Consumer

Orientation

Marketing has been defined as “an organisational function and a set of processes for

creating, communicating, and delivering value to the customer and for managing

customer relationships in ways that benefit the organisation and its stakeholder” (Kotler

et al. 2007, p.7). For decades there have been outcries that the consumer has been

mistreated, but marketing concept, a marketing philosophy that evolved in the late

1950s, not only focuses on consumer requirements, but also protects consumers’ rights

(Bell & Emory 1971). However, different alternative marketing philosophies that have

been followed in the real world; a) production concept, b) product concept, c) selling

concept, d) marketing concept and e) societal marketing concept. The Marketing

concept is crucial and contemporary among these concepts. Marketing management

philosophy holds that achieving organisational goals depends on determining the needs

and wants of target markets and delivering the desired satisfaction more effectively and

efficiently than competitors. Broadly, Marketing concept (market orientation) is

concerned with the processes and activities associated with creating and satisfying

customers by continually assessing their needs and wants, and doing so in a way that

there is a demonstrable and measurable impact on business performance (Uncles 2000).

Most of the literature of marketing concept concentrated on explaining the need of

companies to devote more time and effort to the requirements for their customers. The

extant literature also agreed with this concept because an adequate understanding of

consumers’ needs and wants by organisations could lead to better performance such as

growth in resources, higher customer satisfaction, and growth in reputation (Gainer &

Padanyl 2002; McClymont, Ogunmokun & Akbari 2004).

Overall, development of marketing concept prevails over the consumerism issues that

also cover the point of consumers’ preferences, rights, and above all, interest.

Therefore, understanding consumer behaviour is a pivotal issue for better practice of

marketing concept.

17

2.4 Consumer Behaviour

The extant literature indicates that the marketing concept upholds the consumer benefit

philosophy. Precisely, the classic feature of marketing concept is to focus on ‘benefit

approach’ of consumers (Evans, Jamal & Foxall 2006). Therefore, understanding

consumer behaviour is the prerequisite to offer benefit for consumers. Understanding

the following concepts are rudimentary, before discussing the thoughts and managerial

approaches of consumer behaviour:

• Consumer: The consumer can be defined as “a person who identifies a need or

desire, makes a purchase, and/or disposes of the product” (Solomon 2004, p

596). Furthermore, Foxall (1993) suggested from the theoretical point of view

that consumers are represented in two ways: a) learning history of consumers

that includes the reward and punishment experiences from the past behaviour

and b) state variables that influence consumption decisions.

• Consumption:

Consumption is an activity where people use products and services in a

variety of ways that can be classified in four categories: a) consuming as

experience (an emotional or aesthetic consumption of objects), b)

consuming as integration (learning and manipulating consumption objects

to express aspects of the self or society), c) consumption as classification

(the activities that consumers engage in to communicate their association

with objects, both to self and to others), d) consuming as play (consumers

use objects to participate in a mutual experience and merge their identities

with that of a group) (Solomon 2004).

• Behaviour: Solomon (2004) mentions that, behaviour is consumers’ actions

with regard to an attitude object. Neal, Quester & Hawkins (2004) also added

that consumers learn from previous experiences and behave accordingly, as well

as these; behaviours could be the result of active and passive learning processes.

18

In the extant literature, consumer behaviour is valued as multidisciplinary knowledge,

which is also considered as an applied science drawing from economics (e.g.,

understanding consumers spending, product evaluation skills), psychology (e.g.,

studying consumers’ motivation, perception, learning patterns), sociology (e.g.,

consumers behaviour in group), anthropology (e.g., core beliefs, values, customs of

consumers) and other disciplines (Blackwell et al. 2007; Schiffman et al. 2005). By

supporting this precedent; Hoyer and Maclnnis (2007) proposed four domains of

consumer behaviour.

Figure 2.2 Domains of Consumer Behaviour (Source: Hoyer & Maclnnis 2007)

To make purchasing decisions that affect outcomes like buying a new product or using a

product for symbolic reasons, consumers must first engage in the process explained in

the psychological core. Consumers need to be motivated, able, and have the opportunity

to be exposed to perceive and attend to information. Moreover, consumers need to think

about this information, develop attitudes and form memories about any response.

Similarly, the cultural environment also motivates consumers about how they process

information and the kind of decisions they make.

Physiological core Process of decision making Consumer’s culture Consumer behaviour outcome

19

Overall, the study of consumer behaviour is based on mainly two models a) cognitive

development models and b) a social learning model that also refers to the area of

socialisation and consumer socialisation (John 1999; Moschis & Churchilll 1978). In

general, traditional consumer behaviour theories incorporate the examination, analysis

and the application of perceptual theory of marketing stimuli. Overall, theories of

consumer behaviour can be broadly categorised into three areas (Bareham 2004):

• Consumers are information processors who engages in a rational, scientific,

deliberate and cognitive process leading to a purchase choice.

• Consumer behaviour is learned and as a result much is the result of habit.

• After the post-modern perspective, it is assumed the consumer does not

follow rules, is unlikely to be predictable, and may change their purchase

strategy from occasion to occasion.

In reality, consumer behaviour reflects the totality of consumers’ decisions with respect

to the acquisition, consumption and disposition of goods, services, activities, and ideas

by (human) decision making units. The issues of consumer behaviour are the sequence

of acquisition (refers to the activities leading up to and including the purchase or receipt

of a product), consumption (refers to how, where, when, and under what circumstances

consumers use the product) and disposition (how do consumers dispose the product?)

for the marketers (Blackwell et al. 2007; Chan 2006).

Furthermore, from the marketers’ point of view, understanding consumer behaviour is

like understanding the problem. Marketers keep in mind that consumer behaviour also

varies in terms of consumers’ age differences; family structure, social class and gender

differences (Kamaruddin & Mokhlis 2003). Interestingly, understanding teenaged

consumer behaviour has become critical for marketers these days, as this segment is

growing very fast. Young consumers socialise their consumption related learning in

different ways from various sources, then behave accordingly as a consumer (Dotson &

Hyatt 2005). Since learning is a vital issue of the socialisation of teenagers and their

consumption behaviour, knowing their learning processes is quite important for

researchers and marketers, to understand young consumers’ socialisationes.

20

2.5 Socialisation Theories

Socialisation has already been described as a process that begins at birth, by which an

individual learns the expectations of society, acquires sensitivity to the pressures and

obligations of group life, as well as learn to get along with others (Brown 1976; Melanie

& Locke 2007). Brim and Wheeler (1966) named this learning, ‘socialisation’ that

refers as a process, by which a person acquires knowledge and skills, that makes him or

her more or less an able member of society. Further, Goslin (1971) noted socialisation

as a process which helps human beings to learn from the environment and change their

behaviour accordingly. Elkin and Handel (1972) pointed two issues of socialisation. On

one hand, it helps to explain how a person becomes capable of participating in society.

On the other hand, socialisation helps to explain how society teaches human beings to

learn from society and the environment through a complex process. Similarly, being a

consumer, a person learns basic knowledge, ideas and skills from the environment and

socialisation agents (Cram & Ng 1999) and processes their acquired knowledge into the

market place (John 1999).

Berns (1997) proposed a method of socialisation that follows a specific sequence: a)

affective (effect emerges from feeling-attachment), b) operant (effect emerges from

acting-reinforcement, extinction, punishment, feedback, learning by doing), c)

observational (effect emerges from imitating-modelling), d) cognitive (effect emerges

from thinking - instructions, setting standards, reasoning), e) socio-cultural (effect

emerges from conforming - group pressure, tradition, rites and rituals, symbols), f)

apprenticeship (effect emerges from guided participation - structuring, collaborating,

transferring).

However, learning is a part and parcel of the socialisation and learned behaviour is also

considered to be an outcome of the socialisaion process. Social learning theories have

been focusing on the role of socialisation agents and influences on the socialisation that

are also considered to discuss the essence of socialisation studies.

21

According to Clausen (1968), socialisation may be viewed from the perspective of the

individual or collectively (be it the larger society or a constituent group having distinct

subculture). However, socialisation is a critical concept for all social sciences because

in considering socialisation, researchers are forced to examine the relationship between

the individual and their society.

2.5.1 Socialisation Theories from Multidisciplinary Perspectives

The term of ‘socialisation’ is contributed to and viewed from the perspective of various

disciplines. As is true of many concepts employed in the social sciences, the verb

‘socialise’ and its cognate ‘socialisation’ were current in the language well before they

were used as concepts by sociologists, psychologists or other behavioural scientists.

Brown (1976) mentioned that historically the systematic study of socialisation is rooted

in several fields which are addressed below:

Psychology: The area of socialisation research is always an important field in

psychology. In particular, social psychology focuses on the development of

characteristics relevant to social behaviour and on the basic process through which the

behaviour is acquired. The pathways or normal development of a human being is an

important wing of knowledge. However, psychologists have addressed somewhat

different issues while studying socialisation. While discussing socialisation issues,

psychologists are mostly inclined towards domains of social learning, psycho-analytic,

and cognitive theories.

It is not possible to state which theory is better to discuss the socialisation concept,

rather, that they complement each other. In the case of social learning theories, theorists

have attempted to extend the individual behaviour into society. According to Brown

(1976, p. 19) “social learning is learned behaviour in interpersonal situations and linked

to the needs, that require for their satisfaction, the mediation of other people”. To

understand the social learning theories, ‘stimulus-response’ and ‘trial and error’

concepts also need to be explored. Further, Psychoanalysis technique focuses on the

22

emotional development of the children while researching on socialisation (Larsen, To &

Fireman 2007; Pratt, Skoe & Arnold 2004). Larson, To and Fireman (2007) mentioned

that as children grow up, how they experience with mixed emotions also affects their

socialisation. In addition, parental guidance also helps cognitive development of

adolescents as addressed by Maccoby (1992) and cognitive development occurs as a

result of educational and social changes (Brody et al. 2002).

Sociology: When the term 'socialisation' is defined from the sociologists’ point of

view; then it concentrates on the characteristics of specific groups or institutions in

which socialisation occurs (Brown 1976). Parson and Bales (1955) proposed a ‘general

system theory of society’ that integrates cultural, social, and individual levels of

analysis. Socialisation is one of the major processes in the theory which accounts for the

time dimension for the society to continue across generations which requires that older

members influence younger ones to carry on in similar ways. However, sociologists

have not been seriously concerned about the individual differences among the

organisms that make up a society nor have they been much interested in the nature of

man’s biological development. Yet, in the formative years of sociology, around the end

of the last century, social thoughts were mostly contributed by the biological concepts.

Later on, the researchers brought the issue of environment on adolescent development

and socialisation agents belonging to this environment (Roopnarine 1986). Most

importantly, various socialisation agents are directly or indirectly involved in the

development process of children (Velleman, Templeton & Copello 2005).

Contemporary sociologists are acknowledging the contribution of various socialisation

agents like: the parent-children relationship (Updegraff et al. 2002) and psychological

control (Wang, Pomerantz & Chen 2007), guidance from teachers (Avrahami 1995),

company with a peer group (Rose 2007) and the influence of the media, which became

the most important environmental factor for children socialisation (Fletcher 2006;

Strasburger & Wilson 2002).

23

Anthropology: The discipline of Anthropology views socialisation from the

standpoint of the broader culture which helps to determine the overall boundaries (and

content) of socialisation and enables a comparison of the effects of different cultures

(Brown 1976). The concept of socialisation cannot be understood from an independent

context. In addition to psychological development and the involvement of various

socialisation agents, the context of cultural environment plays an important role in the

socialisation. The cultural dimension socialises people differently, and various

socialisation agents activate the socialisation into different cultures in different ways

(Mcdade & Worthman 2004). Cultural background and the country of origin

affect adolescents socialise differently on various issues (e.g., parental communication,

media exposure) (Choi & Ferle 2004), similarly religious beliefs teach various norms

and values about social behaviour (Seedat 2006). So, definitely and in different ways,

the process of socialisation is shaped and formatted in a cultural context.

Apart from the above mentioned disciplines, there are other areas (e.g., biological

discipline) that also claim their contribution in the socialisation. Nowadays, cultural,

genetic, and biological approaches are considered together, to study the socialisation

discipline. So far, it is needless to mention that socialisation is a multifaceted discipline,

where the consumer behaviourists define the socialisation discipline from the

consumers’ point of view. Consumers these days determine the business existence, and

their purchasing decision making process is quite complicated. Consumer behaviourists

translate the consumers’ role in the market place in terms of psychological growth,

cultural, and social contexts.

24

2.6 Consumer Socialisation Theories

As has already been mentioned, socialisation is a multidisciplinary concept. Consumer

behaviourists have brought some relevant issues of consumer behaviour to the

discipline (Cram & Ng 1999; Granhaug & Venkatesh 1986). As a leading consumer

socialisation researcher, John (1999) addressed that the use of the socialisation approach

in consumer research only began in the 1960s, and has only recently been proposed as a

vehicle for the study of consumer behaviour (Cram & NG 1999; Ward 1974). However,

definitions of consumer socialisation appearing in literature tend to view consumer

behaviour as a ‘sub-set’ of the total socialisation taking place in a person's life. Moschis

(1981) explained the concept of consumer socialisation, as consumer behaviour that is

acquired through interaction between the person and various agents in a specific social

setting. Ward (1974, p.4), somewhat differently defined what is the widely accepted

definition of socialisation that, “consumer socialisation is a process by which young

people acquire skills, knowledge, and attitudes relevant to their functioning as a

consumer in the market place”. Overall, consumer socialisation refers specially to the

process of learning consumer related skills, knowledge, and attitudes, and also focuses

particularly on young consumers’ various patterns of cognition and behaviour (Choi &

Ferle 2004; John 1999; Moschis & Moore 1979).

At the beginning of the 1970s, the study of socialisation became widespread and started

getting the attention of the marketing community (John 1999). At the same time, the

issue of consumer socialisation became the focus of researchers (Moschis & Churchill

1978; Ward & Wackman 1974), and contributions of early researchers led to a new

generation of researchers in the field.

So far, scholars have applied the consumer socialisation perspectives, which addressed

how individuals learn to be consumers (Lueg & Finney 2007). Further, researchers

(Moschis 1985; Moschis & Moore 1979) also explored a wide range of topics,

reflecting children's growing sophistication as consumers including their knowledge of

products, brands, advertising, shopping, pricing and decision-making strategies.

25

However, the field of consumer socialisation has not reached its potential due to its

reliance on stage development theories which limit both the variables and subjects of

inquiry.

So far consumer socialisation researches are typically based on two models of human

learning: a) the cognitive developmental model, and b) the social learning model (John

1999; Lueg & Finney 2007). The cognitive developmental model attempts to explain

the formation of consumer knowledge, skills, and behaviours as a function of qualitative

changes in cognitive development stages. The social learning model attempts to explain

the formation of consumer knowledge, skills, and behaviours as a function of

interactions between socialisation agents and individuals in different social settings

(Moschis & Moore 1979).

However, in the extant literature, consumer socialisation has been defined as a complex

process (Choi & Ferle 2004; Lueg & Finney 2007) that continues from early childhood

to advanced age. The outcomes of consumer socialisation are observed differently in the

various stages of young consumers (John 1999). John (1999) also addressed that these

stages are characterised along a number of dimensions that capture important shifts in

knowledge development, decision-making skills, and purchase influence strategies:

Perceptual Stages (3-7 years): The perceptual stage is characterised by a general

orientation towards the immediate and readily observable perceptual features of the

marketplace, but due to constraints in encoding the message and information, individual

experiences rarely act as rational consumers.

Analytical Stage (7-11 years): In the analytical stage, enormous changes occur in

cognitive and social knowledge. In this stage kids develop their consumption related

knowledge and skills rapidly. Further, the ability to analyse multidimensional stimuli

brings vast changes in the decision making skills of teenaged children.

26

Reflective Stage (11-16 years): The reflective stage is characterised by significant

development in several dimensions of cognitive and social development. On the one

hand, in this stage, consumer decisions are made in a more adaptive manner depending

on the situation and task. On the other hand, influencing friends and family members in

purchasing decisions became quite common at this stage.

Since 1970, the issue of consumer socialisation has become an interesting area for the

consumer behaviourists and researchers. Gradually the perspectives of consumer

socialisation have been developing by conceptual models and theories. Moschis and

Churchill (1978), as well as Moschis and Moore (1978) pioneered one of the conceptual

models of consumer socialisation which is mostly adopted by other researchers for

subsequent studies (John 1999; Lueg & Finney 2007). This model specified and

transferred the issues of socialisation and proposed five variables that constitute: (a) age

or life cycle position, (b) social structure constraints affecting learning are antecedents

to the socialisation, (c) the socialisation agent, (d) the type of learning processes

involved in socialisation, and (e) the content or criterion behaviour (learning properties,

i.e. cognition, attitude and values). This model particularly demonstrates the

significance of socialisation agents and the learning processes between the learner and

the agents.

Figure 2.3 Conceptual Model of Consumer Socialisation (Source: Moschis & Churchill

1978)

Antecedent Socialisationes Outcomes

Social structural variable

Age or life cycle position

Learning properties

Agent-learner relationships • Modelling • Reinforcement • Social Interaction

27

Age or life cycle: Age or life cycle position refers to a person’s life time period during

which learning occurs, and it is also used to index a person’s cognitive development of

life cycle stages.

Social Structural Variables: Social structural variables are factors such as social class

and gender, which help to locate the learner within his/her social environment, where

learning takes place.

Socialisation Agents: Brim and Wheeler (1966) mentioned that the socialisation agents

can be any person or an organisation that is involved in the socialisation, because of

frequency of contact with the individual and control over the reward and punishment

given to the individual. Family, peers, school, and media are the most basic consumer

socialisation agents of teenagers (Dotson & Hyatt 2005; Moschis & Churchill 1978).

Learning Processes: Churchill and Moschis (1979) referred to the learning process

as the way in which the learners acquire specific values and behaviours from

socialisation agents based on interaction with them. This learning process can be

classified: as a) modelling, which is also known as observational learning, and involves

imitation of agent’s behaviour, b) reinforcement that involves either reward (positive

reinforcement) or punishment (negative reinforcement) and c) social interaction, which

is less specific and involves the combination of modelling and reinforcement (Moschis

& Churchill 1978).

Content of Learning: Learning properties refer to a variety of consumer-related

cognition and behaviour that embraces the concept of consumer behaviour, such as

attitude toward the commercials and spending for brands.

Schiffman et al. (2005) also mentioned consumer socialisation has two distinct

components: On one hand, socialisation directly related to consumption, such as the

acquisition of skills, knowledge, and attitude concerned with budgeting, pricing, and

brand attitudes. On the other hand, socialisation indirectly related to consumption, such

as the underlying motivation that spurs youth to buy their desired product. Usually, the

28

direct component of consumer socialisation is the area of the consumer behaviourist,

who has broader goals of understanding all aspects of consumer behaviour. On the other

hand, the indirect component of consumer socialisation specific to individual product

categories is of more interest to marketers, who want to understand why people buy

their offers.

Unfortunately, extant research on consumer buying decisions traditionally dealt with

adult consumer behaviour and comparatively minor attention was given to the young

consumers’ buying decision process (Moschis & Moore 1979). Recently, a number of

researchers (Dotson & Hyatt 2005; Moore et al. 2002) have contributed to this area.

Particularly, advertising and the affects of media have got the attention of the

researchers to understand the young consumer socialisation (Banks & Gupta 1980;

Dotson & Hyatt 2005). These days young consumers are also acquiring appropriate

consumer related skills, behaviour orientations, knowledge, and attitude to participate

effectively in the market place (Cram & Ng 1999). So, teenaged children have become a

new focus for the researchers (Chan & McNeal 2006a; Dotson & Hyatt 2005).

Marketers are also considering these consumers as a profitable segment to explore

(Dammler 2005).

2.7 Learning Theories

Being social creatures, consumers are involved with environmental learning.

Prominently, Brim and Wheeler (1966) named this learning, ‘socialisation’, by which a

person acquires knowledge and skills that make him or her more or less able member of

the society. Most of the leading socialisation research in extant literature is grounded by

learning theories (Cram & NG 1999; Moschis & Moore 1979; Moschis & Churchill

1978).

29

The social learning theory is a stem from several theories including neo-skin-nerian,

neo-hullian, and learning theory (Moschis & Churchill 1978). Social learning is a part

and parcel of the socialisation, which employs the basic explanatory concepts of

learning theories: operant and classical, conditioning, negative and positive

reinforcement, punishment, extinction, suppression, generalisation, discrimination and

others (Windmiller, Lamber & Turie 1980). Most of these learning processes were

conducted experimentally either by ‘stimulus-response’ or ‘trial and error’ methods.

Bandura and Walters (1970), pioneer learning theorists, proposed that learning could

also be transferred by imitation or observation. Pope, Brennan and Voges (2007) noted

from prior researchers indicating that the content of consumer learning can be broken

down into two categories, a) directly relevant learning (directly relevant aspects of

consumer learning are those that are necessary if purchase and use are actually to take

place) and b) indirectly relevant learning (it means everything that has been learned that

motivates purchase and use behaviour).

However, Windmiller, Lamber and Turie (1980) mentioned that social learning theory

is a product of learning theory, which in turn developed out of the behaviouristic

movement in psychology. They also addressed that learning theories consist of

empirical laws that relate the properties of observable events (stimuli) to those of

subsequent observable behavioural responses. Before this stage, most of the theories of

social learning were laboratory based, and in particular, experimented on animals and

were therefore inadequate in explaining human social behaviour properly. Later on,

social learning theories were revised and ‘liberalised’ by learning theorists. Moreover,

the extant social learning theories provide the understanding about why and how a child

learns to become like other members of their society. Initially, most of the theories are

triggered by stimulus-response analysis of behaviour and a tension-reduction approach

in studying the problem of motivation, though this basic orientation has at times been

integrated with a more cognitive approach (Lueg & Finney 2007).

30

Surprisingly, most of the social learning theorists neglected genetic or developmental

pre-formation or in other words, they ignored ‘inner’ factors in contrast to ‘outer’

determinants. McLaughlin (1971) one of the leading learning theorists pointed out that

the key concept of social learning theories are that behaviour learned in interpersonal

situations and is linked to the needs that require for their satisfaction the mediation of

other people. Social learning theorists (Dotson & Hyatt 2005; Lueg & Finney 2007)

therefore had recourse to a new concept; they hypothesised that a great deal of human

learning takes place through observation and imitation, though imitation is not

necessarily deliberately taught. Moschis and Churchill (1978) renowned socialisation

theorists have studied the conditions under which such imitation occurs followed by the

extent of the influence of a ‘model’ on the learner, and the extent to which it is

necessary to hypothesise intervening variables such as needs, motives, expectancies or

other cognitive processes to account for their experimental findings.

While discussing the issue of learning processes of teenagers’ mediation, instrumental

training, and modelling are considered the most common techniques to

influence teenagers’ socialisation (Lueg & Finney 2007; Moschis & Churchill 1978).

Particularly, most of the social learning theorists today tend to deal directly with the

way in which imitation contributes to the child’s socialisation. In social learning

theories imitation is discussed in terms of the relationship of the child with an actual or

symbolic model and the model’s personal qualities or role attributes. Particularly,

learning through imitation is known as vicarious learning or modelling, which is an

important way of consumers’ learning. With this form of learning, it is not necessary for

consumers to experience a reward or punishment directly in order to learn. Instead they

can observe the outcomes of others’ behaviour and adjust their own accordingly.

Krugman (1962) proposed in the passive learning theory that consumers involvement

with media and exposure to media helps in their indirect learning process. Dotson and

Hyatt (2005) also mentioned that this type of learning is transmitted through repeated

exposure of socialisation agents or certain behaviour shown by role models both in low

and high involvement situations.

31

By introducing the concept of imitation, social learning theory moved to a position in

which it explicitly uses cognitive mediating processes in its explanatory repertoire

(Bandura 1986). These cognitive mediation processes indicate that perception, attention,

coding, thinking or memorising are some of the mediators interposed between stimulus

and response. The whole learning process is comprised of not only cognitive mediators,

but also by solid observation. Cognitive learning theory also added that observational

learning occurs when people watch the action of others and note the reinforcements

they receive for their behaviours - learning occurs as a result of vicarious rather than

direct experience. This type of learning is a complex process. People store these

observations in memory as they accumulate knowledge, perhaps using this information

at a later point to guide their own behaviour. The following diagrams show the

components of observational learning.

Figure 2.4 Components of Observational Learning (Source: Solomon 2004)

Overall, the concepts of learning and social learning theories have been applied to the

analysis of many aspects of human behaviour and later on to consumer behaviour. In

particular, the notion of shaping consumer behaviour has led to many practical

applications that also include the study of consumer socialisation.

Attention The consumer focuses on a model’s behaviour

Retention The consumer retains this behaviour in memory

Observational learning The consumer acquires and performs the behaviour earlier demonstrated by a model

Production processes The consumer has ability to perform the behaviour

Motivation A situation arises where in the behaviour is useful to the consumers

32

2.8 Agents of Consumer Socialisation

It has already been mentioned that being a social creature, consumers’ learning takes

place in society and the environment. Similarly, consumers learn some basic

consumption related knowledge, skills and ideas from different socialisation agents that

influence their consumer socialisation (Kamaruddin & Mokhlis 2003; Shim 1996). A

socialisation agent may be a person or an organisation directly involved in socialisation

because of frequency of contact with the individual, primacy over the individual,

control over rewards, and punishment given to the individual (Churchill & Moschis

1979). Furthermore, socialisation agents inevitably transmit norms, attitudes,

motivations and behaviour to the learner and socialisation is assumed to be taking place

during the course of the person’s interaction with different agents in various social

settings.

Furthermore, Moschis (1988) stated a socialisation agent may communicate certain

information to the learner through various mechanisms: Firstly, by performing certain

acts, an agent may consciously or subconsciously communicate certain norms and

expectations. Secondly, a socialisation agent may influence the consumer behaviour of

others by using various reinforcement mechanisms, both positive and negative, in

attempting to communicate to others certain desires. Finally, socialisation agents may

affect the consumer behaviour of the learner through overt communication processes,

often referred to as the ‘social interaction’ mechanism.

Generally, most of the researchers agreed on certain socialisation agents who influence

human behaviour as consumers. Various researchers (Chavda, Haley & Dunn 2005;

Dotson & Hyatt 2005; Lueg & Finney 2007; Moschis & Moore 1979) identified family

(parents), peer groups, schools, and mass media as the four major consumer

socialisation agents. In addition, some other researchers identified religious institutions

and community as consumer socialisation agents (Berns 1997; Lloyd 1985).

33

These socialisation agents play a tremendous role to develop a human being as a mature

person. Similarly consumer socialisation agents bring basic knowledge about

consumption that helps consumers to develop analytical ability and assist them to act as

a consumer in the market place (Granhaug & Venkatesh 1986; Vandell 2000). The basic

socialisation agents have been found to exert both informative as well as normative

influences on consumers (Choi & Ferle 2004). However, the role of various consumer

socialisation agents may vary in different cultural and social settings. Therefore,

significance of consumer socialisation agents has great value for the future researchers

and market practitioners.

2.8.1 Family (Parents)

In the extant literature family is considered as one of the important agents in the

socialisation (Moschis & Moore 1979; Neeley 2005). Particularly, parents as an integral

part of family are the most important role player in this process (Dotson & Hyatt 2005).

Grusec and Devidov (2007) mentioned the necessity of parents in this process is

considered for the following reasons: Firstly, children and parents are part of a biosocial

system that functions to protect offspring and to ensure that they are able to deal with

demands of social life. Secondly, the strong human need for interrelatedness plays a

substantial role in the socialisation and opportunities for such interrelatedness abound in

the parent-child relationship, as parents protect, nurture, and express affection and

warmth to their offspring. Thirdly, in most societies parents are formally assigned the

role of primary agents of socialisation. Fourthly, practical reasons facilitate parents’

motivation to socialise their children given that they must live in close proximity to

these children and that the lives of all are more comfortable when there is some

agreement about the nature of appropriate behaviour. Finally, parents are in a position

in which they can control resources available to their children as well as manage their

environments to ensure that they are either protected from or forewarned about

undesirable influences. The study of family process and children’s development was

focused mainly by the behaviourism and psycho-analytic theories (Maccoby 1992). In

the behaviourism theory, parents are considered as teachers and children are learners.

34

Parents also set the behavioural agenda for their children and teach them accordingly.

Further, parents set contingent behaviour for their children to discriminate various

situations. However, parental socialisation research is typically mentioned within social

learning theories that suggest that parents may use relationships with children to fulfill

their own emotional and social needs (Carlson, Laczniak & Walsh 2001).

Socialisation, often deals with adult-initiated processes by which children acquire the

habits and values congruent with their culture (Carlson, Laczniak & Walsh 2001). They

also added that parents may influence teenaged children by nurturing parent child

relationships, conveying values, social achievement, and directly controlling children’s

behaviour within a culture that develops teenagers’ rational, socially desirable and even

effective consumption related behaviour (Moschis & Moore 1979)

Being a socialisation agent, parental communication with their teenaged children has

been suggested in various socialisation researches with significant weights, which are

generally viewed as being composed of two underlying orthogonal dimensions: a)

socio-oriented, and b) concept-oriented communication, that monitor and control the

children’s consumption related activities (Bristol & Mangleburg 2005a; Caruana &

Vassallo 2003; Moschis 1981; Neeley 2005). On one hand, socio-oriented

communication focuses maintenances of harmonious social relations and defence to

parental authority. On the other hand, concept-oriented communication stresses

acquisition of information and children’s problem solving abilities. When these two

orthogonal dimensions are combined into a typology, then four distinct patterns of

family communication emerge (Bristol & Mangleburg 2005a; Carlson, Grossbart &

Stuenkel 1992; Caruana & Vassallo 2003): a) protective family communication pattern

(high socio-oriented communication and low concept-oriented communication) that

stress obedience and conformity on the part of children with little concern for

developing children’s problem solving abilities, b) consensual family communication

pattern (high socio-oriented communication and high concept-oriented communication)

that encourage children to develop their own views, but to do so without challenging

parental authority, c) pluralistic family communication pattern (high concept-oriented

35

communication and low socio-oriented communication) that encourage children to

develop their own views of the world without placing constraints on children, and

finally d) laissez-faire family communication (low patterns concept-oriented

communication and low socio-oriented communication) that reveal little parent- child

communication. The following Table 2.1 shows the pare-children communication

pattern addressed in consumer socialisation researches:

Table 2.1 Parent-Children Communication Patterns (Source: Chan & McNeal

2003)

Socio-oriented communication

Low (mean <=3.0) High (mean>3.5)

Low

(mean<=3.0)

Concept-oriented

communication

High

(mean>3.0)

Moreover, Moschis (1985) noted that for the following reasons parental communication

and styles with children are very important in the consumer socialisation: First, children

imitate and observe from their parents through conscious and subconscious

communication. Second, parents may influence the consumer behaviour of children by

various reinforcement mechanisms, both positive and negative. Finally, through social

interaction process children learn consumption related behaviour from parents. The

following diagram shows the link socialisation, communications, and offspring

outcomes.

Laissez-Faire

Little communication with children;

Little parental impact on

consumption

Protective

Stress vertical relationships;

Obedience and social harmony;

Children have limited exposure to

outside information

Pluralistic

Stress horizontal relationships; Issue-

oriented communication; Children

are encouraged to explore ideas and

express opinions

Consensual

Maintain control over children’s

consumption; Frequent

communication with children about

consumption

36

Diagram 2.1 Links among Socialisation, Communications, and Offspring Outcomes

(Source: Carlson, Grossbart & Stuenkel 1992).

Parental socialisation Communication pattern Outcomes in children

Further, parental coaching helps their teenagers to acquire certain consumption related

skills: Firstly, consumer affair knowledge refers to the cognitive development of the

human being as a consumer to understand the basic practices in market. Secondly,

consumers’ response to buy any product from the market in a desirable way, and finally

materialism; which refers to the inclination towards the money for personal happiness

and social progress. Moschis (1985) one of the pioneer consumer socialisation

researchers also indicated in his research that teenaged children learn consumption

related behaviour from their parents and this learning occurs through direct and indirect

influences by their parents. Neeley (2005) also mentioned that parents influence

consumer socialisation directly by instruction and indirectly by the observation of teens.

Direct effects of family are grouped into three categories: a) the content of learning

(what young people learn from their parents without reference to specific

communication process, cognitive or overt), b) the effects of various types of family

communication processes (how the family influences consumer learning), and c) the

effects of specific structures and pattern of overt family communication process.

Indirect effects of family influences viewed by the parental exposure towards media

usage and parental views about materialism. Overall, parents influence the consumer

socialisation of their teenaged children in a couple of ways: Firstly, children observe the

parental conversation, attitudes, and pattern of expenses towards purchasing behaviour.

In this way parents act as role models for their children. Secondly, the interaction

between parents and children also transmits the consumption knowledge and that affects

both the parties (Geuens, Pelsmacker & Mast 2002). However, family structures and

patterns are changing differently in some societies. On one hand, parental separation

and family stress significantly influences teenagers’ knowledge, attitudes, and

behaviour regarding consumption behaviour. On the other hand, parental professional

business also makes their teenaged children independent and mature consumer (Geuens,

Pelsmacker & Mast 2002).

37

2.8.2 Peer Group (Friends)

In the social sciences, peers have traditionally been viewed as important social force

acting on adolescents (French, Pidada & Victor 2005). Peers also have been considered

as one of the most important socialisation agents in the teenagers’ socialisation (Bush,

Smith, Martin 1999; Kamaruddin & Mokhlis 2003). According to Elkin and Handel

(1972), peer group as a socialisation agent has certain distinctive characteristics: a) it is

made up of members who have about the same age status, b) within the peer group the

members have varying degrees of prestige and power, c) the peer group is centred on its

own concern and d) in peer group, long-run socialising implications are largely

unintentional. However, Lloyd (1985) identified the following functions of peers and

friends in the socialisation:

• Replacement for family support: In absence of parents, friends provide certain

supports to their peers.

• Stabilising influence: In the midst of the many changes that take place during

adolescence, the peer group provides an anchor (as do parents).

• Social status: Group membership confers social status. Not being a member may

be associated with social stigma.

• Source of self–esteem: Group membership is a sign of peer acceptance.

Adolescents who are positively evaluated by their peers have higher self-esteem

than youth who are less positively evaluated.

• Source of behavioural standards: Peers influence on teenage children decision-

making. In homes, where parents do not get along. or where parents are

unavailable, peer groups assume more importance.

• Source of security: To some degree peer groups protect teenage children from

demands, which parents or other adults attempt to impose on them.

• Opportunities for role-taking and feedback: Participation in group activities

about life, sex, etc., provide opportunities for trying out different roles,

practicing skills important in adulthood, and getting feedback about how one’s

ideas, feelings, and behaviour appears to others.

38

• Opportunity for modelling: Especially in those cases where parents cannot be

useful role models for their teenage children, peers can serve as important

models of behaviour.

Moreover, Lloyd (1985) added that peers may also be a source of challenge, conflict,

victimisation, and rejection. Various adolescents’ researchers (French, Pidada & Victor

2005) refer to the term ‘intergenerational influences’ (measured by frequency and

nature of interaction with friends) that shows peers and friends transmit, information,

beliefs, and resources to their groups and friends. As a socialisation agent, peers have

been found to exert both informative and normative influences on teenaged consumers.

Normative influence is reflected in a consumer's willingness to conform to the

expectations of others (Lachance, Beaudoin & Robitaille 2003). Whereas, informational

influences are reflected in a tendency to learn about products and brands from others or

acquire consumption-related knowledge based on the observation of the behaviour of

others (Choi & Ferle 2004).

Different research of consumer socialisation, considered friends as the most powerful

and important aspect of the peer group (French, Pidada & Victor 2005; Strong &

Eftycia 2006). The significance of friendship is measured by different dimensions such

as: a) friendship closeness, b) instrumental aid, c) enhancement of worth, and d)

exclusivity and extensivity of social contact (French, Pidada & Victor 2005). According

to French, Pidada and Victor (2005), friendship is viewed on the basis of

individualism/collectivism or interdependence/independence from the various cultural

points of view. However from the theoretical perspective, the oldest arguments about

the effects of peer relations can be found in psychoanalytic models (Bukowski,

Brendgen & Vitaro 2007). According to them, the impact of peers would be seen during

the adolescent process of individualisation, whereas adolescents restructure their

childhood relationships with their parents and strive to achieve qualitatively different

relationships with peers. More prominently, social learning theory suggested that peer

group and friends are instrumental in shaping an individual’s behaviour (Bush, Smith &

Martin 1999). Social learning perspective suggests that through modelling and

39

observational learning, children acquire skills and behaviours during interactions with

family members that carry over to their peer relationships (Dotson & Hyatt 2005).

Furthermore, social comparison theory added a new dimension that teenagers need to

evaluate some of their perceived consumption related knowledge acquired from their

parents, by comparing it with the knowledge of other persons who are likely to have

similar value perspectives about consumption such as the peer group and friends

(Dotson & Hyatt 2005).

However, peers’ communication through the social interaction mechanism is less

specified as to the exact type of learning involved. It may involve a combination of

spoken (reinforcement) or unspoken (modelling) messages that peers share with each

other (Lueg & Finney 2007; Mcleod & O’ keefe 1972). Teenagers’ communication with

their peers and friends is defined as overt peer-adolescent interaction concerning goods

and services (Lueg & Finney 2007). Being consumers, peers and friends possess

consumption knowledge that influences the consumer socialisation of teenagers (Choi

& Ferle 2004). Williams and Burns (2001) mentioned that peers have found to influence

affective or expressive aspects of consumption and to be instrumental in teenagers’

consumption behaviour. Peers and friends also influence teenagers’ attitudes towards

advertised products and brand sensitivity, and response to retail shoppers’ offers

(Byoungkwan, Salmon & Hye-Jin 2007; Lachance, Beaudoin & Robitaille 2003;

Williams & Burns 2001). On the contrary, Moore and Bowman (2006) indicated that

peer groups and friends are also influencing deceptive expenditure and bad cash

management forces that are associated with materialism and antisocial behaviour.

40

2.8.3 Schools

In the current literature, school is considered as a very important socialisation source for

teenaged children. School is not only important for academic purposes but also to

develop a social setting for teenaged children. Educational institutions have the role of

educating and socialising citizens so that teenaged children can be productive members

of the society. Lloyd (1985) mentioned that schools have the following general

functions in a society: a) caretaking and managing students, b) teaching required skills

and knowledge, c) transmitting cultural values, beliefs, and tradition, and d) sorting,

classifying, and evaluating students. Though objectives of educational institutions are

primarily academic in nature, pre-schools and later schools also help develop social

values of children (Wentzel & Looney 2007).

However, in many respects schools provide social experiences that are highly similar to,

and overlap with, those provided by families, the broader community, and peer groups.

Since there are no distinctive theories and socialisation models proposed in the context

of schools, social learning theory and the available models of socialisation studies are

also considered in this regard. However, research on family socialisation that also

approaches schools suggested at least three general mechanisms, whereby social

resources and experiences might influence competent functioning (Wentzel & Looney

2007). Firstly, the structure and general feature of social contexts afford opportunities

and resources that can directly support or hinder competence development. Secondly,

ongoing social interactions teach teenaged children about themselves and what they

need to do to become accepted and competent members of their social world. Within

the context of these interactions; teenaged children develop a set of values and

standards or behaviour and goals they strive to achieve. Finally, the qualities of

teenagers’ social relationships are likely to have motivational significance. When

considered with respect to educational settings, these mechanisms reflect the nested

quality of teenage children’s experience at school.

41

Further, teenaged children’s social interaction and dyadic relationships with teachers

and peers in schools describe the more proximal contexts that can influence students’

social and behavioural adjustment. However, as mentioned earlier, the existing

consumer socialisation studies mainly adopted two theoretical models: a) the cognitive

developmental model, and b) the social learning model. The applications of these

models are also observed in the perspective of schools. Teenagers’ cognitive

development includes formatting consumption related knowledge, skills and behaviour

that are shared and developed in schools at different qualitative stages. On the other

hand, teenagers’ social learning processes explain the formation of consumption related

knowledge, skills, and behaviour as a function of interactions with teachers and

classmates in schools. The social learning model implies that the more interaction

between the socialising agents and the individuals, the more likely learning will take

place. As children enter into the school; get the opportunity to mix up with classmates

and start to be influenced by the environment at school (Chan 2006; Kim & Baylor

2006).

According to Bennett (1995), school is the most common place for young teens to

exchange their views, norms, and beliefs, meeting new friends, learning and attending

extra curricular activities. In particular, class mates at school not only instigate the

teenagers’ excessive demand for toys and junk food but are also responsible for some

antisocial behaviour (Lloyd 1985). However, to socialise the teenaged children about

social norms and values, school structure (e.g., level of education, school size) and

course curricula at school play a significant role (Lachance & Choquette-Bernier 2004).

Interestingly, schools are focusing in their business curricular on how to use treasure

goods, develop attitudes regarding frugal use of resources, self –sacrifice to protect

property, discouragement of regrading public, and private goods wastage. Moreover,

these days schools are teaching consumption, expenditure, and economics as text

contents and children start to believe the production of goods is hard work, resources

are scarce, and to protect the benefit of the nation, that ultimately improves teenagers’

consumer behaviour (Stampfl, Moschis & Lawton 1978). Moreover, moral education

about consumption is also getting importance in the text contents (Chan 2006).

42

2.8.4 Other Socialisation Agents

2.8.4.1 Religious Institutions

Various socialisation agents such as; family, school, peers and media have got the

maximum attention of consumer socialisation researchers (Dotson & Hyatt 2005;

Kamaruddin & Mokhlis 2003). According to the ecological concept of Berns (1997)

however, religion is a part of a ‘marco system’, that also play a significant role in the

socialisation of teenagers. Lloyd (1985, p. 128) defined religion as a “unified system of

beliefs and practices relative to sacred things, uniting into a single moral community all

those that adhere to those beliefs and practices”. Every religion has certain implications

on the socialisation that influences the pattern of gender roles, marriage, divorce, sexual

behaviour, and childrearing. Most importantly, basic beliefs and attitudes towards

society are formed by religion (Lam & Shi 2008). Moreover, religion code affects one’s

dress, dietary habits, including alcohol consumption, health care, and social interaction

(Berns 1997). These days in various researches, it is also shown that religious beliefs

render the fair business practices (Lam & Shi 2008). Berns (1997) also mentioned that

religion helps people to have an identity and gives meaning to their lives. On one hand,

many religious activities reflect pride and celebration. On the other hand, a code of

religious belief also discourages excessive wastage of resources and that helps people

not to get involved with undesirable materialistic practices (Flouri 1999). Groome

(2007) also suggested that a coalition between family and school about religious

education provide better results.

43

2.8.4.2 Community

The ecological points of consumer socialisation studies defined the community or

neighbourhood on a smaller scale, but as the main micro system in which children learn

by doing (Berns 1997). The term ‘community’ is derived from the Latin word for

fellowship. Berns (1997) further mentioned that community is more than a group of

people living in the same neighbourhood, city or town under some common laws; it is

also a group of people having fellowship, friendly association, mutual sharing, and

common interests. Berns (1997) also added that the need for community is

psychological, practical, and economical, which also serves the following functions: a)

production, distribution, and consumption, b) socialisation, c) social control d) social

participation, and e) mutual support which plays a significant role in the socialisation of

teenaged children. Community critically socialises human beings in terms of cultural

and social exchange relationships. Community can also have an informal social support

system of the family–relatives, friends and neighbours, who can be counted on to help

in crisis.

2.8.5 Media

As a consumer socialisation agent, the media has got the maximum attention of

researchers in current consumer socialisation research (Bartlett, Griffith & Baddian

2008; McNeal & Ji. 1999). Potter (2005, p. 16) has been defined media as the

“technological means of disseminating the message”. Katz (2006) mentioned media as a

means to convey a specific kind of information. He also added that media mainly

fulfills two basic purposes, a) entertainment and b) information that eventually

socialises consumers. In particular, advertising and editorial program content of the

mass media provides knowledge and guidance that helps consumers’ behavioural

development. Moreover, the mass media educates children about new brands and

products as well as how to use them (McNeal & Ji. 1999). These days, the media is not

only helping to develop social norms and values (Arnett 1995; Attewell, Suazo-Garcia

& Battle 2003; Orlik 2000), but also developing teenagers’ consumption related skills

44

(e.g., attitudes towards shopping and brand product purchase behaviour) (Lachance,

Beaudoin & Robitaille 2003; McNeal & Ji. 1999). Furthermore, the media helps in the

socialisation through entertainment, information, identity formation, and cultural

identification (Arnett 1995). Arnett (1995) also identified a key aspect of media

socialisation that is the main difference between the media and other socialisation

agents, which is, teenagers are mostly bound to adopt other socialisation agents, but

they can choose their preferred media. However, it is important to note that teenagers’

media preferences are also affected by the economic and social structure of society

(Choi & Ferle 2004).

From theoretical perspectives; consumer socialisation by media is predominantly

explained by social learning theory. Social learning theory conceptualises the idea of

modelling that involves imitation either through a conscious attempt to emulate the

socialisation agent (e.g., media) or because the agent’s behaviour is the most salient

alternative open to a person (Lueg & Finney 2007). This type of learning process has

also been referred to as observational and imitation learning (Bandura 1977; Lueg &

Finney 2007). Further, social learning theory states that socialisation agents are

instrumental in shaping an individual’s attitude and behaviour (Cram & NG 1999).

Furthermore, the usages of displacement and cultivation theories are observed in the

process of consumer socialisation by media in the current literature of consumer

socialisation. The displacement theory states that cognitive development of teenaged

children does not depend on the amount of TV viewing, but rather on content of TV

(Anderson et al. 2001; Chernin 2005). On the contrary, cultivation theory suggests that

frequent TV viewing increases the likelihood of the adoption of expressed values and

beliefs that ultimately alter viewers’ behaviour (Brown 1993; Brown & Steele 1995).

Cultivation theory mainly suggests that teenagers’ learning from TV is strongly

associated with the volume of watching and involvement with TV (Gruber & Thau

2003).

45

Recently, electronic media has got maximum attention from researchers, rather than

other media (Bellafante 1995; Bloxham & Prieb 2007), and is becoming the most

important agent into teenagers’ socialisation (Kamaruddin & Mokhlis 2003; Moschis &

Moore 1979). Electronic media are the media that utilize electronics or

electromechanical energy to enable the end user (audience) to access the content. TV in

particular has got maximum attention from various consumer socialisation researchers

(Banks & Gupta 1980; Dotson & Hyatt 2005). In the extant literature, TV has been

addressed not only for children’s’ development, but also for developing their

consumption related behaviour (Chan 2006, Russell, Norman & Heckler 2004).

Though the use of TV is varied on the basis of requirements, the basic function of the

TV is to socialise people as social beings, as well as to cognitively develop consumers

(Banks & Gupta 1980; Churchill Jr & Moschis 1979; Moschis & Moore 1982). On the

one hand, the massive use of TV has brought fabulous informational opportunities for

consumers (Anderson et al. 2001). On the other hand, TV is also accused for expediting

excessive expenditure and an unhealthy lifestyle (Chan, Zhang & Wang 2006; Reading

2008; Yavas & Riecken 1990). Teenagers’ attitudes towards the consumption behaviour

are also affected by TV (Russell, Norman & Heckler 2004). Furthermore, consumption

related values that include materialism and desire towards product consumption are also

influenced by TV (Fletcher 2006; Gruber & Thau 2003).

The impact of TV on consumer socialisation is generally viewed from different

perspectives. Many studies have reported several negative consequences of TV viewing;

including increases in obesity, attention deficit problems, aggression, and decreases in

several measures of academic achievement (Alarcon 2004; Anderson et al. 2001). In

contrast, other studies have shown that some types of TV programming are associated

with teenagers’ developmental outcomes (Keith 1986; Sotirovic 2005). Most

importantly the outcome of TV depends on the content and program vehicle (program).

However, parental control, peers' influences, gender and social class differences,

also have an important role in the consumer socialisation through TV (Belk 1991;

Fujioka & Austin 2002; Keith 1986; López-Sintas & García-Álvarez 2006).

46

2.9 Media Type

The term media covers a wide range of variety and a broad line that includes television

(TV), newspapers, magazines, and radio (Katz 2006). So far, electronic and printed

media are considered as mainstream media by various researchers (Mcquil 1987; Potter

2005). Further, internet, mobile phone and community radio are considered as

alternative media (Senevirante 2007). Today's electronic and printed media, including

billboards, coupons, and advertisements are also utilised in the form of specialised

media (Katz 2006).

Table 2.2 Classification of Media (Source: Rahman 2006)

Print Media Electronic Media

Magazines-consumer, Farm, Business Television, broadcast, cable, syndication

spot

News papers-National, Local Radio-Network , Local

Outdoor bill board Internet

Direct mail

Yellow page

Mcphee (2004) addressed that the past 10-15 years have seen enormous shifts in the

way people interact with mass media, especially with visual media. On the one hand,

magazines and newspapers have always been under their readers’ control, and readers

choose what to read. On the other hand, TV is more passive, and TV networks decide

what programs to air, and when. Recently, due to technological advances (e.g., DVR,

Satellite channel), it is viewers who have become the program schedulers to see what

and when to watch (Katz 2006). So, either way, media would be differentiated as

controlled (e.g., Magazine, News papers) and non-controlled (TV, Radio) (Katz 2006).

47

Further, Mcphee (2004) differentiates media as hot media and cool media: a) A hot

media is one in which a single sense is extended in high definition, (e.g., with a lot of

data). According to the above definition; photographs, printed text, and radio are

characterised by high definition. Hot media, by providing information in detail, are low

in participation, since the audience does not have to ‘fill in’ missing information. b) A

cool media is one where little information is provided to a particular sense (e.g., speech,

telephone, cartoons, television, and computer). They are characterised by low definition

since the audience has to fill in a lot of information like in a conversation or telephone

call, when interpreting a cartoon or when piecing together a picture from the pixels on a

television or computer screen. However, each of the major media type has got

distinctive features shown on Table 2.3:

Table 2.3 Features of Various Media (Source: Kotler et al. 2007)

Medium Advantages Limitations

Newspapers Flexibility, timeliness, good

local market coverage, broad

acceptance, high believability

Short life, poor reproduction

quality, small ‘pass-along’

audience

Television (TV) Combines sight, sound and

motion, appealing to the senses,

high attention, and high reach.

High absolute cost, high clutter,

fleeting exposure, less audience

selectivity.

Radio Mass use, high geographic and

demographic selectivity, low

cost.

Audience presentations only,

lower attention than television,

non-standardised rate structure,

fleeting exposure.

Magazines High geographic and

demographic selectivity,

credibility and prestige, high

quality reproduction, long life,

good ‘pass-long’ readership.

Long ad purchase lead times

some waste circulation, no

guarantee of position.

Online High selectivity, immediacy,

interactive capabilities, easy to

measure number of exposures.

Demographically skewed

audience, audience controls

exposure, less effective at

conveying emotional message.

48

2.9.1 Television

In the current research of consumer socialisation, TV has got the maximum attention of

researchers among other media types. Moreover, TV has been identified as the most

important consumer socialisation agent compared to other media in the consumer

socialisation (Bearison, Bain & Daniele 1982; Cogdon 2007; Dotson & Hyatt 2005).

TV has been considered as the most exciting electronic media since 1950, but initially,

the affects of TV used to be viewed negatively on child development (Abelman 2004).

Recently, TV has been portrayed as an educational tool for pro-social learning (Burke

2005; Troseth, Saylor & Archer 2006). Being an educator, TV certainly affects the

consumer socialisation of teenagers and helps them to act as consumers in the market

(Tufte 2003). The role of TV as a consumer socialisation agent has been widely

researched in the consumer socialisation (Yavas & Riecken 1990; Sullivan et al. 2007).

In particular, outcome variables of the consumer socialisation (e.g., consumption related

cognition, attitude, values) are addressed by the researchers, which are the result of

involvement with TV. So far, consumer socialisation researchers have addressed that

involvement with TV which develops teenagers’ consumption related cognition

(Bartlett, Griffiths & Bardin 2008; Ward & Wackman 1974), influences consumption

related attitudes (Hastings & Aitken 1995) and transmits consumption related values

(Chan 2003; Kamineni 2005).

Overall, as one of the major media types, TV has a significant affect on teenagers’

consumer socialisation in both developed and developing countries (Bhosale & Gupta

2006; Chan 2003). Although, the role of TV in consumer socialisation has not been

addressed adequately in the context of developing countries.

49

2.9.2 Media Vehicle

Media vehicles are specific media within each general media type such as specific

magazines, TV shows or radio programs (Kotler et al. 2007). For example, TV vehicles

include programs such as Big Brother, American Idol, and Amazing Race. Whatever

the medium is, the media vehicle is always telecasted or published to focus on a

particular market segment (audience) (Bushman 2005; Churchill Jr & Moschis 1979).

Katz (2006) mentioned that media vehicles socialise consumers, either by providing

information or entertainment, which are the most common features for any media

vehicle. In particular, a TV vehicle contains a variety of programs like Sesame Street

(informational), American Idol (interactive and entertainment), and Big Brother (reality

show), that are highly popular in the teenaged market (Anderson et al. 2001; Stanley

2003).

TV plays a significant role in teenagers’ social, emotional, and physical development

(Sotirovic 2005) and the TV vehicle particularly socialises teenaged children through

verbal messages and visual cues (Englis, Solomon & Olofsson 1993). Informational TV

vehicles (e.g., Sesame street) help teenaged children to develop social norms, values,

and academic achievement (Anderson et al. 2001). On the one hand, TV vehicles help

teenaged children’s’ cognitive development through information processing skills,

analytical abilities, and creativity (Abelman 2004). On the other hand, street shows like,

‘Intervention’ instigate teenagers’ antisocial behaviour. In particular, some TV vehicles

like ‘America Idol’, ‘Big Brother’ are attracting the attention of teenagers these days,

that also increase and influence teenagers consumption related behaviour (Arnett 1995;

Burke 2005). Currently, interactive shows like 'American Idol' and 'Celebrities Talk

Show' in TV are new sensations that connect the audience directly and provoke

teenagers’ consumption behaviour (Englis, Solomon & Olofsson 1993; Levy &

Nebenzahl 2006). Moreover, musical programs, life style shows and particularly reality

shows are the key sources of teenagers’ consumption imagery (Englis, Solomon &

Olofsson 1993).

50

2.9.3 TV Vehicle - Reality TV (RTV)

Reality TV (RTV) is a generation of TV programming which presents purportedly

unscripted dramatic or humorous situations, documents actual events, and features

ordinary people instead of professional actors. The concept of reality TV (RTV) is quite

contemporary in the electronic media. In the 1992 MTV, the USA experimentally

broadcasted RTV first, and at that time, was known as a documentary program. In fact,

the growth of RTV is in its early stages (Deziel 2004). Therefore, not much academic

work has been found in literature (Yoon & Garma 2006). In this research, mostly text

books, news paper articles, interviews of media experts, and the information from the

media websites were considered to discuss the concept and impact of RTV. The

following Table 2.4 shows the classification of RTV.

Table 2.4 Classification of reality TV (RTV) (Source: reality tv.com 2008)

A) Documentary-Style 1) Special living environment: Some

documentary-style programs place cast members,

who in most cases previously did not know each

other and stay in an artificial living environment

for competition.

2) Celebrities: Another subset of fly-on-the-

wall-style shows involves celebrities. Often these

show a celebrity going about their everyday life:

examples include The Anna Nicole Show, The

Osborne’s.

3) Professional activities: Some documentary-

style shows portray professionals either going

about day-to-day business or performing an entire

project over the course of a series. No outside

experts are brought in (at least, none appear on

screen) to either provide help or to judge results

(e.g., Miami Ink, American Chopper).

51

B) Elimination/Game Shows 1) Dating based competition: Dating-based

competition shows follow a contestant choosing

one out of a group of suitors (e.g., Rock of Love)

2) Job Search: In this category, the competition

revolves around a skill that contestants were pre-

screened for. Competitors perform a variety of

tasks based around that skill, are judged, and are

then kept or removed by a single expert or a panel

of experts (e.g., Apprenticeship)

3) Fear Centric: fear-centric shows place people

in situations or locations aimed at generating

emotions of fright, panic, or revulsion (e.g., Fear

Factor).

4) Sports: These programs create a sporting

competition among athletes attempting to

establish their name in that sport (e.g., The

Ultimate Fighter).

5) Talent Hunt show: Another type of reality

TV is "reality-competition", or so-called "reality

game shows", in which participants are filmed

competing to win a prize, often while living

together in a confined environment. In many

cases, participants are removed until only one

person or team remains, who/which is then

declared the winner. Voting is done by the

viewing audience, the show's own participants, a

panel of judges, or some combination of the three

(e.g., American Idol, Big Brothers).

C) Others Self-improvement/makeover, Renovation, Dating

shows, Talk shows.

52

So far, teenagers are the fastest growing viewers and are considered to be the prime

audience of RTV (Tedesco 2007). Mellor (2003) also addressed that since 2000, RTV is

the top preference of teenagers.

RTV is also considered to be a source of joy, entertainment, and happiness in teenagers’

daily life (Mellor 2003). On the contrary, some critics are observing that the craze of

RTV is not like it was in its early stages (Tedesco 2007) and teenagers are not as keen

on RTV programs anymore. However, TV channels are still bombarded by the various

RTV programs. In particular, musical talent hunt programs and interactive realty games

shows are the most popular among teenagers (e.g., American Idol, Close Up

Bangladesh, Who Wants to be a Millionaire) (Berman 2008; Tedesco 2007). Initially,

mostly celebrities were the participants of RTV, but recently the trend has changed and

ordinary people are the prime focus of RTV (Hibberd 2005). In fact the reality is,

teenagers are still big followers of RTV celebrities and have emotional attachment to

them (Dolliver 2006; Idato 2007).

However, the affect of RTV has received a mixture of appreciation and criticism. At the

participants’ end, RTV has the potential to turn its participants into national celebrities

at least for a short period. This is the most notable in talent-search programs such as the

'Idol' series, which has spawned music stars in many of the countries in which it has

aired, as viewer teenagers are taught some knowledge, skills, and values by the RTV

that affects their life-style, including consumption behaviour (Dolliver 2006;

Middendorp 2008). RTV is also maturing the teenagers as consumers regarding their

tastes, preferences, and brand item selection (Mellor 2003). It has also been observed

(Berman 2008; Keenan 2008; Scardino 2005) that teenagers are quite keen on new hair

styles, dress design, outfits, cosmetics and the body image of RTV celebrities. RTV also

influences the sense of accomplishment and freedom of teenagers as consumers, while

they take their purchase decisions (Adams, Sutherland & Johnson 2007; Schwartz,

Newsome & Mathieson 2007). To some extent, some popular RTV (e.g., Biggest

Looser) helps teenagers to understand real life problems and help them to find out the

possible solutions (Dale 2007). Moreover, RTV develops creativity and rationality of

53

teenagers as consumers (Johnson, Cohen & Brook 2007). Further, it is not unlikely that

some teenaged participants express their desire to become celebrities of RTV and want

to make it their career (Armstrong 2006; Mellor 2003). Interestingly, some parents

believe RTV is a good educator to guide their children and they really appreciate

information based RTV (Pursell & Hibberd 2007; Robinson 2005). Also, some RTV

celebrities like Opera Winfrew is influential even in the national level (Lowry 2008).

On the contrary, critics blame RTV for bringing unwanted situations into consumers’

lives (Yoon & Garma 2006). Critics are also indicating that RTV is badly affecting

teenagers’ social and consumption behaviour. Unfortunately, much of the RTV shows

excessive sexual content negatively, which affects teenagers’ outfits, hair styles, and

cosmetics purchase decisions (Idato 2007). Moreover, RTV involvement really hampers

participants' academic careers (Middendorp 2008; Wilder 2008). Unfortunately, RTV is

also accused of showing violence, conflict, and sexual tension that also triggers

teenagers’ aggressive and anti-social behaviour (Dubecki, Goodings & Mathieson

2008). Even the life style of RTV celebrities instigates the alcohol and drug

consumption of teenagers (Mcwhirter 2007). However, the gimmick of RTV is still on

and a lot of newspapers, websites, and channels are covering this unscripted new trend

of TV genres (Frutkin 2008).

It has already been mentioned that RTV is a new generation of TV programming.

Recently, RTV has got attention of viewers because of its special features like program

layout (Frutkin 2008), reward options for audience and participants (Bown 2008),

format (James 2007), various content (Lundy & Jacobson 2008) and nature of

participation (Jacobs 2008). These key features make RTV different than other general

TV programs. Unlike traditional TV programs, in RTV not only celebrities but also

ordinary people, often teenagers’, are participants. Also, unlike traditional TV

programs, mostly two-way communication is observed between RTV and its audience

(James 2007). This interactive nature of RTV, as opposed to traditional non-interactive

types, ensures greater involvement of audience who are often teenagers. Widespread

use of reward systems in RTV, coupled with its content which are often varied and

54

close to participants day-to-day life, i.e., ‘reality competition’, ‘reality game shows’,

ensure greater audience involvement with RTV. Unscripted nature of RTV, as opposed

to traditional scripted ones, also makes it more dramatic and realistic to its audience.

The fact that RTV is often live also makes it more realistic and ensures greater audience

involvement than traditional TV. As viewers and participants, teenagers gain

consumption related knowledge, skills, and values through different content and their

interaction with RTV (Dolliver 2006; Middendorp 2008). RTV also influence teenagers

as consumers regarding their tastes, preferences, and brand item selection (Mellor

2003). Past researches show (Berman 2008; Keenan 2008; Scardino 2005), teenagers

are quite keen on new hair styles, dress design, outfits, cosmetics and the body image of

RTV celebrities, as viewers can join in the live shows. RTV also influence teenagers’

sense of accomplishment (Adams, Sutherland & Johnson 2007; Schwartz, Newsome &

Mathieson 2007). This happens due to widespread use of ‘contests’ and ‘rewards’ for

participants in RTV. Participation and wining rewards in these contests often provide

the sense of accomplishment to the teenagers’. To some extent, some popular RTV

(e.g., Biggest Looser) helps teenagers to understand real life problems close to their

hearts and help them to find out the possible solutions (Dale 2007). Overall, RTV has

more options than ordinary TV programs to influence teenagers’ involvement in it and

thereby influence their consumer socialisation. Generally, the set of identifiable

components associated with RTV that are based on current literature are given bellow:

55

Table 2.5 Domain of RTV Observables and Their Dimensions (Developed for this

Research)

Domain of RTV Observables Dimensions

A) Participants

Not only celebrities but also ordinary people are

participants of RTV. Sometimes mature people are

also observed in the RTV shows along with

teenagers (Jacobs 2008; Schrobsdorff 2008)

1) Celebrities & Ordinary people

B) Content

Content of RTV is quite varied and has got a

mixture of responses from the media critics

(Lundy & Jacobson 2008)

1) Suspense

2) Violence

3) Sexually appealing

4) Talent hunt program

5) Information

C) Format

Mostly two way communication

is observed between RTV and it’s audience while

others are of a one way type (James 2007)

1) Interactive

2) Non-interactive

D) Reward System

Teenagers have keen interest in the reward given

in the RTV. It might be a monetary or non-

monetary reward (Bown 2008; Driscoll 2007)

1) Gift item

2) Prize money

E) Program layout

Mostly anchor and participants play major role in

RTV which is not directed through script (Frutkin

2008).

1) Non scripted

56

2.10 Conclusion

In the above sections of this chapter, literatures of parent and immediate disciplines

linked to consumer socialisation have been addressed. Furthermore, theoretical grounds

of consumer socialisation research were discussed, followed by the role of various

consumer socialisation agents, which are mainly discussed in the extant literature from

the perspective of developed countries. So far, media as one of the consumer

socialisation agents has got the maximum attention of researchers in the current

consumer socialisation research. Particularly, the role of TV as a consumer socialisation

agent has been widely researched. Overall, TV commercials (TVC) and the different

content of TV programs are mostly considered while conducting consumer socialisation

research, which has been rarely examined from the perspective of developing countries

(Bhosale & Gupta 2006; Choi & Ferle 2004). Recently, RTV has become hugely

popular among teenagers and they have involvement with different RTV shows.

However, as a new generation of TV vehicle, RTV has never been focused on as a

consumer socialisation agent in the current literature of consumer socialisation (Yoon &

Garma 2006). Furthermore, RTV as a consumer socialisation agent has so far been

overlooked in consumer socialisation research, not only in developing countries, but

also in developed countries (Deziel 2004). Moreover, Yoon and Garma (2006)

addressed that there is no empirically tested model of consumer socialisation by RTV in

current literature. The following chapter will primarily address how TV, particularly

RTV, does affect consumer socialisation of teenagers in the context of a developing

country.

It has been addressed that teenagers’ involvement with RTV is largely observed in

developed countries (Berman 2008). There is also anecdotal evidence that suggests that

teenagers in developed countries are big followers of different life-style based shows;

celebrities’ talk shows and talent hunt shows that affect their consumption related

behaviour (Cajueiro 2008; Lowry 2008). Teenagers in developed countries have higher

accessibility and various options of using different media than their developing

countries counterparts. So, they show a relatively higher involvement with RTV

57

through different sources of media because of it’s interesting content (e.g.,

documentaries, game shows, celebrity talk shows, lifestyle shows, talent hunt shows)

(Lundy & Jacobson 2008). On the other hand, teenagers in developing countries are

getting involved with electronic media, particularly TV (Bhosale & Gupta 2006; Nath

2006). More recently, teenagers in developing countries are showing their involvement

with RTV (Rahman 2008). So, RTV as one of the consumer socialisation agents can be

considered in consumer socialisation research.

In this chapter, the parent and immediate disciplines have been discussed and their

relevance to this research has been analysed. Furthermore, the role of the media has

been discussed specifically to help understand the consumer socialisation of teenagers

by TV. Particularly, as one of the latest TV vehicles, the role of RTV has been

focused on conceptualising the issues of consumer socialisation. In the next chapter, the

current literature relevant to the key constructs of this research will be discussed,

analysed and a proposed conceptual model developed.

58

Chapter 3 Conceptual Framework and Hypotheses

3.1 Introduction

In the previous chapter, the parent and immediate disciplines of consumer

socialisation have been discussed and analysed. The key focus of this chapter will be to

develop a conceptual model of consumer socialisation by RTV and how as a consumer

socialisation agent, RTV affects the consumer socialisation of teenagers in the context

of a developing country. Furthermore, the role of social structural variables will be

addressed to understand their role in the consumer socialisation.

3.2 Involvement with Media

The focus of this research is to assess the impact of teenagers’ involvement with RTV

on their consumer socialisation. Zaichkowsky (1985) defined involvement as “A

person’s perceived relevance to the object based on inherent needs, values, and

interest”. So, teenagers may get involved in watching, thinking and participating in

RTV for getting fun, pleasure and in the process develop consumption related

knowledge and skills.

Zaichkowsky (1985) defined involvement as, “a person’s perceived relevance to the

object based on inherent needs, values, and interest.” This definition can be applied to

RTV studies. Literature also indicates that a person can be involved with advertisements

(Krugman 1966) and RTV (Yoon & Garma 2006). In media research, involvement

pertains to media users’ relationships with the message conveyed by media or with

media (Levy & Windahl 1984), media personalities (Rubin & Perse 1987), or other

media users (Rubin & Step 1997). Rubin and Perse (1987) conceptualised media

involvement as “cognitive, affective, and behavioural perceptions during and because of

exposure” (p. 247). To understand involvement, one needs to understand the audience

and the role of involvement between media motive and media effects (Perse 1998).

59

Involvement and Audience activity: Lin (1993) explained that audience activity is a

construct describing audience’s involvement when using the media. Activity is mostly

manifested in people’s media motives, selectivity, and involvement with the message

provided by media (Blumer 1979). Levy and Windhal (1984) focused on three types of

activity: a) selective before exposure, b) involvement during exposure, c) and use after

exposure. Diagram 3.1 shows the links between ‘Media Motives, Involvement and

Media Use and Effects.

Diagram 3.1 Links between Media Motives, Involvement and Media Use and Effects

(Perse 1990a, 1990b)

Media motives Involvement Media use and effects.

Media involvement as a mediator: Several researchers have investigated relationships

between media motives, involvement, and effects. Perse (1990a) found instrumental

viewing motivation positively related to elaboration, whereas ritualistic viewing

motivation positively related to engaging in distracting behaviour. Perse (1998) also

found instrumental motivation predicted cognitive and affective involvement (e.g.,

positive reactions) with television viewing. In line with this thinking and research, the

relationship between media motives, cognitive and affective involvement, and media

use and outcomes can be linked and illustrated as follows (Perse 1990a, 1990b).

Diagram 3.2 shows the links between Media Motives, Cognitive and Affective

Involvement, Media Use and Outcomes.

Diagram 3.2 Links between Media Motives, Cognitive and Affective Involvement,

Media Use/Effects (Source: Perse 1990a, 1990b)

Media motives Cognitive and affective involvement Media use/ effects.

60

3.2.1 Involvement with Reality TV (RTV)

The idea of RTV originated from documentary TV programs and the growth of RTV is

still in its early stage (Biressi 2005; Deziel 2004; Hill 2005). However, RTV has got the

serious attention of teenagers because of it’s interesting content (e.g., documentary,

game show, celebrity talk show, lifestyle show, talent hunt show) (Lundy & Jacobson

2008). Furthermore, the nature of participants (e.g, ordinary people, celebrities) (Jacobs

2008), format (e.g., interactive, non interactive) (James 2007), and reward system (e.g.,

prize money, gift items) (Driscoll 2007), makes RTV different and exclusive from other

TV programs.

However, on theoretical grounds, cultivation theory suggests that frequent viewing with

TV increases the likelihood of the adoption of expressed values and beliefs that will

ultimately alter teenagers’ actual behaviour (Gruber & Thau 2003). Moreover, higher

consumption of TV viewing brings greater changes in values, beliefs, and resulting

behaviour. In fact, cultivation theory starts with the idea that people who watch TV for

many hours a day think differently about the social reality than people who rarely or

never watch TV (Nacos 2000). Cultivation theorists suggest that TV programming

cultivates a mainstream world view that reflects and perpetuates the social view

(Hetsroni & Tukachinsky 2006; Nacos 2000).

So far, researches on consumer socialisation in the context of RTV have been rarely

examined (Yoon & Garma 2006). Therefore, scale items in this research have been

adopted mostly from current research findings on consumer socialisation by media in

general Table 3.2.

As mentioned earlier, cognition, attitudes, and value formation towards consumption

are the three key constructs of the consumer socialisation of teenagers. In the following

sections, these three constructs have been discussed in relation to the involvement with

RTV.

61

3.3 Key Constructs of Consumer Socialisation

In the extant literature it has been generally agreed that consumption related cognition,

attitude, and values are the common outcome components of the consumer socialisation

(Chan 2003; Moschis & Moore 1979; Schmoll et al. 2006). Generally, the consumer

socialisation is mostly examined from the perspectives of vicarious learning, cultivation

and displacement theories.

From theoretical perspectives, consumer socialisation by the media, particularly TV, is

predominantly explained by social learning theory. Social learning theory

conceptualises the idea of modelling that involves imitation, either through a conscious

attempt to emulate the socialisation agent (e.g., media), or because the agent’s

behaviour is the most salient alternative open to a person (Lueg & Finney 2007).

According to social learning theory socialisation agents are instrumental in shaping an

individual’s attitude and behaviour (Cram & NG 1999; Moschis & Churchill 1978).

Furthermore, usages of displacement and cultivation theories are observed in the

process of consumer socialisation by the media in the current literature of consumer

socialisation. According to displacement theory teenagers’ exposure to TV content

expedites their cognitive development (Anderson et al. 2001; Chernin 2005). On the

contrary, cultivation theory suggests that the consumer socialisation takes place due to

involvement with the media (Brown 1993; Brown & Steele 1995). Cultivation theory

mainly suggests that teenagers’ learning from TV is strongly associated with the volume

of watching and involvement with TV (Gruber & Thau 2003).

Overall, extant literature and consumer socialisation research identifies three elements

of the consumer socialisation, like socialisation agents (e.g., family, media), social

structural variables (e.g., social class, gender) and outcomes (e.g., cognition, attitude,

values) (Shim 1996; Moschis & Churchill 1978). According to these theoretical and

conceptual findings, RTV is considered to be a socialisation agent and other social

structural variables such as social class differences, gender differences, parental control

and peer group influences are instrumental in shaping an individual’s consumption

related cognition, attitude and values outcomes (John 1999; Kamaruddin & Mokhlis

62

2003; Lueg & Finney 2007).

So, in this research, cognition (through knowledge and skills), attitudes formation, and

values formation towards the consumption have been identified as 3 key outcome

constructs of the consumer socialisation of teenagers. In the following section, these 3

outcome constructs have been discussed in relation to their involvement with RTV.

Figure 3.1 Outcomes of Consumer Socialisation

3.3.1 Cognition

Cognition is defined as “knowledge that is acquired by a combination of direct

experience and information from various sources” (Schiffman et al. 2005, p. 617).

Wagner (2008) has defined cognition as the mental processes involved in gaining

knowledge and comprehension, including thinking, knowing, remembering, judging,

and problem solving. So, consumption related cognition can be conceptualised as the

mental processes involved in gaining knowledge and comprehension; including

thinking, knowing, remembering, judging, problem solving and skill development

regarding products and brands consumption (Bartlett, Griffiths & Badian 2008; Bush,

Martin & Bush 2004; Moschis & Moore 1978). Famous psychologist Piaget believed

that cognitive development occurs in the different stages of children exclusively as

observed in different researches (John 1999; Solomon 2004) like, a) limited: bellow the

age of 6, children do not employ storage and retrieval strategies. b) cued: children

between the ages of 6 and 12 employ this strategies, but only when prompted and c)

strategies: children 12 and older spontaneously employ storage and retrieval strategies.

However, other psychologists did not agree with this cognitive development stage

exclusively, rather they think cognitive development takes place throughout the life

cycle.

Outcomes of consumer socialisation

Consumption related cognition

Consumption related attitude

Consumption related values

63

3.3.1.1 Consumption Related Cognition Development by Television (RTV

Involvement)

The cognitive development of teenaged children is significantly affected by electronic

media, particularly Television (TV) (Anderson et al. 2001; Bearison, Bain & Daniele

1982). These days, teenagers are highly involved with the different content of TV and

various TV vehicles (Acevedo-Polakovich et al. 2005; Choi & Ferle 2004). The extant

literature has indicated that TV plays a vital role in teenagers’ cognitive development

(Abelman 2004). Moreover, teenagers’ cognitive development through TV has been

explored from a social perceptive as well as from the context of consumer socialisation

(Bearison, Bain & Daniele 1982; Dotson & Hyatt 2005).

Theoretically, cognitive development as an outcome of teenagers’ socialisation by the

media is mainly explained from the perspectives of displacement theory, cultivation

theory and observational learning theory. According to displacement theory the

cognitive development of teenaged children does not depend on the amount of TV

viewing but rather on the content of TV (Anderson et al. 2001). From the findings of

different research on displacement theory, Chernin (2005) noted that as children spend

more time watching TV, they spend less time participating in more ‘valuable’ activities.

On the contrary, cultivation theory suggests that frequent TV viewing increases the

likelihood of the development of consumption related cognition that ultimately alters

viewers’ behaviour (Brown 1993; Brown & Steele 1995). Cultivation theory mainly

suggests that teenagers’ learning from TV is strongly associated with the volume of

their watching and involvement with TV (Gruber & Thau 2003). Furthermore, cognitive

development of the consumer through the media also has been discussed by social

learning theory, particularly by observational learning theory (Gruber & Thau 2003).

Observational learning theory posits that people learn from observing others’ behaviour

and practice accordingly if rewarded and reinforced (Bandura 1977; Ward &

Rivadeneyra 1999). In particular, teenaged children through observation or by imitating

their favourite characters from TV develop their consumption related knowledge, ideas

and skills that influence their consumption related behaviour (Lueg & Finney 2007;

64

Moschis & Churchill 1978).

The influences of TV on the cognitive development of teenaged children are viewed by

researchers as very significant (Bearison, Bain & Daniele 1982; Brand & Greenberg

1994). Abelman (2004) noted that young and talented children watch significantly more

TV per week and are engaged in more active and selective tele-viewing. Furthermore,

their information processing capacity, creativity, and narrative skills are viewed as a

result of involvement with TV (Zimmerman & Christakis 2005). Research findings

indicate that not only is the time given to TV, but also the TV content is significant for

the cognitive development of teenaged children (Abelman 2004; Zimmerman &

Christakis 2005).

TV as a consumer socialisation agent consolidates consumption related knowledge,

skills, capacity and consumer role perception in to the consumer socialisation. In

particular, teenagers’ information seeking experiences about various products and

services are triggered by the TV (Eastin 2005). Furthermore, TV provides information

about the market condition which creates teenagers’ awareness and helps their recalling

capacity of various products and brands (Bartlett, Griffiths & Badian 2008; Ward &

Wackman 1974). TV not only develops product and brand recalling capacity of

teenagers but also develops product and brand recognition ability (Zuckerman, Ziegler

& Stevenson 1978). It is also likely that teenagers get product usage instruction from

TV (Moschis & Moore 1978). Furthermore, Moschis and Moore (1978) added that TV

provides a clue to understand the cost benefit analysis of using any product and brand

that sometimes help teenagers’ consumption decisions.

In this research, involvement with RTV is considered as an antecedent variable of

teenagers’ consumption related cognition development. As one of the most

contemporary forms of TV vehicles, RTV brought a lot of exclusive features with it

(e.g., nature of participants, content, format, rewards, program layout) (Bown 2008;

Frutkin 2008; Lundy & Jacobson 2008; Jacobs 2008; James 2007). Therefore, the more

teenagers’ involvement in it, it is expected the more it will lead to their consumption

related cognitive development. In particular, from the perspective of cultivation theory,

65

teenagers’ involvement (e.g., time spent, attention, likings) with RTV may develop their

consumption related cognition, as RTV shows provide product related knowledge and

help to recall and understand product usage instructions (Ward & Wackman 1974;

Zuckerman, Ziegler & Stevenson 1978). The same can be said of TV shows in general.

Furthermore, from the perspective of displacement theory, various content of RTV (e.g.,

informational, entertainment) may also influence teenagers’ brand related knowledge,

consumption related skills and activities. Discussion in this section on cognition

theories and specifically cognitive development occurs as a result of vicarious rather

than direct experience, and through a process of imitating the behaviour of others called

modelling. Therefore, imitating the role models from RTV and observing different

contents, as well as getting involved in RTV are assumed to be the solid ground of

teenagers’ cognitive development that may reflect their consumption related behaviour.

Furthermore, teenagers’ interest towards electronic media, particularly TV, is observed

in developing countries (Bhosale & Gupta 2006; Nath 2006). As mentioned earlier,

previous research, though mostly developed country based, shows that TV directly

affects teenagers’ consumption related cognition (Bartlett, Griffiths & Bardin 2008;

Ward & Wackman 1974). Similarly, development of consumption related cognition by

TV is also expected among teenagers of developing countries. Furthermore, there is

teenagers’ involvement in TV in developing countries (Chan 2003; Nath 2006).

Accordingly, for this research we can hypothesise that:

H1: There is a positive impact of RTV involvement on consumption related

cognition.

Since the research on consumer socialisation within the context of RTV is rarely

examined, the domain of scale items to measure consumption related cognition has been

mostly adopted from current research findings on consumer socialisation research by

the media.

66

3.3.2 Attitude

In psychology, attitude is defined as a “psychological tendency that is expressed by

evaluating a particular entity with some degree of favour or disfavour” (Eagly &

Chaiken 2007, p.1). Solomon (2004) referred to attitude as the lasting general

evaluation of people (including one self), objects or issues. Eagly and Chaiken (2007,

p.5) also added as attitude, “an evaluative judgement or affective or evaluative

response”. Current definitions of attitude possess some key features like evaluation,

attitude to object, and tendency (Eagly & Chaiken 2007). Attitude towards the act of

buying is the perceived consequences of purchase which is heavily discussed in

consumer behaviour discipline (Solomon 2004). Overall, consumption related attitude

can be conceptualised as an enduring combination of motivational, emotional,

perceptual and cognitive process with respect to imitation, fashion, inspiration and

product placement (Hirschman & Thompson 1997; Russell, Norman & Heckler 2004;

Schmoll et al. 2006).

However, from theoretical grounds, according to social learning theory teenagers’

attitude is an outcome of the socialisation where the individual is viewed as a passive

participant in the learning process (Bush, Smith & Martin 1999). Attitude is also viewed

as an inner tendency of human beings that can be expressed implicitly and explicitly

(Eagly & Chaiken 2007). Neal, Quester & Hawkins (2004) identified three components

of attitude from existing research: a) cognitive component (an attitude consists of a

consumers’ beliefs and knowledge about an object), b) affective component (a

consumers’ feelings or emotional reaction to an object), and c) behavioural component

(the tendency of the attitude holder to respond in a certain manner towards an object or

activity). However, the formation of attitudes through affective, cognitive, or

behavioural processes establishes associations that are linked to the attitude object and

can become part of the mental residue that is attitude. These associations can reflect one

or a mixture of the affective, cognitive, and behavioural precursors of attitude (Eagly &

Chaiken 2007).

67

3.3.2.1 Consumption Related Attitudes Influenced by Television (RTV

Involvement)

The current literature of consumer socialisation reveals that teenagers’ attitudes are

highly influenced by TV (Barr & Hayne 1999; Eagly & Chaiken 2007; Russell, Norman

& Heckler 2004). TV influences a wide range of attitudinal domains that include from

macro issues (e.g., environment, economics) to particular consumption related events

(e.g., product evaluation) (Haron, Paim & Yahaya 2005). Moreover, teenagers’ attitudes

towards health issues, lifestyle and personal life are also influenced by TV (Banks &

Gupta 1980). TV also plays a vital role in influencing teenagers’ attitudes toward daily

life events, and to a greater extent, in consumption related behaviour (Bush, Martin &

Bush 2004; Donohue 1975). Generally, teenagers’ attitudinal development as a result of

involvement with TV is viewed from a social perspective followed by the context of

consumer socialisation (Banks & Gupta 1980; Hastings & Aitken 1995).

From theoretical grounds, social learning theory indicates that TV is one of the

important socialisation agents that causes teenagers’ attitudinal development and

influences their behavioural intentions (Haron, Piam & Yahaya 2005; Bush, Martin &

Bush 2004). Social learning theory mainly highlighted the observational learning of

teenagers from TV that influence teenagers’ consumption related attitude (Lueg &

Finney 2007; Sullivan et al. 2007). These attitudinal and behavioural intentions are

considered as an integral output of socialisation (Bush, Martin & Bush 2004; Moschis

& Moore 1979). According to social learning theory; teenagers mostly observe from

socialisation agents and develop their perceptions followed by particular actions (John

1999; Moschis & Moore 1979; Moschis & Churchill 1978). Furthermore, from the

perspective of displacement theory, teenagers watch various content and role models on

TV. In particular, the content shown on TV, and celebrities are the key sources of

teenagers’ attitudinal development (Hastings & Aitken 1995; Kocher & Lalos 2008;

Sullivan et al. 2007). Teenagers’ consumption related attitude development is also

viewed from the perspective of cultivation theory. Cultivation theory suggests that the

higher involvement with TV brings greater changes in values, beliefs, and resulting

68

behaviour (Hastings & Aitken 1995; Harrison & Cantor 1997; Nacos 2000). Therefore,

teenagers’ consumption related attitude can be explained through their involvement

with TV.

So far, various researchers (Choi & Ferle 2004; Churchill & Moschis 1979) noted that

teenagers’ attitude towards consuming products and services in the market places are

largely influenced by the knowledge acquired through TV. TV as a consumer

socialisation agent also influences teenagers’ perceptions about fashion, style and

clothing behaviour (Lachance, Beaudoin & Robitaille 2003; Russell, Norman &

Heckler 2004). Furthermore, teenagers’ tendency to imitate any particular role model on

TV is also influenced by TV (Russell, Norman & Heckler 2004). It is also very

common that TV celebrities create fashion consciousness among teenagers (Russell,

Norman & Heckler 2004). Even perceptions towards an attractive figure like celebrities

(e.g., slim, muscle, plastic surgery), are also triggered by TV (Wan & Faber 2003).

Moreover, teenagers’ desires towards product consumption or views on prohibiting any

unethical products are also influenced by TV (Kocher & Lalos 2008; Sullivan et al.

2007).

On the contrary, researchers often blame TV for its negative effects on teenagers’

attitudinal development (Abelman 2004). In particular, there is disagreement regarding

the credibility of information provided by TV (Barkin 2006). Various content of TV

through different vehicles is often accused of distorting teenagers’ attitudes towards the

consumption practice that includes eating disorder, unhealthy life style and excessive

expenditure. Further, previous research linked the positive relationship between TV

advertisements and unhealthy consumption habits that lead to a positive perception

towards prohibited items (Hastings & Aitken 1995; Schmoll et al. 2006). Further,

repeated exposures of unethical products (e.g., cigarette, alcohol) on TV create interest

as well as excessive thin models in different life style shows also trigger teenagers’

association with unhealthy life style (Hastings & Aitken 1995; Lawrie et al. 2006).

69

In this research, involvement with RTV is considered as an antecedent variable of

teenagers’ consumption related attitudinal development. As one of the most

contemporary forms of TV vehicles, RTV has brought a lot of exclusive features into it

(e.g., nature of participants, various content, format, rewards, program layout) (Bown

2008; Frutkin 2008; Lundy & Jacobson 2008; Jacobs 2008; James 2007) Therefore,

with more teenagers’ involvement in it, it is expected it will lead to their development

of consumption related attitudes. In particular, from the perspective of cultivation

theory, teenagers’ involvement (e.g., time spent, attention, likings) with RTV may help

develop their fashion consciousness and motivate them to imitate RTV celebrities

(Mellor 2003; Scardino 2005). Furthermore, from the perspective of displacement

theory, various content of RTV (e.g., life style shows, celebrity fashion shows) may also

influence teenagers’ perception towards using products and brands (Idato 2007; Stevens

2007). However, most of this learning occurs as a result of vicarious, rather than direct

experience, and through a process of imitating behaviour of others called modelling.

Therefore, imitating the role models from RTV and observing different content as well

as getting involved in RTV are assumed to be grounds of teenagers’ consumption

related attitudinal development.

Furthermore, teenagers’ interest towards electronic media, particularly TV is observed

in developing countries (Bhosale & Gupta 2006). As mentioned earlier, previous

research, though mostly developed country based, shows that TV directly affects

teenagers' consumption related attitudes (Hirschman & Thompson 1997; Schomoll et al.

2006). Similarly, development of consumption related attitudes through TV is also

expected among teenagers of developing countries. Furthermore, there is teenagers’

growing involvement with TV in developing countries (Chan 2003; Nath 2006).

70

In this research, teenagers’ consumption related attitudinal development was addressed

from the perspective of TV involvement. As previously identified, teenagers' attitude

formation towards consumption is one of the outcomes of their consumer socialisation.

Discussions in this section also show that researchers in the field agree that an attitude

has three components: cognition (knowing), affect (feeling) and behavioural intension

(doing). An intention to behave in a certain way is influenced by teenagers’ knowledge

and feeling towards the object, and a liked spokesperson can enhance teenagers’

feelings towards the object. Furthermore, a common and effective approach to changing

attitude is to focus on the cognitive component.

Accordingly, based on the above discussion, for this research we can hypotheses that:

H2: There is a positive impact of RTV involvement on consumption related

attitude.

H3: There is a positive impact of consumption related cognition on consumption

related attitude.

The extant literatures of consumer socialisation have so far, hardly made any

contribution regarding consumption related attitude development through RTV

involvement. Therefore, scale items to measure consumption related attitude are mostly

adapted from current research findings on consumer socialisation by the media.

Consumption related attitude is a vital outcome of the consumer socialisation, which is

one of the vital constructs in this research, and in the proposed model. Furthermore, the

interrelationships between consumption related cognition and attitude are assumed in

this research.

71

3.3.3 Values

Schiffman et al. (2005, p. 637) defined values as “relatively enduring beliefs that serve

as guides for what is considered ‘appropriate’ behaviour and are widely accepted by the

members of the society”. Values are widely held beliefs about what is acceptable and/or

desirable (Neal, Quaster & Hakins 2004). So, consumption related values can be

conceptualised as shared beliefs among group members as to what behaviours are

desirable or undesirable toward products, prohibited product consumption, materialism

and anti-social behaviour. Blackwell, et al (2007) also added values as the terminals

(goals) or instrumental (behaviour) beliefs of individuals. However, in various

socialisation research learning has been addressed as part and parcel of socialisation

(Chan 2003; John 1999; Moschis & Moore 1979). The process by which people develop

their values, motivations and habitual activities called socialisation. This learning

develops individuals’ values that are also considered as an outcome of socialisation

(Chan 2003; Moschis & Churchill 1978).

However, various researches (Homar & Kahale 1988; McCarty & Shrum 1993)

indicated that theoretical contributions of values are mediated by attitude. McCarty and

Shrum (1993) showed the interrelationship among values, attitude and behaviour. They

also indicated that attitude plays a mediating role in the value-behaviour relationship.

Moreover, McCarty and Shrum (1993) conferred that values have an indirect affect on

behaviour: Values influence attitude, and attitude in turn influences behaviour.

Blackwell et al. (2007) summarised from current literature that like attitude, values

represent consumer beliefs about life and acceptable behaviour. Unlike attitude, values

transcend situations or events and are more enduring because they are more central in

the personality structure. Values are particularly important in the need recognition stage

of consumer decision making. However, values are enduring motivations or the ‘ends’

people seek in their lives. In a sense, Marketing often provides the ‘means’ to reach

these ends.

72

3.3.3.1 Consumption Related Values Transmitted by Television (RTV

Involvement)

According to Saint-Cyr (2003), TV exists in the value system of an individual that is

also reflected in their consumption related behaviour. In the extant literature, it is

found that TV transmits not only moral values, norms and social codes, but also

develops consumption related values of individuals (Chan 2003; Kamineni 2005).

Overall, Saint-Cyr (2003) addressed that TV helps to shape the opinions, beliefs and

values of an individual, which ultimately leads to their consumption related behaviour.

Most of the studies of the role of the media in the socialisation are based on cultivation

theory and social learning theory (Gruber & Thau 2003). In particular, Gerbner’s

cultivation theory is one of the most crucial and widely applied theories addressing the

effect of media on beliefs and values (Gerbener 1990; Morgan & Shanahan 1998).

According to Gerbener (1990), “cultivation means the specific independent contribution

that a particularly consistent and compelling symbolic stream makes to the complex

process of socialisation and enculturation” (P.249). Cultivation theory particularly

suggests that frequent viewing increases the likelihood of the adoption of expressed

values and beliefs that ultimately adjust actual behaviour. From the perspective of

cultivation theory, children’s higher rate of TV consumption viewing brings greater

change in consumption related values and beliefs and resulting consumption relating

behaviours (Gruber & Thau 2003). Furthermore, social learning theory posits that

teenagers learn from observing others’ behaviour that influences their values and beliefs

(Bandura 1977; Ward & Rivadeneyra 1999). In particular, observing TV characters and

different content affects teenagers’ passive learning that ultimately influences their

consumption related values (Bushman 2005; Russell, Norman & Heckler 2004).

Teenagers’ consumption related values are also examined from the perspectives of

displacement theory (Gruber & Thau 2003; Clark, Nabi & Emily 2007). Teenagers’

interest towards consuming new brands, product consumption imagery and pester

power are largely addressed as a result of involvement with TV (Chan 2003; Malthouse

& Calder 2007).

73

Teenagers’ values towards consuming products and services in the market-place are

largely influenced by knowledge acquired through TV (Gruber & Thau 2003). On the

one hand, personal values systems affect TV consumption habits (McCarty & Shrum

1993). On the other hand, TV also shapes the consumption values of consumers (Burke

2005). Saint-Cyr (2003) mentioned that overall consumption values of teenagers are

largely transmitted by TV these days, and that has both pros and cons. In particular,

teenagers’ material values are largely influenced by the content and products shown on

TV (Shrum, Burroughs & Rondfleisch 2005; Chan & Prendergast 2007). Teenagers’

interest in getting the best and latest products shown on TV, as well as their desire to

have more pocket money from parents, are considered to measure teenagers' material

values (Chan 2003). Moreover, teenagers often use their pester power to get a particular

product shown on TV (Chan 2003; Ghani 2005).

However, TV’s role in influencing teenagers’ consumption related values can be

negative as well (Peterson et al. 2007). Chan (2003) mentioned that the grey side of

materialism has been treated as negative values connected to possessiveness, envy, and

lack of generosity, greed and jealousy. Moreover, beliefs towards deceptive

expenditure, such as consuming unethical products (e.g., cigarette, alcohol) are strongly

triggered by TV (Bristol & Mangleburg 2005b). Further, various researchers (Burke

2005; Gruber & Thau 2003; Hennessy 2006) also found that TV is responsible for

teenagers’ anti-social behaviour that includes aggression, violence and distorted

consumption behaviour (e.g., scanty outfits, eating disorders, provocative clothing).

Furthermore, TV is accused of transmitting irrelevant cultural values among

teenagers which also triggers their consumption related values (Zhang & Harwood

2002).

74

In this research, involvement with RTV is considered as an antecedent variable of

transmitting and developing consumption related values among teenagers. It has already

been mentioned that RTV offers lot of exclusive features (e.g., nature of participants,

various content, format, rewards, and program layout) (Bown 2008; Frutkin 2008;

Lundy & Jacobson 2008; Jacobs 2008; James 2007). Therefore, with more teenagers’

involvement in it, it is expected that it will lead to their consumption related values. In

particular, from the perspective of cultivation theory, teenagers’ involvement (e.g., time

spent, attention, participation) with RTV may develop their consumption related values.

Information based RTV shows different products and brands that might influence

teenagers’ desire towards consuming new brands and their overall sense of materialism

(Chan 2000; Ghani 2005). Furthermore, the rewards system of RTV has great appeal to

teenagers which might influence their sense of materialism (Gould & Gupta 2006).

Displacement theory also states that various content of RTV (e.g., elimination game

shows, lifestyle shows) may also influence teenagers’ interest towards prohibited

products (Gruber & Thau 2003; Hastings & Aitken 2005), since lot of prohibited items

(e.g., alcohol, cigarettes) are placed on lifestyle based game shows. However,

teenagers’ consumption related values are mostly transmitted by TV through

observation and the passive learning method. Teenagers mostly feel attached to

celebrities of RTV, and get influenced by their fashion, style and personalities (Bush,

Martin & Bush 2004; Berman 2008; Scardino 2005). Therefore, imitating the role

models from RTV and observing different content, as well as getting involved in RTV,

are assumed to be solid ground for influencing and transmitting consumption related

values among teenagers.

Teenagers’ interest towards the media and the growth of electronic media is observed in

developing countries (Bhosale & Gupta 2006; Nath 2006). As mentioned earlier,

previous research, though mostly developed country based, shows that TV directly

affects teenagers’ consumption related value development (Burke 2005; Peterson et al.

2007). Similarly, consumption related values transmitted by TV are also expected

among the teenagers of developing countries. Furthermore, there is teenagers’

involvement with TV in developing countries (Bhosale & Gupta 2006; Nath 2006). As

75

previously identified, values formation on consumption is an outcome of the consumer

socialisation of teenagers. From the discussion in this section, it can be established that

values are the enduring beliefs that serve as guide for what is considered appropriate

behaviour, in this case consumption behaviour, by the society. Moreover, values

influence consumers’ attitudes and attitudes in turn, the consumer behaviour.

Accordingly, based on the existing literature, for this research, we can hypothesise that:

H4: There is a positive impact of RTV involvement on consumption related values.

H5: There is a positive impact of consumption related values on consumption

related attitude.

In this research, the domain of scale items to measure consumption related values have

been mostly adapted from current research on consumer socialisation by the media.

Consumption related values are a vital outcome of the consumer socialisation, which is

one of the critical constructs in this research. Furthermore, the interrelationships

between consumption related values and attitude are assumed in this research.

76

3.4 Social Structural Variables in Consumer Socialisation

The roles of various social structural variables are observed while studying consumer

socialisation research. In particular, parental control (Chan & McNeal 2006b; Cogdon

2007), peer group influences (French, Pidada & Victor 2005), gender differences

(Yavas & Riecken 1990; Jung & Peterson 2007) and social class differences

(Kamaruddin & Mokhlis 2003; Shim 1996) are mostly considered to be social structural

variables in current consumer socialisation research. However, like any other domain of

consumer socialisation research, these social structural variables are largely examined

from the perspectives of developed countries. In particular, the role of these social

structural variables has not been addressed either in developing or in developed

countries, in relation to RTV involvement, to understand the consumer socialisation of

teenagers. In the following sections, the above mentioned social structural variables are

discussed to understand their role on consumer socialisation and to develop a model of

consumer socialisation by RTV.

3.4.1 Peer Group Influences on RTV Involvement of Teenagers

In the extant literature, peer groups, particularly friends, are considered to be one of the

crucial socialisation agents that play a significant role in the teenaged consumer

socialisaion process (Churchill & Moschis 1979; French, Pidada & victor 2005).

Moreover, various researchers indicated that peer groups play an important role in

teenagers’ media usage and involvement, which ultimately affects their consumer

socialisation (Lueg & Finney 2007). It has been shown in various consumer

socialisation researches that teenagers’ media usage behaviour is highly influenced by

their peers (Lueg & Finney 2007, Moschis & Moore 1979). However, compared to

parental control of teenagers’ media involvement, peer group influences on the same

has received less attention from researchers, particularly in developing countries.

77

Within the limited range of current literature, peer groups influence on teenagers’ media

consumption are measured by reinforcement, modelling, motivation, co-viewing and

mediation (Edward & Grantham 2009; Nathanson 2001). Peer communication is

conceptualised as encouragement or approval of certain behaviour and intention, either

spoken (reinforcement) or unspoken (modelling) messages, which peers send to each

other. Peer communication also influences teenagers’ attitudes toward the TV vehicle

and content, which leads to the development of their material values and provides social

motivation for consumption (Bush, Smith & Martin 1999). Moreover, teenagers’

shopping behaviour through media (e.g., e-shopping) is largely influenced by their

peers’ reinforcement and modelling (Lueg & Finney 2007). According to Bush, Smith

and Martin (1999), reinforcement happens when peers encourage making purchases

through TV and the internet. Also, teenagers seek advice from their peers about buying

through the media (e.g., internet). Furthermore, modelling can indirectly influence

teenagers’ media usage by their peers’ media consumption behaviour, as they want

to have some attachment with their peers (Edward & Grantham 2009).

Peer groups serve the function of social comparison, self-evaluation, and co-learning

(Lloyd 1985), influencing teenagers’ motivation of media usage. Motivation occurs

when an individual is exposed to an external or internal stimulus, and is then pulled or

pushed in a direction to act upon the stimulus. Peers as co-learners play a role of

extrinsic motivation for using any particular media. Additionally, teenagers with similar

values tend to reinforce each other through supporting, sharing of resources, and

clarifying tasks for each other. These processes guide a teenagers’ motivation to act

while using any media.

78

As already mentioned, peer group influence on teenagers’ media involvement is a

relatively less researched area. Adolescence is a period where peers' influence is strong

and growing, so there should be evidence of them in viewing behaviour. Perhaps then,

communication research would benefit from a construct such as peer mediation that

would capture the interaction that peers have with one another regarding media.

Overall, as a mediating factor, peers play a significant role between the media and

teenagers. Peers' mediation can be found through media discussion with peers, peers co-

viewing and perception of peers’ attitudes towards the media (Lueg & Finney 2007;

Nathanson 2001).

As a developing county, Bangladesh is a country with a large population. A large

number of teenagers socialise with their peers and friends, either in school or within

the community. Furthermore, teenagers of Dhaka city have involvement with TV (Nath

2006). Accordingly, based on the above discussion, the following hypothesis is made

for this research.

H6: There is a positive impact of peer groups influences on teenagers’ RTV

involvement.

So far, researches on peer group influence on teenagers’ TV have been mostly carried

out in developed countries or western cultural settings (Lueg & Finney 2007).

Therefore, scale items to measure peer group influences on teenagers’ media (TV)

involvement for this research has been mostly adapted from current literature. In this

research, peer group influences on teenagers’ RTV involvement is considered not only

as a construct but also as one of the social structural variables in this research.

79

3.4.2 Parental Control on RTV Involvement of Teenagers

As has been already discussed (Sections 2.8.1 and 2.8.5) in the extant literature of

socialisation, parents and media have got the maximum attention of researchers. Among

diverse types of media, however, electronic media have got the greater attention of the

researchers in this field (Lachance, Beaudoin & Robitaille 2003; Zhang & Harwood

2002). TV has been considered to be the most influential consumer socialisation agent

among all media (Moscardelli & Liston-Heyes 2005). It has also been established by the

researchers that parents play a mediating role on teenagers’ media consumption habits

(Hennessy 2006; Keith 1986; Shifrin 2006). These days’ teenaged children have

involvement with a variety of media but still parents play a significant role in

influencing teenagers’ media consumption habits, which ultimately affect their

consumer socialisation. Though parental control on teenagers’ media usage has received

attention, but very little attention has been given by researchers about the outcome of

this parental media mediation process (Bulck & Bergh 2000).

From a theoretical point of view, the role of the media in relation to teenaged children’s

development and parental control on teenagers’ media usage are defined from various

perspectives. Various researches (Jordan 2004) have identified the theory of ecological

perspective to understand the role of the media in children’s development. In particular,

the micro-system perspective of ecological theory focuses on the role of the media and

the family’s domestic environment in children’s' development (Jordan 2004).

Furthermore, according to the exo-system perspective of ecological theory, parental

stresses might encourage or discourage the use of the media in a particular way.

Parents also socialise their teenaged children through the way they communicate

regarding TV involvement and that in turn also influences teenagers’ socialisation

(Chan & McNeal 2006c; Cogdon 2007; Fujioka & Austin 2002). The existing literature

indicates that family communication patterns are closely related to the parental control

on the media that influences teenagers’ media usage behaviour (Carlson Grossbart &

Stuenkel 1992). Family communication patterns provide a way of assessing the

80

interaction between parents - children and their media consumption behaviour (Chan &

McNeal 2003). Family communication patterns can also be conceptualised through two

dimensions like ‘socio–orientation’ and ‘concept-orientation’ (Chan & McNeal 2003).

The first dimension, socio-orientation, measures vertical or relationship-oriented

patterns of communication and puts emphasis on parental control and the children’s

deference to authority. The second, concept-orientation, measures issue-oriented

communication and focus on the establishment of an independent evaluation of an issue

by the children. These two family communication patterns may be expanded into four

categories (e.g., protective, pluralistic, consensual, laissez–faire) (Bakir, Rose &

Shoham 2005; Fujioka & Austin 2002).

Moreover, Carlson, Lacznik and Walsh (2001) categorised parental style as ‘warm vs

hostile’, and ‘restrictive vs permissive’, which are also supported by some other

researchers (Bakir, Rose & Shoham 2005; Fujioka & Austin 2002). On the basis of the

parental communication dyad, pluralistic (low socio–oriented and high concept-

oriented) and consensual parents (high socio–oriented and high concept-oriented)

discuss and share various TV content and vehicles with their children. Whereas, laissez-

faire (neither concept nor socio oriented) and protective family communication patterns

(low concept-high socio-oriented) reveal the least bilateral communication between

parents and children (Carlson, Lacznik & Walsh 2001; Chan & McNeal 2003).

Overall, Moschis (1981) noted that the influence of socio-oriented communication

structures appear to be indirect, by influencing TV viewing motives, which in turn

affect the development of materialistic orientations. On the other hand, concept-oriented

family communication structures apparently lead to differential exposure to media,

which in turn may lead to consumer knowledge; the influence does not appear to be

direct. However, consumer activity relates strongly to concept-orientation, suggesting a

direct family impact on such behaviours. Overall, parental communication styles affect

the teenagers’ media consumption habits, which ultimately develop the teenagers’

consumption related behaviour (Isler, Topper & Ward 1987; Wiman 1983).

81

On the one hand, consumer socialisation of teenagers’ through media depends on

parental style. On the other hand, teenagers’ perception about parental attitudes towards

the media also affects their involvement with the media (Carlson, Laczniak & Walsh

2001). Various researchers have found that mothers play the most important gate

keeping role regarding the time, content, and programs of TV (Carlson, Laczniak &

Walsh 2001; Chan & McNeal 2003). Furthermore, Carlson, Laczniak and Walsh (2001)

mentioned that mothers’ verbal interaction about TV and co-viewing, along with giving

an opinion, monitoring, and the controlling of TV, varies across cultures. Fujioka and

Austin (2002) indicated that the parental mediating role has a huge impact on teenagers’

media usage behaviour. Parent-child discussion patterns related to the media are often

referred to as parental mediation, defined as, the active discussion of the TV content

with the child (Fujioka & Austin 2002). Parental mediation can also be distinguished as

a) positive mediation (parental endorsement of TV message and content), b) negative

mediation (parental counterarguments or contextualisation of TV messages).

Interestingly, from the teenagers point of view; parental media mediation particularly,

mothers’ TV mediation is portrayed as mothers’ TV co-viewing frequency, mothers’

opinions about TV shows, and mothers’ knowledge about TV (Carlson, Laczniak &

Walsh 2001). Overall, parents have four major routes by which they can modify the

effects of television messages like, a) making restrictions on a child’s TV viewing, b)

active mediation via critical discussion of television messages with a child, c) the

general style of communication patterns that exist within the family and d) the example

parents set as models for children to imitate (Fujioka & Austin 2002).

Researchers (Cardoza 2002; Kuhl 2006) have diverse views about parental control on

media in developed and developing countries. However, like any other socialisation

research, parental control in developing countries teenagers’ media usage is also

generally less researched (Cardoza 2002). Comparatively, parents in developed

countries consider TV as an affective academic and entertainment tool for their

children, also, they give their teenaged children more freedom to use various media

(Kuhl 2006) than their developing countries' counterparts. Interestingly, co-viewing and

verbal interaction between parents and children are largely observed in developed

82

countries while watching TV (Carlson, Grossbart & Stuenkel 1992; Kim & Shin 1996).

On the contrary, in developing countries, parental control on TV is measured by

encompassing mediation, regulation, restriction, criticism, reinforcement, rules (explicit

and implicit), and censorship (Cardoza 2002). Due to social and cultural reasons, most

of the developing countries' teenaged children are sheltered by their parents (Cardoza

2002). Most of the time parents in developing countries decide for their children what to

watch and when (Kennedy 1996). Furthermore, parents provide moderate control over

TV programs and content selection, followed by escaping sensitive issues (e.g. sexual

content, violence) shown on TV (Cardoza 2002). Consequently, teenagers of

developing countries follow parental direction regarding involvement with TV more

than their counterparts in developed countries (Cardoza 2002). Accordingly, based on

the above discussion, for this research, the following hypothesis may be developed:

H7: Parents through communication, mediation and instruction positively

influence teenagers’ RTV involvement.

So far, researchers on parental control of teenaged children’s media usage have been

mostly carried out in developed countries or western cultural settings (Barkin 2006;

Fujioka & Austin 2002). For this research, scale items to measure the parental control of

teenagers’ media (TV) involvement have been mostly adapted from current literature.

Parental control on teenagers’ RTV involvement is considered not only as a construct,

but also as one of the social structural variables in this research.

83

3.4.3 Gender Differences of Teenagers and RTV Involvement

As one of the social structural variables, gender issues have received major

attention from researchers in consumer socialisation (Belk 1991; Kamaruddin &

Mokhlis 2003). As children grow up, they start considering the media a vital source of

information and entertainment that ultimately influences their consumption related

behaviour (Belk 1991). On the one hand, the media influences teenagers’ socialisation.

On the other hand, teenagers’ perception and values connect themselves with various

media (Anderson et al. 2001; Yavas & Riecken 1990). Similarly, teenagers’ gender

differences affect the nature and level of their involvement with various media. The

socialisation outcome of media involvement also varies due to gender differences of

teenagers (Chang 2006; Jung & Peterson 2007).

In the extant literature, TV has been considered a vital consumer socialisation agent in

the consumer socialisation, but by and large, it has been viewed from the perspective of

TV commercial (TVC) (Schomoll et al. 2006; Wolin 2003). The reciprocal relationship

of gender differences with TV Involvement are largely overlooked in the consumer

socialisation research (Bushman 2005). However, researchers have started focusing

lately on teenagers’ involvement with TV that affects the consumer socialisation (John

1999; Tiggemann 2005).

Among the electronic media, TV has got the maximum attention of the consumer

socialisation researchers. Teenagers’ tastes and preferences towards involvement with

TV vary due to gender differences. Various theories like ‘comparison theory’

and ‘mood management theory’ discuss the boys' and girls' involvement differences

towards the media (Jung & Peterson 2007; Knobloch-Westerwick 2007). Comparison

theory focuses on teenagers’ upward and downward comparison with others regarding

beauty consciousness and body image, which is largely influenced by TV. On the

contrary, Mood management theory conceptualises selections of media messages as

motivated by affect optimisation goals, which is originally called the theory of affect-

dependent stimulus arrangement. Its theoretical claims which pertain to enhancement of

both emotions and moods regarding media involvement vary due to gender differences.

84

However, research findings reveal that boys and girls show their involvement

differences towards various media. In particular, boys prefer more violence and musical

TV programs (Tufte & Halling 2002; Tiggemann 2005). Also, boys generally have

a higher inclination toward non-academic programs. Further, athletes in TV programs

provide a sense of body image to the teenaged boys (Jung & Peterson 2007; Tiggemann

2005). On the other hand, teenaged girls are strong followers of life style based shows

(e.g., style and fashion related shows, glamour based reality shows) which create their

beauty consciousness as well as influence their sense of sexual appeal (Clark &

Tiggemann 2006; Harrison & Cantor 1997; Lawrie et al. 2006). Interestingly, teenaged

girls process more information from TV than boys (Mangleburg & Grewal 1997), but

show lower responses towards the marketing stimuli from TV (Kamaruddin & Mokhlis

2003; Schmoll et al. 2006). Teenaged girls also show greater interest towards academic

programs, non-violent content and drama shows on TV (Tiggemann 2005). However,

Tiggemann (2005) also addressed that girls show similar interest in musical programs

like boys. Overall, teenagers’ gender differences have a large impact on TV

involvement (Anderson et al. 2001; Yavas & Reichken 1990). Boys are

inclined towards action and fantasy, while girls lean towards romance, gossip, and

glamour in TV programs (Grimes et al. 2004). Overall, sports based programs are

preferred by the teenagers followed by musical and drama serials (Cardoza 2002;

Kennedy 1996). Table 3.1 shows the teenagers’ involvement towards different TV

programs that vary due to gender differences.

Table 3.1 Comparison between Boys and Girls’ Involvement with TV shows (Source:

Tiggemann 2005)

Nature of TV Shows Boys Girls

Total TV viewing (hrs per

week)

24.5 21.1

Sports 5.0 1.7

Information 1.2 0.7

Soaps 1.6 3.8

Music Video 1.3 2.2

Entertainment 14.6 12.2

85

Research shows variations in their tastes and preferences towards TV programs also

influence teenaged boys and girls’ socialisation differently. As mentioned earlier, TV

commercial (TVC) has so far got the most research attention to understand the affect of

the media on teenagers’ consumer socialisation (Brand & Greenberg 1994; Donohue

1975; Gorn & Florsheim 1985; Rossiter 1979). Teenagers’ preference towards the

model, TV personalities, brand consumption attitudes, and information provided by

advertisements (TVC) are perceived differently due to gender differences (Jung &

Peterson 2007; Mangleburg & Grewal 1997; Suri & Phillip 2004). Moreover, learning

from TVCs and their influence on the consumer socialisation (e.g., purchasing brand

items, demand for fashionable apparel, reading the product label) varies between boys

and girls as addressed by different researchers (Kamaruddin & Mokhlis 2003;

Mangleburg & Grewal 1997).

Teenagers’ consumption related behavioural outcomes affected by TV also reflect

differences between boys and girls. Differences are also observed in teenagers’

consumption related cognition through development of product information processing

capacity, product recalling capacity, the influence of role models, and development of

consumption related skills from TV (Jung & Peterson 2007). Furthermore, teenaged

boys and girls perceive the products and services differently displayed on TV, which

leads to their consumption related attitude (Jung & Peterson 2007; Kamaruddin &

Mokhlis 2003). Moschis and Churchill (1978) also mentioned that teenaged boys and

girls respond differently towards consuming products and expressing their desire

towards the social motivation for consumption.

Based on the limited research conducted in developing countries so far, it can be

concluded that teenagers show different levels of involvement with the media than their

developed countries' counterparts (Cardoza 2002; Nath 2006). However, media

accessibility determines the teenagers’ preferences towards media consumption

behaviour in developing countries (Nath 2006). In particular, teenagers irrespective of

their gender have more or less similar access to media in urban areas which are not

observed in rural areas in developing countries (Nath 2006). However, in developing

86

countries, boys enjoy slightly more access to TV than girls (Nath 2006). Accordingly,

based on the above discussion, for this research, the following hypothesis may

be developed:

H8: There is a difference between male and female teenagers’ involvement with

RTV and its effect on their consumer socialisation.

3.4.4 Social Class Theories

Social class differences, as a vital social structural variable, have received major

attention from researchers in consumer socialisation research (Durham 1999; Granhaug

& Venkatesh 1986). Therefore, social class differences of teenagers have also been

considered in this research, to explore their impact on the consumer socialisation.

Social class is defined as the hierarchical division of a society into relatively distinct

and homogeneous groups with respect to attitude, values and lifestyle (Neal, Quester &

Hawkins 2004). Furthermore, this attitude, values and lifestyles differ between the

members of different social classes (Hoyer & Maclnnis 2007). Social class as a

construct relates to various dimensions of social position, which includes power,

prestige, and economic well being (Conger & Dogan 2007). Social classes are not

formal groups with a strong identity, but rather loose collections of individuals with

similar life experiences. Social class structures are important because they strongly

affect norms and values, and therefore behaviour. Given that members of a social class

interact regularly with each other (both formally and informally); people are more likely

to be influenced by individuals in their own social class than by those in other classes.

87

Social class cannot be measured directly, but indicators of social class status are

typically perceived by tangible (e.g., income, occupation, education) and intangible

factors (e.g., lifestyle, ties and political connections (Pope, Brennan & Voges 2007).

Historically, the most widely used index, W. Lloyd Warner’s Index of status

characteristics (ISC) is a method of measuring social class that evaluates occupation,

source of income, house type and location of home. The ISC is based on the following

socio-economic indicators: a) occupation (ranging from unskilled workers to

professional), b) source of income (range from public relief to inherited wealth) c)

education (no academic education to graduate) and, d) dwelling area (ranging from

slums to expensive areas).

According to Solomon et al. (2007), measuring of social class is a complex and difficult

task because of following reasons:

a) Most measurement of the social classes was designed to accommodate the

traditional nuclear family, with a male wage earner in the middle of his career

and a female full time homemaker. Such measures have trouble accounting for

two income families, young singles living alone, or households headed by

women, which are so prevalent in today’s society.

b) Another problem with measuring social class is attributable to the increasing

anonymity of our society. Earlier studies relied on the reputation method in

which extensive interviewing was done within a community to determine the

reputations and backgrounds of individuals. This approach is virtually

impossible to implement in most communities today.

c) The accuracy of questionnaires relies largely on the interviewers’ judgement,

especially regarding the quality of the respondent’s neighbourhood. These

impressions are in danger of being biased by the interviewer’s own

circumstances, which may affect his or her standard of comparison.

88

d) Assigning any group of people to a social class while they may not be equal

in their standing in all of the relevant dimensions. A person might come from a

low status ethnic group, but have a high status job, whereas another may live in

a fancy part of town but may not have finished high school.

Most countries have different kinds of social class structure depending on its relative

prosperity. For example, compared with the USA, Japan and Scandinavia have an even

larger and more predominant middle class with a much smaller group above and below.

In the developing countries, most of the concentrations are in the middle and lower

classes. In China, as one of the relatively affluent developing countries, 130 million

people belong to the middle class and will grow to 400 million within the next 10 years.

Interestingly, the upper classes in most societies are more similar to each other then

they are to other classes within their own countries because the upper classes tend to be

more cosmopolitan and international in orientation.

3.4.4.1 Social Class Differences of Teenagers and their RTV Involvement

In the extant literature of socialisation, social class has been considered as one of the

major social structural variables (Kamaruddin & Mokhlis 2003; Shim 1996). Social

class has also been considered while proposing conceptual models of consumer

socialisation (Moschis & Churchill 1978; Moschis & Moore 1979). Moreover, social

class is considered as mediating factor that affects the socialisation outcomes (Bradley

& Corwyn 2002; Conger & Dogan 2007). Generally, the nature of socialisation and its

affects vary in terms of social class and cultural variations (Kamaruddin & Mokhlis

2003).

Socialisation researchers so far have been contributing towards the relationship between

the media and adolescent together, but very few of them carried the issue of teenagers’

media usage from the perspective of social class differences (Durham 1999). However,

from a theoretical perspective, gratification theory (media and ways of life model)

discusses how audience selection of TV content depends on whether it is satisfying their

informational and entertainment needs (López-Sintas & García-Álvarez 2006), and

89

ecological theories reveal the ideas regarding the affect of the socio-economic status of

teenagers and its relationship with the way the media impacts on the consumer

socialisation of teenagers (Erametsa 1990; Mark 1998). These theories either describe

the socio-economic status of teenagers reflecting their media consumption behaviour, or

how different media are perceived in different social classes. These days, sociologists

are also researching the social class differences of teenagers and their media

consumption habits, which are extending the knowledge of socialisation (López-Sintas

& García-Álvarez 2006). Generally, teenagers’ social class is measured by their

occupation and education (Erametsa 1990). Interestingly, from the point of view of

media consumption habits, social class is also measured by the educational level and

income level of teenagers (López-Sintas & García-Álvarez 2006).

So far, teenagers’ involvement with TV from the perspective of social class

differences is relatively less researched (Lachance, Beaudoin & Robitaille 2003;

Tiggemann 2005). Within the last few decades, various researchers (Erametsa 1990;

Granhaug & Venkatesh 1986; Mark 1998; Van Eijck 2001) have found teenagers’

social class differences as a determinant variable of teenagers’ TV involvement, which

in turn affect their socialisation. Furthermore, Erametsa (1990) mentioned that media

content and vehicle selection is also influenced by different social class, because values,

appreciation, and consumption habits vary between social classes. In addition,

teenagers’ involvement with the media in different social classes is measured by

accessibility, affordability, availability and home environment (Griswold, Mcdonnell &

Wright 2005).

Interestingly, López-Sintas and García-Álvarez (2006) mentioned that social class status

can also be stratified by media consumption habits and choice (e.g., TV fans, occasional

film goers, light film goers, film fans, audio-visual fans). Social-class differences of

teenagers strongly affect their preferences in selecting particular content and vehicles of

TV (Mark 1998). Robinston et al. (2004) mentioned that the academic status of

teenagers significantly influences their TV program selection. TV usage preferences

also vary between different social classes, not only for informational purposes, but also

for entertainment purposes (López-Sintas & García-Álvarez 2006).

90

Teenagers’ social class differences reflect their TV usage behaviour in terms of vehicle

and content selection (Traeen, Nilsen & Stigum 2006; Van Eijck 2001). Researchers

have found that upper-class teenagers are culturally much more conscious regarding

various media than their lower social class counterparts (Van Eijck 2001). Teenagers

from a middle class social background also show more impulse interest towards various

media (Butsch 2001). In particular, TV is observed as an important medium for the

middle class teenagers to meet their informational and entertainment needs (Butsch

2001). However, the findings of media preferences of the middle class teenagers often

vary. Lopez-Sintas and Garcia-Alvarez (2006) found that middle class teenagers with

moderate income and education show interest towards TV. On the other hand, Shim

(1996) mentioned that upper-middle class teenagers show their inclination towards

reading magazines more than watching TV. Butsch (2001) observed those middle-class

norms as ‘work, activity and self-betterment,’ and the middle-class ‘work ethic’ results

in ‘less TV and a larger proportion of realistic, non-entertaining, self-betterment

programs.’ Middle class teenagers also show interest in libraries, home diversion, and

lecture studies, followed by involvement with cultural activities (Van Eijck 2001).

It is also supported by various researchers from the perspectives of developed countries,

that middle class teenagers have less inclination towards TV than their lower social

class counterparts (Butsch 2001; Dubow, Huesmann & Greenwood 2007). Furthermore,

teenagers from lower social classes show interest in TV content selection and TV

vehicle preferences (Butsch 2001; White 1995). Teenagers from the working class

mostly watch TV for entertainment purposes which also influences their consumption

fantasies (Butsch 2001). Greenberg (1974) also mentioned that teenagers from the lower

social class also prefer too much violence, sexual content and offensive language used

by the TV personalities. Moreover, teenagers from the lower social class consider

media to be a status symbol (Butsch 2001). Butsch (2001) also identifies middle social

class teenagers as the ‘reality group’ (mature users) and teenagers from the lower social

class as ‘fantasy group’ (immature audiences).

91

Teenagers’ social class background differences not only influence teenagers’ nature of

involvement with TV, but also affect their consumption socialisation. Generally,

teenagers from the middle social class have higher academic and economic status than

their lower social class counterparts. Moreover, they have higher access to different

media than their lower social class counterparts. Therefore, they understand and process

information better than their lower social class counterparts’ (e.g., social information,

consumption related information) (Durham 1999; Robinston et al. 2004). As a result,

the outcome of the consumer socialisation, such as consumption related cognition,

attitude and values, take different shapes between the two social classes (Granhaug &

Venkatesh 1986). According to Graunhaug and Venkatesh (1986), buying decision

styles, saving patterns and motives for product consumption are part of the cognitive

and affective development that also vary between the two social classes due to different

media exposure. Teenagers’ consumption related attitudes and values are also closely

related to their parents' resource allocation for their teenaged children, whereas media

influences teenagers’ expenditure patterns. Furthermore, consumption related values

such as eating habits, household goods consumption, and dietary habits also vary

between the two social classes due to media exposure (Mark 1998; Robinson 2004).

Shim (1996) also identified some reverse affects of TV on the consumer socialisation,

such as utilitarian, social/conspicuous and undesirable orientation that also vary due to

social class differences.

Most countries have different kinds of social class structures depending on the relative

prosperity of a particular country. Income, occupation, house type, education and

dwelling area are the most common determinants to define social class, both in

developed and developing countries. Social class differences affect the media usage of

developing countries' teenagers differently where most of the concentrations are

observed in middle and lower social classes. Recently, growth of both the middle class

and media is observed in developing countries (Bhosale & Gupta 2006; Nath 2006).

Furthermore, they mostly keep themselves updated, with higher access to a wide range

of local and foreign TV channels. Whereas, due to a low literacy rate, teenagers of the

lower social class show low interest in reading magazines and newspapers (Nath 2006).

92

H9: There is a difference between middle and lower social class teenagers’

involvement with RTV and its effect on their consumer socialisation.

3.5 Conclusion

In this chapter, the relevant literatures concerning the parent and immediate disciplines

for this research have been evaluated. Based on this evaluation, some key hypotheses

have been developed. Also, a proposed model, as shown in Figure 3.2 has been

developed considering the hypotheses. On the basis of the extant literature of consumer

socialisation research, ‘RTV Involvement’ has been considered as a connecting

construct between socialisation outcome and social structural variables in the proposed

model. Whereas, ‘Peer group influences on RTV involvement’ (Section 3.4.1) and

‘Parental control on RTV’ (Section 3.4.2) and are considered as social structural

variables. Furthermore, ‘Consumption related cognition’ (Section 3.3.1.1),

‘Consumption related attitude’ (Section 3.3.2.1), and ‘Consumption related values’

(Section 3.3.3.1) are examined as an outcome of the consumer socialisation. Table 3.2

below shows domains of multi-item measurement scales of the key outcome constructs

and social structural variables identified throughout the literature review above. From

the existing literature review, a number of variables are identified that were useful to

test the hypotheses for this research. Furthermore, these variables were tested to develop

the proposed model.

93

Figure 3.2 Proposed Model of the Consumer Socialisation of Teenagers’ by RTV

(CSRTV)

H6 H1 H3

H7 H2 H2

H8 H5

H4 H5

H9

Table 3.2 List of Constructs, Domains and Scale Items Adapted from Literature

Constructs Domains Scale Items Cognition

(Consumption related cognition)

a) Product knowledge

b) Consumer role perception

c) Purchasing intentions

d) Information processing

e) recalling capacity

f) consumption related skills

(Bartlett, Griffiths & Badian

2008; Bush, Martin & Bush

2004; Moschis & Moore

1978; Zuckerman, Ziegler &

Stevenson 1978; Ward &

Wackman 1974)

1) RTV make me aware about

particular brands

2) I understand product usage

instruction from RTV

3) RTV helps me to recall about

certain products

4) RTV helps me to recognise

particular products and brands

5) RTV helps me to decide what

brands to buy

6) RTV is a good way to learn

about what products/bands are

available in market

7) The opinion of my favorite

RTV celebrities influence me to

say positive things about products

or brands to other people

8 ) RTV helps me to do value

analysis (compare price and

benefit of brands) before buying

Peer group influences

Parental control

Gender differences

RTV Involvement

Consumption related cognition

Consumption related attitude

Consumption related values Social class differences

94

Attitude (Consumption related attitude)

a) Fashion

b) Imitation

c)Inspiration from media

personalities

d) Attitude towards product

placement

(Hirschman & Thompson

1997 Lawrie 2006; Russell,

Norman & Heckler 2004;

Schmoll et al. 2006).

1) I get fashion tips from RTV

2) I like the product shown on

RTV

3) I often buy clothing styles that I

have seen in RTV

4) My favorite RTV celebrities

make me fashion conscious

5) I want to get good figure like

RTV celebrities (e.g. slimmer,

muscular)

6) Thin models in RTV creates my

interest (e.g., plastic surgery, diet

habit, figure shape)

7) I see my favorite RTV celebrity

as my idol

8) I would like to have lunch with

my favorite RTV celebrity

9) I feel very strongly about

banning unethical products (e.g.,

tobacco, beer, alcohol) shown in

RTV

Values (Consumption related values)

a) Desire towards products

b) Prohibited product

consumption

c) Materialism

d) Antisocial behaviour

(Chan 2003; Ghani 2005;

Gruber & Thau 2003; Wyllie,

Zhang & Casswell 1998)

1) I like to own the newest things

shown on RTV

2) I like to own the best things

shown on RTV

3) It is better to have more

allowance from parents to buy the

brand shown on RTV

4) I often request my parents to get

the products shown on RTV

5) I often request my parents to get

the products shown on

advertisements in RTV shows

6) Some RTV shows promote use

of prohibited items (e.g., alcohol,

cigarette) among teenagers

7) Consumption of prohibited

95

items (e.g., alcohol, cigarette)

makes life more fun and exciting

8) I like the sexually appealing

outfit worn by RTV participants

and judges

Peers group influences a) Reinforcement

b) Modelling

c) Extrinsic motivation

d) Peer co-viewing

e) Peers mediation

(Edward & Grantham 2009;

Lueg & Finney 2007;

Nathanson 2001)

1) My friends and I talk about

content of RTV shows (e.g.,

entertainment, sex, information)

2) My friends and I watch often

RTV together

3) My friends think the content of

RTV shows can provide useful

information

4) My friends think content of

RTV are entertaining

5) My peers encourage me to send

SMS to RTV

6) My peers’ opinions on RTV

shows are important to me

7) My friends and I participate

together live on the RTV shows

Parental control on TV a) Parental communication

styles

b) Parental opinion

c) Co-viewing

d) Parental knowledge

(Carlson, Laczniak & Walsh

2001; Chan & McNeal 2003;

Fujioka & Austin 2002; Rose,

Bush & Kahle 1998)

1) I talk to my parents about RTV

content

2) My parents and I discuss about

RTV show selection

3) My parents and I watch RTV on

weekdays

4) My parents and I watch RTV

during weekends

5) My parents know what I watch

on RTV

6) My parents think some content

at RTV are really bad for

teenagers

7) My parents turn off channel

while unsuitable contents are

shown on RTV

96

8) My parents prohibit me to

watch certain RTV shows

RTV Involvement a) Information search

b) Time spent, likings

c) Attention

d) Desire towards RTV

(Dolliver 2006; Higie &

Feick 1989; Pecheux &

Derbaix 2002; Perse 1998;

Yoon & Garma 2006)

1) I watch RTV

2) I pay close attention when I

watch RTV

3) I read about RTV in news

papers

4) I regularly vote to RTV (e.g.,

SMS, Phone)

5) I would attend a public

appearance if my favorite RTV

character from the show were

there

6) I imagine how I would act in a

similar situation shown on RTV

7) I watch other programs that

include contestants from the RTV

show

8) I wish I could be on RTV

shows

9) RTV is just like real life

10) I would like to participate in

RTV if I like it.

11) RTV provides me a lot of

information

97

Chapter 4 Research Methodology

4.1 Introduction

In the previous chapter an extensive review of the literature involving the parent and

immediate disciplines of consumer socialisation as well as the focus topic of this

research, role of reality TV (RTV) as a consumer socialisation agent of teenagers, and

some other relevant topics has been conducted. This has helped to establish the

importance of the research objectives for this study and led to the development of

specific hypotheses. This chapter will describe and justify the methodology selected to

collect and analyse the data to address the research objectives and test the underlying

hypotheses. In this research both qualitative and quantitative methods were used.

Qualitative research was conducted to help refine the constructs and respective

measurement variables that have been developed based on the literature review, and

quantitative methods to test the constructs. Structural Equation Modelling (SEM)

technique was used to test the constructs and confirm the model of consumer

socialisation of teenagers in a developing country by RTV.

The methodological objective of this research was to identify how TV Vehicle, RTV as

a consumer socialisation agent, influence the teenagers’ consumer socialisation.

Further, the objective was to assess the effects of socio-structural variables like; social

class differences and gender differences of teenagers on the consumer socialisation.

Parental control and peers group influences on the role of RTV as a consumer

socialisation agent also were examined.

98

4.2 Justification of the Methodology - Quantity or Quality

The metaphor of the ‘paradigm wars’ described by Gage (1989) is undoubtedly

overdrawn, as ‘wars’ paint a picture of confrontation. In building and testing new

theories, paradigm differences can be resolved through the emergence of a new

paradigm, more informed and sophisticated than any existing one. A paradigm is a basic

set of beliefs that guide action (Guba 1990, p.17). It represents a worldview that defines,

for its holder, the nature of the ‘world’, the individual's place in it, and the range of

possible relationships to that world and its parts. A paradigm encompasses three

elements: ontology, epistemology, and methodology. a) ontology (ontology raises basic

questions about the nature of reality), b) epistemology (epistemology asks, how do we

know the world? What is the relationship between the inquirer and the known?), and c)

methodology (methodology focuses on how we gain knowledge about the world).

Questions of methods are secondary to questions of paradigm, not only choices of

method but in ontologically and epistemologically fundamental ways (Guba & Lincoln

1994). Guba and Lincoln (1994) identified four competing paradigms in informing and

guiding inquiry: positivism, postpositivism, critical theory, and constructivism.

Undoubtadely, the positivist and postpositivist paradigms provide the backdrop against

which other paradigms operate.

Conventional positivist social science applies four criteria to disciplined inquiry:

internal validity, the degree to which findings correctly map the phenomenon in

question; external validity, the degree to which findings can be generalised to other

settings similar to the one in which the study occurred; reliability, the extent to which

findings can be replicated, or reproduced by another inquirer, and objectivity, the extent

to which findings are free from bias. Postpositivism represents efforts of the past few

decades to respond in a limited way to the most problematic criticisms of positivism.

The positivist and postpositivist views have recently come under considerable attack

(Guba & Lincoln 1994), supporting arguments that these paradigms are unable to deal

adequately with the issues surrounding the etic, emic, nomothetic, and idiographic

dimensions of inquiry. Too many local (emic), case-based (idiographic) meanings are

99

excluded by the generalising (etic) nomothetic (Denzin & Lincoln 1994), positivist

position. At the same time, the nomothetic, etic approaches fail to address satisfactorily

the theory-and value-laden nature of facts, the interactive nature of inquiry, and the fact

that the same set of ‘facts’ can support more than one theory.

Constructivism (Guba & Lincoln 1994) or interpretivism (Perry 1995; Schwandt 1994),

adopts relativist ontology, a transactional epistemology, and a hermeneutic, dialectical

methodology. The inquiry aims of this paradigm are oriented to the production of

reconstructed understandings, wherein the traditional positivist criteria of internal and

external validity are replaced by the terms trustworthiness, and authenticity. Proponents

of positivism and postpositivism still dominate the scene. But, over the past decade, it

has become more and more possible for proponents of critical theory and constructivism

to achieve acceptance with the growing acceptability of ‘qualitative’ dissertations.

Method is context specific. From historical perspective, there has been heavy emphasis

on quantification in science. Scientific maturity is commonly believed to emerge as the

degree of quantification found within a given field increases. This view of science

focuses on efforts to verify (positivism) or falsify (postpositivism) a priori hypotheses,

most usefully stated as mathematical (quantitative) propositions or propositions that can

easily converted into precise mathematical formulas expressing functional relationships.

There exists a widespread conviction that only quantitative data are ultimately valid, or

of high quality (Schreast 1992).

The intraparadigm problems offer a weighty challenge to conventional methodology,

but could be eliminated, or at least ameliorated, by greater use of qualitative data.

However, the findings of qualitative research are misused when they are regarded as

conclusive and are used to make generalisation to the population of interest (Colwell

1990; Gabriel 1990). It is sound research principle to view qualitative and quantitative

research as complementary, rather than in competition with each other (Burrell &

Morgan 1979; Hunt 1991).

100

Seldom is enough known about a marketing problem or situation for the researcher to

be able to proceed directly to the design of a structured study yielding representative

and quantifiable results. As has been stated in the literature review in the previous

chapter, most of the researches on consumer socialisation of teenagers have been

conducted in the developed countries that discourage the use of existing constructs and

their measurement variables for the purpose of this research. While some measurement

items and constructs have been identified through the literature review, reported in the

previous chapter, it has also been established that these constructs and their

measurement items are not directly applicable to a developing country without some

refinement. Consequently, it is important to become mindful about complexity of

research problem on hand and identify likely methodological problems.

Qualitative method was primarily used to refine the constructs and their multi-item

measurement scales that may be adapted to a developing country. The combination of

multiple methods, empirical materials, perspectives, and observers in a single study is

best understood, then, as a strategy that adds rigour, breadth, and depth to any

investigation (Flick 1992). “Because different ‘lenses’ or perspectives result from the

use of different methods, often more than one method may be used within a project so

the researcher can gain a more holistic view of the setting” (Morse 1994, p. 224).

“Increasingly, quantitative researchers seem dissatisfied with purely quantified results

and are turning to supplementary qualitative analyses, while qualitative researchers

have become less defensive about their modes of analysis and more open to working

with quantitative researchers on research projects. Sometimes, they combine

quantitative methods with their qualitative ones” (Strauss & Corbin 1994, p.277-78).

101

4.3 Research Procedure

4.3.1 Qualitative

Qualitative research is used in many separate disciplines. It does not belong to a single

discipline. Nor does qualitative research have a distinct set of methods that are entirely

its own. Qualitative research is inherently multi-method in focus (Brewer & Hunter

1989).

Initially this research involved qualitative research to understand the perspectives of

consumer socialisation of teenagers in a developing country. So far, significant

researches (Shoham & Dalakas 2003; Moschis & Moore 1979; Vakratsas & Ambler

1999) have been conducted particularly in the developed countries to understand

consumer socialisation of teenagers. Unfortunately, the applicability of their findings is

unknown in a developing country like Bangladesh. This eliminates the direct

application of existing constructs for the purpose of this research. In the previous

chapter of this thesis; research problem, objectives, constructs and some measurement

variables were identified from existing literature, based on previous consumer

socialisation researches in developed countries. These needed further refinement to

make them appropriate for a developing country. Also, exploratory research is used

when researchers must define the problem more precisely; identify relevant course of

action, and gain additional insights before an approach can be developed (Malhotra

2003; Tufte & Halling 2002).

A variety of qualitative methods can be used for exploratory purposes. These methods

are less structured and more intensive than standardised questionnaire based interviews,

and result in context specific information, which is directly relevant to attitudes, rather

than remote from actual behaviour (Johansson & Nonaka 1987). In this research, a

focus group interview (FGI) method was used for exploratory purposes.

102

4.3.1.1 Sampling for Qualitative Research

A good informant is one who has the knowledge and experience the researcher requires,

has the ability to reflect, is articulate, has the time to be interviewed, and is willing to

participate in the study (Morse, 1986, 1991). Patton (1990) provides guidelines for

sampling and suggests that the logic and power behind purposeful selection of

informants is that the sample should be information rich. Patton (1990) suggested three

types of sampling: a) extreme or deviant case sampling is used to select participants

who exemplify characteristics of interest, b) intensity sampling has less emphasis on

extremes but emphasises selection of participants who are authorities about a particular

experience, and c) maximum variety sampling is the process of deliberately selecting a

heterogeneous sample and observing commonalities in their experiences. For this

research intensity sampling and maximum variety sampling were used with the view to

finding out shared patterns of commonalities among the participants experiences

regarding RTV involvement and its effects on their consumer socialisation. To ensure

this, it was considered that all the 20 FGI participants have concepts and ideas about as

well as involvement with RTV. During the participant selection process teenagers’ who

watch RTV for at least for an hour in a week were considered. Where as, maximum

variety sampling was ensured by choosing participants from two different social classes

(e.g., middle, lower), different age groups within the broad category of teenagers, and

gender (e.g., boys, girls).

4.3.1.2 The Focus Group Interviews (FGI)

The interview is the favourite methodological tool of the qualitative researcher.

Interviewing has a wide variety of forms and a multiplicity of uses. In this research,

FGIs were conducted. Focus group is an interview technique that is conducted in an

unstructured and natural manner by a trained moderator among a small group of

respondents (Fontana & Frey 1994). The main purpose of focus groups is to gain

insights by listening to a group of people from the appropriate respondents about issues

of interest to the researchers. “Unstructured interviewing provides a greater breadth than

103

the other types, given it’s qualitative nature” (Fontana & Frey 1994, p. 365).

Unstructured interviewing techniques have been reported extensively in the literature

(Lofland & Lofland 1984).

Malinowski (1989) captures the differences between structured and unstructured

interviewing. The former aims at capturing precise data of a code-able nature in order to

explain behaviour within pre-established categories, whereas the latter is used in an

attempt to understand the complex behaviour of members of society without imposing

any prior categorisation that may limit the field of inquiry.

In this research, to gain further insight and understanding on the research problem, two

separate FGIs involving 10 teenagers in each were conducted among Bangladeshi

teenagers representing urban middle and lower social classes of Dhaka city (capital of

Bangladesh). These FGIs were conducted in the respective campuses of two different

schools e.g., with lower social class teenagers in Nilkhet School and middle social class

in Udayan School, both based in Dhaka. A flyer for volunteers (Appendix 2) was

circulated among the potential participants and consent from each participant was taken

using a consent form (Appendix 4). During the time of respondent selection,

participants from 13-19 years representing class 5 to 11 were considered and both the

focus groups had equal representation of the gender. The first focus group was

conducted at Nilkhet School on 16th July 2009 between 12:00 noon to 3:00 pm at the

school common room and the second focus group was conducted at Udayan School on

23rd July between 12:00 to 3:00 pm at the school common room. A UWS Research

Committee approved research protocol providing background information about the

research (Appendix 3) was distributed among the participant at the beginning of the

FGIs. During these FGI sessions participants were encouraged to describe events; draw

linkages, and give explanations about their consumer socialisation by their involvement

in RTV. Particularly, respondents were asked to name different RTV shows and their

features. Teenagers’ involvements with different RTV shows also were addressed. Most

importantly, how teenagers connect different ideas, clues and information from RTV for

getting knowledge about product, recalling and usages brands were the focus of

104

discussion. Similarly, how RTV celebrities and various features of RTV affect

teenagers’ consumption related attitude and values were explored.

Both the FGIs were moderated by this researcher. During conducting the FGIs; a pre-

developed discussion guideline (Appendix 5) was used. As the study progressed,

insights of the researcher about consumer socialisation of teenagers increased, making

the process exciting as ‘what is going on’ finally become clearer and more obvious.

Both the FGI’s were audio recorded with prior approval from each participant and

following the ethics protocol (a discussion of ethical issues follows in section 4.4).

Audio recording was used as a method of recording the discussion because of the

following advantages of its use:

1. Provides a completely accurate record of what each respondent said.

2. Frees the interviewer to listen and respond more rapidly.

3. Allows better voice modulation and hence better development of rapport between

interviewer and interviewee.

Further, hand written notes were taken to record statements and non-verbal gestures

e.g., nodding to indicate agreement or disagreement of others participants’ comments

regarding RTV relating to consumer socialisation. These FGIs were conducted in local

language (Bengali) which is also this researcher’s native language. So, no difficulties

were faced while transcribing the interview sessions from the audio tapes.

The recorded information from the FGIs were coded and classified for qualitative

analysis by the researcher. Participant boys were identified as ‘B’ and girls as ‘G’ from

both the social groups and then were given numeric identifiers. Each participant’s

comments were then coded with reference to their individual identifier number (Table

5.3).

105

Findings of these FGIs were quite important to refine and further develop the constructs

and identified variables making them more relevant for Bangladesh. This also helped to

properly identify the data that were needed to address the research problem, objectives,

and hypotheses.

4.3.1.3 Ensuring Rigour

There are numerous methods of ensuring rigour in qualitative work. The major methods

for ensuring rigour are intricately linked with issues of reliability and validity. In

qualitative research the main methods to ensure this are criteria of adequacy and

appropriateness of data and verification of data with informants (Morse 1994).

In this study criteria of adequacy and appropriateness were established. In qualitative

research, adequacy refers to the amount of data collected, rather than to the number of

subjects, as in quantitative research. Adequacy is attained in this research by ensuring

that sufficient data have been collected and there is saturation as well as variations are

both accounted for and understood. Appropriateness refers to selection of information

according to the theoretical needs of the study (Morse 1994). In this research FGI

samples were purposefully drawn from only those teenagers who are experienced RTV

viewers, who watch RTV at least one hour a week and have involvement in it.

As a further test of validity, the outcomes of the discussions as they were becoming

evident were presented to the participants for discussion and review. Often informants

are able to confirm immediately the accuracy and validity of the study, and sometimes,

offer additional information (Glaser 1978). In both the FGI sessions of this research, the

outcomes of the discussions on consumer socialisation by RTV were presented to the

participants towards the end of the sessions with the help of a white board. Participants

were then asked to make comments on addressed variables related to RTV involvement

and its affects on consumer socialisation. Participants’ comments on different variables

and how they are related with other variables were audio taped as well as noted.

106

4.3.1.4 Data Management and Analysis

Application of data management methods during the study is essential for the success of

the study. In this research, audio tapes and hand notes were transcribed, corrected and

edited before analysis. Transcripts and notes must be easily retrieved, easily cross-

referenced, and easily separated from and linked with their original sources.

The data was analysed using content analysis method (Weber 1990). This method is

often theory driven e.g., theory determines what to look for (Weber 1990). In this

research the conceptual model, its constructs and variables developed through the

literature review were the basis for what to look for in the FGI data. Looking at the

transcripts the themes and how these themes relate to each other were identified

analysing each sentence spoken by the participants. The themes were systematically

assigned numeric codes. What respondents talk about most were noted.

How data are stored and retrieved is the heart of data management. In this research,

coding of the data gathered at the qualitative stage was done immediately after each

FGI, importance of which has been emphasised by Strauss (1987). Werner and

Schoepfle (1987) noted that a system needs to be designed prior to actual data

collection, and stressed the importance of a clear indexing system. In this research, the

two groups were coded as 1 (lower social class) and 2 (middle social class). Participant

boys were coded as B and girls as G followed by numerical number to identify each of

them. So, record of discussions at FGI sessions can be traced later on to relevant

participants for reporting purposes.

Miles and Huberman's (1994) definition of data analysis contains three linked sub

processes: data reduction, data display, and conclusion/verification. With data

reduction, the potential universe of data is reduced in an anticipated way on the basis of

research objectives and instruments (Miles & Hubermans 1994). In this research, once

actual field notes, FGI tapes became available, data summaries, coding, finding themes,

clustering, and writing summaries were done manually for selection of data that was

107

relevant to the conceptual model and its constructs earlier developed through literature

review. Data display, defined as an organised, compressed assembly of information that

permits conclusion drawing, was the second part of analysis. This was done by

summarising the findings (Fischer & Wertz 1975) and developing matrices with text

(Miles & Huberman 1994) with each cell representing a construct and the cell contents

their respective variables.

Figure 4.1 Qualitative Data Analysis Process (Source: Miles & Huberman 1994)

Conclusion drawing and verification involve the researcher in interpretation: drawing

meaning from display data. The tactics used in this research, ranged from the use of

comparison/contrast, noting of patterns and themes, clustering and checking results with

respondents (Miles & Huberman 1994).

4.3.2 Need for Quantitative Research

In this research, qualitative method has been used to refine the measurement variables

that were earlier identified through literature review making them relevant to a

developing country. The limitations of the qualitative methods are: (1) The results are

not necessarily representative of what would be found in the population, and hence are

not project able, and (2) There is typically ambiguity in the results (Aakar & Day 1990).

Data collection

Data display

Data reduction

Conclusions drawing/ verification

108

In view of these shortcomings as well as to test and validate the conceptual model and

its constructs developed through literature review quantitative method was used.

4.3.3 Scale Development and Confirmatory Factor Analysis

As recommended by Churchill Jr. (1979), a widely used process for developing

measurement scales in Marketing involves the following steps:

1) Defining theoretical constructs (Austin, Plouffe & Peters 2005)

2) Generating a list of items from literature and/or qualitative research that relate to

these constructs (Attewell, Suazo-Garcia & Battle 2003; Austin, Plouffe &

Peters 2005) and

3) Purifying these measures using exploratory factor analysis (Sweeney & Soutar

2001) and coefficient alpha (Chan & Prendergast 2007).

In this research all of the above steps were followed and the third step was

supplemented with confirmatory factor analysis (Bristol & Mangleburg 2005b; Shrum,

Burroughs & Gainesville 2005)

The use of multi-item scales to measure a construct is considered superior to single item

scales as it increases reliability and decreases measurement error (Churchill Jr.1979).

These sets of items are generally reflective in that they all measure the same theoretical

construct. Coefficient alpha was used to measure the reliability (Malhotra & Peterson

2006) of the scale at exploratory factor analysis (EFA) stage. Exploratory factor

analysis was considered a test of dimensionality, with the aim to produce a set of items

that reflect a single underlying factor or construct. Purifying the set of indicators is an

iterative process whereby indicators with low factor loading are dropped (Conway &

Huffcutt 2003; Garver & Mentzer 1999; Hair et al. 1998).

109

This traditional approach has since been expanded with the confirmatory factor

analysis. Confirmatory factor analysis using AMOS was carried out to give a truer

estimation of reliability and formally test the uni-dimensionality of a scale (Hoyle

1995). It is therefore considered a more rigorous scale development procedure that was

followed.

4.3.4 Constructs, Domains and Scale Items

In this research, six constructs are addressed from current literature to develop a model

of consumer socialisation by RTV. Furthermore, domains and scale items relating to

respective constructs also been addressed to develop and test the above mention model.

These constructs including the domains and scale items are adopted from the current

literature of consumer socialisation by media and later on refined through exploratory

research. These domains and measurement scales items of respective constructs are

mentioned in the following sections.

4.3.4.1 RTV Involvement

In this research impact of RTV on consumer socialisation of teenagers has been

measured by teenagers’ involvement in it.

Zaichkowsky (1985) defined involvement as “a person’s perceived relevance to the

object based on inherent needs, values, and interest.” This definition can be applied to

RTV studies. Literature also indicate that a person can be involved with advertisements

(Krugman 1962), and RTV (Yoon & Garma 2006). In media research, involvement

pertains to media users’ relationships with message conveyed by media or with media

(Levy & Windahl 1984), media personalities (Rubin & Perse 1987), or other media

users (Rubin & Step 1997). Rubin and Perse (1987) conceptualised media involvement

as “cognitive, affective, and behavioural perceptions during and because of exposure”

(p. 247). To understand involvement, one needs to understand audience and role of

involvement between media motive and media effects (Perse 1998).

110

Various researchers have agreed and proposed different domain of scale items to

measure the involvement with media. Liking (e.g., watching RTV, reading about RTV)

is one of the common domains to measure the involvement with RTV (Pecheux &

Derbaix 2002; Yoon & Garma 2006). Yoon and Garma (2006) added that involvement

with RTV is also measured by how much time audience spend to watch different shows

relating to RTV celebrities. Furthermore, amount of attention given to RTV is

considered as a measurement domain of RTV involvement (Perse 1998; Yoon & Garma

2006). Interestingly, audience desire to attend RTV shows also indicates their

involvement with it (Dolliver 2006). At the qualitative stage of this research, teenagers

from both the groups including boys and girls have shown desire to watch RTV shows

on web, though none of them have regular access of it. On the contrary, involvement

with RTV is not only addressed as a source of fun and fantasy but also as a source of

information (Higie & Feick 1989). On the basis of above discussion following Table 4.1

shows the domains of scale items to measure RTV involvement.

Table 4.1 Domain of Scale Items of RTV Involvement

Construct Domain of Scale Items

RTV involvement Time spent, Likings, Attention, Desire

towards RTV (Dolliver 2006; Higie &

Feick 1989; Pecheux & Derbaix 2002;

Perse 1998; Yoon & Garma 2006)

4.3.4.2 Consumption Related Cognition

Cognition is defined as “knowledge that is acquired by a combination of direct

experience and information from various sources” (Schiffman et al. 2005, p. 617).

Wagner (2008) has indicated cognition as the mental processes involved in gaining

knowledge and comprehension; including thinking, knowing, remembering, judging,

and problem solving. So, consumption related cognition can be conceptualised as the

mental processes involved in gaining knowledge and comprehension; including

111

thinking, knowing, remembering, judging, problem solving and skill development

regarding products and brands consumption (Bartlett, Griffiths & Badian 2008; Bush,

Martin & Bush 2004; Moschis & Moore 1978).

In the current literature, researchers have agreed and proposed different domain of scale

items to measure consumption related cognition. Product knowledge (e.g., information

about product, recognising product, brands) is considered as one of the vital domains to

measure consumption related cognition (Bartlett, Griffiths & Badian 2008; Zuckerman,

Ziegler & Stevenson 1978). Bartlett, Griffiths and Badian (2008) also added that

developing awareness about products and brands are also considered as product

knowledge that measure consumption related cognition. At the focus groups discussions

sessions, teenagers addressed very strongly that advertisement shown in RTV also give

them products and brands ideas. Furthermore, consumer learning capacity is an

important measurement scale of product knowledge (Ward & Wackman 1974).

Zuckerman, Ziegler and Stevenson (1978) also added that recognising and recalling

capacity is vital domains to measure consumption related cognition. Consumer’s

purchasing intention is also viewed as domain of measurement variables in this regard

(Bush, Martin & Bush 2004). Teenagers’ also responded towards RTV sponsoring

brands were quite positively that reflecting teenagers’ purchasing intentions found from

focus groups discussions. Consumers’ role perception (e.g., understanding instruction,

developing creativities) considered as a domain to measure consumption related

cognition (Moschis & Moore 1978). RTV shows also develop teenagers’ creativity

regarding their fashion and style consciousness that are addressed from the focus groups

discussions sessions. On the basis of above discussion following Table 4.2 shows the

domains of scale items to measure consumption related cognition.

112

Table 4.2 Domain of Scale Items of Consumption Related Cognition

Construct Domain of Scale Items

Cognition (Consumption related

cognition)

Product knowledge, Consumer role

perception, Purchasing intentions,

(Bartlett, Griffiths & Badian 2008; Bush,

Martin & Bush 2004; Moschis & Moore

1978)

4.3.4.3 Consumption Related Attitude

Attitude is defined as a “psychological tendency that is expressed by evaluating a

particular entity with some degree of favour or disfavour” (Eagly & Chaiken 2007, p.1).

Solomon (2004) referred attitude as lasting general evaluation of people (including one

self), objects or issues. Eagly and Chaiken (2007, p.5) also added as attitude “an

evaluative judgment or affective or evaluative response”. Current definitions of attitude

posses some key features like evaluation, attitude object, and tendency (Eagly &

Chaiken 2007). Overall, consumption related attitude can be conceptualised as an

enduring combination of motivational, emotional, perceptual and cognitive process with

respect to imitation, fashion, inspiration and product placement (Hirschman &

Thompson 1997; Lawrie 2006; Russell, Norman & Heckler 2004; Schmoll et al. 2006).

Existing literature shows that perception towards fashion (e.g., fashion tips from RTV,

liking fashion) is a domain to measure consumption related attitude (Russell, Norman &

Heckler 2004). Further, imitation of RTV celebrities’ behaviour also is a domain of

consumption related attitude development (Russell, Norman & Heckler 2004). Lawrie

et al. (2006) added that audience attitude towards TV celebrities body image can be

considered as a domain for consumption related attitude measurement. Harrison and

Cantor (1997) agreed with the mentioned measurement domain and pointed that

audience likings towards TV celebrities’ attractive figure also indicate measuring

113

consumption related attitude. Further, emotional attachment towards RTV celebrities

also considered while measuring consumption related attitude (Hirschman & Thompson

1997). Apart from that emotional attachment with RTV celebrities, perception towards

the product placement (e.g., perception towards unethical product showing) are also

noted to measure consumption related attitude (Schmoll et al. 2006). In this research,

teenagers from both focus groups have mentioned that they prefer multinational brands

to buy if is shown by RTV. On the basis of above discussion following Table 4.3 shows

the domains of scale items to measure consumption related attitude.

Table 4.3 Domain of Scale Items of Consumption Related Attitude

Construct Domain of Scale Items

Attitude (Consumption related attitude) Imitation, Inspiration from media

personalities, Attitude towards product

placement (Hirschman & Thompson 1997;

Lawrie 2006; Russell, Norman & Heckler

2004; Schmoll et al. 2006).

4.3.4.4 Consumption Related Values

Schiffman et al. (2005, p. 637) noted values as “relatively enduring beliefs that serve as

guides for what is considered ‘appropriate’ behaviour and are widely accepted by the

members of the society”. Values are widely held beliefs about what is acceptable and/or

desirable (Neal, Quaster & Hakins 2004). So, consumption related values can be

conceptualised as shared beliefs among group members as to what behaviours are

desirable or undesirable towards products, prohibited product consumption, materialism

and anti-social behaviour.

In the extant literatures, sense of materialism (e.g., desire towards best things, newest

things) has been found as one of the most common domains to measure consumption

related values (Chan 2003). Kamineni (2005) also referred to materialism as a crucial

domain of measuring consumption related values which are strongly influenced by

114

media. Sense of materialism also triggers desire towards consuming particular products.

Various researchers (Ghani 2005; Hastings & Aitken 1995) considered desire towards

particular product consumption as a domain to measure consumption related values as

well. Hastings and Aitkens (1995) also supported this domain by adding consumption

fantasy to measure consumption related values which are highly influenced by RTV. On

the contrary, consumption related values are also measured by belief towards prohibited

product consumption (e.g., beliefs towards drinking alcohol, smoking cigarettes)

(Wyllie, Zhang & Casswell 1998). On the basis of above discussion following Table 4.4

shows the domains of scale items to measure consumption related values.

Table 4.4 Domain of Scale Items of Consumption Related Values

Construct Domain of Scale Items

Values (Consumption related values) Desire towards products, Prohibited

product consumption, Materialism (Chan

2003; Ghani 2005; Wyllie, Zhang &

Casswell 1998)

4.3.4.5 Parental Control

These days’ teenaged children have involvement with varieties of media but still parents

play significant role to influence teenagers’ media consumption habit that ultimately

affects their consumer socialisation. Particularly, parental control on teenagers’ media

usage can be defined from the contexts of parental communication styles, parental

opinion, co-viewing, parental knowledge (Carlson, Laczniak & Walsh 2001; Chan &

McNeal 2003; Fujioka & Austin 2002; Rose, Bush & Kahle 1998). Generally, parental

control can be conceptualised as parental communication pattern, mediation style,

influence and instruction capacity to their children (Bakir, Rose & Soham 2005; Barkin

et al. 2006; Cardoza 2002; Fuzioka & Austin 2002).

115

Various researchers have identified different domains of scale items to measure parental

control on teenagers’ media, particularly TV involvement. Parental knowledge (e.g.,

knowledge about RTV shows, perception about content) about TV is most commonly

identified domain to measure parental control on teenagers’ TV usage (Carlson,

Lacznaik & Walsh 2001; Cardoza 2002). Rose, Bush and Kahle (1998) also identified

parental knowledge as a vital domain to measure parental control on RTV involvement

of teenagers. Furthermore, parental communication pattern is considered as a crucial

measurement domain which is agreed by various researchers (Carlson, Lacznaik &

Walsh 2001). Fujioka and Austin (2002) also mentioned that communication pattern

between teenagers’ and their parents could be expressed through discussion about TV

shows and selection process of TV shows. Furthermore, researchers have pointed out

that parental opinion regarding RTV show selection and they putting restriction on their

children watching RTV also are vital issues in this regard (Carlson, Lacznaik & Walsh

2001). Interestingly, co-viewing between parents and teenaged children are widely

accepted by the researchers to measure parental control on teenagers’ RTV usages

behaviour (Chan & McNeal 2003). On the basis of above discussion following Table

4.5 shows the domains of scale items to measure parental control on RTV involvement.

Table 4.5 Domain of Scale Items of Parental Control on TV

Construct Domain of Scale Items

Parental control on TV

Parental communication styles, Parental

opinion, Co-viewing, Parental knowledge.

(Carlson, Laczniak & Walsh 2001; Chan

& McNeal 2003; Fujioka & Austin 2002;

Rose, Bush & Kahle 1998).

116

4.3.4.6 Peer Group Influence

Various researchers indicated that peer groups particularly friend play an important role

on teenagers’ media usages and involvement that ultimately affects their consumer

socialisation (Lueg & Finney 2007). It is shown in various consumer socialisation

researches that teenagers’ media usage behaviour is highly influenced by their peers

group influences (Lueg & Finney 2007, Moschis & Moore 1979). However, peer group

influences can be defined form the perspectives of reinforcement, modelling, extrinsic

motivation, peer co-viewing, peers mediation (Edward & Grantham 2009; Lueg &

Finney 2007; Nathanson 2001)

According to current literatures, peers group reinforcement (e.g., taking advice from

friend to use media) regarding media has been most commonly identified domain to

measure peer group influences (Lueg & Finney 2007). Edward and Grantham (2009)

also agreed with the domain of peer group reinforcement along with teenagers’ extrinsic

motivation by their peer group to use any media. Most commonly, teenagers

participated in the focus groups discussions regularly talk and discuss about RTV shows

with their friends. Teenagers’ also consider their peer group perception towards TV

shows (e.g., entertaining, useful) influence their involvement with RTV (Nathanson

2001). Overall, peer group play a mediating role (e.g., watching TV shows together,

discussing about TV shows) in teenagers’ mind regarding involvement with media and

particularly RTV (Nathanson 2001). On the basis of above discussion Table 4.6 shows

the domain of scale items to measure peer group influences on RTV involvement.

Table 4.6 Domain of Scale Items of Peers Group Influences

Construct Domain of Scale items

Peer group influences Reinforcement, Extrinsic motivation, Peer

co-viewing, Peers mediation (Edward &

Grantham 2009; Lueg & Finney 2007;

Nathanson 2001)

117

4.3.4.7 List of Constructs, Domains and Scale Items

So far, researches on consumer socialisation in the context of RTV have been rarely

examined (Yoon & Garma 2006). Therefore, scale items in this research have been

adapted mostly from current research findings on consumer socialisation by media in

general. The following table (4.7) shows the constructs, domains and scale items that

are adapted from the current literature:

Table 4.7 List of Constructs, Domain and Scale Items

Constructs Domains Scale Items Cognition

(Consumption related cognition)

a) Product knowledge

b) Consumer role perception

c) Purchasing intentions

d) Information processing

e) recalling capacity

f) consumption related skills

(Bartlett, Griffiths & Badian

2008; Bush, Martin & Bush

2004; Moschis & Moore

1978; Zuckerman, Ziegler &

Stevenson 1978; Ward &

Wackman 1974)

1) RTV make me aware about

particular brands

2) I understand product usage

instruction from RTV

3) RTV helps me to recall about

certain products

4) RTV helps me to recognise

particular products and brands

5) RTV helps me to decide what

brands to buy

6) RTV is a good way to learn

about what products/bands are

available in market

7) The opinion of my favorite RTV

celebrities influence me to say

positive things about products or

brands to other people

8 ) RTV helps me to do value

analysis (compare price and benefit

of brands) before buying

Attitude (Consumption related attitude)

a) Fashion

b) Imitation

c)Inspiration from media

personalities

d) Attitude towards product

1) I get fashion tips from RTV

2) I like the product shown on RTV

3) I often buy clothing styles that I

have seen in RTV

4) My favorite RTV celebrities

118

placement

(Hirschman & Thompson

1997 Lawrie 2006; Russell,

Norman & Heckler 2004;

Schmoll et al. 2006).

make me fashion conscious

5) I want to get good figure like

RTV celebrities (e.g. slimmer,

muscular)

6) Thin models in RTV creates my

interest (e.g., plastic surgery, diet

habit, figure shape)

7) I see my favorite RTV celebrity

as my idol

8) I would like to have lunch with

my favorite RTV celebrity

9) I feel very strongly about

banning unethical products (e.g.,

tobacco, beer, alcohol) shown in

RTV

Values (Consumption related values)

a) Desire towards products

b) Prohibited product

consumption

c) Materialism

d) Antisocial behaviour

(Chan 2003; Ghani 2005;

Gruber & Thau 2003; Wyllie,

Zhang & Casswell 1998)

1) I like to own the newest things

shown on RTV

2) I like to own the best things

shown on RTV

3) It is better to have more

allowance from parents to buy the

brand shown on RTV

4) I often request my parents to get

the products shown on RTV

5) I often request my parents to get

the products shown on

advertisements in RTV shows

6) Some RTV shows promote use

of prohibited items (e.g., alcohol,

cigarette) among teenagers

7) Consumption of prohibited items

(e.g., alcohol, cigarette) makes life

more fun and exciting

8) I like the sexually appealing

outfit worn by RTV participants

and judges

Peers group influences a) Reinforcement 1) My friends and I talk about

119

b) Modelling

c) Extrinsic motivation

d) Peer co-viewing

e) Peers mediation

(Edward & Grantham 2009;

Lueg & Finney 2007;

Nathanson 2001)

content of RTV shows (e.g.,

entertainment, sex, information)

2) My friends and I watch often

RTV together

3) My friends think the content of

RTV shows can provide useful

information

4) My friends think content of RTV

are entertaining

5) My peers encourage me to send

SMS to RTV

6) My peers’ opinions on RTV

shows are important to me

7) My friends and I participate

together live on the RTV shows

Parental control on TV a) Parental communication

styles

b) Parental opinion

c) Co-viewing

d) Parental knowledge

(Carlson, Laczniak & Walsh

2001; Chan & McNeal 2003;

Fujioka & Austin 2002; Rose,

Bush & Kahle 1998)

1) I talk to my parents about RTV

content

2) My parents and I discuss about

RTV show selection

3) My parents and I watch RTV on

weekdays

4) My parents and I watch RTV

during weekends

5) My parents know what I watch

on RTV

6) My parents think some content

at RTV are really bad for teenagers

7) My parents turn off channel

while unsuitable contents are

shown on RTV

8) My parents prohibit me to watch

certain RTV shows

RTV Involvement a) Information search

b) Time spent, likings

c) Attention

d) Desire towards RTV

(Dolliver 2006; Higie & Feick

1) I watch RTV

2) I pay close attention when I

watch RTV

3) I read about RTV in news papers

4) I regularly vote to RTV (e.g.,

120

1989; Pecheux & Derbaix

2002; Perse 1998; Yoon &

Garma 2006)

SMS, Phone)

5) I would attend a public

appearance if my favorite RTV

character from the show were there

6) I imagine how I would act in a

similar situation shown on RTV

7) I watch other programs that

include contestants from the RTV

show

8) I wish I could be on RTV shows

9) RTV is just like real life

10) I would like to participate in

RTV if I like it.

11) RTV provides me a lot of

information

4.3.5 Scale Evaluation

A multi-item scale should be evaluated for accuracy and applicability (Greenleaf 1992).

This involves an assessment of reliability, validity, and generelisibility of the scale.

Reliability refers to the extent to which a scale produces consistent results if repeated

measurements are made (Perreault & Leigh 1989). In more formal terms, reliability is

the degree to which a set of two or more indicators shares in their measurement of a

construct. The indicators of highly reliable constructs are highly inter-correlated,

indicating that they are all measuring the same latent construct. As reliability decreases,

the indicators become less consistent and thus are poor indicators of the latent construct.

Approaches for assessing reliability include the test-retest, alternative forms, and

internal consistency methods. In this research the internal consistency method was used.

To assess the internal consistency reliability, a popular approach, coefficient alpha was

used, at the exploratory factor analysis stage. Coefficient Alpha “A measure of internal

consistency reliability, which is the average of all possible split-half coefficients

resulting from different splitting of the scale items” (Malhotra 2003, p. 282).

121

Validity is the ability of a construct's indicators to measure accurately the concept under

study. Validity is measured to a great extent by the researcher, because the original

definition of the construct or concept is proposed by the researcher and must be

matched to the selected indicators or measures. The validity of a scale also may be

defined as the extent to which differences in observed scale scores reflect true

differences among objects on the characteristic being measured, rather than systematic

or random error (Malhotra 2003). Perfect validity requires that there be no measurement

error. Researchers may assess content validity, criterion validity, or construct validity.

Construct validity includes convergent, discriminant, and nomological validity. In this

research, scales were tested for convergent and discriminant validity. Convergent

validity is the extent to which the scales correlate positively with other measures of the

same construct. This was achieved as 1) the model receives a satisfactory level of fit,

and 2) the regression co-efficient (factor loadings) of all the indicators are statistically

significant, e.g., greater than twice its standard error (Anderson & Gerbing 1998; Dunn

et al. 1994). Where as, discriminant validity estimates the degree to which a measure

does not correlate or converge with other constructs from which it is supposed to be

different. This was achieved as the co-efficient alpha of all the constructs was above

.70. The value of average variance extracted for all constructs were very close to or

exceeding 0.50 also indicates discriminant validity. There were also no significant cross

loading of individual variables.

Validity does not guarantee reliability, and vice versa. A measure may be accurate

(valid) but not consistent (reliable). Also, it may be quite consistent but not accurate.

Thus validity and reliability are two separate but interrelated conditions.

Generalisability refers to the extent to which one can generalise from the observations

at hand to a universe of generalisations. Traditionally, reliability methods can be viewed

as single-facet generalisibility studies. A test-retest correlation is concerned with

whether scores obtained from a measurement scale are generalisable to the universe

scores across all times and places of possible measurement. Even if the test-retest

correlation is high, nothing can be said about the generalisability of the scale to other

universes.

122

4.3.6 Questionnaire - the Research Instrument

The first step in questionnaire design was to specify the information or plan what to

measure. The answer for this question in this research has come from the extensive

literature review and focus group based qualitative research done earlier. Those two

steps helped to identify the constructs and variables to understand the role of RTV as

consumer socialisation agent of teenagers. Special care has been taken to translate the

desired question content into words that respondents clearly and easily understand.

Deciding on question wording is perhaps the most critical and difficult task in

developing a questionnaire. If a question is worded poorly, respondents may refuse to

answer it or answer it incorrectly. The first condition, known as item non-response, can

increase the complexity of data analysis (Omura 1983). The second condition leads to

response error. Unless the respondents and the researcher assign exactly the same

meaning to the question, the results will be seriously biased (Morgan 1990). To avoid

these problems, Malhotra (2003) recommends the importance of using ‘ordinary’ words

that have clear meanings, expressing statements both in positive and negative terms and

avoiding the inclusion of assumptions and estimates. In this research, a structured

survey instrument was developed to facilitate this. Pilot survey was then conducted

using the draft survey instrument among 20 Bangladeshi teenagers to establish content

validity of the survey instrument and focused to evaluate and determine the applicability

of questionnaire in Bangladesh. The questionnaire was tested; revised and than

finalised. Academic experts in the field were also consulted before finalising the

instrument. The final questionnaire was designed on the basis of the results of the pre-

test and the observations, comments, and recommendations of the academic experts.

The final instrument was developed in English, which was than translated to Bengali

(local language) and retranslated to English. The Bengali version was administered for

data gathering purposes. Mostly closed ended questions were used.

123

4.3.7 Sample and Sample Size

Information about the characteristics of a population may be obtained by conducting

either a sample or a census. Budget and time limits, large population size, and small

variance in the characteristic of interest favour the use of a sample.

In this research both qualitative and quantitative data was gathered from Bangladesh, a

typical developing country. In Bangladesh, 23% of total population falls is the 10-19

age groups, which is the largest segment of the country’s population (Quraishi, Bhuiya

& Mohammad 2004). Rapid growth of media and teenagers’ involvement with various

media is also observed in Bangladesh, particularly in Dhaka (Nath 2006). Teenagers in

Dhaka city have high involvement with electronic media, TV (Nath 2006). This was the

primary reason why data for this research has been collected from the teenaged children

population representing urban middle and lower social classes of Dhaka city. Also, the

concentration of the middle class population of Bangladesh is mainly in urban areas

(Islam 2005).

Target population for the study was the urban teenagers (13-19 age groups) representing

both genders. However, due to non-availability of a sample frame of the research

population, use of simple random probability sampling was particularly difficult.

Instead a multi-stage stratified sampling method was used. Accordingly, the research

was conducted in the two different schools representing two social classes of Dhaka city

corporation (municipal) area, which has the largest concentration of urban population of

the country with many working class slum dwellers. A map showing all the suburbs and

slums within the city was collected from the corporation. From this map, the entire

middle social class suburbs as well as lower social class slums were identified and listed

(Appendix 14). From list of middle class suburbs one suburb and from the list of lower

social class slums one slum area were randomly selected through lottery. This was done

by putting the names of each of the suburbs in a separate piece of paper which were

then folded and put in a basket and one piece of paper drawn. Same process was

followed for the other group (slums). There was one school in the selected suburb and

124

the slum, both of which were selected for the survey. Teenaged students from each of

the selected schools were further stratified into two genders. Using class lists, all

students from the 13-19 age groups representing academic classes 5 to 11 of the two

schools were identified. At Udayan school there were 360 such students and at Nilikhet

school there were 350 representing both genders. A total of 200 students were than

selected from each school through lottery using student identification numbers and

contacted in their respective classes. A total of 400 respondents were surveyed with

equal representation of each social class and genders.

Social class cannot be measured directly but indicators of social class status are

typically perceived by tangible factors (e.g., income, occupation, education) (Pope,

Brennan & Voges 2007). In this research, respondents’ social class was verified by

using composite variable technique (Pope, Brennan & Voges 2007). Historically, the

most widely used index is W. Lloyd Warner’s Index of status characteristics (ISC) is a

method of measuring social class that evaluates occupation, income, house, education

and location of home. The ISC is based on the following socio-economic indicators that

have been used to verify the respondents’ social status in this research:

• Occupation (lower class mostly blue collar, whereas middle class mostly white

collar)

• Income (lower class bellow national average, whereas middle class average and

above – current Bangladesh national average Taka).

• Education (lower social class mostly HSC and bellow, whereas middle social

class mostly above HSC).

• Dwelling area (lower social class from slums, whereas middle social class in

buildings).

Bangladesh is a developing country with a large population (152.6 million) and a

teenager population of 76.8 million (UNICEF 2005). Though, 45% of total population

live under the poverty level, urbanisation has been rapid and largely imbalanced with

concentration in Dhaka, the capital city of Bangladesh. According to World Bank

(2007), 25% of the population now lives in urban areas, while in 1960 the number was

125

just 5%. Urbanisation has been skewed toward Dhaka, making it among the fastest

growing metropolitan cities in the world. To support this newly migrated family in the

urban city, child labour is observed mostly in lower social classes. This segment of the

urban population is mainly concentrated in the slum areas of Dhaka. There is

government and NGO (non-government agency) run schools in Dhaka city for these

children. Nilkhet School, one of the two schools from which respondents with lower

social class background were selected is one such school. Whereas, the other school

selected, Udayan School which is centrally located caters for teenagers of middle social

class background.

4.3.8 Administration of Questionnaire

In line with the research objectives of this study, the translated (Bengali version)

questionnaires (Appendix 13) were self-administered by the teenagers from the two

selected schools in presence of this researcher, two research assistants and the

respective class teachers (no incentives were offered to students or teachers). Prior

permission from the two school principals (Appendix 7) were taken before

administering the questionnaires. Teenaged students from seven classes who were

selected through lottery (refer to section 4.3.7) and agreed to participate were then

brought together in their respective schools auditoriums to complete the survey at the

same time. The survey was conducted during the schools hours to maximise the

response rate. Discussion sessions with the respondents and school teachers on the

purpose and background of the research were also conducted in each of the seven

classes on the day prior to the survey, along with a copy of the translated questionnaire.

Any questions and concerns that the participants raised during these information

sessions were addressed and their consent taken using the participant consent form

(Appendix 4).

126

4.3.9 Data Preparation

In total 400 questionnaires were completed by the respondents from the two schools.

Data preparation began with a preliminary check of all questionnaires, followed by

more thorough checking of any missing values by this researcher with the help of two

research assistants. Editing consisted of screening questionnaires to identify illegible,

incomplete, inconsistent, or ambiguous responses. After screening, 32 questionnaires

were found either unanswered or for some of the questions more than one answer

chosen. These 32 incomplete and inconsistent questionnaires were dropped from further

data analysis. Rest of the 368 questionnaires were considered valid and used for

subsequent analysis.

4.3.10 Data Entry and Screening

Initially all collected data for the category questions from the 368 valid questionnaires

was codified and then all data entered into a SPSS for WINDOWS release 17.0 spread

sheet which was previously constructed and tested checking variable names, labels and

codes for category questions. Each completed questionnaire was given reference codes

to facilitate cross checking. Measures included the examination of the value of each

data cell independently by two research assistants who cross checked the data set

against the questionnaires. Furthermore, using the questionnaire reference codes, this

researcher personally cross-checked the questionnaires with the data set. Exploratory

factor analysis was conducted on this data set. Based on the results obtained variables

with low factor loadings were eliminated. The remaining data set was entered into

AMOS for final analysis.

127

4.3.11 Structural Equation Modelling and Hypotheses Testing

In this research, the conventional way of SEM involving evaluation of the measurement

and structural models together in one stage or at a time was followed (Diamantopolous

1994). One factor measurement models were estimated first in order to assess the

quality of the measurement items before the structural model was estimated. At this

stage, this research used CFA model to maximize the reliability of latent variables and

to reduce the variables to a manageable level in fitting structural equation models to

data at the subsequent stage (Holmes-Smith & Rowe 1994).

4.4 Ethical Considerations

Because the objects of inquiry in interviewing are human beings, extreme care has been

taken to avoid respondent concern about any aspect of the research. Traditional ethical

concerns revolve around the topics of informed consent (consent has been taken from

the respondents after they are carefully and truthfully informed about the research),

right to privacy (protecting the identity of the subject), and protecting from harm

(physical, emotional, or any other kind) (Fontana & Frey 1994). Consent from

principals of schools where interviews were conducted was taken and class teachers

were present while interviewing their students in this study.

For this research, a research protocol was developed to cover the ethical aspects of both

qualitative and quantitative research and approval (H 6890) was obtained from

University of Western Sydney’s Human Research Committee and all the relevant

documents are attached in the appendix section. Accordingly, at the beginning of the

study the respondents were promised about anonymity for their participation. Special

care has been taken while reporting the research findings that none of the respondents

are recognisable even through some contextual references.

128

Chapter 5 Refinement of Measurement Scales by Qualitative

Measures

5.1 Introduction

In the previous chapter, it was mentioned that both qualitative as well as quantitative

methods of research have been used in this study. Traditionally, qualitative and

quantitative approaches have been positioned as if they were polar and mutually

exclusive alternatives. More recently however, a more balanced approach has emerged

in which both types of researcher admit the contribution of the other. Overall,

qualitative research is best suited to areas calling for a flexible approach, while

quantitative research is necessary to define more precisely the issues identified through

qualitative methods. As a broad generalisation, qualitative research is an essential

prerequisite to most quantitative research in that it helps clarify the issue to be

addressed (Baker 1991).

Exploratory research is mostly carried out by qualitative research. An extension of the

exploratory intent is the use of group interviews for the purpose of refining and pre-

testing measurement scales, or other elements of a survey design, which is very

common in survey research (Desvousges & Frey 1989). Group interviewing may be

conducted questioning several individuals simultaneously in formal or in-formal

settings (Fontana & Frey 1994). In this research, two FGIs were conducted to carry out

qualitative research as a prelude to the quantitative research. This was done in order to

refine the constructs of the model for consumer socialisation by reality TV (CSRTV)

and their measurement variables that have been developed through literature reviews as

well as being reported (Table 5.1). This will later be tested using a quantitative method.

129

In this research, understanding teenagers’ consumer socialisation through involvement

with RTV is the pivotal issue. Moreover, as social structural variables, teenagers’ social

class differences, gender differences, parental control and peer group influences on

RTV involvement have been considered while exploring the key research issues. In this

qualitative study, students of two schools; Nilkhet school and Udayan school from

Dhaka city were selected for the focus group interviews. During the participant

selection process for the two FGIs; teenagers’ age groups (13-19 years), equal gender

participation, representing both lower and middle social classes who watch RTV for at

least an hour a week were considered. To conduct the FGIs, ten students, five from each

gender from both the schools were selected, and all the participants’ viewpoints were

coded for qualitative analysis. As participants, boys were coded as ‘B’ and girls as ‘G’

from both the groups, and then coded numerically while analysing and reporting focus

group interview findings. As mentioned earlier, social class differences and gender

difference are two important social structural variables. Therefore, during the time of

school selection, both the issues were given consideration. Most of the students in

Nilkhet school come from the lower social class. Whereas, the students of Udayan

School are mainly from the middle social class. However, both the schools offer co-

education. In the following section, the key findings of the qualitative phase of the

research are presented, focusing on the key constructs developed in chapter 3.

5.2 RTV Involvement

In this research, the impact of RTV on the consumer socialisation of teenagers is

determined by teenagers’ involvement with it. It has already been mentioned that the

concept of RTV is at an early stage of its development and very little research has been

carried out to measure involvement with RTV (Yoon & Garma 2006). Accordingly,

measurement items of RTV involvement in Table 5.1 were adopted; refined and new

scale items were developed based on the focus group interviews of the qualitative phase

of this research.

130

Findings of this phase of the research show that teenaged boys and girls representing

both the social classes watch RTV. However, it has been found that teenagers from the

middle social class watch relatively more RTV than their lower social class counterparts

due to their better access to local and satellite TV channels. Generally, girls watch RTV

more regularly than the boys within the same social class. Overall, eighteen participants

of the two social classes regularly (at least 5 hours a week) and two participants

occasionally watch RTV. Teenagers from the two social classes have different as well

as some common favourite RTV shows. Overall, most of the boys and girls from both

the social classes regularly watch only their favourite RTV shows, rather than watching

RTV for a particular number of hours. Moreover, they pay close attention when they

watch their favourite RTV shows. Interestingly, boys spend lesser time than girls

watching RTV and claim to pay more attention while watching RTV shows. Teenagers

watch RTV not only for entertainment but also for getting information about among

other things, products and services in the market. Teenagers also keep themselves

updated about RTV shows and it’s celebrities from various sources including

newspapers. Teenagers from the lower class social background have limited access to

newspapers, but still follow with interest, the news of RTV and it’s celebrities through

newspapers whenever they get the chance. Whereas, middle class teenagers have more

access to newspapers and know more than their lower social class counterparts about

RTV and it’s celebrities from that source. Overall, both boys and girls of both social

classes mention that they selectively read about their favourite RTV shows in

newspapers. Interestingly, teenagers do not watch everything that they read in

newspapers about RTV and its celebrities. Teenagers of both genders and social classes

know that the internet is a powerful medium for watching RTV, although none of them

watch RTV on the web regularly. In the middle social class, some teenagers have access

to the internet. They think the internet is expensive and because of the server’s slow

speed, they only occasionally watch RTV on the internet. On the other hand, lower

social class teenagers do not watch RTV on the net because none of them have a

computer or an internet connection in their homes. However, both the groups expressed

their interest in watching RTV on the internet, provided they could afford it and get the

chance.

131

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1):

1) Teenagers watch RTV.

2) Teenagers pay close attention when they watch RTV.

3) Teenagers read about RTV in news papers.

In addition, the following new scale item has been developed based on the findings of

the FGIs of this research:

4) Teenagers wish to watch RTV on the internet.

Generally, both the boys and girls are involved in RTV. They also show interest in

participating in RTV. However, they are also concerned about the standard of RTV

shows. Most of the boys and girls of the middle social class have shown interest in

participating in RTV. Furthermore, they have also shown interest in participating in

RTV, if their favourite RTV celebrities are in the program. Whereas, teenagers in the

lower social class have an interest in joining RTV shows, but have limited opportunity

to do so. Teenagers of this group are quite selective with the content of RTV and show

moderate interest in participating in RTV shows if their favourite celebrities are there.

Some of the respondents from both the social classes follow their favourite celebrities’

style and fashion. Teenagers however, have different motives behind attending RTV but

both the groups are also concerned about the content of the RTV that they wish to join.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

5) Teenagers wish they could be on RTV shows.

6) Teenagers would attend a public appearance if their favourite character from the

show were there.

132

Sometimes teenagers also play imaginative roles while watching particular RTV

shows. They often critically analyse complex situations like game shows, quiz shows

of RTV and like to imagine how they would react in that situation. Particularly,

respondents of middle social class tries to analyse a particular critical situation shown in

RTV then play an imaginative role more than their lower social class counterparts.

Further, this seems to be more common among boys than girls in this regard. However,

respondents have mixed feelings about RTV celebrities who participate in others

multiple types of programs. RTV celebrities who are participating in different programs

do not carry much appeal for the boys and girls of the lower social class. Most of the

participants of this group want to see the celebrities’ original performance, not

excessive media presence. Still though, most of them follow the RTV celebrities’ who

participate in multiple featured programs.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

7) Teenagers imagine how they would act in a similar situation shown on RTV.

8) Teenagers watch other programs which include contestants from the RTV show.

Finally, teenagers were asked about the credibility of RTV content. Most of the

teenagers from the lower social class are not very clear about the features of RTV, but

they still think that the content of RTV is real as ordinary people participate there.

Whereas, teenagers from the middle social class analyse content of RTV better than

their lower social class counterparts. The majority of them also think that RTV is real.

They consider these shows as more realistic, as the audience votes for participants and

programs are telecast as live shows.

On the basis of the above discussions, the following scale item is adopted and refined

from current literature (Table 5.1) as it is supported by the findings of the FGIs of this

research:

9) RTV is just like real life.

133

After the focus group interviews, previous scales identified from current literature were

refined and adopted for this study. Furthermore, the following scale items from current

literature were found irrelevant to a developing country's teenagers, based on the FGI

findings.

Teenagers from both the social class backgrounds watch RTV, but neither of them

allocates any particular number of hours to watching RTV. So, certain programs rather

than any particular number of hours viewing are important to them. Interestingly also,

teenagers from both the social classes have shown low interest in voting, either by short

message service (SMS), or making phone calls for RTV shows. None of the teenagers

of the middle social class vote regularly for RTV shows, though, most of them have

access to a mobile phone. Both the groups think it is unnecessary expenditure. Lower

social class teenagers are even more reluctant to vote for RTV shows. Overall, both

boys and girls from the two social classes are selective regarding sending votes for RTV

shows and none of them cast their vote regularly for the RTV shows and its celebrities.

Moreover, none of the teenagers in the middle class think about RTV situations when

they are not watching RTV, be it in leisure periods or in busy moments. Similar

responses were also found with both the genders of the lower class social background.

On the basis of above discussions, the following scale items are dropped from current

literature as they are not supported by the findings of the FGIs of this research:

1) Teenagers watch RTV at certain hours of the week.

2) Teenagers regularly vote in RTV (e.g., SMS, Phone calls).

3) Teenagers think about RTV shows when they do not watch them.

134

5.3 Consumption Related Cognition

Consumption related cognition is an important outcome variable of consumer

socialisation (Moschis & Moore 1982; Guilfoy 1998) which is considered as a construct

in this research. In the following section, FGI findings on teenagers’ consumption

related cognition through involvement with RTV has been discussed.

Teenagers from both the social classes acknowledge that RTV provides a lot of

information. Generally it has been found though, teenagers from the middle class social

backgrounds are relatively more aware of the features of RTV than their lower social

class counterparts. As a result, the FGI results show that the teenagers of this social

class are comparatively better than the lower social class teenagers regarding

information processing and recalling capacity provided by RTV. Both these groups

however, acknowledge that information provided by RTV about products and services

keep them updated about the market condition.

However, sometimes it does fail to provide sufficient and exact information which is

required by the teenagers. Most of the teenagers of the middle social class mentioned

that during the time of RTV shows, the name of various products and brands are shown

on the TV screen. Furthermore, anchors and participants often provide information

about products, as do judges of various RTV shows expose particular products and

brands. Most of the teenagers agreed that RTV creates brand awareness that generates

their interest to know more about new products. Participants from the lower social class

consider RTV as a source of information about new products and brands like their

middle class counterparts. However, teenagers from this social class background have a

lower understanding of the concept of brands, but still can recall the different names of

various products shown on RTV. It has already been mentioned that teenagers pay close

attention while watching their favourite shows. Therefore, RTV often develops

teenagers’ product usage instruction capacity and creativity about new fashion and

styles. Most of the participants from the middle social class agreed that RTV helps them

to understand product usage instruction by mentioning features and the nature of

135

products. Sometimes anchors also describe how to use particular products such as a

mobile phone, and MP3 etc. Moreover, teenagers from the middle social class pick

more product usage instruction information from RTV than their lower social class

counterparts. On the other hand, teenagers in the lower social class are also aware of

these products and services related instruction, which is often shown on RTV, but do

not follow them as strongly as their middle social class counterparts. Whereas, girls

understand product usage instruction better than boys as they spent more time on RTV.

Furthermore, FGI findings show, irrespective of their gender and social class, RTV

helps teenagers recall product and brand names. However, it has been observed that

although teenagers from both the social classes and genders can recall the name of

products and brands from RTV, the extent of that varies between social classes.

Teenagers from the middle social class understand the concept of brands and can

differentiate from generic products quite well. They can also recall different brand

names from RTV better than their lower social class counterparts. Whereas, teenagers in

the lower social class are not quite clear about the concept between brands and

products; they often recall the products and brands interchangeably. Overall, although

to a lesser extent, RTV also helps teenagers in the lower social class to recall different

brands.

RTV is also a good source of information about products and brands that are available

in the market. Teenagers from the middle social class can get an idea about the sources

(e.g., shopping mall, outlets) of products and brands that are shown on RTV.

Furthermore, boys have more opportunities than girls to move around different

shopping malls together with their friends. However, teenagers from the lower social

class mostly rely on RTV advertisements rather than the programs to know and learn

about the products and brands that are available in the market. However, it is observed

that overall, girls are to some extent better than boys in understanding and processing

products and brands related information from RTV.

136

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

1) RTV provides lot of information (e.g., products, brands) for teenagers.

2) RTV makes teenagers aware about particular brands.

3) Teenagers understand product usage instruction from RTV.

4) RTV helps teenagers to recall certain products.

5) RTV is a good a way to learn about what products/brands are available in the

market.

Moreover, FGI findings show that teenagers from both the social classes keenly observe

and follow what brands and products are used by the participants, anchors and judges of

RTV. In particular, judges or any special guest of RTV create awareness about new

products among the teenagers. Moreover, teenagers like to share and discuss with others

what participants, anchors and judges are wearing, commenting on and promoting in

RTV. Therefore, teenagers sometimes say positive things about products and services to

others which are recommended by their favourite celebrities. In particular, participants

from the middle social class have more access to local and satellite TV channels than

their lower social class counterparts. So, they have more information to share with

others. Moreover, some of the participants from this group think RTV celebrities

influence their buying decisions as they like certain music and film celebrities. In this

social class RTV celebrities create some initial awareness that leads teenagers to say

positive things to others about products and brands. Furthermore, sharing information

with peers and friends is quite common among the teenagers of both the genders. So

teenagers, irrespective of their gender, also influence moderately their friends and

others about brands advised by their favourite RTV celebrities.

137

On the basis of the above discussions, the following scale item is adopted and refined

from current literature (Table 5.1) as it is supported by the findings of the FGIs of this

research:

6) The opinions of teenagers’ favourite RTV celebrities influence them to say positive

things about products or brands to other people.

It is needless to mention that most of the multinational and local big companies sponsor

RTV to promote their brands. Teenagers from the middle social class consider not only

RTV content, but also consider advertisements in RTV for new product ideas. Whereas

in the lower social class, teenagers mostly rely upon advertisements shown on RTV for

new product ideas. They mentioned that it is easier for them to follow new products and

brands ideas from advertisements on RTV, as they are straightforward. Moreover

teenagers, irrespective of their gender and two social classes, have similar opinions

about the brands that sponsor RTV. Teenaged boys and girls can easily recall the

sponsoring brands of RTV that ultimately create brand image. Moreover, these RTV

sponsoring brands create awareness among the teenagers. Most of the participants

mentioned that, ‘We believe in big brands’. Overall, these big brands are more

successful in creating awareness about new products and services in the market.

Teenagers from the middle social class selectively follow fashion and styles shown on

RTV and then they adopt it according to their tastes and personalities. However,

teenagers from the lower social class show interest in the creative features that RTV

offers, but often they fail to fulfill it in their personal lives due to social and economic

constraints.

On the basis of the above discussions, the following new scale items has been

developed based on the findings of the FGIs of this research:

7) Advertisements shown in RTV give teenagers new product ideas.

8) Teenagers prefer the brands that sponsor RTV as they increase their brand

awareness.

9) RTV makes teenagers do something new and creative.

138

It has already been mentioned that RTV helps teenagers recall different products and

brand names. Generally however, they do not differentiate much between recalling and

recognising products and brands, rather, they mean it interchangeably. Overall,

respondents from both the social class backgrounds are comfortable with the concept of

recalling any brands or products instead of product and brand recognition. However,

recalling different products’ names does not necessarily help them to decide strongly

what brands to buy. Teenagers from a middle class social background think that RTV is

a good source of knowledge about new products and services in the market, but they

strongly consider their peer group’s opinion and knowledge before making buying

decisions. Teenagers from the lower social class background also agreed with their

middle class counterparts in this regard. They mostly consider their affordability and

parental consent before making any buying decisions instead of being influenced by

RTV.

Teenagers from both the social class backgrounds have agreed that RTV is a good

source of information because it creates awareness about different products. However,

quite often RTV fails to provide the desired information to make buying decisions.

Teenagers of both the groups think RTV provides a lot of information, but not always

accurate information. Furthermore, teenagers of both the groups failed to understand the

concept of value analysis, as they are not well aware about this term. So, value analysis

is quite unusual among them while making buying decisions.

On the basis of the above discussions, the following scale items are dropped from

current literature as they are not supported by the findings of the FGIs of this research:

1 RTV helps teenagers to recognise particular products and brands.

2 RTV helps teenagers to decide what brands to buy.

3 RTV helps teenagers to do value analysis.

139

5.4 Consumption Related Attitude

In the consumer socialisation discipline (Hastings & Aitken 1995; Russell, Norman &

Heckler 2004), consumption related attitude was found to be an important outcome

variable of the consumer socialisation, which is also a construct in this research. In the

following section, FGI findings on teenagers’ consumption related attitudes through

their involvement in RTV are discussed.

Generally, teenagers from both the social classes watch RTV for entertainment and

information. Teenagers from the middle social class more closely observe outfits, hair

styles and the latest fashions of the anchors and participants of RTV. Moreover, they get

fashion and style related tips from lifestyle based RTV which develops their fashion

consciousness. Furthermore, teenagers’ favourite celebrities on RTV provide ideas

about recent trends of style and fashion. Generally, girls are more involved with fashion

and lifestyle based shows. However, teenagers of both the groups think rationally when

they think of buying products and brands shown on RTV. In particular, girls try to

closely follow the fashion tips from RTV celebrities in lifestyle based RTV shows.

However, social, religious and economic issues are also key variables while they make

their buying decisions. Overall, teenaged boys and girls in the middle social class like

the attractive figures of RTV celebrities. Some of them want to have a good figure or

physique like RTV celebrities. Particularly boys like the celebrities who have muscular

physiques. Moreover, most of the teenaged boys think celebrities should have a good

figure and physical fitness. Girls on the other hand appreciate the good looking

celebrities, particularly the good figure of actresses and singers.

Teenagers from the middle social class are comparatively more involved with RTV.

They mostly watch RTV for fun and pleasure. They also get pleasure from seeing what

their favourite celebrities wear on RTV and pay attention to their favourite shows that

satisfy their style and fashion related information needs. Girls also pay attention to

products shown on RTV. Particularly, girls strongly consider RTV to be a good source

of their fashion related product information. Moreover, the fondnesses towards the

140

products shown on RTV are also felt strongly in this social class irrespective of gender,

although they are quite conscious and selective about their tastes and personalities.

Interestingly, teenaged boys from this social class are more proactive than their female

counterparts in following the instructions from technology based RTV (e.g., MP3, I

phone). However, teenagers from this social class only show their interest towards their

favourite celebrities on RTV if it matches their taste and personality. They also

generally agree with their lower social class counterparts that social, cultural, and

religious issues are considered while making any purchasing decision regarding

products and brands that are shown on RTV.

Whereas, unlike their middle social class counterparts, teenagers from the lower social

class also show interest towards physically attractive and good looking RTV celebrities.

In particular, teenaged boys have a strong liking for physically attractive models.

Teenaged girls also have shown similar interest in physically attractive RTV celebrities.

However, none of them actively try to achieve this because they know it is impossible,

rather they fantasise.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

1) Teenagers get fashion tips from RTV.

2) Teenagers’ favourite celebrities on RTV make them fashion conscious.

3) Teenagers like the products shown on RTV.

4) Teenagers often buy clothing styles that they have seen on RTV.

5) Teenagers want to have a good figure like RTV celebrities.

Teenagers from both the social classes, irrespective of gender differences have a

particular liking toward some RTV celebrities, but sometimes they are quite concerned

about their RTV celebrities’ activities and personalities. Teenagers from the middle

class social background critically evaluate RTV celebrities’ personalities and outlooks.

141

Interestingly, when they consider any RTV celebrities as their idol, they hardly consider

the celebrities' outlook; rather they consider celebrities' knowledge and personality.

Irrespective of gender, both groups have a fondness towards their favourite RTV

celebrities and want to show strong attachment to them. They would even not mind

having lunch with their favourite celebrities.

Interestingly, a number of teenagers from the lower social class consider actresses and

singers as their favourite idols. However, while both boys and girls from this social

class think that some RTV celebrities are their idols, none of them dreams of following

RTV celebrities. However, they are also interested in meeting their favourite RTV

celebrities. They even desire to visit an expensive restaurant if they get the chance to

have lunch with their celebrities. Most of them think having lunch with their favourite

RTV celebrities is also a good opportunity to take a picture and get an autograph!

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1), as they are supported by the findings of the FGIs of

this research:

6) Teenagers see their favourite RTV celebrities as their idols.

7) Teenagers would like to have lunch with their favourite RTV celebrity.

It has already been mentioned that RTV makes teenagers aware about various products

and brands. As teenagers have a favorable attitude towards famous brands, whenever

RTV promotes any brand of a famous company, it has a positive affect on the teenagers

minds. Most of the teenagers from both the social classes think that the importance

attached to a brand by RTV helps them to know about new brands. Interestingly,

teenagers from both the groups have shown similar preferences towards the foreign

brands that sponsor RTV. Teenagers from the middle social class understand the

concept of the multinational brand and they think credibility of multinational brands is

higher than the local or unknown brands that sponsor some RTV too. Therefore,

multinational brands that sponsor RTV create a greater awareness among the teenagers

142

than the local or unknown brands. Whereas, teenagers from the lower social class do not

have such an understanding regarding multinational brands like their middle social class

counterparts. They can only understand about foreign products. Furthermore, they show

interest in watching RTV which is sponsored by foreign products. Overall, both the

groups show interest in knowing about foreign brands shown on RTV. Interestingly,

both the groups also have common views about showing unethical products (e.g.,

alcohol, beer) on RTV, whether through an advertisement or as a product placement.

Teenagers from both the social classes strongly believe these unethical products (e.g.,

alcohol, beer) may be common in western society but it should not be promoted in the

local media. However, to some extent, teenagers of the middle social class are more

lenient than their lower social class counterparts regarding showing unethical products

at RTV.

On the basis of the above discussions, the following scale item is adopted and refined

from current literature (Table 5.1):

8) Teenagers feel very strongly about banning unethical products (e.g., tobacco,

alcohol) from being shown in RTV.

In addition, the following new scale item has been developed, based on the findings of

the FGIs of this research:

9) Teenagers prefer to buy multinational brands if they are shown by local RTV.

Teenagers from both the social class backgrounds have shown some attachment to RTV

celebrities. Particularly, good looking celebrities are well appreciated by most of them.

However, none of them appreciate these celebrities when they get plastic surgery to

enhance their attractiveness. Most of the participants appreciate natural beauty, rather

than the artificial beauty of RTV celebrities. Furthermore, not only good looking, but

also the physically attractive RTV celebrities are warmly accepted by the teenagers

from both the social classes. Teenagers, particularly from the middle social class know

143

that movie celebrities maintain a special diet for an attractive figure. Interestingly none

of them like skinny RTV models. Similarly, teenaged boys and girls in the lower social

class do not like excessively thin RTV models. None of the respondents think it is

important that they should have figures like RTV celebrities, but rather that they should

be in good health.

On the basis of the above discussions, the following scale item is dropped from current

literature as it is not supported by the findings of the FGIs of this research:

1) Thin models in RTV create teenagers’ interest in plastic surgery.

5.5 Consumption Related Values

Consumption related values have been indicated as an outcome of consumer

socialisation research (Chan 2003; Gruber & Thau 2003). In the following section, FGI

findings regarding teenagers’ consumption related values developed through their

involvement in RTV have been discussed.

Teenagers from both the social classes expressed some similarities and distinctive

features regarding consumption related values. Most of them consider affordability and

parental consent as important while making purchasing decisions. Generally, they think

products and brands shown on RTV are quite expensive. Therefore, some of them do

not show a strong interest to get the new or expensive products shown on RTV.

However, ‘pester’ power is observed in both the social classes. In particular, teenagers

from the middle social class have more affordability, but they are quite selective

regarding any products and brands quality and features that are shown on RTV.

Interestingly, if any promoted brands on RTV catch the attention of teenagers, then they

try to get that. Moreover, they also express strong desire to get the newest brands (e.g.,

MP3, mobile phone) if they like them. Overall, teenagers irrespective of both genders

are quite selective about products shown on RTV and they always think of the best

brands whenever they try to buy. However, most of them get bursary from their parents,

and their parents also keep monitoring how this money is used. However teenagers,

144

irrespective of gender, wish to have more financial freedom to make their own buying

decisions. Therefore, whenever they like any product they try to convince their parents.

Others also express their desire to their parents about preferred brands shown on RTV

or even in advertisements on RTV. Overall, teenage boys of the middle social class are

comparatively more demanding than the girls in the same social class.

Whereas teenagers from the lower social class are less demanding than their middle

social class counterparts about having products shown on RTV. Two boys and three

girls wish to have the newest and best things shown on RTV. One participant (B 2) said,

‘Every one likes good products’. Furthermore, three boys and one girl added that new

things are always appreciated. However, they also realise products shown on RTV are

expensive, as well as not always matching their taste and choice. Furthermore,

sometimes they wish to get the best products shown on RTV, but due to financial

constraints they compromise. Teenagers from this social class also pay attention to what

products and brands are shown in RTV advertisements. Similarly, like their middle

social class counterparts, they are fully dependent on their parental bursary and mostly

request their parents to buy any product for them. Therefore, most of them wish to have

more financial freedom, to have their favourite products that are shown on RTV.

However in general, parents mostly take their teenaged kids with them for shopping for

whatever is required.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

1) Teenagers like to own the newest things shown on RTV.

2) Teenagers like to own the best things shown on RTV.

3) It is better to have a bigger allowance from parents to buy the brand shown on RTV.

4) Teenagers often ask their parents for money to buy products that are promoted in

RTV programs.

5) Teenagers often ask their parents to get the products shown on advertisement in

RTV shows.

145

Teenagers of both the social classes are not only aware of different brands but also are

concerned with unethical products shown on RTV. Sometimes prohibited items such as

alcohol and cigarettes are promoted by the participants as product placement in foreign

RTV. Furthermore, sponsors promote their brands through RTV. Teenagers irrespective

of gender think that some RTV content and advertisement promotion of prohibited

items may deceive teenagers and lead them to consume them. Furthermore, parents are

particularly strict regarding these unethical product promotions through RTV. Even

from a societal point of view this unethical product consumption is also discouraged

heavily in the context of Bangladesh.

On the basis of the above discussions, the following scale item is adopted and refined

from current literature (Table 5.1) as it is supported by the findings of the FGIs of this

research:

6) Some RTV shows promote the use of prohibited items (e.g., alcohol, cigarettes)

amongst teenagers.

Teenagers irrespective of gender from both the social classes are quite conservative

regarding prohibited items (e.g., alcohol, cigarettes) promoted on RTV, either as part of

program content or as advertisements. Interestingly, a couple of boys do not mind

having a sip of an alcoholic beverage on a special occasion, but girls strongly oppose it.

Most of them believe that consuming these prohibited items can not be a source of fun

and excitement; rather they bring trouble in life. Furthermore, teenagers from both the

groups have also shown similar attitudes towards sexually appealing outfits worn by

participants, or by judges. These days, not only in foreign RTV, but also in local ones,

there are lifestyle based fashion shows where participants and sometimes judges are

wearing sexually appealing outfits. Most of the respondents believe that RTV

participants who wear sexually revealing outfits want cheap publicity. Further, they also

believe judges should maintain their image, not wearing unnecessarily sexually

revealing outfits.

146

On the basis of the above discussions, the following scale items are dropped from

current literature as they are not supported by the findings of the FGIs of this research:

1) Consumption of prohibited items (e.g., alcohol, cigarettes) makes life more fun and

exciting.

2) Teenagers like sexually appealing outfits worn by RTV participants and judges.

5.6 Social Structural Variables and RTV Involvement of

Teenagers

In the consumer socialisation discipline, various social structural variables received

significant attention from researchers (Kamaruddin & Mokhlis 2003; Mangleburg &

Grewal 1997). In various consumer socialisation research; social class differences,

gender differences, parental controls and peer group influences on media usage were

observed as social structural variables (Cardoza 2002; Kuhl 2006; Lueg & Finney 2007)

that influence the consumer socialisation of teenagers (Carlson, Lacznaik & Walsh

2001; Cardoza 2002). Accordingly, the above mentioned variables have been

considered as social structural variables in this research as a means to understand the

consumer socialisation of teenagers by RTV in a developing country.

Television plays an important role as a consumer socialisation agent in the consumer

socialisation of teenagers (Arnett 1995; Choi & Ferle 2004). Not only in developed

countries but also in developing countries, teenagers are involved with TV (Bhosale &

Gupta 2006). Bangladesh is one of the developing countries in South Asia. Teenagers in

Dhaka, the capital of Bangladesh, have strong involvement with TV (Nath 2006).

However, their choice of television programs and the nature of their involvement in the

programs vary due to the above-mentioned social structural variables. Recently, RTV

has been receiving the attention of teenagers in Dhaka city. Particularly, teenaged boys

and girls from the middle and lower social classes of Dhaka city have shown

involvement with RTV. In the following section, based on the findings of the qualitative

stage of this research, teenagers’ involvement with RTV is explained from the

147

perspective of each of the social structural variables to understand the consumer

socialisation of teenagers.

5.6.1 Peer Group Influences on RTV Involvement

Various researches (Churchill & Moschis 1979; French, Pidada & Victor 2005) have

shown that peers play a significant role in the consumer socialisation as a socialisation

agent. Furthermore, it has also been shown that peer groups play an important role in

teenagers’ media, particularly TV involvement, which ultimately affects their consumer

socialisation (Lueg & Finney 2007; Nathanson 2001). Peer group influence on

teenagers’ TV involvement was also considered as a construct in this research.

School is not only a place for education but also a source of socialisation with peers and

friends. Every weekday, teenaged boys and girls from both the social class backgrounds

spend a long time with their peers and friends at school. Based on the findings of FGIs

for this research, it is clearly evident that the teenagers from both groups share and

influence each other with their TV program selection. However, generally, teenagers

from both the social classes have shown attachments towards their friends.

Furthermore, teenagers discuss with their friends what they watch on TV whatever the

content is (e.g., entertainment, sexual issues and information). They feel more

comfortable sharing with their peers and friends about what they watch on RTV, than

with their parents.

Generally, teenaged boys and girls keep in touch with their friends and peers even at

weekends. Most of the teenagers said it is very difficult to spend time with peers outside

of classes during the weekdays. Some of the respondents said they visit and meet with

their peers and friends at their homes during the weekends. Sometimes, teenagers of the

middle social class watch RTV on the internet while they meet together. On the other

hand, both boys and girls in the lower social class watch RTV together with their peers

and friends when they meet each other at weekends or during any family occasions.

Furthermore, it is not unusual that they also watch RTV together, as some of them live

within the same suburb.

148

The following new scale item has been developed, based on the findings of the FGIs

(table 5.1) of this research:

1) Teenagers talk to their friends about RTV shows.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

2) Teenagers talk to their friends about RTV content.

3) Teenagers watch RTV together with their friends.

Generally, teenagers from the middle social class get more opportunity to meet their

friends. They often meet with their friends even outside the school (e.g. coaching class,

social groups). Therefore, they really care about their friends’ ideas and opinions about

different issues, including RTV show selection. Particularly, teenaged girls share more

commonly, as well as prefer the ideas of their peers regarding selection of RTV shows

like celebrity shows, fashion shows and information based RTV shows. They also think

that some RTV shows; particularly lifestyle based shows provide ideas about style and

fashion related products and during that program selection they consider their peers’

opinions. Furthermore, some RTV shows provide useful career tips to teenagers, though

a general finding is that teenagers watch RTV particularly to have fun and

entertainment. In particular, talent hunt shows, quiz shows and fashion shows are a real

source of fun and joy for them. Teenagers of both the genders from the two social

classes also share about this experience mostly with their friends. Peers and friends

often give their opinions and suggestions regarding RTV program and content

selection which ultimately influences purchasing decisions (e.g., buying MP3, DVD).

As teenagers have a reciprocal role of influencing each other, so they really care about

their counterparts’ opinions about different RTV shows. Generally, teenaged girls are

more proactive in sharing and making buying decisions together.

149

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

4) Teenagers’ peers think the content of RTV shows can provide useful information.

5) Teenagers’ peers think the content of RTV is entertaining.

6) Peers’ opinions of RTV shows are important to teenagers.

Teenagers of both the social classes watch RTV for fun and entertainment, but very few

of them have strong commitment towards any particular shows. None of the participants

from the middle social class send SMS or make phone calls to RTV, though most of

them have personal mobile phones. On the other hand, teenagers of the lower social

class show some desire to send SMS and make phone calls to their favourite RTV

shows, but none of them have a personal mobile phone. It has been found that

teenagers’ peers do not influence or encourage them to vote with RTV either. However,

the lower social class teenagers have shown their interest to participate as an audience

in RTV but hardly get the opportunity either alone or with their friends. Similarly,

teenagers from the middle class also have some interest in attending RTV as an

audience, but they are very selective in their decision whether to attend or not.

On the basis of the above discussions, the following scale items are dropped from

current literature as they are not supported by the findings of the FGIs of this research:

1) Teenagers encourage their peers to send SMS to RTV.

2) Teenagers and their peers participate together live on the RTV shows.

150

5.6.2 Parental Control on RTV Involvement

According to current literature, parental communication styles (Fujioka & Austin 2002),

parental co-viewing of TV with teenagers (Kim & Shin 1996), and parental instructions

regarding TV influence teenagers’ TV consumption behaviour (Keith 1986), which

ultimately affect teenagers’ consumption decisions behaviour. In this research, Parental

control on RTV involvement of teenagers is considered as a construct.

Generally, it was found that teenagers of both the social classes and genders have close

relationships with their parents. Most of the teenagers of both genders discuss and share

with their mothers the content of RTV. Sometimes their mothers show interest towards

a particular RTV show. Furthermore, though to a lesser extent, some boys and girls

discuss and share views with their fathers about RTV content. Interestingly girls show

more attachment to their fathers than boys. They think sometimes, that their fathers

understand them better than their mothers. In addition, sometimes their fathers ask them

about RTV shows. Similarly, teenagers in the middle social class irrespective of gender

discuss the content of RTV with their parents, but not very frequently like their lower

class social counterparts. However, regarding the RTV program selection, parents and

teenagers play a joint role in the middle social class. Sometimes, boys and girls play an

active role in RTV show selection. Whereas, in the lower social class, it is mostly the

parents who play the active role in RTV show selection though both parents and

teenagers discuss their favourite shows selection.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

1) Teenagers talk to their parents about RTV content.

2) Teenagers and their parents discuss RTV show selection.

151

In both the social classes, not only discussing the RTV shows, but also co-viewing is a

common practice between the parents and their teenaged children. However, it is not a

common phenomenon. In the lower social class, parents have a more regular presence

during the time of watching RTV with their teenaged children. Most of the respondents

in the lower social class said they do not get the chance to watch RTV alone at home.

One or the other parent is always there to watch RTV with them. Whereas, teenagers in

the middle social class, get considerably more freedom to watch RTV as in the middle

social class, often both parents work. However, more than half of the participants from

both social classes insist that they love to watch RTV with their parents. Interestingly,

teenagers’ co-viewing of RTV with parents varies in the two social classes during the

weekdays and weekends. In the middle social class, the majority of the teenagers watch

RTV with their parents during the weekends. Parents also give them time during the

weekend since some working parents are busy during weekdays. Furthermore, during

weekdays, if they watch RTV with their parents, the duration of it is much shorter.

Similarly, teenagers from the lower class social background also watch RTV with their

parents during weekends. They said, during the weekdays, they do not get enough time

to watch TV. During the weekend, TV is the main source of their joy and entertainment.

Generally, both boys and girls spend a moderate amount of time co-viewing RTV with

their parents.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

3) Teenagers and their parents watch RTV at weekends.

4) Teenagers and their parents watch RTV on weekdays.

Parental guidance, monitoring and knowledge about their teenagers’ lives are key

features of the Bangladeshi culture. In the lower social class, parents have a better

knowledge about their teenagers’ preferred shows as co-viewing of TV is more

common in their homes. On the contrary, in the middle social class, teenagers have

152

relatively more freedom due to their parental professional business and availability of

various media. However, parents of both the social classes think, due to cultural and

social differences, some RTV content is not really suitable for their kids. Moreover,

they also think some RTV content is unacceptable to some extent. Interestingly, parents

are more strict and rigid towards their teenaged boys than their teenaged girls regarding

TV show selection.

On the basis of the above discussions, the following scale items are adopted and refined

from current literature (Table 5.1) as they are supported by the findings of the FGIs of

this research:

5) Teenagers’ parents know what teenagers watch on RTV.

6) Teenagers’ parents prohibit them from watching certain RTV shows.

Overall, parents of both the groups are more concerned about the content of RTV that

their teenaged children watch than the amount of time spent viewing RTV. Normally,

parents discourage their teenagers from watching RTV at hours of the day that they

consider to be study time. However, respondents generally reported that, they do not

face too much restriction when watching their favourite shows at prime times.

Interestingly, in the middle social class, parents are comparatively more lenient towards

teenagers’ media freedom than lower social class parents. Parents from both the social

classes rarely turn off, rather than shift TV channels while unsuitable content is shown.

On the basis of the above discussions, the following scale items are dropped from

current literature as they are not supported by the findings of the FGIs of this research:

1) Teenagers’ parents think some content on RTV is really bad for their teenagers.

2) Teenagers’ parents turn off the T.V. while unsuitable content is shown on RTV.

153

5.7 Social Class and Gender Differences of Teenagers and

their RTV Involvement and Consumer Socialisation

From the current literature, not only parental control and peer group influences are

addressed as social structural variables, but also social class as well as the gender

differences of teenagers considered in this research. In the following section, it has been

identified that social class and the gender differences of teenagers influence their

involvement with RTV differently which ultimately affects their consumer socialisation.

5.7.1Social Class Differences and RTV Involvement of Teenagers

It has already been mentioned that separate FGIs were conducted among the teenaged

respondents selected from the lower and middle class social backgrounds. Each group

of teenagers was asked similar types of questions in order to understand their

involvement with RTV and how it influences their buying behaviour. Initially,

teenagers’ knowledge was discussed, to explore the range of ideas about RTV and then

how it relates to their consumer socialisation.

All the participants of both the social class backgrounds mentioned that they regularly

watch different TV shows. Recently, RTV has got the keen attention of teenagers of

Dhaka city. Both teenaged boys and girls of the two social classes watch RTV quite

regularly. Furthermore, they also pay close attention to their favourite RTV shows.

Most of the participants from both the social class backgrounds watch RTV for

entertainment, excitement, updated information and to gain knowledge. The most

watched RTV by both the groups are musical talent hunt shows, quiz shows, game

shows and lifestyle based shows. Furthermore, they also have shown interest in RTV

celebrities.

154

Teenagers from the middle social class get more information about RTV and its

celebrities than lower social class teenagers, as they have more access to news papers.

Similarly, teenagers from the lower social class show interest in reading about RTV

shows and celebrities from news papers, but not as strongly as their middle social class

counterparts, as they have limited access to newspapers. Generally, teenagers from the

middle social class would be interested in meeting their favourite celebrities. So, they

strongly express their desire to attend selective RTV shows. Whereas teenagers from a

lower class social background also express their interest towards RTV celebrities, but

they have limited opportunity and access to attend RTV live shows.

Teenagers from both the social classes follow some favourite RTV shows, and that is

why sometimes as an audience, they imagine how they would act in a particular

challenging and critical situation shown on different RTV shows like ‘Junior Horlicks’.

However, most of them from both the groups just watch it for fun and are not much

attached to it. Most of the boys and girls from the middle social class have favourite

RTV celebrities and for that reason they give attention to other programs where their

favourite RTV celebrities are also performing. Also, they like to see how talented they

are with other programs and they enjoy watching them as well. Teenagers from the

lower social class follow moderately what RTV celebrities are doing with different

programs, but some of them think that RTV celebrities are unnecessarily involved with

other programs for shortcut success.

Interestingly, RTV has some distinctive features that make it more exclusive than other

TV shows. RTV shows are naturally unscripted and interactive with viewers. All the

participants from the middle social class school think the content of RTV is real, as they

watch live shows and have the chance to vote for participants. Whereas, teenagers of the

lower social class moderately consider that the content of RTV is real, as ordinary

people are participating in RTV shows. However, popularity and originality of RTV

came from the western media to Bangladesh. Information about RTV and their

celebrities is regularly available in various websites. Unfortunately, respondents from

the lower social class have no personal computers and no access to the internet as well.

155

Whereas, middle class teenagers have some access to the internet and they watch RTV

on the internet occasionally. However, both the groups express a desire to get

information and search websites about RTV.

Overall, FGI findings show that there are some significant similarities as well as

differences between the two groups regarding their RTV involvement and the way it

effects their consumption related cognition, attitude and values.

5.7.2 Gender Differences and RTV Involvement of Teenagers

Teenagers’ gender differences have been found to be an important social structural

variable in the consumer socialisation of teenagers (Belk 1991; Kamaruddin & Mokhlis

2003) and have also been considered as an important social structural variable in this

research. Furthermore, some consumer socialisation studies have shown that

involvement with TV varies between boys and girls and that influences teenagers’

consumer socialisation differently (Grimes et al. 2004; Knobloch-Westerwick 2007).

Qualitative findings in this research show that boys and girls from both the social class

backgrounds watch RTV quite regularly. Generally, girls are relatively more regular

viewers of RTV than boys. Furthermore, girls are comparatively regular viewers of

certain RTV programs of their choice than boys. Moreover, they watch a variety of

RTV (e.g., musical talent hunt shows, quiz shows, life style based shows, fashion

shows). Whereas, teenaged boys watch particular RTV shows quite regularly (e.g., Quiz

shows, Musical talent hunt shows). Interestingly, boys often are more attentive than

girls when they watch their favourite shows. However, most of the boys and girls from

both the social class backgrounds watch RTV for fun, entertainment and information

about new products in the market. Furthermore, both the groups also read about RTV

shows and RTV celebrities in newspapers with enthusiasm. The boys read mostly about

sports and quiz based RTV, whereas the girls are more interested in entertainment and

glamour based RTV shows. Teenaged girls have a variety of favourite RTV shows and

they also closely observe what fashion and styles are followed by RTV celebrities.

156

Moreover, girls also express their desire to meet their favourite celebrities in RTV

shows. On the other hand, boys are keener of sports stars rather than RTV stars.

Generally however, boys show slightly more interest than girls in participating in RTV

shows, because in a country like Bangladesh, girls are not encouraged to be seen in

public places unaccompanied by a family member. Generally, both boys and girls are

quite selective about participating in RTV.

Since girls watch more varieties of RTV, they pay attention to the challenging (e.g.,

elimination shows) situations shown on RTV. Furthermore, they also play imaginative

roles and think about what they would do in a challenging and critical situation in RTV.

On the contrary, boys are quite selective when watching RTV, and still think about

critical and challenging situations on RTV, but only when they strongly like it.

However, teenaged girls from both the social class backgrounds watch what their

favourite RTV celebrities are doing with other programs as well, and to some extent

they follow the fashion and style of their favourite celebrities. Whereas boys are not

strong followers of RTV celebrities regular activities with other TV shows.

RTV shows are quite popular on various TV channels these days. RTV offers some

exclusive features (e.g. interactivity, unscripted programs). So, both boys and girls

believe the content and structure of RTV is real since ordinary people participate, and

they can make phone calls during the shows. Furthermore, both the groups expressed

their desire to watch RTV on the internet and like to get RTV related information from

the websites. Unfortunately, none of them watch RTV on the internet regularly due to

low speed, unavailability and other economic reasons.

Overall, FGI findings show that there are some significant similarities as well as

differences between the two groups regarding their RTV involvement and the way it

effects their consumption related cognition, attitude and values.

157

5.8 Conclusion

From the existing literature, various social structural variables (e.g., social class

differences, gender differences) along with parental control and peer group influences

were addressed in this research to understand their affect on the consumer socialisation.

Furthermore, through RTV involvement, consumption related cognition, attitude and

values were considered as outcome variables of the consumer socialisation from extant

literature. After conducting the FGIs, the measurement domains and variables of these

constructs were refined, adopted and rejected for this study, as these were developed

from the perspective of developed countries or in a western cultural setting, and

refinement was needed so that they are relevant for a developing country. Furthermore,

6 new scale items were developed on the basis of the FGI findings in this research for

further analysis. Table 5.1 shows the refined constructs, domain and measurement

variables. Furthermore, Table 5.2 shows different RTV programs that are watched by

teenaged boys and girls, as well as the frequencies of watching those RTV shows from

the two social classes.

Table 5.1 List of Constructs, Domains and Scale Items

Constructs Domains Scale Items from

Existing Literature

Adapted Scale Items

from Literature and

New Scale Items

from FGI

a) Time spent (Yoon & Garma 2006)

1) I watch RTV.

1) I watch RTV. RTV

Involvement b) Likings (Dolliver 2006; Pecheux &

Derbaix 2002; Yoon & Garma

2006)

1) I would attend a public

appearance if my favorite

RTV character from the

show were there. 2) I watch other programs

that include contestants

from the RTV show. 3) Reality TV shows are

just like real life. 4) I read about RTV shows

1) I would attend a public

appearance if my favorite

RTV character from the

show were there. 2) I watch other programs

that include contestants from

the RTV show. 3) RTV is just like real life. 4) I read about RTV in news

papers.

158

in newspapers. c) Attention (Perse 1998; Yoon & Garma 2006)

1) I imagine how I would

act in a similar situation

shown on RTV shows. 2) I pay close attention

when I change TV

channels.

1) I imagine how I would act

in a similar situation shown

on RTV. 2) I pay close attention when

I watch RTV.

d) Desire towards RTV (Dolliver 2006)

1) I wish I could be on

reality TV

shows.

1) I wish I could be on RTV

shows. 2) I wish to watch RTV on

the internet (New scale from

FGI). a) Product Knowledge (Bartlett, Griffiths & Badian 2008; Zuckerman, Ziegler & Stevenson 1978;

Ward & Wackman 1971)

1) TV advertisements

(TVC) provide me with a

lot of information. 2) TVC makes me aware

about particular brands. 3) TVC helps me to recall

certain products. 4) TVC is a good a way to

learn about what

products/brands are

available in the market.

1) RTV provides me with a

lot of information. 2) RTV makes me aware

about particular brands. 3) RTV helps me to recall

certain products. 4) RTV is a good a way to

learn about what

products/brands are available

in the market. 5) Advertisements shown in

RTV give me new product

ideas (New scale from FGI) b) Consumer Role Perception (Moschis & Moore 1978)

1) I understand product

usage instructions from

TVC

1) I understand product

usage instruction from RTV. 2) RTV makes me creative

to do something new (New

scale from FGI).

Consumption

Related

Cognition

c)Purchasing intentions (Bush, Martin & Bush 2004)

1) The opinions of my

favourite athlete celebrities

influence me to say positive

things about products or

brands to other people.

1) The opinions of my

favourite RTV celebrities

influence me to say positive

things about products or

brands to other people. 2) I prefer the brands that

sponsor RTV to increase my

brand awareness (New scale

from FGI). Consumption

Related

a) Imitation (Russell, Norman & Heckler

1) I get fashion tips from

TV shows. 1) I get fashion tips from

RTV.

159

2004) 2) I like the products shown

on TV programs. 3) I often buy clothing

styles that I have seen on

TV programs. 4) My favorite TV

celebrities make me fashion

conscious.

2) I like the products shown

on RTV. 3) I often buy clothing styles

that I have seen on RTV. 4) My favorite RTV

celebrities make me fashion

conscious.

b) Inspiration from media personalities (Hirschman & Thompson 1997;

Lawrie et al. 2006)

1) I want to look like my

favourite celebrities 2) I believe my favorite TV

celebrity is a good idol 3) I would like to have

lunch with my favourite TV

celebrity

1) I want to get a good figure

like RTV celebrities 2) I see my favorite RTV

celebrity as my idol 3) I would like to have lunch

with my favourite RTV

celebrity

Attitude

c) Attitude towards product placement (Schmoll et al. 2006).

1) I feel very strongly about

banning unethical products

shown in TVC.

1) I prefer to buy

multinational brands if

they are shown by RTV

(New scale from FGI). 2) I feel very strongly about

banning unethical products

shown in RTV. a) Materialism (Chan 2003)

1) I like to own the newest

things shown on TVC. 2) I like to own the best

things shown on TVC. 3) It is better to have more

of an allowance from

parents to buy the brand

shown on TVC

1) I like to own the newest

things shown on RTV. 2) I like to own the best

things shown on RTV. 3) It is better to have more of

an allowance from parents to

buy the brand shown on

RTV. b) Desire towards products (Ghani 2005; Hastings &

Aitken 1995)

1) 1) I often request of my

parents to get the products

shown on TV. 2) I often request of my

parents to get the products

shown on advertisements in

TV.

1) I often request of my

parents to get the products

shown on RTV. 2) I often request of my

parents to get the products

shown on advertisements in

RTV shows.

Consumption Related Values

c) Prohibited product consumption

1) TVC promote awareness

about beer drinking. 1) RTV shows promote use

of prohibited items among

160

(Wyllie, Zhang & Casswell

1998)

teenagers.

a)Parental communication styles (Carlson, Lacznaik &

Walsh 2001)

1) I talk to my mom about

content of TV shows. 2) My parents and I discuss

TV show selection.

1) I talk to my parents about

RTV content. 2) My parents and I discuss

RTV show selection.

b) Co-viewing (Chan & McNeal 2003)

1) My parents and I watch

TV on weekdays. 2) My parents and I watch

TV on Saturday and

Sunday.

1) My parents and I watch

RTV on weekdays. 2) My parents and I watch

RTV on weekends.

c) Parental knowledge (Cardoza

2002; Carlson, Lacznaik &

Walsh 2001)

1) My mom knows what I

watch on TV. 1) My parents know what I

watch on RTV.

Parental

Control

d) Parental Opinion (Carlson,

Lacznaik & Walsh 2001) 1) My mom says evening

TV shows are bad for kids

1) My parents prohibit me

from watching certain RTV

shows a) Extrinsic motivation (Edward & Grantham 2009)

1) My friends and I talk

about any website that is

visited.

1) I talk to my friends about

the RTV shows (New scale

from FGI) 2) My friends and I talk

about the content of RTV

shows b) Peers Co-viewing (Nathanson 2001)

1) My friends and I often

watch TV together 1) My friends and I often

watch RTV together c) Peers Mediation (Nathanson 2001)

1) My friends think the

content of TV shows can

provide useful information. 2) My friends think the

content of TV is entertaining.

1) My friends think the

content of RTV shows can

provide useful information. 2) My friends think the

content of RTV is

entertaining.

Peer Group

Influences

d) Reinforcement (Lueg & Finney 2007)

1) My peers’ opinion on e-

shopping is important to

me.

1) My peers’ opinions on

RTV shows are important to

me.

161

Table 5.2 Teenagers’ Favourite RTV shows (Number within parentheses indicating the

number of times mentioned by respondents)

RTV Shows Middle Social Class Lower Social Class

Boys Girls Boys Girls

Quiz Shows Roddies (2) , Dus

Ka Dam (2) ,

KBC (1)

KBC (1) ,

Roddies (1)

Junior Horlicks

(1)

N/A

Musical shows Live Rock Shows

(1)

Phone & Live

musical shows (2)

Khude Gan Raj

(1)

Khude Gan Raj

(2)

Talent Hunt

Shows

Indian Idol (1) ,

Close up

Bangladesh (1)

Indian idol (1) ,

Close up

Bangladesh (1)

Close up

Bangladesh (1)

Close up

Bangladesh (1),

Dhum Ta Na (1),

Nach Balyee (1)

Informational

Shows

Nokia Music Jam

next generation

(1)

Elimination game

Shows

Roddies (1) ,

Amazing Race (1)

Roddies (1), Fan

club tournament

(1)

Super hero-

heroine (1)

Life Style Shows N/A Look @ to me

(3), Lakme

Fashion (2),

Sunsilk hair

expert (2)

N/A P & G you got the

look (1)

Celebrity Talk

Shows

Amar Ami (1),

MTV celebrity

shows (1)

Amar Ami (1),

MTV celebrity

show (3)

Big Boss (1) Amar Ami (1)

162

Table 5.3 The Compressed Assembly of Data based on FGIs

Constructs/Variables Comments made by Participants

Parental Control 1) ‘It is easier for me to ask for any product if my mother likes it from RTV’ (msc g 3). 2) ‘It is easier to influence my father to ask for any preferred product shown on RTV’ (msc g 5). 3) ‘It is really good fun to watch RTV at the weekend with my parents’ (lsc g 5). 4) ‘My father asks me which RTV shows are entertaining like Close Up Bangladesh’ (msc g 4).

5) ‘My mom gets annoyed if I watch something culturally irrelevant on RTV like Nach Balyee’ (lsc g 5).

Peer Group Influences 1) ‘I can not wait to share with my friends about TV shows like ‘Indian Idol’ ’(msc g 2). 2) ‘I trust my friends most’ (msc g 1). 3) ‘I go to my friend’s home to watch our common favourite shows like Close up Bangladesh’ (lsc b 2). 4) ‘I guide my friend, and she does me as well to follow some useful RTV shows like Lakme fashion shows’ (msc g 1). 5) ‘I do share my pocket money with my friends to buy our favourite things’ (msc g 2). 6) ‘My friend and I bought our MP3 together which we liked from RTV

shows’ (msc b 8).

RTV Involvement 1) ‘I watch and think when it is a quiz show e.g., Dus Kadam’ (msc b 2). 2) ‘I read newspapers that help me to know about RTV e.g., KBC, Roddies’ (msc g 1). 3) ‘My parents have a negative attitude towards the internet but I know that they are wrong’ (lsc g 4). 4) ‘I wish I could meet with my favourite singer at RTV shows e.g., Fan club tournament’ (msc g 3). 5) ‘I would like to meet my favourite hero at ‘Super hero and heroin’ shows’ (lsc g 4). 6) ‘We do not go with all RTV shows, we are quality viewer e.g., Amar Ami, Amazing Race’ (msc b 8, g 3). 7) ‘RTV celebrities want quick fame’. Still though, most of us follow the RTV celebrities’ (lsc b 8). 8) ‘Live shows like ‘Phone & Live concert’ give me real pleasure’ (msc g 2). 9) ‘RTV participants are real but judges are commercial’ (msc b 11).

163

Consumption Related Cognition 1) ‘RTV makes a brand famous then we know about that product’ (msc b 9).

2) ‘I got to know about ‘Airtel’ as a mobile phone, when started watching ‘Dus KaDam’ (msc b 9).

3) ‘We know about new product from RTV’ (lsc b 2, g 4).

4) ‘We got to know about ‘Meridian chips’ from musical reality show like ‘Khude Ganraj’’ (lsc b 2, g 4). 5) ‘I have learnt how to use blue tooth of my mobile phone from RTV shows like ‘Nokia next generation’’ (msc b 8). 6) ‘Fashion shows help me to understand how to use make up ‘Look @ to me’’ (lsc g 3). 7) ‘Roddies rings me the bell of Hero Honda’ (msc b 8). 8) ‘I watch celebrity talk shows like ‘MTV celebrity talk shows’ (msc g 1). 9) ‘Often I suggest my friends about buying new products which are endorsed by RTV celebrities’ (msc b 8). 10) ‘I asked my friends that Habib (local popular singer) wear shirt from Cats’ eye’ (msc b 8). 11) I got to know about Airtel because I watch Big Boss’ (lsc b 2). 12) ‘RTV celebrities help me to think differently about my fashion’ (msc g 3). 13) ‘Fashion shows like ‘Sunsilk-Hair expert’ help me to think about new hair style’ (msc g 1).

Consumption Related Attitude 1) ‘If I could afford it, I would buy some of the clothes that my favourite actress wears in shows like Indian idol’ (msc g 5). 2) ‘I like John Abraham because he is so manly’ (msc b 6). 3) ‘I just imagine how my favourite actress maintains her figure!’ (msc g 4).

4) ‘I like what my favourite celebrities are wearing at ‘MTV celebrity

shows’ (msc g 2).

5) ‘I like Katrina’s movies, because she is so attractive’ (lsc b 8).

6) ‘Jafar Iqbal is our inspiration’ (msc b 7, b 9).

7) ‘We can get inspiration from RTV shows judges e.g., Roddies’.

8) ‘Having lunch with RTV celebrities is a life time experience e.g., lunch

with MTV celebrity shows’ (lsc g 2).

164

9) ‘I do not spend for unknown brands’ (msc b 8).

10) ‘We prefer Lakme or Sunsilk than local cosmetics brand who are also

sponsoring fashion shows’ (msc g 1, g 2).

11) ‘RTV should not allow any unethical product sponsors’ (lsc b 1).

Consumption Related Values 1) ‘By the time I save money to buy a product shown in RTV, that model gets outdated’ (msc b 8). 2) ‘My mom takes my money when she knows I have good savings for buying a new brand new tops shown in ‘Look @ to me’’(msc g 3). 3) ‘I do not mind to cry to pursue my parents to get my favourite product shown in RTV’ (msc g 2). 4) ‘We often play mental games with our parents’ (msc b 10, b 11). 5) ‘Every one likes good products’ (lsc b 2). 6) ‘People will praise me if I get new products shown on RTV’ (lsc g 3). 7) I really like the outfit shown at ‘P & G you got the look’ (lsc g 3). 8) ‘We should not forget our culture’ (lsc g 5).

165

Chapter 6 Assessment and Refinement of Measurement

Scales by Quantitative Measures

6.1 Introduction In this research both qualitative and quantitative methods have been used. Qualitative

research has been conducted to help develop the constructs and refining their multi-item

measurement items. Quantitative methods were applied to test the constructs and their

relationships. In the previous chapter results of the qualitative research have been

reported. The results of the qualitative research have been subjected to further

quantitative testing for model development of consumer socialisation by RTV. In this

chapter the results of the quantitative research will be reported. Before reporting

detailed quantitative analyses the sample characteristics are presented.

6.2 Sample Characteristics

In the quantitative phase of the research, a total of 400 teenagers representing both

genders participated in this survey. After careful editing data related to 368 of these

participants were found valid. Editing consisted of screening questionnaires to identify

illegible, incomplete, inconsistent, or ambiguous responses. After screening, 32

questionnaires were found either unanswered or some of the questions answered more

than once. These 32 incomplete and inconsistent questionnaires were dropped from

further data analysis. Rest of the 368 questionnaires were considered valid and used for

subsequent analysis.

In this research, gender differences of teenagers have been considered as one of the

social structural variables that influence their consumer socialisation. So, during sample

selection, proportional representation of both genders was ensured. Table 6.1 shows the

distribution of respondents by gender differences.

166

Table 6.1 Distribution of Respondents by Gender Differences Gender Frequency Percent (%)

Male 181 49.2

Female 187 50.8

Total 368 100.0

Age group wise there were higher representation (75.3%) of late teenagers (16-19 age

group) then young teenagers (13-15 age groups) who had 24.7% representation. Table

6.2 shows the distribution of respondents by age groups. On the other Table 6.3 shows

the distribution of respondents by academic class (Equivalent to academic year in

Australia). Teenaged respondents were selected from class 5 to class 12 for survey in

this research.

Table 6.2 Distribution of Respondents by Age Groups Age groups Frequency Percent (%)

13-15 91 24.7

16-19 277 75.3

Total 368 100.0

Table 6.3 Distribution of Respondents by Academic Class

Class Frequency Percent (%)

5 1 .3

6 1 .3

7 43 11.7

8 46 12.5

9 118 32.1

10 114 31.0

12 45 12.2

Total 368 100.0

167

In this research, social class difference of teenagers has also been considered as one of

the antecedent variables that influence their consumer socialisation. Like gender

distribution, proportional representation of teenagers from the two social classes

considered for this study, e.g., lower and middle was ensured. As reported earlier a total

of 368 valid responses were received in this survey from initial 400 respondents. 185

(50.3%) of these respondents were from lower social class and 183 (49.7) from middle

social class school. Table 6.4 shows the distribution of respondents from the two social

classes.

Table 6.4 Distribution of Respondents List by Social Classes Sample List Number of Respondents Percent (%)

Lower social class 185 50.3

Middle social class 183 49.7

Total 368 100.0

It is already mentioned that teenagers’ social class differences has been measured by

considering their parental education, income, profession and suburbs they live in.

Tables 6.5 to 6.7 below show the distribution of respondents’ parental education,

occupation and income levels by using descriptive statistics and cross tabulation (Chi2).

Teenagers’ parental education was considered as one of the variables to measure

teenagers’ social class. Results, as shown in Table 6.5, there is significance difference

between the education levels of parents of the two social classes. In the lower social

class, teenagers’ fathers’ education mainly fall in between no formal education to

bellow graduation (95%), whereas, in middle social class 75% of the fathers’ are

graduates or above. Similarly, in the lower social class, teenagers’ mothers’ education

mainly fall in between no formal education to bellow graduation (93%), whereas, in

middle social class 59% of the mothers’ are graduates or above.

168

Table 6.5 A Cross-Tabulation of Teenagers’ Parents Social Class and Their Education

Father Mother

Level

of education

Lower Social Class Middle Social Class Lower Social Class Middle Social Class

Actual

Count 08 00 07 01

Expected

Count 04 04 04 04

No Formal

Education

% 04 00 04 01

Actual

Count 18 03 29 05

Expected

Count 11 10 17 17

Bellow SSC

% 10 02 16 03

Actual

Count 94 12 106 24

Expected

Count 53 53 66 64

SSC

% 51 07 57 13

Actual

Count 55 29 30 46

Expected

Count 42 42 38 38

HSC

% 30 16 16 25

Actual

Count 10 139 13 107

Expected

Count 75 74 60 60

Graduate &

Above

% 05 75 07 59

Chi-Square 201.876 150.160

df 4 4

Sig .000 .000

169

Teenagers’ parental occupation was considered as one of the variables to measure

teenagers’ social class. Results, as shown in Table 6.6, there is significance difference

between parental occupations in the two social classes. In the lower social class, 53%

teenagers’ fathers’ occupation are mainly observed by blue collar jobs and 47% for

white collar as well as others. Whereas, in the middle social class teenagers’ fathers are

mainly involved with white collar jobs (94%). Similarly, in the lower social class,

teenagers’ mothers’ are mainly involved with blue collar jobs and house-hold duties

(91%). Whereas, in middle social class 69% are involved with house-hold duties and

rest (31%), are involved with white collars as well as other jobs.

Table 6.6 A Cross-Tabulation of Teenagers’ Parents Social Class and Their Occupation Father Mother Occupation

Lower Social Class Middle Social Class Lower Social Class Middle Social Class

Actual

Count 04 01 144 127

Expected

Count 03 02 159 147

House Duties

% 02 00 85 69

Actual

Count 92 11 07 00

Expected

Count 52 51 13 07

Blue Collar

Jobs

% 51 06 07 00

Actual

Count 40 118 21 38

Expected

Count 79 79 06 22

White Collar

Jobs

% 22 65 04 22

Actual

Count 49 53 13 18

Expected

Count 51 51 07 12

Others

% 27 29 04 09

Chi-square 104.154 44.684

df 3 3

sig .000 .000

170

Teenagers’ parental income was also considered as one of the variables to measure

teenagers’ social class. Results, as shown in Table 6.7, there is significance difference

between parental incomes in the two social classes. Teenagers’ parents in the lower

social class mostly earn less than 70 thousand taka in a year (65%), rest of them 16 %

and 19% earn more than 70 and 90 thousand in a year respectively. Whereas, in the

middle social class (53%) teenagers’ parental income fall above 90 thousand. Further,

21% are observed under 50 thousand in a year and rest of the portion is followed by

other income levels.

Table 6.7 A Cross-Tabulation of Teenagers’ Parents Social Class and Their Annual

Income

Yearly Income

Level (Taka)

Lower Social Class Middle Social Class

Actual

Count 83 39

Expected

Count 61.3 60.7

30-50 Thousand

% 45 21

Actual

Count 37 21

Expected

Count 29.2 28.8

51-70 Thousand

% 20 13

Actual

Count 30 24

Expected

Count 27.1 26.9

71-90

Thousand

% 16 13

Actual

Count 35 99

Expected

Count 67.4 66.6

90 Thousand +

% 19 53

Chi-square 51.507

df 3

sig .000

171

Dhaka is one of the fastest growing metropolitan cities in Asia. Urbanisation has been

observed largely in skewed manner. It is very difficult to segregate any particular area

as either lower or middle social class area. The slums areas and expensive suburbs co-

exist in Dhaka city. However, there are no official sampling frames in Dhaka.

Therefore, a city map has been considered to find out lower and middle social class

suburbs, where two schools represent these teenaged respondents from lower and

middle social class.

In the above mentioned tables descriptive statistics followed by cross tabulation shown

clearly that there are significance differences between two social classes on the basis of

teenagers’ parental education, income and occupation.

6.3 Assessment of Measurement Scales

As has been discussed in section 4.2 both exploratory and confirmatory factor analysis

techniques have been used in this research. Exploratory factor analysis has been used to

test dimensionality of data with the aim to produce a set of scale items that reflect a

single underlying domain/ factor or construct. Further confirmatory factor analysis

using SEM program AMOS has been used to gain a more precise estimation of

reliability, and formally test the uni-dimensionality of the scales. SEM has also been

used to test a proposed model that has been developed on the basis of literature review

(Figure 3.2) and qualitative research findings (Table 5.1).

172

6.4 Results of Exploratory Factor Analysis

Factor analysis was applied using principal component analysis extraction method. The

most common rule is to base the numbers of factors on the number of eigenvalues

greater than 1 (Zikmand & Babin 2007). However, when a large number of variables

are being factor analysed, many unimportant factors will be associated with eigenvalues

as large as 1.0, making it especially crucial to consider other indicators of the "correct"

solution. Determining the correct number of factors is a matter of balancing

comprehensiveness against parsimony. Further, variables to be included in the factor

analysis should be specified based on past research, theory and judgments of the

researcher (Malhotra 2003). In practice, one is usually happy with a factor solution that

explains 50-75% of the variance in the original variables (Diekhoff 1992). In this case,

the cut off point of eigenvalues of 1.0 gave factor solution that explained 56% of the

variance overall for all the 45 indicators. Another consideration in determining the cut

off point was interpretability of factors. Factors are interpreted by examining their

correlations, called loading. Garver and Mentzer (1999) recommended that variables

with poor factor loadings (less than .50) should be dropped from further analyses. In

addition, Hair et al. (2006) suggested that smaller number of variables (less than 4) that

are unable to make a factor should be dropped from further analysis as well. This

interpretation is often facilitated by factor rotation, the second stage of factor analysis,

in which original factors are redefined. In this research varimax rotation method was

used. Factor loadings of all the underlying constructs were inspected after applying

varimax rotation method.

A total of 45 variables earlier identified through literature review and refined through

qualitative research making those adoptable to a developing country (Table 5.1) were

submitted for the exploratory factor analysis (EFA). Table 6.8 shows the factors

extracted with the variables that explain each of the factors, at this stage. From the

table, number of variables that had poor factor loading (less than .50) and factors with

smaller number of variables (less than 4) were dropped from further analysis. This

included variables cog 6 (opinion of RTV celebrities influences), cog 8 (prefer RTV

173

sponsoring brands), att 6 (RTV celebrities are idol), att 7 (teenagers like to have lunch

with celebrity), pc 6 (parents prohibit RTV shows). Moreover, as suggested by Garver

and Mentzer (1999) items with low factor loading (< 0. 50) were eliminated because

they do not converge properly with the latent constructs they were designed to measure.

Further, four variables, val 4 (teenagers request parents for RTV products), val 5

(teenagers request parents for products shown in RTV), val 6 (RTV promote prohibited

items) and att 8 (banning unethical products shown in RTV) were dropped at this stage

as they shown in column 4 and 5 (factor) with two measurement variables each and are

not able to represent independent factor (Hair et al. 2006). Table 6.9 shows the retained

7 factors and there 36 measurement variables for the further analysis. To test reliability

of each set of factors co-efficient Alpha was used. Each retained constructs/factors

achieved more than Alpha .70 as recommended by Hair et al. (2006).

Table 6.8 Factors, Variables and Factor Loading

Variables Factor 1 Factor 2 Factor 3 Factor 4 Factor 5

Product instruction from RTV (cog 3) .816

RTV helps recalling (cog 4) .773

RTV makes aware of brands (cog 2) .715

RTV provides product information (Cog 1) .658

RTV helps market learning (Cog 5) .645

RTV ads provide product ideas (Cog 7) .608

Opinion of RTV celebrities influence (Cog 6) .398

RTV celebrities make fashion conscious

(Att 4)

.737

Fashion tips from RTV (Att 1) .685

Buy cloths shown in RTV (Att 3) .654

RTV develops teenagers’ creativity (Cog 9) .613

Teenagers like celebrities’ good figure (Att 5) .573

Teenagers like product shown in RTV (Att 2) .562

Prefer RTV sponsoring brands (Cog 8) .429

RTV Celebrities are idol (Att 6) .411

MNC brand in RTV (Att 9) .672

Teenagers like newest thing in RTV (Val 1) .664

Teenagers like best thing in RTV (Val 2) .604

Teenagers request for allowance (Val 3) .564

174

Teenagers request parents for RTV products

(Val 4)

.854

Teenagers request parents for products shown

in RTV ads (Val5)

.807

Teenagers like to have lunch with celebrity

(Att 7)

.402

RTV promote prohibited items (Val 6) .785

Banning unethical products shown in RTV

(Att 8)

.749

Variables Factor 6 Factor 7 Factor 8 Factor 9

Pay attention to RTV (Inv 2) .752

RTV from news papers (Inv 3) .735

Watch RTV (Inv 1) .726

Watch other programs relate to RTV celebrities (Inv 6)

.582

RTV on web (Inv 9) .766

RTV like real life (Inv 8) .686

Wish to attend RTV (Inv 7) .670

RTV with public appearances (Inv 4) .531

Imagine about RTV shows (Inv 5) .513

Teenagers and parents jointly select RTV

show (pc 2)

.845

Co-viewing at weekends (pc 4)

.800

Co-viewing at weekdays (pc 3)

.743

Teenagers talk to parents about RTV content

(pc 1)

.728

Parents know about RTV (pc 5)

.545

Parents prohibit RTV shows (pc 6)

.254

Teenagers and friend talk about RTV content

(pi 2)

.699

Friends think RTV provide useful

information (pi 4)

.689

Friends think RTV entertaining (pi5)

.686

Talk to friends about RTV shows (pi 1)

.667

Friends watch RTV together (pi 3)

.604

Peers opinion important ( pi 6)

.598

175

Next, EFA with, eigenvalue � 1 and varimax rotation was applied separately to each of

the 6 constructs under investigation as suggested by Conway and Huffcutt (2003). The

main purpose of this step is to see whether the measurement items of each constructs

under investigation is uni-dimensional or multi-dimensional. A total of 7 valid factors

representing the 6 constructs with their measurement items were identified at this stage.

Among these, factor 1 clearly indicates six measurement items for consumption related

cognition each with factor loading of more than .50. Further, co-efficient alpha for this

factor is acceptable .81. Factor 2 indicates five of the measurement items of

consumption related attitude variables with co-efficient alpha .81. However, one of the

measurement items of consumption related cognition (cog 9) has shown its loading with

this factor. This is not surprising as theoretically, attitude is an enduring combination of

motivational, emotional, perceptual and cognitive process (Neal, Quester and Hawkins

2004). Accordingly, cog 9 was retained as part of attitude for subsequent analysis.

Similarly, for factor 3, one of the attitude variables ‘att 9’ have loading with rest of the

three measurement items of consumption related values with the co-efficient alpha .75.

This is also theoretically supported as consumption related values as an out-come of

consumer socialisation influence consumers’ attitudes (McCarty & Shrum 1993). This

item was retained as part of values for subsequent analysis.

However, measurement items of RTV involvement created two factors with acceptable

co-efficient alpha value. Factor 4 has four items of involvement each with factor

loading more than .50 that strongly relate to the domain of attention towards RTV. Co-

efficient alpha of this factor was .71. Whereas, factor 5 relating to the domain of desire

to getting involved with RTV has 5 variables with factor loading of more than .50. This

factor has the co-efficient alpha of .70.

Factors 6 and 7 were extracted as expected on the basis of previous literature and

qualitative studies. Factor 6 confirmed all the five measurement items of parental

control each with factor loading of more than .50 and factor co-efficient alpha .81.

Factor 7, has 6 items of peer group influences each with factor loading of more than .50

and factor co-efficient alpha .76.

176

Using the eigenvalue grater than 2 the variance explained by the extracted factors

ranges from 45.43% to 58.12%, while factor loadings are all above of the threshold of

.50, except 2 factors. If eigenvalue is taken as 1 then variances of all factors fall above

the threshold level. Further, uni-dimensionality of six factors is measured in this stage.

Uni-dimensionality is defined as the existence of one construct underlying a set of items

(Garver & Mentzer 1999). It is the degree to which a set of items represent one and only

one underlying construct. In this study only RTV involvement is multi-dimensional at

EFA stage as most of its measurement items strongly related to two measurement

domains such as attention towards RTV and desire towards RTV. However, the current

RTV involvement literature shows these two domains are part of the single construct

(Perse 1998; Pecheux & Derbiax 2002; Yoon & Garma 2006). Further, as suggested by

Tabachnick and Fidell (2001) that the SEM model has to be driven by theory rather than

survey data. If substantive theory suggests any variable or a relationship should be

included in a model, as well as it should be left there even if it does not contribute to the

over all fit of the model (Joreskog & Sorbom 1993). Also a test was run combining

these 2 factors (4, 5) as measurement items of RTV involvement shows .77 co-efficient

alpha. Therefore, in the subsequent analysis stage these measurement items are

considered as common items for the underlying construct, RTV involvement. After

establishing that these scales are uni-dimensional, the reliability of the composite score

was assessed. As shown in Table 6.9, Co-efficient alpha of all the 7 retained factors are

.70 or above. Further, the item-total co-relation values range from .513 to .841 and are

all above the required threshold of .50. All items comprising these 6 factors were

therefore retained for subsequent analysis. Table 6.9 shows the retained factors and their

measurement variables and factor loadings.

177

Table 6.9 Retained Factors, Variables and Factor Loading

Measurement

Items

Factor

Loading

% Variance

Extracted

Eigenvalue Co-efficient

Alpha

Inter-item

Correlation

Cognition Cog 3

Cog 4

Cog 2

Cog 1

Cog 5

Cog 7

.816

.773

.715

.658

.645

.608

56.18

3. 37

. 81

.719

.747

.819

.787

.706

.711 Attitude Att 4

Att 1

Att 3

Cog 9

Att 5

Att 2

.737

.685

.654

.613

.573

. 562

52.03 3. 12

. 81 .779

.725

.732

.774

.647

.688 Values Att 9

Val 1

Val 2

Val 3

.672

.664

.604

.564

58.12 2. 32 . 75

.823

.774

.715

.727

Involvement (Attention)

Inv 2

Inv 3

Inv 1

Inv 6

.752 .735

.726

.582

53. 72

2. 14

.71

.579

.622

.595

.653

Involvement (Desire)

Inv 9

Inv 8

Inv 7

Inv 4

Inv 5

.766

.686

.670

.531

.513

45. 43

2. 27

.70

.598

.629

.685

.513

.561

Parental Control Pc 2

Pc 4

Pc 3

Pc 1

Pc 5

.845

.800

.743

.728

.545

57. 96

2. 89

.81

.748

.841

.771

.816

.603

Peers’ Influences Pi 2

Pi 4

Pi 5

Pi 1

Pi 3

Pi 6

.699

.689

.686

.667

.604

.598

46. 62

2. 80

.76

.721

.649

.643

.712

.677

.680

178

To assess and refine the measurement scales in terms of reliability, uni-dimensionality

and validity, there are two main approaches like; exploratory factor analysis (EFA) and

confirmatory factor analysis (CFA) are commonly used in the literature (Hurley et al.

1997). The issue on which type of factor analysis (e.g., EFA or CFA) to use in a

particular situation is the subject of a debate among organizational researchers (Hurley

et al. 1997). EFA is often considered to be more appropriate than CFA in the early

stages of scale development because CFA does not indicates how well the item load on

the non-hypothesised factors (Kelloway 1995). However, the majority of researchers

agree that EFA is more appropriate for scale development/refinement and CFA is for

scale validation. Hurley et al. (1997) mentioned that “EFA may be appropriate for scale

development while CFA would be preferred where measurement models have a well-

developed underlying theory for hypothesized patterns of loadings. A line of research

would start out with studies utilising EFA, while later work would show what can be

confirmed” (p. 668).

The current research employed a combination of both EFA and CFA to form two-phase

approaches. The first phase involved employing EFA for scale assessment and

refinement and the second phase involves employing CFA for scale validation (Fabrigar

et al. 1999)

6.5 Measurement Model Evaluation and Specification

In this section different measures for SEM model evaluation have been discussed. There

are two stages in this step: a) deciding on the goodness-of-fit criteria and b) testing the

measurement model fit.

179

6.5.1 Deciding Goodness of Fit Criteria

Two details were considered in evaluating both measurement models and structural

models: their uni-dimensionality and their fit. In this research, both coefficient alpha

and standardized regression weights were used to measure uni-dimensionality of

measurement models. Further, a variety of goodness-of-fit indices was used in this

research because no single or particular statistical test of significance identifies a correct

model from the sample data (Byrne 2001; Maruyama 1998). Therefore, the assessment

of model fit for this research was based on multiple criteria; they are summarised in

Table 6.10.

Table 6.10 Summary of Reliability, Weights, and Fit Indices Used in this Research

Name Abbreviation Type Accepted Level

in this

Research

Goodness-of-fit model of this study

Coefficient Alpha

Standard

regression weight

�

Beta

Uni-

dimensionality

� > 0.60 adequate

� >0.40

� 0.75 – 0.81 � 0.392 - .737

Chi-square (with

associated degrees

of freedom and

probability of

significantly

different)

X2 (df,,p)

Model fit P>0.05 (at �

equals to 0.05

level)

0.000

Normed chi-

square

X2�df Absolute fit and

model persimony

1.0< X2/df/ <5.0 2.010

Root mean square

error of

approximation

RMSEA

Absolute fit

<0.08

.052

Goodness-of-fit

index

Adjusted

Goodness-of-fit

GFI

AGFI

Absolute Values close to

0.9and above

indicate

satisfactory fit

.845 .824

180

index

Normed fit index

Tucker Lewis fit

index

Comparative fit

index

NFI

TLI

CFI

Incremental fit Values above 0.8

and close 0.9

indicate

acceptable fit

.773 .861 .870

HOELTER >200 critical value 201 Source: Developed from Baumgartner and Homburg (1996), Byrne (2001), Carmines and McIver (1981),

Hair et al. (1998), Hulland, Chow and lam (1996), Kline (1998), Holmes-Smith (2002).

Co-efficient alpha measures the internal consistency reliability, which is the average of

all possible split-half coefficients resulting from different splitting of the scale items

(Malhotra 2003). Chi-square is the only measure that has a direct statistical test as to its

significance and it forms the basis for many other goodness-of-fit measures (Hair et al.

2006). Normed Chi-square refers to Chi-square divided by degrees of freedom that

indicates an acceptable fit between the hypothetical model and the sample data

(Carmines & McIver 1981). The Chi-square value should not be significant if there is a

good model fit (Garson 2009). However, Chi-square P is considered as conservative

measure of goodness of fit (prone to type two error), that is why many researchers who

use SEM with reasonable sample size (e.g., > 200) determine acceptability of fit using

goodness-of-fit measures like NFI, CFI and RMSEA (Garson 2009). As an absolute fit

indicator that attempts to correct the tendency of the goodness fit test statistics to reject

models with a large samples or a large number of observed variables is the root mean

square error of approximation (RMSEA) (Hair et al. 2006). Goodness-of-fit index (GFI)

was also an early attempt to produce a fit statistics is still indirectly sensitive to sample

size due to the effect of N on sampling distribution. Whereas, adjusted good-ness-of-fit

index tries to take in to account differing degrees of model flexibility. It does by

adjusting GFI by a ratio of the degrees of freedom used in a model to the total degrees

of freedom available (Hair et al. 2006). The normed fit index (NFI) is one of the

original incremental fit indices. It is a ratio of the differences in the X2 value for the

fitted model and a null model divided by the X2 value for the null model. Comparative

fit index (CFI) is an incremental fit index that is an improved version of normal fit

index. Tucker Lewis Index (TLI) predates the CFI and is conceptually similar in that it

181

also involves a mathematical comparison of a specified theoretical measurement model

and a baseline null model (Hair et al. 2006). At last HOELTER’s critical N for

significance level of .05 is 200 and more. The Table 6.10 also indicates the cut-off

points that are followed in this study.

6.5.2 Testing the Measurement Model Fit

Not only the above mentioned criteria but also other standard estimates associated with

the evaluation of CFA models were used in this study. The following Table 6.11

provides a summary of these criteria used in this analysis. Each of these model

evaluation criteria will be discussed next.

Table 6.11 Summary of Standardised Estimates used for CFA Models (Holmes-Smith

2002)

Standardised Estimates Value Criteria

Factor loading >0.7 good, >0.5 acceptable

Covariance (cr) > 1.96

Variable reliability good >0.3

Firstly, standardised factor loading (or standard regression weight in SEM output)

should have value greater than 0.5 to show a good association between the factor and

variable (Churchill jr. 1979; Holmes-Smith 2002). Secondly, covariance of two

variables is the estimate of amount of correlation between two variables because some

unmeasured latent variable may influence both variables (Schumacker & Lomax 1996).

Critical ratio (cr) of the estimate should be greater than 1.96 for the factor loading or

variance that it is estimating. In this research, the covariance was considered significant

if the critical ratio (cr) of the measure is greater than 1.96 (Byrne 2001; Holmes-Smith

2002). Finally, variable reliability is the squared multiple correlation for the observed

variables and reflected how well the variable measures the underlying latent trait. For

this research variable reliability was considered acceptable if above 0.3 (Holmes-Smith

2002).

182

6.6 Assessment of Measurement Scales Using CFA

Traditional approaches such as Co-efficient Alpha and EFA are useful for the

assessment and refinement of measurement scale items. However, they just serve as

first round tools (Hair et al. 2006). In the confirmatory approach, it is necessary to

employ confirmatory factor analysis (CFA) (Garver & Mentzer 1999; Hurley et al.

1997). Hurley et al. (1997, p. 674) also mentioned that “EFA provides important

diagnostics which should be considered along with the results of CFA in judging scale

and its items”. This is because CFA provides a test of hypotheses about population

factor structures based on sample data e.g., relationship of the construct with its

measures; while EFA attempts to describe, reduce data to make them easily understood

(Hurley et al. 1997).

To be more specific, for this research fitting CFA models to large data sets served few

functions. First, and foremost, a fitted CFA model allowed large number of observed

variables to be reduced to single composite scale, thus reducing the number of variables

in subsequent structural equation model. In turn, fitting a CFA model allowed for

differences in the degree to which each individual measure contributes to the over all

composite scale. That is CFA models provided a more realistic representation of the

data (Fleishman & Benson 1987). Secondly, CFA models are also considered to be

sufficiently constrained to represent good measurement properties. Finally, the fit

statistic for the CFA model was a quasi test of validity.

However, the method of using a series of one-factor CFA models to reduce large data

set down to a manageable level has a disadvantage. The procedure can lead to potential

loss of information in the measurement part of the model when only the composite

variables are used in structural equation models (Holmes-Smith & Rowe 1994). To

keep balance, the advantages of CFA modelling outweigh its disadvantages.

183

In short, this research used confirmatory factor analysis (CFA) to purify the initial

measures and test for the internal consistency of the scale items. In this research EFA

were used to reduce 45 observed variables (Table 6.8) down to smaller number 36

variables (Table 6.9) which were then used for CFA. Table 6.12 shows the list of 36

variables that were tested using CFA. CFA is similar to EFA in some respects, but

philosophically it is quite different. Instead of allowing the statistical method to

determine the number of factors and loading as in EFA, CFA statistics tell us how well

our specification of the factor matches reality (the actual data). In a sense, CFA is a tool

that enables us to either confirm or reject preconceived theory (Hair et al. 2006). The

highest number of composite variables in any one factor CFA was nine and minimum

was four. Six composite variables were used in the structural models that investigated

the interrelationships amongst these composites. The sample size of 368 was sufficient

for parameter estimation and model fitting.

Table 6.12 List of Variables

Variables I watch RTV I pay close attention when I watch RTV I read about RTV in news papers I would attend a public appearance if my favorite RTV character from the show were there I imagine how I would act in a similar situation shown on RTV I watch other programs that include contestants from the RTV show I wish I could be on RTV shows RTV is just like real life I wish I could watch RTV on web RTV provides me a lot of information RTV make me aware about particular brands I understand product usage instruction from RTV RTV helps me to recall about certain products RTV is a good way to learn about what products/bands are available in market Advertisements shown in RTV gives me new product ideas RTV makes me creative to do some thing new (e.g., makeup, hair style) I get fashion tips from RTV I like the product shown on RTV I often buy clothing styles that I have seen in RTV My favorite RTV celebrities make me fashion conscious I want to get good figure like RTV celebrities (e.g. slimmer, muscular) I prefer to buy multinational brands if is shown by RTV I like to own the newest things shown on RTV I like to own the best things shown on RTV It is better to have more allowance from parents to buy the brand shown on RTV I talk to my parents about RTV content My parents and I discuss about RTV show selection My parents and I watch RTV on weekdays

184

My parents and I watch RTV during weekends My parents know what I watch on RTV I talk to my friends about the RTV shows My friends and I talk about content of RTV shows (e.g., entertainment, sex, information) My friends and I watch often RTV together My friends think the content of RTV shows can provide useful information My friends think content of RTV are entertaining My peers’ opinions on RTV shows are important to me

6.6.1 CFA results of Parental Control

The first latent variable, parental control was measured by five observed variables. The

structure of this measurement model is presented in Figures 6.1 scale items that

measured the parental control are adopted from current literature and refined through

qualitative research. Further, the following scales items are tested through EFA relating

to different domains of scale items have distinct meaning for measuring ‘parental

control’. The results of CFA of the measurement items of the ‘parental control’ are

summarised in Table 6.13.

Figure 6.1 CFA Model of Parental Control

Parental

control

pc 5e 5

1

1

pc 4e 4

1

pc 3e 3

1

pc 2e 2

1

pc 1e 1

1

Table 6.13 Regression Weights of Parental Control

Regression Weights Estimate S.E. C.R. P

value

Parents know about RTV <--- 1.000

Teenagers talk to parents about RTV content <--- 1.608 .212 7.571 0.000

Co-viewing at weekdays <--- 1.690 .218 7.767 0.000

Co-viewing at weekends <--- 1.750 .217 8.077 0.000

Teenagers and parents jointly select RTV show <--- 1.968 .242 8.117 0.000

185

The Table 6.13 of regression weight refers to the un-standardised parameter estimates

for the factor loadings. No critical ratios (t-values) are stated for the respective first

factor loadings of parental control and Parents know about RTV were fixed to unity to

scale the latent variables. All remaining factor loading were significant.

Table 6.14 Standard Regression Weights, Goodness-of-Fits Estimates and Composite

Reliability of the Parental Control

Standardised Regression Weights

Parental Control

Estimate Composite Reliability

Parents know about RTV <--- .447

Teenagers talk to parents about RTV content

<--- .673

Co-viewing at weekdays <--- .718

Co-viewing at weekends <--- .766 Teenagers and parents jointly select RTV show

<---

.827

.88

Reliability-Co-efficient alpha � 0.81

Chi-square 16.543

Degree of freedom 5

P .005

Normed Chi-square (CMIN/df) 3.309

Root mean square of error of estimation (RMSEA) .079

Goodness of fit index (GFI) .983

Adjusted of goodness-of-fit index (AGFI) .948

Normed fit index (NFI) .974

Tucker-Lewis Index (TLI) .963

Comparative fir index (CFI) .981

In the Table 6.14 shows that the coefficient alpha for the ‘parental control’ of CFA

model is 0.81 indicating that the variables are reasonable measures of ‘parental control’.

Only one variable ‘pc 6’ was dropped at this stage due to poor factor loading (Garver &

Mentzer 1999). Results suggest that three of the variables have a standardized

regression weight of more than 0.7 and others are more than 0.5 indicating a good fit for

those items. Goodness-of-fit indices also indicate that the model fitted data well, with

the GFI, AGFI, NFI, TLI, RMSEA and CFI all within acceptable level of criteria.

186

Composite reliability .88 also indicates the reliability the underlying variables of

‘parental control’.

However, Chi-square P value .005 does not satisfy the threshold requirement of greater

than .05. However, as mentioned earlier chi-square P is considered as conservative

measure of goodness-of-fit (prone to type two error), that is why many researchers who

use SEM with reasonable sample size (e.g., > 200) determine acceptability of fit using

goodness-of-fit measures like NFI, CFI and RMSEA (Garson 2009), which will be

followed to report rest of the CFA results in this research.

6.6.2 CFA results of Peer Group Influences

Peer group influences were measured by six observed variables. The structure of this

measurement model is presented in Figures 6.2 scale items that measured the ‘peer

group influences’ are adopted from current literature and refined through qualitative

research. Further, the following scale items are tested through EFA relating to different

domains of scale items have distinct meaning for measuring ‘parental control’. The

results of CFA of the measurement items of the ‘peer group influences’ are summarised

in Table 6.15.

Figure 6.2 CFA Model of Peer Group Influences

Peers'

Influences

pi 6e 6

1

1pi 5e 5

1pi 4e 4

1pi 3e 3

1pi 2e 2

1pi 1e 1

1

187

Table 6.15 Regression Weights of Peer Group Influences

Regression Weights Estimate

S.E. C.R. P value

Peers opinion important <--- 1.000

Friends watch RTV together <--- .926 .122 7.575 0.000

Talk to friends about RTV shows <--- 1.013 .114 8.861 0.000

Friends think RTV entertaining <--- .949 .109 8.694 0.000

Friends think RTV provide useful information <--- 1.059 .113 9.359 0.000

Teenagers and friend talk about RTV content <--- .953 .128 7.456 0.000

The Table 6.15 of regression weight refers to the unstandardised parameter estimates

for the factor loadings. No critical ratios (t-values) are stated for the respective first

factor loadings of peer group influences and peers opinions important were fixed to

unity to scale the latent variables. All remaining factor loading were significant.

Table 6.16 Standard Regression Weights, Goodness-of-Fits Estimates and Composite

Reliability of the Peer Group Influences

Reliability-Co-efficient alpha � 0.76 Chi-Square 38.764 Degree of freedom 9 P .000 Normed Chi-Square (CMIN/DF) 4.307 Root mean square of error of estimation (RMSEA) .095

Goodness of fit index (GFI) .965

Adjusted of goodness-of-fit index (AGFI) .919

Normed fit index (NFI) .921

Tucker-Lewis Index (TLI) .896

Comparative fir index (CFI) .938

Standardised Regression Weights

Peer group Influences Estimate

Composite Reliability

Peers opinion important <--- .599 Friends watch RTV together

<--- .500

Talk to friends about RTV shows

<--- .634

Friends think RTV entertaining

<--- .648

Friends think RTV provide useful information

<--- .697

Teenagers and friend talk about RTV content

<---

.510

.92

188

In the Table 6.16 The coefficient Alpha for the ‘peer group influences’ of CFA model is

0.76 indicating that the variables are reasonable measure of ‘peer group influences’.

None of the variables’ standard regression weights were less that 0.5. Further, goodness

–of-fit indices also indicates that the model fitted data well, with GFI, AGFI, NFI, TLI,

RMSEA and CFI all within acceptable level of criteria. Composite reliability .92 also

indicates the reliability the underlying variables of ‘peer group influences’.

6.6.3 CFA results of RTV Involvement

RTV involvement was measured by nine observed variables. The structure of this

measurement model is presented in Figures 6.3. Measurement items that measured the

‘RTV Involvement’ are adopted from current literature and refined through qualitative

research. Further, the items were tested using EFA. However, at the EFA stage ‘RTV

Involvement’ reflected in to two different factors relating to the domain of ‘attention’

and ‘desire’ towards RTV but at the CFA stage these two factors have been considered

as a common factor of ‘RTV Involvement’ construct as they are strongly supported by

the existing literature as one construct. The results of CFA of the measurement items of

the ‘RTV Involvement’ are summarised in Table 6.17.

Figure 6.3 CFA Model of RTV Involvement

RTV

Involvement

inv 1 e1

1

1

inv 2 e2

1

inv 3 e3

1

inv 4 e4

1

inv 5 e5

1

inv 6 e6

1

inv 7 e7

1

inv 8 e8

1

inv 9 e9

1

189

Table 6.17 Regression Weights of RTV Involvement

Regression Weights Estimate S.E. C.R. P value

Pay attention to RTV 1.000

RTV from news papers <--- 1.069 .135 7.909 0.000 Watch RTV <--- .912 .114 7.988 0.000 Watch other programs relate to RTV celebrities

<--- 1.132 .133 8.486 0.000

Wish to attend RTV <--- .900 .140 6.422 0.000 RTV like real life <--- .695 .111 6.269 0.000 Wish to attend RTV <--- 1.236 .149 8.289 0.000 RTV with public appearances <--- 1.049 .144 7.292 0.000 Imagine about RTV shows <--- 1.050 .132 7.955 0.000

The Table 6.17 of regression weight refers to the un-standardised parameter estimates

for the factor loadings. No critical ratios (t-values) are stated for the respective first

factor loadings of RTV involvement and pay attention to RTV were fixed to unity to

scale the latent variables. All remaining factor loading were significant.

Table 6.18 Standard Regression Weights, Goodness-of-Fits Estimates and Composite

Reliability of the RTV Involvement

Standardised Regression Weights

RTV Involvement

Estimate

Composite Reliability

Pay attention to RTV <--- .577

RTV from news papers <--- .528 Watch RTV <--- .520 Watch other programs relate to RTV celebrities

<--- .619

Wish to attend RTV <--- .443 RTV like real life <--- .423 Wish to attend RTV <--- .636

RTV with public appearances

.503

Imagine about RTV shows <---

.567

.90

Reliability-Cronbach alpha � 0.78 Chi-Square 105.914 Degree of freedom 27 P .000 Normed Chi-Square (CMIN/DF) 3.923 Root mean square residual (RMR) .259

190

Standardised Regression Weights

RTV Involvement

Estimate

Composite Reliability

Root mean square of error of estimation (RMSEA) .089

Goodness of fit index (GFI) .931

Adjusted of goodness-of-fit index (AGFI) .885

Normed fit index (NFI) .845

Tucker-Lewis Index (TLI) .837

Comparative fir index (CFI) .878

In the Table 6.18, coefficient alpha for the RTV involvement of CFA model is .78,

indicating that the variables are a reasonable measure of RTV involvement. Results

show that all the nine items except two have standardized regression weight more than

0.5. Moreover, goodness-of-fit indices also indicate that the model fitted data well, with

GFI, AGFI, NFI, TLI, RMSEA and CFI all within acceptable level of criteria.

Composite reliability .90 also indicates the reliability the underlying variables of ‘RTV

Involvement’.

6.6.4 CFA results of Consumption Related Cognition

Consumption related cognition was measured by six observed variables. The structure

of this measurement model is presented in Figures 6.4. Scale items that measured the

‘consumption related cognition’ are adopted from current literature and refined through

qualitative research. Further, the following scales items are tested through EFA relating

to different domains of scale items have distinct meaning for measuring ‘consumption

related cognition’. The results of CFA of the measurement items of the ‘consumption

related cognition’ are summarised in Table 6.19.

191

Figure 6.4 CFA Model of Consumption Related Cognition

Consumption

related

cognition

cog 1 e1

1

1

cog 2 e2

1

cog 3 e3

1

cog 4 e4

1

cog 5 e5

1

cog 6 e6

1

Table 6.19 Regression Weights of Consumption Related Cognition

Regression Weights Estimate S.E. C.R. P value

Product instruction from RTV <--- 1.000

RTV helps recalling <--- .993 .066 15.127 0.000

RTV makes aware of brands <--- .845 .063 13.337 0.000

RTV provides product information <--- .778 .063 12.442 0.000

RTV helps market learning <--- .732 .065 11.210 0.000

RTV ads provide product ideas <--- .761 .067 11.316 0.000

The Table 5.19 of regression weight refers to the un-standardised parameter estimates

for the factor loadings. No critical ratios (t-values) are stated for the respective first

factor loadings of cognition and product instruction from RTV were fixed to unity to

scale the latent variables. All remaining factor loading were significant.

Table 6.20 Standard Regression Weights, Goodness- of- Fit Estimates and Composite Reliability of the Consumption Related Cognition

Standardised Regression Weights

Consumption Related Cognition

Estimate Composite Reliability

Product instruction from RTV <--- .823

RTV helps recalling <--- .755

RTV makes aware of brands <--- .686

RTV provides product information <--- .644

RTV helps market learning <---

.602

.96

192

Standardised Regression Weights

Consumption Related Cognition

Estimate Composite Reliability

RTV ads provide product ideas <--- .606

Reliability-Co-efficient alpha � 0.81 Chi-Square 16.730 Degree of freedom 9 P .053 Normed Chi-Square (CMIN/DF) 1.859 Root mean square of error of estimation (RMSEA) .048

Goodness of fit index (GFI) .985

Adjusted of goodness-of-fit index (AGFI) .965

Normed fit index (NFI) .979

Tucker-Lewis Index (TLI) .983

Comparative fir index (CFI) .990

The coefficient alpha for the consumption related cognition of CFA model is 0.81 in the

Table 6.20 indicating that the variables are a reasonable measure of level of

consumption related cognition. However, couple of the measurement items, ‘cog 6’ and

‘cog 8’, were dropped at this stage due to poor factor loadings. Table 6.20 shows that

standard regression weights of all the variables were more than 0.5 with some are even

more than 0.7. Goodness-of-fit indices also indicate that the model fitted data well with

value GFI, AGFI, NFI, TLI, RMSEA and CFI all within acceptable level of criteria.

Composite reliability .96 also indicates the reliability the underlying variables of

‘consumption related cognition’.

6.6.5 CFA results of Consumption Related Attitude

Consumption related attitude was measured by six observed variables. The structure of

this measurement model is presented in Figures 6.5. Scale items that measured the

‘consumption related attitude’ are adopted from current literature and refined through

qualitative research. Further, the following scales items are tested through EFA relating

to different domains of scale items have distinct meaning for measuring ‘consumption

related attitude’. The results of CFA of the measurement items of the ‘consumption

related attitude’ are summarised in Table 6.21.

193

Figure 6.5 CFA Model of Consumption Related Attitude

Consumption

related

attitude

att 1 e1

1

1

att 2 e2

1

att 3 e3

1

att 4 e4

1

att 5 e5

1

att 6 e6

1

Table 6.21 Regression Weights of Consumption Related Attitude

Regression Weights Estimate S.E. C.R. P value

RTV celebrities make fashion conscious <--- 1.000

Fashion tips from RTV <--- .943 .083 11.407 0.000

Buy cloths shown in RTV <--- .995 .085 11.643 0.000

RTV develops teenagers’ creativity <--- .826 .083 9.981 0.000

Teenagers like celebrities’ good figure <--- .777 .085 9.125 0.000

Teenagers like product shown in RTV <--- .869 .075 11.635 0.028

The Table 6.21 of regression weight refers to the un-standardised parameter estimates

for the factor loadings. No critical ratios (t-values) are stated for the respective first

factor loadings of consumption related attitude and RTV celebrities make fashion

conscious were fixed to unity to scale the latent variables. All remaining factor loading

were significant.

194

Table 6.22 Standard Regression Weights, Goodness-of-fits Estimates and Composite

Reliability of the Consumption Related Attitude

Standardised Regression Weights

Consumption Related Attitude

Estimate Composite Reliability

RTV celebrities make fashion

conscious <--- .741

Fashion tips from RTV <--- .693

Buy cloths shown in RTV <--- .667

RTV develops teenagers’ creativity <--- .595

Teenagers like celebrities’ good figure <--- .521

Teenagers like product shown in RTV <---

.683

.94

Reliability-Co-efficient alpha � 0.81 Chi-Square 17.851 Degree of freedom 9 P .037 Normed Chi-Square (CMIN/DF) 1.983 Root mean square of error of estimation (RMSEA) .052

Goodness of fit index (GFI) .984

Adjusted of goodness-of-fit index (AGFI) .963

Normed fit index (NFI) .972

Tucker-Lewis Index (TLI) .976

Comparative fir index (CFI) .986

The coefficient alpha for the consumption related attitude of CFA model is 0.81, shown

in the Table 6.22 indicating that the variables are a reasonable measure of consumption

related attitude. However, couple of the measurement items, ‘att 6’ and ‘att 7’, were

dropped at this stage due to poor factor loadings as well as ‘att 7’ was dropped not

qualify to represent a factor (Hair et al. 2006). Further, one of the measurement items of

consumption related cognition ‘cog 9’ shows loading with consumption related attitude

shown at Table 6.9 as have got theoretical supports. Table 6.22 shows that standard

regression weight of all the scale items are clearly more than 0.5. Goodness-of-fit

indices also indicate that the model fitted data well with GFI, AGFI, NFI, TLI, RMSEA

and CFI all within acceptable level of criteria. Composite reliability .94 also indicates

the reliability the underlying variables of ‘consumption related attitude’.

195

6.6.6 CFA results of Consumption Related Values

Consumption related values were measured by four observed variables. The structure of

this measurement model is presented in Figures 6.6. Scale items that measured the

‘consumption related values’ are adopted from current literature and refined through

qualitative research. Further, the following scales items are tested through EFA relating

to different domains of scale items have distinct meaning for measuring ‘consumption

related values’. The results of CFA of the measurement items of the ‘consumption

related values’ are summarised in Table 6.23.

Figure 6.6 CFA Model of Consumption Related Values

Values

val 1 e11

1

val 2 e2

1

val 3 e3

1

val 4 e4

1

Table 6.23 Regression Weights of Consumption Related Values

Regression Weights Estimate S.E. C.R. P values

MNC brand in RTV <--- 1.000

Teenagers like newest thing in RTV <--- 1.508 .154 9.802 0.000

Teenagers like best thing in RTV <--- 1.270 .128 9.961 0.000

Teenagers request for allowance <--- 1.011 .132 7.683 0.000

The Table 6.23 of regression weight refers to the unstandardised parameter estimates

for the factor loadings. No critical ratios (t-values) are stated for the respective first

factor loadings of consumption related values and MNC brand in RTV were fixed to

unity to scale the latent variables. All remaining factor loading were significant.

196

Table 6.24 Standard Regression Weights, Goodness-of-Fit Estimates and Composite

Reliability of the Consumption Related Values

Standardised Regression Weight

Consumption Related Values

Estimate

Composite Reliability

MNC brand in RTV <--- .591

Teenagers like newest thing in

RTV <--- .833

Teenagers like best thing in RTV <--- .736

Teenagers request for allowance <---

.497

.92

Reliability-Co-efficient alpha � .75 Chi-Square 3.277 Degree of freedom 2 P .194 Normed Chi-Square (CMIN/DF)

1.639

Root mean square of error of

estimation (RMSEA)

.042

Goodness of fit index (GFI) .996

Adjusted of goodness-of-fit

index (AGFI)

.978

Normed fit index (NFI) .991

Tucker-Lewis Index (TLI) .990

Comparative fir index (CFI) .997

The coefficient alpha for the consumption related values of CFA model is 0.75 in the

above mentioned Table 6.24, indicating that the variables are a reasonable measure of

consumption related values. However, number of the measurement items, ‘val 4’ and

‘val 5’, and ‘val 6’ were dropped at this stage as they are unable to represent a factor

(Hair et al. 2006). Further, one of the variables of consumption related attitude ‘att 9’

shows loading with consumption related values shown at Table 6.9 have got theoretical

supports. Table 6.24 shows that standard regression weights of all four variables are

more than 0.5. Goodness-of-fit indices also indicate that the model fitted data well, with

the P value, GFI, AGFI, NFI, TLI, RMSEA and CFI all within acceptable level of

criteria. Composite reliability .92 also indicates the reliability the underlying variables

of ‘consumption related values’.

197

6.7 Reliability and Validity of Constructs

Before testing the full model in relation to the research proposition; the final and next

step was to examine theoretical constructs of the model for reliability and validity.

Reliability is the degree to which measures are free from random error, that is,

systematic error does not affect reliability because it implies on measurement in a

consistent way, whereas random error produces inconsistency in measurement and

lower reliability (Malhotra 2003). Thus reliable measures of unobservable theoretical

constructs assist researchers in obtaining a clear picture of true relationships between

variables whereas random errors tend to minimize relationships. Whereas, validity is the

ability of a construct’s indicators to measure accurately the construct, that is, whether a

variable measures what it is supposed to measure (Hair et al. 1998; Malhotra 2003).

Both reliability and validity were considered in this research using SEM, because in

contrast to the limitations of traditional assessment techniques, SEM allows researchers

to examine the effects of latent variables on observed variables and partition error

variance in the observed variables (Baumgartner & Homburg 1996; Bollen 1989).

6.7.1 Reliability

Reliability of this research’s instrument was checked to assure that it was free from

random error and to provide consistent results (Malhotra 2003). In this research four

approaches were considered: split-sample, test-retest, alternative forms, and internal

comparison method (Carmines & Zeller 1979) and the fourth approach were chosen.

The first two approaches require that the same test is given to the same group of people

in a different period and this method has not been considered due to time and budget

constraints. The third approach requires a larger sample than this study to examine the

differences between correlations of the first and second halves of the sample and the

correlation of even and odd items (Zikmund 2000). In this research, the fourth approach

was applied to administrate a single test of scale and item reliability (Carmines & Zeller

1979). Furthermore, composite reliability has been assessed of the scale items in this

198

research. Reliability has been evaluated in CFA using the composite reliability.

Composite reliability is a better indicator than Co-efficient Alpha because it is free from

the assumption of equal item reliabilities (Anderson & Gerbing 1988; Hair et al. 1998).

Composite reliability of variable is calculated by the following (Hair et al. 1998):

Composite reliability =( � standardized loading) 2 / ( � standardized loading) 2 = � € j).

6.7.2 Reliability of Constructs

At first, the coefficient alpha, one of the most commonly used statistics, was used to

assess the reliability of the measuring scales with multiple items (Malhotra 2003). It

indicates the degree of consistency of the observed variables (Carmines & Zeller 1979).

In exploratory research such as this research, modest reliability is recommended where

Coefficient alpha 0.5 and above indicates satisfactory internal consistency reliability

(Hair et al. 1998; Malhotra 2003). In this research, as shown in Table 6.25, co-efficient

alpha of all the variables of underlying constructs are quite high more than .70

indicating good reliability of the scales.

Table 6.25 Construct Reliability of Scale Items

Variables Original Reliability

Parental control 0.81

Peers group Influences 0.76

RTV involvement 0.77

Consumption related cognition 0.81

Consumption related attitude 0.81

Consumption related values 0.75

199

6.7.3 Item Reliability

Item reliability was assessed through squared multiple correlation coefficients (SMCs)

for the observed or measured variables, where the measured variables are the dependent

variable and the latent variable is the independent variable (Bollen 1989; Tabachnick &

Fidell 2001). Each SMC is interpreted as the reliability of the measured variable, and as

the proportion of variance in the measured variable that is accounted for by the latent

variable. An SMC of › 0.3 correspondents to an� error variance that exceeds the

variation in the measured variable explained by the latent variable (Byrne 2001;

Holmes-Smith 2002). Table 6.26 shows the SMCs for the variables in the main model.

Table 6.26 Multiple Correlation Coefficients

Variables Estimate Talk to friends about RTV shows .238

Friends think RTV provide useful information .484

Friends think RTV entertaining .408

Teenagers and friend talk about RTV content .404

Friends watch RTV together .260

Peers opinion important .382

Teenagers and parents jointly select RTV show .670

Co-viewing at weekends .596

Co-viewing at weekdays .521

Teenagers talk to parents about RTV content .452

Parents know about RTV .199

Imagine about RTV shows .269

RTV with public appearances .217

Wish to attend RTV .322

RTV like real life .206

RTV on web .194

Watch other programs relate to RTV .318

Watch RTV .239

RTV from news papers .235

Pay attention to RTV .272

Teenagers request for allowance .248

Teenagers like best thing in RTV .517

Teenagers like newest thing in RTV .653

MNC brand in RTV .379

Teenagers like product shown in RTV .497

200

Variables Estimate Teenagers like celebrities’ good figure .258

RTV develops teenagers’ creativity .355

Buy cloths shown in RTV .407

Fashion tips from RTV .458

RTV celebrities makes fashion conscious .494

RTV ads provide products ideas .372

RTV helps market learning .364

RTV provides product information .448

RTV makes aware of brands .454

RTV helps recalling .528

Product instruction from RTV .616 Overall, Table 6.26 shows that there is good item reliability for the observed variables.

No more than 12 of the 36 variables have square multiple correlation (SMCs) less than

0.3. As it is mentioned already, most of scale items in this research were adopted from

other researches and some of them are established for the first time in the context of

consumer socialisation through reality TV in a developing country. However, most of

the scale items in these latent variables have good item reliability because the items

were carefully adopted and developed from extensive literature review and refined

through qualitative research (Table 5.1), it will be accepted for this research but with

caution about its applicability in later research (Rao 2002).

6.7.4 Validity

Convergent validity is a measure of construct validity (Malhotra 2003) and is the extent

to which a measure of some characteristic correlates or converges with other measures

of the same characteristic (Aasker, Kumar & Day 2001). Convergent validity of a

measure is the degree to which multiple attempts to measure the same construct are in

agreement (Hair et al. 1998). It is achieved if 1) the model receives a satisfactory level

of fit, and 2) the regression co-efficient (factor loadings) of all the indicators are

statistically significant, e.g., greater than twice its standard error (Anderson & Gerbing

1998; Dunn et al. 1994). In this research, each variable has a critical ratio larger than

critical value and all the CFA models receives satisfactory fit indicating that convergent

validity was achieved.

201

On the other hand, discriminant validity estimates the degree to which a measure does not

correlate or converge with other constructs from which it is supposed to be different

(Malhotra 2003). It involves demonstrating a lack of co-relation among different constructs.

There are two types of discriminant validity; 1) within-construct validity and 2) across

construct validity. This research satisfies the within-construct validity as the measurement

items are highly co-related with the underlying construct. Further, this research

demonstrates across-construct validity as the model acheieved a satisfactory level of fit

(Anderson & Gerbing 1988).

In summary, in the first stage of this research, EFA were used to reduce the 45 variables

down to a smaller number of variables (36) which were then used for CFA and structural

equation models in the next step. Nine scale items as discussed earlier were dropped due

poor factor loadings at EFA stage and some others did not qualify to represent as an

individual factor. The greatest number of composite variables in any CFA model was nine

and the lowest were four in the main structural model which investigated the inter-

relationships amongst these composites. For any of these models, the sample size of 368

was sufficient for parameter estimation and model fitting. In this first stage of SEM,

clarifications of the measurement models were made before proceeding to test the main full

model (Anderson & Gerbing 1998; Bagozzi 1994).

One important issue for estimating a model is to decide on a one-or-two stage approach.

The conventional way of SEM involves estimating the measurement and structural models

together in one stage or at a time (Diamantopolous 1994). The one-step approach’s strength

lies in its rigor and lack of estimating bias because it is more consistent with the ‘the spirit

of causal modeling’ (Hulland, Chow & Lam 1996). Measurement models were estimated

first in order to assess the quality of the measurement items. This research used CFA to

maximize the reliability of latent variables and to reduce the variables to a manageable level

in fitting structural equation models to data at the subsequent stage as recommended by

Holmes-Smith & Rowe 1(994). Subsequently, the measurement and structural models were

estimated together in one model (one-step approach). Fornell and Yi (1992) strongly argued

for the superiority of this approach in which the over all fit of a model is tested without

regard to separate measurement and structural models.

202

Chapter 7 Testing the Theoretical Model and Hypotheses

7.1 Structural Model Evaluation

The development of CFA in the previous section has validated data and obtained a

manageable number of valid and more reliable variables which can be used in the final

step of evaluating structural models (Anderson & Gerbing 1988). This section reports

the test results of the subsequent stage of SEM, full structural model (main model) as

well as proposed structural model. Particularly, three issues were considered in this

process: distribution normality, estimating the main model, estimating the proposed

model and modification of the model. Above mentioned issues will be discussed in the

following sections.

Further, along with other social structural variables in the main model; teenagers’

gender differences and social class differences were not measured by underlying

variables in this research. Since, these two variables are categorical variables therefore

in section 7.2 multi-group analysis has been carried out to understand the affect of

gender differences and social class differences on consumer socialisation of teenagers

by RTV.

7.1.1 Assessment of the Distribution of the Variables

Following the estimation of the CFA models, this section assesses the distributional

properties of the measurement variables that were used to estimate the structural

models. The univariate summary statistics for the variables are presented in Table 7.1.

On the basis of standard deviation, there appears to be a substantial amount of variation

in each of the variables and many of them departed from normality. However, because

the distribution result is asymptotic and not exact, some allowance for non-normality

can be made (Rao 2002). In this case, all the data negatively skewed and negative

kurtosis value indicates flatter than normal distribution. In addition, the use of

maximum likelihood (ML) moderates its effects in this research (Anderson & Garbing

1988; Kline 1998). Researchers have shown that when the data has slight or moderate

203

deviation form multi-normality, ML has been the preferable method (Anderson &

Garbing 1988; Kline 1998). ML is based on the assumption that the observed variables

are normally distributed. To repeat all 36 composite indicators have kurtosis values

ranging from -1.491 to -.080 and skew ness values ranging from ranging from .983 to +

.077. All indicators can be considered as normally distributed because all the values are

less than 3.0 for skew ness and 10 for kurtosis (Kline 1998). Another indication is that

ML requires a large number sample size (Byrne 2001), this study satisfy this

requirement with 368 samples. In this model estimation, this requirement has also been

met because HOELTER index threshold is above required 200.

Table 7.1 Assessment of Normality

Variable min max skew c.r. kurtosis c.r. Talk to friends about RTV shows 1.000 7.000 -.692 -5.421 -.980 -3.837 Friends think RTV provide useful information

1.000 7.000 -.579 -4.534 -.483 -1.893

Friends think RTV entertaining 1.000 7.000 -.835 -6.542 -.080 -.311 Teenagers and friend talk about RTV content 1.000 7.000 -.983 -7.697 -.170 -.668 Friends watch RTV together 1.000 7.000 -.148 -1.160 -1.380 -5.405 Peers opinion important 1.000 7.000 -.225 -1.763 -1.096 -4.292 Teenagers and parents jointly select RTV show

1.000 7.000 -.294 -2.305 -1.237 -4.844

Co-viewing at weekends 1.000 7.000 -.707 -5.536 -.750 -2.938 Co-viewing at weekdays 1.000 7.000 -.061 -.480 -1.235 -4.837 Teenagers talk to parents about RTV content 1.000 7.000 .119 .929 -1.283 -5.025 Parents know about RTV 1.000 7.000 -.861 -6.746 -.439 -1.718 Imagine about RTV shows 1.000 7.000 -.367 -2.876 -1.013 -3.967 RTV with public appearances 1.000 7.000 -.450 -3.525 -1.287 -5.039 Wish to attend RTV 1.000 7.000 -.595 -4.656 -1.002 -3.924 RTV like real life 1.000 7.000 -.504 -3.951 -.625 -2.447 RTV on web 1.000 7.000 -.517 -4.051 -1.149 -4.500 Watch other programs relate to RTV 1.000 7.000 -.540 -4.233 -.858 -3.360 Watch RTV 1.000 7.000 -.940 -7.359 -.228 -.892 RTV from news papers 1.000 7.000 -.144 -1.130 -1.407 -5.508 Pay attention to RTV 1.000 7.000 -.526 -4.119 -.802 -3.142 Teenagers request for allowance 1.000 7.000 .005 .042 -1.491 -5.838 Teenagers like best thing in RTV 1.000 7.000 -.790 -6.184 -.506 -1.981 Teenagers like newest thing in RTV 1.000 7.000 -.344 -2.696 -1.089 -4.266 MNC brand in RTV 1.000 7.000 -.036 -.282 -1.001 -3.922 Teenagers like product shown in RTV 1.000 7.000 -.436 -3.414 -.731 -2.861 Teenagers like celebrities’ good figure 1.000 7.000 -.696 -5.449 -.910 -3.565 RTV develops teenagers’ creativity 1.000 7.000 -.837 -6.552 -.580 -2.273 Buy cloths shown in RTV 1.000 7.000 -.131 -1.027 -1.310 -5.128

204

Variable min max skew c.r. kurtosis c.r. Fashion tips from RTV 1.000 7.000 -.748 -5.859 -.598 -2.341 RTV celebrities makes fashion conscious 1.000 7.000 -.295 -2.314 -1.028 -4.025 RTV ads provide products ideas 1.000 7.000 -.750 -5.875 -.525 -2.056 RTV helps market learning 1.000 7.000 -.766 -5.995 -.272 -1.067 RTV provides product information 1.000 7.000 -.854 -6.686 -.209 -.818 RTV makes aware of brands 1.000 7.000 -.701 -5.488 -.381 -1.492 RTV helps recalling 1.000 7.000 -.436 -3.414 -.857 -3.355 Product instruction from RTV 1.000 7.000 -.637 -4.988 -.442 -1.731 Talk to friends about RTV shows 163.540 29.989

7.1.2 Estimating the Proposed Model

A proposed model was developed and tested to ensure that the model with the best

explanatory power was accepted (Bollen & Lennox 1991; Hair et al. 1998). The

proposed model developed on the basis of current literature which has been reported in

chapter 3 (Figure 3.2) and illustrated in Appendix 8, was tested at first. The proposed

model is shown in Figure 7.1 and results of the proposed model evaluation are shown in

Table 7.2.

In Table 7.2 shows that fitness measures of proposed model does not satisfy the set

criteria as suggested by researchers mentioned in Table 6.10. Overall, the indicators of

goodness-of-fit criteria and other estimated values are under the suggested ranges. So,

modification was needed to develop a better model fit. Therefore, the proposed model

will not be considered in the rest of this thesis.

Table 7.2 Fitness of Measure for the Proposed Model

Fit Measure Proposed Model

Chi-Square 2170.099

Degree of freedom 938

P .000

Normed Chi-Square (CMIN/DF) 2.314

Root mean square of error of estimation (RMSEA) .060

Goodness of fit index (GFI) .779

Adjusted of goodness-of-fit index (AGFI) .756

Normed fit index (NFI) .683

Tucker-Lewis Index (TLI) .778

205

Comparative fir index (CFI) .790

HOELTER 177

Figure 7.1 Proposed Model of CSRTV

Cognition

cog1 e1

1

1

cog2 e2

1

cog3 e3

1

cog4 e4

1

cog5 e5

1

cog6 e6

1

cog7 e7

1

cog8 e8

1

cog9 e9

1

Attitude

att1 e10

1

1

att2 e11

1

att3 e12

1

att4 e13

1

att5 e14

1

att6 e15

1

att7 e16

1

att8 e17

1

att9 e18

1

Values

val1 e19

1

1

val2 e20

1

val3 e21

1

val4 e22

1

val5 e23

1

val6 e24

1

RTV

Inv

inv1 e25

1

1

inv2 e26

1

inv3 e27

1

inv4 e28

1

inv5 e29

1

inv6 e30

1

inv7 e31

1

inv8 e32

1

inv9 e33

1

Peers'

Influences

pi6e39

1

1pi5e38

1pi4e37

1pi3e36

1pi2e35

1pi1e34

1

Parental

control

pc6e45

1

1pc5e44

1pc4e43

1pc3e42

1pc2e41

1pc1e40

1

res 11

res 2

1

res 31

res 4

1

7.1.3 Modifying the Model (Refinement of Scale Items)

At this stage, it was considered to modify the model according to the modification

indices to improve model fit after the model assessment (Hair et al. 1998). However,

modifying the model has to be driven by theory rather than survey data

(Diamantopoulus 1994; Tabachnick & Fidell 2001). If substantive theory suggests any

variable or a relationship should be included in a model, as well as it should be left there

even if it does not contribute to the over all fit of the model (Joreskog & Sorbom 1993).

In this research the main model which is discussed next has got both theoretical as well

as data supports.

206

7.1.4 Estimating the Main Model and Test Hypotheses

After evaluating the data against the assumption of SEM, next stage was to estimate the

main model and assess over all model fit (Hair et al.1998). This section presents the test

results of the full hypothesised model in Table 7.3. Further, differences between the

result of main and proposed model evaluation are shown in Table 7.4. The main

structural model shows the latent constructs: parental control, peers group influences,

RTV involvement, consumption related cognition, attitude and values. The diagram of

the main model is illustrated in Figure 7.2 and Appendix 9 shows over all path

diagrams. The result shows that there is no difference between the proposed and the

main structural model. However, there is difference between the measurement variables

of the two models. Accordingly, the proposed model did not require any modification

except refinement of the scale items.

Keesling (1972) and Wiley (1973) proposed that general structural equation model

combine regression technique with factor analysis. In this research, exploratory and

confirmatory factor analysis has been carried out followed by developing SEM. Further,

absolute and incremental fit indices (Hu & Bentler 1999) as well as goodness-of-fit

reported that the survey data fit quite well with model. Overall, goodness-of-fit statistics

were within acceptable fit as shown in Table 7.3. Marsh and Hocevar (1988)

recommended that CMIN/DF of 2 to 5 is indicative of an acceptable fit between

hypothetical model and the sample data. The cmin/df is 2.010 in this case along with

RMSEA .052 respectively which is recommended as <.08 (RMSEA) by Hair et al

(2006). Further, Bentler and Bonett (1980) as well as Mcdonald and Marsh (1990)

considered fit index NFI and CFI close to 1 is really desirable which is clearly

indicating support for the model in this research. Rest of the incremental fit indices like

GFI, AGFI and TLI also indicates reasonable fit in this model, which is recommended

as much close as 1 (Bentler & Bonett 1980; Tanaka & Huba 1985). HOELTER value

in this case is 201 which also satisfy the acceptable fit requirement. However, Chi-

square P value has got ‘.000’ in this case which is one of the indicators of goodness-of -

fit. The chi-square value should not be significant if there is a good model fit (Garson

207

2009). A significant chi-square indicates lack of satisfactory fit (Garson 2009), which is

not the case in this research. Moreover, chi-square P is considered as conservative

measure of goodness-of-fit (prone to type two error), that is why many researchers who

use SEM with reasonable sample size (e.g., > 200) indicating good approximate fit test

by (e.g., NFI, CFI, RMSEA) (Garson 2009). Overall, these results confer that the main

model can be assessed as being adequate.

Table 7.3 Fitness of Measure for the Main Model

Fit Measure

Chi-Square 1179.943 Acceptable

Degree of freedom 587 Acceptable

P .000 Lower than critical value

Normed Chi-Square (CMIN/DF) 2.010 Acceptable

Root mean square of error of estimation (RMSEA)

.052 Acceptable

Goodness of fit index (GFI) .845 Acceptable

Adjusted of goodness-of-fit index (AGFI) .824 Acceptable

Normed fit index (NFI) .773 Acceptable

Tucker-Lewis Index (TLI) .861 Acceptable

Comparative fir index (CFI) .870 Acceptable

HOELTER 201 Acceptable

208

Figure 7.2 Main Model of CSRTV

Cognition

cog 1 e1

1

1

cog 2 e2

1

cog 3 e3

1

cog 4 e4

1

cog 5 e5

1

cog 6 e6

1

Attitude

att 1 e7

1

1

att 2 e8

1

att3 e9

1

att 4 e10

1

att5 e11

1

att 6 e12

1

values

val 1 e 1311

val 2 e14

1

val 3 e 15

1

val 4 e 16

1

RTV

Inv

inv 1 e 17

1

1

inv 2 e 18

1

inv 3 e 19

1

inv 4 e 20

1

inv 5 e 21

1

inv 6 e 22

1

inv7 e 23

1

inv 8 e 24

1

inv 9 e 25

1Parental

control

pc 5e 36

1

1pc 4e 35

1pc 3e34

1pc 2e 33

1pc 1e 32

1

Peers

Inf luences

pi 6e 31

1

1

pi 5e 30

1

pi 4e 29

1

pi 3e 28

1

pi 2e 27

1

pi 1e 26

1

res 1

1

res 2

1

res 3

1

res 4

1

Table 7.4 Comparisons of Fit Measures for Main and Proposed Model

Fit Measure Main Model Proposed Model

Chi-Square 1179.943 2170.099

Degree of freedom 587 938

P .000 .000

Normed Chi-Square (CMIN/DF) 2.010 2.314

Root mean square of error of estimation (RMSEA) .052 .060

Goodness of fit index (GFI) .845 .779

Adjusted of goodness-of-fit index (AGFI) .824 .756

Normed fit index (NFI) .773 .683

Tucker-Lewis Index (TLI) .861 .778

Comparative fir index (CFI) .870 .790

HOELTER 201 177

209

Models demonstrate two types of fit: overall fit and the local fit of individual

parameters. Overall fit is evaluated by how well the model explained all of the data in

the entire analysis. Local fit is determined examining how specific parameters may have

achieved statistical significance within the model. The chi-square statistic is the

generally recognised fit index for assessing overall model fit. In this case, the chi-square

is statistically significant indicating poor model fit. However, for models with more

cases (above 200), the chi-square is almost always statistically significant. As this study

involved much higher than 200 cases, the chi square statistic is not a suitable fit

indicator. It has been suggested that the ratio of chi-square to degrees of freedom be

examined as an alternative realistic indicator of fit when sample sizes are larger than

200 (with ratio of 5:1 to 2:1 considered to be within the acceptable range (Tanaka,

1993). In this study, this ratio was 1:2.66 which falls within the acceptable range

(Tanaka 1993). A number of fit indices, as identified in the above table are also used to

evaluate overall model fit. Fit indices achieved in this research are mostly above the

accepted level, except for GFI, AGFI and NFI which are marginally below their

respective acceptable levels. GFI and AGFI are also affected by sample size

(MacCullum & Hong 1997). On the other hand, NFI was one of the earliest fit indices

developed by Bentler & Bonnett in 1980. However, later on they found that NFI did

have some inherent weaknesses and created CFI as an alternate measure. In this

research, CFI is well within acceptable range, indicating acceptable fit. Overall, model

fit may be poor, but there are other alternatives for determining the value of a model.

Specifically, local model fit focuses on the value of individual parameters in the model.

Support regarding local fit is indicated when significant paths are found to be in the

hypothesised direction, and the magnitude of the item loadings is greater than .45

(Bentler & Wu 1993; Joreskog & Sorbom 1989). All the item loadings in this model

were above this threshold confirming local fit of the model. In spite of all that, however,

Overall model fit in this case should be accepted with caution due to some of the below

threshold fit indices achieved.

210

7.1.5 Testing Hypotheses of the Main Model

In the following section, the standardized regression coefficient obtained from the

modified main model is used to test the hypotheses discussed in chapter 3. The

following empirical data standard regressions weights are significantly greater than

.000. P values fall within the range <.05 and critical rations all are above 1.96. As

shown in Table 7.5, all the 7 hypotheses are fully supported (not rejected).

Table 7.5 Standard Estimates for the Main Model

Estimates S.E C.R P values

Hypothesis test

H1 Cognition

<---

Involvement .737 .118 8.818 .000 Supported

H2 Attitude

<---

Involvement .301 .138 3.015 .003 Supported

H3 Attitude

<---

Cognition .211 .085 2.441 .015 Supported

H4 Values

<---

Involvement .596 .104 6.779 .000 Supported

H5 Attitude

<---

Values .392 .087 5.246 .000 Supported

H6 RTV Involvement

<---

Parental control .249 .075 3.656 .000 Supported

H7 RTV Involvement

<---

Peers group influences

.665 .075 7.077 .000 Supported

211

7.1.6 Interpretation/Discussion of Empirical Results

The following section shows the detail explaination and interpretation of the empirical

results:

The following section shows the detail explanation and interpretation of the Empirical

results:

H1) There is a positive impact of RTV involvement on consumption related

cognition.

The standardized co-efficient of the path from ‘RTV involvement’ to ‘consumption

related cognition’ is significantly greater than zero (�=.737, p=.000) as shown in Table

7.5. H 1 is also supported by empirical data. Teenagers’ involvement with RTV plays a

significant role in their consumption related cognition development.

Hypothesis 1 for research objective 2 relates to the relationship between RTV

involvement and teenagers’ consumption related cognition development. Current

literature shows, as part of mainstream media, TV helps to develop product related

knowledge, consumer role perception and influences purchasing intentions (Bartlett,

Griffiths & Badian 2008; Moschis & Moore 1978). Findings of this study confirm these

issues from the perspective of RTV involvement. Furthermore, findings of this research

support that RTV advertisements also provide products and brand related knowledge

and ideas to teenagers. However, this study also shows that RTV does not directly help

teenagers to make specific buying decisions or help them to do value analysis as

proposed by the current literature. In other words, teenagers’ RTV involvement helps

them in their consumption related cognition providing knowledge and information but

do not directly influence their buying behaviour.

In summary, the findings of this research support the hypothesis that there is a positive

impact of RTV involvement on consumption related cognition. The empirically tested

CFA model, standard regression weights and goodness-of-fit shown in Chapter 6,

(Section 6.6.4) support this hypothesis.

212

H2) There is a positive impact of RTV involvement on consumption related

attitude.

The standardized co-efficient of the path from ‘RTV involvement’ to ‘consumption

related attitude’ is greater than zero (�=.301, p=.003) as shown in Table 7.5. H 2 is also

supported by empirical data. The involvement with RTV plays a significant role in

developing teenagers’ consumption related attitude.

Hypothesis 2 for research objective 2 relates to the relationship between RTV

involvement and teenagers’ consumption related attitude. Current literature shows, as a

part of mainstream media, TV influences teenagers’ imitating tendencies, inspiration

from media personalities and attitudes towards product placement (Hirschman &

Thompson 1997; Russell, Norman & Heckler 2004; Schmoll et al. 2006). Findings of

this study also confirm these issues from the perspective of RTV involvement.

Furthermore, findings of this research support that teenagers prefer to buy multinational

brands if they are shown on RTV. However, this study also shows that teenagers do not

always blindly follow RTV celebrities regarding style. For example, they do not like to

do plastic surgery like their favourite RTV celebrities because they perceived them as

artificial.

In summary, the findings of this research support the hypothesis that there is a positive

impact of RTV involvement on consumption related attitudes. The empirically tested

CFA model, standard regression weights and goodness-of-fit shown in Chapter 6,

(Section 6.6.5) support this hypothesis.

213

H3) There is a positive impact of consumption related cognition on consumption

related attitude.

The standardized co-efficient of the path from ‘consumption related cognition’ to

‘consumption related attitude’ is greater than zero (�=.211, p=.015) as shown in Table

7.5. H 3 also is supported by empirical data, that there is a positive association between

consumption related cognition and consumption related attitude.

The literature suggested that formation of attitudes through affective, cognitive, or

behavioural processes establishes associations that are linked to the attitude object and

can become part of the mental residue that is attitude (Neal, Quaster & Hawkins 2006).

These associations can reflect the cognitive precursors of attitude individually, or as a

whole (Eagly & Chaiken 2007). So, changes of cognition through the media can affect

the attitudes of teenagers. The findings of this research suggest that involvement with

RTV results in consumption related cognition, that in turn affects consumption related

attitudes, as shown in Chapter 3, (Section 3.3.1.1) thus agreeing with the theory.

The interrelationship between cognition and attitude as proposed in the current literature

is also supported by the findings of this research. Factor loading has been observed for a

couple of consumption related cognition measurement variables (e.g., RTV develops

teenagers’ creativity, preference towards RTV sponsoring brands) with consumption

related attitude (Table 6.9). It also confirms this theoretical agreement and indicates that

there is a positive impact of consumption related cognition on consumption related

attitude.

H4) There is a positive impact of RTV involvement on consumption related values.

The standardized co-efficient of the path from ‘RTV involvement’ to ‘consumption

related values’ is significantly greater than zero (�=.596, p=.000) as shown in Table 7.5.

H 4 is also supported by empirical data. Teenagers’ involvement with RTV plays a

significant role in the development of consumption related values.

214

Hypothesis 4 for research objective 2 relates to the relationship between RTV

involvement and teenagers’ consumption related values development. Current literature

shows, as a part of mainstream media, TV plays a significant role in teenagers’ value

systems (Fletcher 2006). Furthermore, TV influences teenagers’ material values and

desires toward product consumption (Chan 2003; Ghani 2005; Wyllie, Zhang &

Casswell 1998). Findings of this study confirm these issues from the perspective of

RTV involvement. However, this study also shows that teenagers do not feel attracted

towards the prohibited items shown on RTV (e.g., alcohol, cigarettes). Furthermore,

contradicting the current literature, the findings show teenagers dislike the scanty outfits

worn by RTV participants and judges which they perceive as contradictory to their

culture.

In summary, the findings of this research support the hypothesis that there is a positive

impact of RTV involvement on consumption related values. The empirically tested CFA

model, standard regression weights and goodness-of-fit shown in Chapter 6, (Section

6.6.6) support this hypothesis.

H5) There is a positive impact of consumption related values on consumption

related attitude.

The standardized co-efficient of the path from ‘consumption related values’ to

‘consumption related attitude’ is clearly greater than zero (�=.392, p=.000) as shown in

Table 7.5. H 5 is also supported by empirical data, that development of teenagers’

consumption related values also affects their consumption related attitude.

Hypothesis 5 for research objective 2 relates to the relationship between consumption

related values and consumption related attitudes. Findings of this research confirm that

involvement with RTV results in consumer socialisation through consumption related

values and in turn, the consumption related attitudes supporting the current literature

(McCarty & Shrum 1993). The literature suggested that consumption related values,

215

which are widely held beliefs about what is acceptable and/or desirable, are an

important outcome of consumer socialisation (Neal, Quester & Hawkins 2004). Further,

Solomon (2004) refers values affect attitude to alter the consumption socialisation

(McCarty & Shrum 1993).

Findings of this research, as discussed in chapter 3 (Section 3.3.3.1), confirm this

relationship between values and attitudes. Findings of this research have also

empirically shown that there is a positive impact of consumption related values on

consumption related attitude. As discussed earlier in chapter 6 (Table 6.9), one of the

consumption related attitude items, teenagers’ perceptions toward multinational brands

has loading with consumption related values.

H6) There is a positive impact of peer groups influences on teenagers’ RTV

involvement.

The standardized co-efficient of the path from ‘peer group influences’ to ‘RTV

involvement’ is significantly greater than zero (�=.665 p=.000) as shown in Table 7.5.

H 6 is also supported by empirical data. Peer groups have significant influence on

teenagers’ involvement with RTV.

The literature suggested that as one of the socialisation agents, peer groups influence

teenagers’ socialisation (Churchill & Moore 1979). Moreover, peers’ influence is also

considered as a vital social structural variable that affects teenagers’ involvement with

the media (Lueg & Finney 2007). In particular, peer groups influence teenagers’ media

usage behaviour by sharing knowledge, giving preferences and opinions about different

TV programs (Lueg & Finney 2007; Nathanson 2001). Findings of this study confirm

these issues from the perspective of RTV involvement.

216

In summary, the findings of this research support the hypothesis that there is a positive

impact of peer groups influences on teenagers’ RTV involvement. The empirically

tested CFA model, standard regression weights and goodness-of-fit shown in Chapter 6,

(Section 6.6.2) support this hypothesis.

H7) Parents through communication, mediation and instruction positively

influence teenagers’ RTV involvement.

The standardized co-efficient of the path from ‘parental control’ to ‘RTV involvement’

is greater than zero (�=.249, p=.000) as shown in Table 7.5. Therefore, H 7 is also

supported by empirical data. So, parents through communication, mediation and

instructions influence teenagers’ involvement with RTV.

Hypothesis 7 for research objective 3 relates to the parental supervision of teenagers’

RTV involvement. Current literature shows that parents control their teenaged

children’s media involvement through communication, mediation, and instruction

(Bakir 2005; Barkin 2006, Cardoza 2002; Fuzioka 2002). The literature suggested that

as one of the social structural variables, parents in developing countries substantially

supervise teenagers' TV viewing behaviour (Cardoza 2002). Findings of this study also

confirm these issues from the perspective of RTV involvement. However, it has been

found in this research that due to teenagers’ parental business and their co-viewing

habits, they often provide moderate freedom to their teenaged kids to watch different

RTV shows. So, parents in Bangladesh do not control teenagers’ RTV involvement as

much as proposed by the current literature (Cardoza 2002).

Overall, findings of this research suggest that parents in Bangladesh control moderately

their teenagers’ involvement with RTV. The empirically tested structural model,

standard regression weights, goodness of fit shown in Chapter 6, (Section 6.6.1)

supports this hypothesis.

217

H8) There is a difference between male and female teenagers’ involvement with

RTV and its effect on their consumer socialisation.

As this was one of the categorical variables, it was not tested with the main model.

Rather, effect of teenagers’ gender difference was tested using multi-group analysis,

results of which will be presented bellow. Male group was considered as the original

sample and the female group as the validation sample. Table 7.6 shows that RTV has

positive affect on consumer socialisation of both male and female teenagers, as both the

groups satisfy the good-ness-of fit criteria. Further, each progressive test (factor

structure equivalence, factor loading equivalence) supports cross validation of the

results (Table 7.7). The loose validation test produces adequate fit statistics. The first

multiple sample test (Factor Structure Equivalence) also provides adequate fit.

Furthermore, each progressive test produces insignificant change in fit as well as

change in Chi-square over the subsequent test.

Table 7.6 Goodness-of -Fits Estimates of Male and Female

Fit Measures Male Female

Chi-Square 900.660 1088.097 Degree of Freedom 587 587 P .000 .000 Normed Chi-square (CMIN/df) 1.534 1.854

Root mean square of error of estimation (RMSEA) .054 .068

Goodness of fit index (GFI) .789 .763

Adjusted of goodness-of-fit index (AGFI) .760 .732

Normed fit index (NFI) .698 .642

Tucker-Lewis Index (TLI) .857 .776

Comparative fir index (CFI) .866 .792

218

Table 7.7 Cross Validation Statistics - Gender

Model

Description

Df X2 CFI RMSEA Comments

Loose cross

validation

587 1088.097 .792 .068 Reasonably

good fit

Factor

structure

equivalence

1174 1988.743 .821 .044 Reasonably

good fit

Factor loading

equivalence

1204 2016.82 .829 .043 Change in

X2 (30) = 2 8

Insignificant

However, differences of goodness-of-fit between two groups also show that RTV affect

their consumer socialisation marginally differently. Table 7.8 shows standard estimation

differences between male and female teenagers’ regarding consumer socialisation by

RTV.

Table 7.8 Multi-Group Analysis of Gender Differences on Consumer Socialisation

Process

Standard Estimate

Male Female

H1 Cognition <--- Involvement .756 .729

H2 Attitude <--- Involvement .425 .186

H3 Attitude <--- Cognition ..058 .339

H4 Values <--- Involvement .609 .598

H5 Attitude <--- Values .420 .374

H6 Involvement <--- Parental control .269 .133

H7 Involvement <--- Peers group influences .630 .754

219

Current literatures show that teenaged male and female have different tastes and

preferences towards different types of TV programs that ultimately affect the outcomes

of consumer socialisation. Findings of this study somewhat confirm these issues from

the perspective of RTV involvement. Table 7.8 shows teenagers’ consumption related

cognition is highly influenced by RTV involvement for both the groups. However, there

is marginal difference between male (.756) and female (.729) regarding their product

knowledge, role perceptions and purchasing intentions. Teenagers’ consumption related

attitude is also affected by RTV and shows significant differences between male (.425)

and female (.186) in terms of imitating tendencies, getting inspiration from celebrities

and perception towards products. On the contrary, it appears that teenaged females’

attitude towards consumption are more likely to be influenced by their cognition (.339)

compared to their male counterparts (.058). Further, consumption related values of

teenaged male (.609) and female (.598) are also affected by RTV involvement. The two

groups show minor differences as both the groups have similar beliefs towards

materialism, desire towards products and prohibited product consumption. Further,

consumption related values affect both males (.420) and females (.374) consumption

related attitude.

As one of the social structural variables, parents also control their teenaged boys and

girls differently. It has been addressed in current literatures that teenaged girls enjoy

more freedom to watch TV than boys, as girls have more attachment with their parents.

This study also confirms this finding as parents control their teenaged boys RTV

involvement (.269) more than girls (.133). Further, both boys (.630) and girls (.754)

show strong attachment with peer groups regarding RTV involvement but it appears

that teenaged females are more influenced by their peer groups than their male

counterparts.

Overall, the structural model fits for both the genders though there are some differences

in the standard estimation for some of the measurement items between the genders.

Accordingly, the hypothesis is only partially supported.

220

H9) There is a difference between middle and lower social class teenagers’

involvement with RTV and its effect on their consumer socialisation.

As social class was another categorical variable, it couldn’t be tested with the main

model. Rather, effect of social class difference was tested using multi-group analysis,

results of which will be presented in the following section. Middle social class was

considered as the original sample and the lower social class as the validation sample.

Table 7.9 shows that RTV has positive affect on consumer socialisation of middle and

lower social class teenagers, as both the groups satisfy the good-ness-of fit criteria.

However, differences of goodness-of-fit between two groups also show that RTV affect

their consumer socialisation somewhat differently. Further, each progressive test (factor

structure equivalence, factor loading equivalence) supports cross validation of the

results (Table 7.10). The loose validation test produces adequate fit statistics. The first

multiple sample test (Factor Structure Equivalence) also provides adequate fit.

Furthermore, each progressive test produces insignificant change in fit as well as

change in Chi-square over the subsequent test.

Table 7.9 Goodness-of-Fits Estimates of Middle and Lower Social Class

Fit Measures Middle Social Class Lower Social Class

Chi-Square 996.690 991.583 Degree of Freedom 587 587 P .000 .000 Normed Chi-square (CMIN/df) 1.698 1.689

Root mean square of error of estimation

(RMSEA)

.062 .061

Goodness of fit index (GFI) .776 .774

Adjusted of goodness-of-fit index (AGFI) .746 .744

Normed fit index (NFI) .724 .576

Tucker-Lewis Index (TLI) .852 .746

Comparative fir index (CFI) .862 .763

221

Table 7.10 Cross Validation Statistics - Social Class

Model

Description

Df X2 CFI RMSEA Comments

Loose cross

validation

587 991.583 .763 .061 Reasonably

good fit

Factor structure

equivalence

1174 1988.273 .826 .044 Reasonably good fit

Factor loading

equivalence

1204 2024.125 .825 .043 Change in

X2 (30) = 35.8

Insignificant

However, differences of goodness-of-fit between two groups also show that RTV affect

their consumer socialisation marginally differently. Table 7.11 shows standard

estimation differences between male and female teenagers’ regarding consumer

socialisation by RTV.

Table 7.11 Multi-Group Analysis of Social Class Differences on Consumer

Socialisation Process

Standard Estimate

Middle Social Class Lower Social Class

H1 Cognition <--- Involvement .740 .697

H2 Attitude <--- Involvement .274 .336

H3 Attitude <--- Cognition .251 .170

H4 Values <--- Involvement .554 .563

H5 Attitude <--- Values .410 .363

H6 Involvement <--- Parental control .231 .230

H7 Involvement <--- Peers group influences .715 .578

222

Current literatures show that teenagers from different social classes get involved with

media differently. Further, they have different tastes and preferences towards different

types of TV programs that ultimately affect the outcomes of consumer socialisation.

Findings of this study also partially confirm these issues from the perspective of RTV

involvement. Table 7.11 shows that teenagers’ consumption related cognition is highly

influenced by RTV involvement. However, due to higher access to TV, teenagers from

middle social class (.740) process product related knowledge and instruction better than

their lower social class counterparts (.697). Whereas, standard estimates of consumption

related attitude are stronger with lower social class teenagers (.336) than their middle

social class teenagers (.274). Relationship between consumption related cognition and

attitude of middle social class teenagers are more apparent (.251) than lower social class

teenagers (.170). Teenagers from middle social class (.554) and lower social class (.563)

show minor differences as both the groups have similar beliefs towards materialism,

desire towards products and prohibited product consumption. However, the effect of

consumption related values on attitude are stronger for middle social class (.410) than

the lower social class (.363).

Parents in both middle (.231) and lower social classes (.230) moderately control their

teenaged children RTV involvements. It has been found at qualitative stage of this

research and confirmed at quantitative stage that due to teenagers’ parental business and

their co-viewing habits, they often provide moderate freedom to their teenaged kids to

watch different RTV shows. On the other hand, teenagers from both the social classes

are highly influenced by their peer groups regarding their involvement with RTV.

Teenagers from middle social class are comparatively more influenced by their peers

(.715) than their lower social class counterparts (.578) as they have more chances to

meet their peers and friends.

Overall, the structural model fits for both the social classes though there are some

differences in the standard estimate for some of the measurement items between the

social classes. Accordingly, the hypothesis is only partially supported.

223

In chapter 3 (Figure 3.2) a model of consumer socialisation by RTV (CSRTV) was

proposed on the basis of current literature. The variables that are considered in this

research were refined by qualitative analysis (Table 5.1) and tested by quantitative

analysis shown the following model of consumer socialisation of RTV (CSRTV). The

following empirically tested model supports hypotheses 1 to 7 that were developed on

the basis of current literature of consumer socialisation. However, hypotheses 8 and 9

were not fully supported, particularly no differences were found in the structural model

regarding between the two gender and also the two social classes. Accordingly, Figure

7.3 shows that empirically tested main model.

Figure 7.3 The Empirically Tested Model of CSRTV (Main Model)

(H1) yes

(H6) Yes (H3) Yes

(H2) yes

(H4) Yes (H5) Yes

(H7) yes

7.2 Conclusion

This chapter reported the results of the quantitative data analysis for this research.

Particularly, this chapter addressed the test result of structural model and hypotheses.

Structural equation modelling using AMOS was used to test the proposed as well as the

main model. The results indicated that the main model which excluded gender and

social class differences as social structural variables from the proposed model and a few

measurement variables on the basis of CFA results generate acceptable fit. The

summary of the research findings are shown in Table 7.12. The implication and

significance of these results will be discussed in the final chapter.

Peer group influences

Parental control

RTV Involvement

Consumption related cognition

Consumption related attitude

Consumption related values

224

Table 7.12 Summary of the Research Findings

Research Objectives Research Hypotheses Results

1) To develop and test a model

of consumer socialisation of

teenagers of a developing

country by RTV.

Developed SEM by AMOS Supported

H1: There is a positive impact of

RTV involvement on consumption

related cognition.

Supported

H2: There is a positive impact of

RTV involvement on consumption

related attitude.

Supported

H3: There is a positive impact of

consumption related cognition on

consumption related attitude.

Supported

H4: There is a positive impact of

RTV involvement on consumption

related values.

Supported

2) To determine how RTV affect

in the consumer socialisation of

teenagers in a developing

country.

H5: There is a positive impact of

consumption related values on

consumption related attitude.

Supported

H6) There is a positive impact of

peer groups influences on teenagers’

RTV involvement.

Supported

3) To determine how teenagers’

social class differences, their

gender differences, peer group

and parental control influence the

role of RTV in their consumer

socialisation.

H7) Parents through communication,

mediation and instruction positively

influence teenagers’ RTV

involvement.

Supported

225

H8) There is a difference between

male and female teenagers’

involvement with RTV and its effect

on their consumer socialisation.

Partially Supported

H9) There is a difference between

middle and lower social class

teenagers’ involvement with RTV

and its effect on their consumer

socialisation.

Partially Supported

226

Chapter 8 Conclusions

8.1 Overview

This research addressed the research problem: How does teenagers’ involvement with

reality television (RTV) affect their consumer socialisation? This final chapter

reports the outcomes generated in response to this research problem analyses the

implication of their outcomes and draws conclusions.

Chapter 1 covered the background of the research issue, setting the prospect for this

research. The research problem was defined and research issues were outlined. The

significance of this research was discussed in terms of the potential for and the gap of

prior research about the topic.

In turn, Chapter 2 and 3 reviewed the existing literature relating to the research problem

and identified gaps in theory. Chapter 2 began with an overall discussion on consumer

behaviour and consumer socialisation. The two parent theories of socialisation,

cultivation theory and vicarious learning theory were reviewed that found a lack of

empirical examination of the affect RTV has on consumer socialisation in a developing

country. Based on this extant literature, the theoretical foundations for this research

were established in chapter 3. Further, a model of CSRTV was proposed and, three

‘research objectives’ followed by hypotheseses were derived in this chapter (Table 8.1).

Table 8.1 List of Research Objectives for this Research

Research Objectives

1) To develop and test a model of consumer socialisation of teenagers of a developing country

by RTV.

2) To determine how RTV plays a role in the consumer socialisation of teenagers in a

developing country.

3) To determine how teenagers’ social class differences, their gender differences, parental

control and peer group influences influence the role of RTV in their consumer socialisation.

227

Next, Chapter 4 justified the methodology followed for this research. Both qualitative

and quantitative methods have been used in this research. In chapter 5, qualitative

findings were reported. Two FGIs were conducted to refine the current measurement

scale items of consumer socialisation and make them relevant for a developing country.

Then, chapter 6 and 7 analysed the two stage quantitative data: preliminary analysis, the

structural equation modelling for the measurement and structural models in first stage,

and testing of hypotheses in the main model in the second stage respectively.

In the previous chapter 7, findings of the hypotheses and their implications were

discussed relating to the empirical findings. This chapter reports on the contributions of

this research both in academic, policy and practice terms. Furthermore, it identifies the

limitations of the research and implications for the further investigations.

8.2 Contribution of this Research

According to the extant literature, consumer socialisation have three major components;

consumer socialisation agents, social structural variables and consumer socialisation

outcomes (John 1999). Media, particularly TV is considered as one of the influential

agents in the consumer socialisation (Bhosale & Gupta 2005; Orlik 2000). The process

of consumer socisalisation is multi-faceted and difficult to measure for the following

reasons: Firstly, outcome of consumer socialisation must be measured with same

sample group at a different time period. Secondly, the outcome of consumer

socialisation can not isolate the role of any particular socialisation agent.

However, this research comes along with the contribution by plugging gaps of the

above mentioned issues. Firstly, this research critically adopted and modified variables

through qualitative and quantitative analysis from the relevant discipline to measure

consumer socialisation by RTV in a developing country like Bangladesh.

228

Secondly, academics of consumer socialisation have identified the role of TV in the

consumer socialisation. However, most of this research does not indicate any structured

model of consumer socialisation by TV that has been empirically tested. Furthermore,

there is very little contribution in the extant literature about the role of RTV in the

consumer socialisation. This research work represents one of the valued empirical

investigations of consumer socialisation by RTV.

Thirdly, this research addressed the influences of various social structural variables such

as parental control, peer group influences, gender differences and social class on

consumer socialisation outcomes which was largely overlooked so far from developing

countries perspectives. Further, addressing the significance of these social structural

variables on consumer socialisation by RTV and empirically testing their influences is

not only a contribution from developing countries perspectives but also in general.

8.2.1 Implication for Theory

In one sense there is nothing much new in the findings of this research. This research

involves refurbishing many of the variables which are described as commonsensical,

and have been previously established in other areas of consumer socialisation research.

However, the implications of such findings for the CSRTV have thus far been

overlooked by researchers. Critically, this research fills this gap. This research

integrates and applies existing knowledge in learning theories, consumer socialisation

theories and involvement theories to this research, as well as empirically testing the

research hypotheses. Consumer behaviourists are increasingly emphasising the need for

an integrated approach to Marketing, particularly, consumer behaviour. It is timely that

consumer behaviourists take such an integrated approach towards research in these

fields.

As a part of this integrated approach, this research establishes the importance of social

structural variables with RTV involvement in the consumer socialisation by RTV

(CSRTV) which has been overlooked by previous researchers. Moreover, it is observed

that no model has been found to consider such factors, though variables of consumer

229

socialisation by TV have been extensively researched as part of consumer socialisation

theories. Not only does this research institute the importance of such factors but also

addresses the outcome results. Furthermore, consumption related cognition; attitudes

and values are considered as outcomes of the consumer socialisation in the extant

literature but have not been empirically tested regarding RTV involvement and its

consumer socialisation outcomes. This research contributes to filling this gap by

showing interrelationships of socialisation outcomes.

One major criticism of previous consumer socialisation research is the lack of focus on

audiences’ involvement with the media. Empirically tested models in this research take

into account the involvement of teenagers with RTV which expedites teenagers’

consumer socialisation. Furthermore, the extant literature recommended that the media

plays a significant role in the consumer socialisation of teenagers. This research has

established the role of RTV in teenagers’ consumer socialisation.

This research has identified and tested the specific variables and constructs, and has

developed multi-item measurement scales for those constructs. It has already

been mentioned that variables and constructs have been adopted from extant literature.

These constructs, and their measurement items, can now form the basis for further

research in the area, both in Bangladesh and relevant developing countries. Compared

to consumer socialisation theories, CSRTV is a completely unexplored research area,

not only in developing countries but also in developed countries. This research is a

major study in an important, but overlooked, area of business importance.

Furthermore, social structural variables are one of the crucial parts of extant consumer

socialisation research. However, not many studies highlighted the integration of the

number of social structural variables, such as social class differences, gender

differences, parental control and peers’ influences on the consumer socialisation. The

shortcomings have been explored through an investigation of theory, and the findings of

this research. The empirically tested, new model developed by this research may now

be presented as published work, or in future research reference as representative

of reality in the consumer socialisation.

230

Another major contribution of this research is the successful integration of qualitative,

as well as quantitative methodologies of research. The use of SEM techniques for

quantitative testing of the proposed model is also unique in this area of research.

Multiple regressions, factor analysis, and multi-variate analysis of variance analysis

provide the researcher with powerful tools for addressing a wide range of managerial

and theoretical questions. But, they all share one common limitation: each technique

can examine only a single relationship at a time. SEM, as an extension of several multi-

variate techniques, resolves this issue.

8.2.2 Implication for Methodology

This research has one methodological implication. In this research FGIs were conducted

at the stage of qualitative data collection. In this research two different FGIs were

conducted with two different social classes which were different by social class

measurement criteria, but homogeneous within the group. In this research, FGIs were

focused to discuss the issues of teenagers’ involvement with RTV and to understand the

effect of consumer socialisation. Additionally, participants’ insights and opinions led to

the confirmation and refinement of the initial framework developed in Chapter 3. Thus,

the conceptual framework upon which the primary data collection stage is based was

not just from sparse literature, but indeed captured insights from practitioners (Rao &

Perry 2004). This practical orientation strengthened the foundations for this research.

8.2.3 Implications for the Policy and Practice

The previous section discussed the implications of this research for theory. This section

in turn, discusses implications for policy and practice.

Understanding consumers’ taste and preferences is the key issue for any marketers.

Most of the international and local businesses give maximum priority and effort to

understanding customers’ consumption behaviour through consumption related

cognition, attitudes and values. Accordingly, the findings of this research and the

231

resulting model will be of interest to brand marketers and marketing communication

planners in Bangladesh and other developing countries. This model will be practically

significant to both experienced and aspiring brand marketers and marketing

communication planners. One major feature of this model is its flexibility. Marketers

can either consider any particular variable, or they can consider this process by

integrating all the specified variables. Furthermore, marketers can address the issues of

consumer socialisation from the angle of gender and social class differences. For

example: this research indicates that teenagers in middle social class are more involved

with RTV and it also varies due to their gender differences. Therefore, brand marketer

can get ideas about how to develop their positioning strategy among the teenagers from

different gender and social class. Further, market communication planners can also

consider teenagers way of involvement with RTV that might be helpful for reaching

their target customers. Media strategist and sponsors also can get a clear indication of

what are the different ways that teenagers get involved with RTV, which might be

fruitful for their marketing strategies. Finally, the model of this research will be useful

for the research and development wing of an organisation to assist with new products

that may support the brand marketers.

The CSRTV model developed and tested as a part of this thesis, will encourage

practitioners in how they structure any specific decisions. In addition to promoting

changes to structuring the process, this new decision model includes flexibility in the

decision process. In considering this new model, market practitioners will be able to add

specific insights, qualified judgments, and intuition towards their target markets that are

particularly crucial for media strategists and sponsors.

This research will also be a good point of reference for the academics and researchers

who are keen in using specialised software for a structural equation model. AMOS has

been used to develop and test the proposed model in this research. One of the major

implications of usage AMOS has is that it allows the flexibility of including and

excluding variables to get desired results. Therefore, researchers and academics can

understand the relative weights of variables that are considered in structural equation

modelling.

232

Overall, this research has both academic and practical implications. For practitioners, its

findings are directly appropriate to their business activity. For academics and

researchers, it provides empirical evidence and opens up potential new areas of further

research.

8.3 Limitations

This research has been conducted in Bangladesh only. There are so many other

developing countries in the world. As only one of the developing countries, Bangladesh

does not represent all the economic and cultural indicators of all the developing

countries. Accordingly, the findings of this research may not be treated as completely

applicable to the rest of the developing countries, and needs to be tested further from the

perspective of individual countries.

Another shortcoming of this research is that it only develops the model of CSRTV,

representing the teenagers from middle and lower social class backgrounds in Dhaka

city. In Bangladesh more than 40 percent of the population lives at the bottom of a

pyramidal class system, and a significant number of people live in the middle class,

followed by a small number of upper social class people. However, this research does

not represent teenagers from upper social class.

Further, there was higher representation (75%) of 16-19 years old teenagers compared

to 13-15 years (25%) age group in the data set. This may have incurred a bias in results

because of potential differences in the cognitive ability and RTV involvement of

different age groups.

Furthermore, this research only studied consumer socialisation of teenagers from Dhaka

city (the capital of Bangladesh). There are six metropolitan cities in Bangladesh where

the population size of teenagers is quite large. Teenagers in other cities are also

involved with RTV. However, this research does not consider those populations as well.

233

Finally, this research also did not consider a large number of non-resident Bangladeshi

teenagers living overseas. According to the Bangladesh Bureau of Statistics, nearly 20

million non-Bangladeshi residents are living overseas and a large number of them are

teenaged. This research did not examine the affect of acculturation and consumer

socialisation of these expatriate teenagers.

From the methodological point of view this research suffers from a limitation. In the

current literature some researches suggest that understanding consumer socialisation

outcomes through TV involvement may be tested by experimental design (e.g, pretest-

posttest control group design) (Bushman 2005) or longitudinal study (Moschis & Moore

1982). However, due to time and resource constraints this research did not follow those

methods.

8.4 Implications for Further Research

This research has the potential to open up a new area of empirical research. This

research has been conducted among teenagers who have involvement with RTV.

Similar research may be conducted among the other options of the TV vehicle (e.g.,

Drama, musical shows) and electronic media (e.g., radio, film) which also have a

significant role in the consumer socialisation.

Particularly, findings from this research are relevant only for the TV vehicle, RTV.

RTV being a most contemporary vehicle in the electronic media area, such findings

may or may not be representative of other TV vehicles. For the sake of generalisation,

similar research needs to be conducted among other TV vehicles to get the notion of

consumer socialisation. Generally, RTV is a specific form of electronic media and a TV

vehicle. This research has developed factors or constructs that contribute to CSRTV and

their measurement scales. Further research is required to re-test such scales for possible

refinement and future usage for other media studies and confirm its applicability in the

consumer socialisation discipline.

234

Furthermore, findings of this research have identified the factors that are considered in

CSRTV. Also, it has identified the observed variables that measure those factors.

However, this research did not measure the relative weight and importance of each of

those factors and variables. In reality, although each of those factors and associated

variables are included in CSRTV, it may well be the case that different consumer

sociologists and market practitioners put different weighting on the factors and

variables. While it was not an objective of this research to find out such relative

importance of variables, if any, further research may be conducted to establish whether

individual variables will need to assign specific numerical weights to each of the factors

to indicate their relative importance in the CSRTV.

It was mentioned earlier that most of the research in consumer socialisation by media

was carried out in the developed countries. Further, RTV is also an invention in

electronic media in developed countries. The findings of this research may contribute in

the development of a new model, and guide data definition for future research in

developed countries. Given the rapid change in electronic media and socialisation

research, it is surprising that there has been little research on the empirically tested

model of consumer socialisation through TV. Currently, there is a significant gap in the

area of consumer socialisaion through TV, particularly RTV, even in the developed

countries. Further research needs to be conducted to assess such implications.

235

8.5 Specific Conclusions from the Findings of This

Research Regarding the Research Objectives and

Hypotheses

If considered against the specific research hypotheses in chapter 3, the findings of this

research lead to the following conclusions and empirically tested model:

1) There is a positive impact of RTV involvement on consumption related

cognition.

2) There is a positive impact of RTV involvement on consumption related attitude.

3) There is a positive impact of consumption related cognition on consumption

related attitude.

4) There is a positive impact of RTV involvement on consumption related values.

5) There is a positive impact of consumption related values on consumption related

attitude.

6) There is a positive impact of peer groups influences on teenagers’ RTV

involvement.

7) Parents through communication, mediation and instruction positively influence

teenagers’ RTV involvement.

8) There is a difference between male and female teenagers’ involvement with

RTV and its effect on their consumer socialisation.

9) There is a difference between middle and lower social class teenagers’

involvement with RTV and its effect on their consumer socialisation.

236

8.6 Conclusion In this chapter, contributions of this research has been addressed, implications for the

theory, policy and practice were then offered, followed by limitations of this research.

At last, implications for further research were suggested.

In brief, this research highlights a structure of understanding consumer socialisation of

teenagers by RTV and how different social structural variables are also involved in this

process in a developing country. Further, the empirically tested model is a contribution

because it is the first extensively and rigorously researched steps towards understanding

the role of RTV as consumer socialisation agents of teenagers in a developing country.

Finally the model was built from theory and empirical research to present this process

and provides a foundation for the further research.

237

References

Aakar, DA & Day, GS 1990, Marketing Research. John, Wiley & Sons Inc., Canada.

Abelman, R 2004, "TV Literacy and Academic/Artistic Giftedness: Understanding

Time Leaps and Time Lags", Roeper Review, vol.26, no.2, p. 85-89, viewed 27/12/07,

EBSCO Host Research Database. Psychology and Behavioural Science Collection.

About.com 2010, Number of Countries in World.

Acevedo-Polakovich, ID, Lorch, EP, Milich, R, Ashby, RD 2005, "Disentangling the

Relation Between Television Viewing and Cognitive Processes in Children With

Attention-Deficit/Hyperactivity Disorder and Comparison Children" Archives of

Pediatrics and Adolescent medicine vol. 160, no. 4, p. 354-360.

Aasker, DA, Kumar, V, Day, GS 2001. Marketing Research. New York, John Willey &

Sons.

Adams, C, Sutherland, C & Johnson, N 2007, "Teens Find Daddy Cool", Herald Sun,

12 July.

Alarcon, C 2004, "Is TV fattening kids?", B&T Weekly, Viewed 3rd March.

Anderson, DR, Huston, AC, Schmitt, KL, Linebarger, DL, Wright, JC & Wright, JC

2001, "Academic Achievement", Monographs of the Society for Research in Child

Development, vol.66, no.1, p. 36-66, viewed 04/01/08, EBSCO Host Research

Database. Psychology and Behavioural Science Collection.

Anderson, JC & Gerbing, DW 1988, "Structural Equation Modelling In Practice: A

Review and Recommended Two-Step Approach." Psychology Bulletin, vol. 103, no.3.

238

Armstrong, J 2006, "Laguna Beach: The Next Generation", Entertainment Weekly, 25

August.

Arnett, JJ 1995, "Adolescents' USAGE of the Media for Self-Socialisation", Journal of

Youth and Adolescence, vol.24, no.5, p. 519-533, viewed 16/12/07, EBSCO Host

Research Database. PsycINFO.

Attewell, P, Suazo-Garcia, B & Battle, J 2003, "Computers and Young Children: Social

Benefit or Social Problem?", Social Forces, vol.82, no.1, p. 277-296, viewed 27/12/07,

EBSCO Host Research Database. Psychology and Behavioural Science Collection.

Austin, CG, Plouffe, CR & Peters, C 2005, "Anti-Commercial Consumer Rebellion:

Conceptualisation and Measurement", Journal of Targeting, Measurement & Analysis

for Marketing, vol.14, no.1, viewed 04/01/08, p. 62-78 EBSCO Host Research

Database. Business Source Premier.

Avrahami, A 1995, "Orientation and Behaviour of Youth in the Kibbutz: Roots in

Adolescent Socialisation", Journal of Moral Education, vol.24, no.3, p. 307-327.

Bagozzi, RP 1994, "Structural Equation Models in Marketing Research: Basic

Principles", Bagozzi, RP (eds), Principles of Marketing Research, Blackwell Publisher,

Massachusetts.

Baker, MJ 1991, Research for Marketing, Macmillan Education Ltd., Hong Kong.

Bakir, A, Rose, GM & Shoham, A 2005, "Consumer Communication and Parental

Control of Children's Television Viewing", Journal of Marketing Theory and Practice,

vol.13, no.2, p. 47-58, viewed 24/01/08, ProQuest. ABI/INFORM Global.

Bandura, A & Walters, RH 1970, Social Learning and Personality Development

Redwood Press Limited, Great Britain.

239

Bandura, A 1986, Social Foundations of Thought and Action: A Social Cognitive View.

New Jersy, Prentice Hill.

Bandura, A 1977, Social Learning Theory. United States of America, Prentice-Hall, Inc.

Banks, S & Gupta, R 1980, "Television as a Dependent Variable, for a Change", Journal

of Consumer Research, vol.7, no.3, p. 327-330, viewed 04/01/08, EBSCO Host

Research Database. ProQuest Social Science Journal.

Barr, R & Hayne, H 1999, "Developmental Changes in Imitation from Television

during Infancy", Child Development, vol. 70, no. 5.

Bareham, JR 2004, "Can Consumers be Predicted or are they Unmanageable?",

International Journal of Contemporary Hospitality Management, vol.16, no.3, p. 159-

165, viewed 12/12/07, Emerald Group Publishing Limited. Emerald Fulltext.

Barkin, S 2006, "Parental Media Mediation Styles for Children Aged 2 to 11 Years",

Archives of Paediatrics and Adolescent Medicine, vol.160, no.4, p. 395-401, viewed

07/01/08, ProQuest Social Science Journal.

Bartlett, E, Griffiths, G & Badian, V 2008, Digital advertising in 2142: measuring the

effectiveness of advertising to the PlayStation® generation. Market Research Society.

Baumgartner, H & Homburg, C 1996, "Applications of Structural Equation Modelling

in Marketing and Consumer Research: A review." International Journal of Research in

Marketing, vol. 13, no. 2.

Bearison, DJ, Bain, JM & Daniele, R 1982, "Developmental Changes in How Children

Understand Television", Social Behaviour & Personality: An International Journal,

240

vol.10, no.2, p. 133-144, viewed 04/01/08, EBSCO Host Research Database.

Psychology and Behavioural Sciences Collection.

Berns, RM 1997, Child, Family, School, Community-Socialisation and Support. United

States of America, Harcourt Brace college publisher.

Belk, SR 1991, "The Influence of Information Source on Brand Loyalty and Consumer

Sex Roles of the Elderly", Advances in Consumer Research, vol.18, no.1, p. 673-680,

viewed 20/01/08, EBSCO Host Research Databases. Business Source Premier.

Bell, ML & Emory, CW 1971, "The Faltering Marketing Concept", Journal of

Marketing, vol. 35, no.4, p. 37-42, viewed 08/12/07, EBSCO Host Research Database.

Business Source Premier.

Bellafante, G 1995, "Their so-called lives", Time, 31 July.

Bellafante, G 2008, 'Gladiators of the Dance In the Arena of Reality TV', New York

Times, 3 April.

Bennett, D 1995, Growing Pains, Double Day, Australia.

Bentler, P.M. 1980, Multivariate Analysis with Latent Variables: Causal Modelling,

Annual Review of Psychology, vol. 31, p. 419-456.

Bentler, PM & Bonett, DG 1980, "Significance Tests and Goodness of Fit in the

Analysis of Covariance Structures", Psychological Bulletin, p. 588-606.

Bentler, PM & Wu, EJC 1993, EQS/Windows Users' guide, Los Angeles, BDMP,

Statistical Software.

Berman, M 2008, "Point of View", Media-week, 10 March.

241

Berns, RM 1997, "Child, Family, School, Community-Socialisation and Support",

Harcourt Brace college publisher United States of America.

Bhosale, S & Gupta, S 2006, "Tomorrow is a New Consumer! Who is the Future Asian

Consumer?", ESOMAR Asia-Pacific Conference, Mumbai, viewed 16/05/07, European

Association of Communications Agencies, WARC.

Biressi, A 2005, Reality TV: Realism and Revelation. London, Wallflowers Press.

Blackwell, R, D'Souza, C, Taghian, M, Miniard, P & Engel, J 2007, Consumer

Behaviour-An Asia Pacific Approach, Thomson, Australia.

Bloxham, M & Prieb, M 2007, "Ad avoidance and the TV audience", Admap Magazine.

Blumer, JA 1979, "The Role of Theory in Sses and Gratification Studies."

Communication Research, vol. 6, p. 9-36.

Bollen, KA 1989, Structural Equation with Latent Variables, Wiley, New York.

Bollen, K & Lennox, R 1991, "Conventional Wisdom on Measurement: A Structural

Equation Perspective." Psychological Bulletin, vol. 110, no. 2, p. 305-314.

Bown, I 2008, 'Sisters Take act on Stage in Quest for $1m salon', Herald, 29 Jan.

Bradley, RH & Corwyn, RF 2002, "Socioeconomic Status and Child Development",

Annual Review of Psychology, vol.53, no.1, p. 371-399, viewed 05/03/09, EBSCO

Host Research Databse. Psyc INFO.

242

Brand, JE & Greenberg, BS 1994, "Commercials in the classroom: The impact of

Channel One advertising", Journal of Advertising Research, vol.34, no.1, p. 18-27,

viewed 05/07/07, EBSCO HOST Research Database PsycINFO.

Brewer, J & Hunter, H 1989, Multi-Method Research: A Synthesis of Styles, SAGE,

Newbury Park.

Brim, OG & Wheeler, S 1966, Socialisation after Childhood: Two Essay, John Wiley &

Sons, Inc., USA.

Bristol, T & Mangleburg, TF 2005a, "Expanding the Boundaries of Consumer

Socialisation Research", Advances in Consumer Research, vol.32, no.1, p. 78-95.

Bristol, T & Mangleburg, TF 2005b, 'Not Telling the Whole Story: Teen Deception in

Purchasing', Academy of Marketing Science, vol.33, no.1, p. 79-95, viewed 12/01/08,

ProQuest ABI/INFORM Global.

Brody, GH, McBride Murry, V, Kim, S & Brown, AC 2002, 'Longitudinal Pathways to

Competence and Psychological Adjustment among African American Children Living

in Rural Single-Parent Households', Child Development, vol.73, no.5, p. 1505-1516.

Brown, H 1976, Socialisation: The Social Learning Theory Approach, The Open

University Press, Great Britain.

Brown, JD 1993, "Media Sex and the Adolescent ", Theoretical Overview, (ed) B.S

Gernberg, BS, Brown JD, Buerkel-Rothfuss, NL, New Jersey, Hampton Press.

Brown, JD & Steele, JR 1995, Sex and the Mass Media. California, Henry J. Kaiser

Family Foundation.

243

Bukowski, WM, Brendgen, M & Vitaro, F 2007, "Socialisation Outside the Family",

Bukowski, WM, Brendgen, M & Vitaro, F (eds), Handbook of Socialisation-Theory and

Research, The Guilford Press, New York.

Bulck, JVd & Bergh, BVd 2000, "The Influence of Perceived Parental Guidance

Patterns on Children's Media Use: Gender Differences and Media Displacement",

Journal of Broadcasting & Electronic Media, vol.44, no.3, p. 329-349, viewed

050/02/08, ProQuest. ABI/INFORM Global.

Burke, MG 2005, "Reasons to Turn off the Screen", Encounter, vol.18, no.1, p. 49-53,

viewed 27/12/07, EBSCO Host Research Database. Psychology and Behavioural

Science Collection.

Burrell, G & Morgan, G 1979, Sociological Paradigms and Organisational Analysis,

UK.

Bush, AJ, Smith, R & Martin, C 1999, "The Influence of Consumer Socialisation

Variables on Attitude Toward Advertising: A Comparison of African-Americans and

Caucasians", Journal of Advertising, vol.28, no.3, p. 13-44, viewed 16/01/08, ProQuest.

ABI/INFORM Global.

Bush, VD, Martin, CA & Bush AJ 2004. "Sports Celebrity Influence on the Behavioral

Intentions of Generation Y.", Journal of Advertising Research, vol. 44, no. 1, p. 108-

118, viewed 27/09/08, WARC.

Bushman, BJ 2005, "Research Article; Violence and Sex in Television Programs Do

Not Sell Products in Advertisements", Psychological Science, vol.16, no.9, p. 702-708,

viewed 24/01/08, ProQuest. ABI/INFORM Global.

244

Butsch, R 2001, "Class and Audience Effects: A History of Research on Movies, Radio,

and Television", Journal of Popular Film & Television, vol.29, no.3, p. 112-120, viewed

10/02/08, ProQuest. ProQuest Social Science Journals.

Butterbaugh, L 1999, 'Fiji put TV on a Diet', Off Our Backs-a Women's News Journal,

vol.29, no.6, p. 5-7, viewed 04/01/08, ProQuest. Social Science Journal.

Byoungkwan, L, Salmon, CT & Hye-Jin, P 2007, "The Effects of Information Sources

on Consumer Reactions to Direct-to-Consumers (DTC)", Journal of Advertising, vol.36,

no.1, p. 107-119, viewed 20/01/08, EBSCO Host Research Database. Business Source

Premier.

Byrne, BM 2001, Structural Equation Modelling with AMOS? Windows: Basic

Concepts, Applications and Programming SAGE Publications.

Cajueiro, M 2008, "Endemol Changes Partners", Variety, 4 February.

Cardoza, K. 2002. "Parental Control Over Children's Television Viewing in India"

Contemporary South Asia, vol. 11, no, 2, p. 135-161, viewed 18/07/08, ProQuest.

Carlson, L, Grossbart, S & Stuenkel, JK 1992, "The Role of Parental Socialisation

Types on Differential Family Communication Patterns Regarding Consumption",

Journal of Consumer Psychology, vol.1, no.1, p. 31-52, viewed 20/12/07, EBSCO Host

Research Database. Business Source Premier.

Carlson, L, Laczniak, RN & Walsh, A 2001, "Socialising Children about Television: An

Intergenerational Study", Academy of Marketing Science, vol.29, no.3, p. 276-288,

viewed 16/01/08, ProQuest. ABI/INFORM Global.

Carmines, E & McIver, J 1981, Analysing models with unobserved variables: Analysis

of covariance structure. Social Measurement. Beverly Hills, SAGE Publications

245

Carmines, E & Zeller, J 1979, Reliability and Validity Assessment. Beverly Hills,

SAGE Publications.

Caruana, A & Vassallo, R 2003, "Children's Perception of their Influence Over

Purchases: The Role of Parental Communication Patterns", Journal of Consumer

Marketing, vol.20, no.1, p. 55-66, viewed 24/07/07, EBSCO Host Research Databases.

Business Source Premier.

Chan, K 2003, "Materialism among Chinese Children in Hong Kong", Young

Consumers, vol.4, no.4, p. 1-10, viewed 06/01/08, WARC.

Chan, K 2006, "Consumer Socialisation of Chinese Children in Schools: Analysis of

Consumption Values in Textbooks", The Journal of Consumer Marketing, vol.23, no.3,

p. 125-132, viewed 21/01/08, ProQuest. ProQuest Social Sciences Journal.

Chan, K & McNeal, JU 2003, "Parent-Child Communications about Consumption and

Advertising in China", The Journal of Consumer Marketing, vol.20, no.4, p. 125-132,

viewed 16/01/08, ProQuest. ABI/INFORM Global.

Chan, K & McNeal, JU 2006a, "Children and Media in China: An Urban and Rural

Comparative study", The Journal of Consumer Marketing, vol.23, no.2, p. 79-90,

viewed 17 June 2007, ProQuest. ABI/INFORM Global.

Chan, K & McNeal, JU 2006b, "Chinese Children's Understanding of Commercial

Communications: A comparison of Cognitive Development and Social Learning

Models", Journal of Economic Psychology, vol.27, no.1, no. 36-56, viewed 07/01/08,

Science Direct. ProQuest Social Science Journal.

246

Chan, K & McNeal, JU 2006C, 'Rural Chinese Children as Consumers: Consumption

Experience and Information sourcing', Journal of Consumer Behaviour, vol.5, no.3, p.

182-192, viewed 04/01/08, EBSCO Host Research Databases. Psychology and

Behavioural Sciences Collection.

Chan, K & Prendergast, G 2007, “Materialism and Social Comparison among

Adolescents", Social Behaviour & Personality: An International Journal, vol.35, no.2, p.

213-228, viewed 04/01/08, EBSCO Host Research Database. Psychology and

Behavioural Sciences Collection.

Chang, C. 2006, "Cultural Masculinity/Femininity Influences on Advertising Appeals."

Journal of Advertising Research, vol. 46, no. 3, p. 1-8, viwed 05/02/08, WARC.

Chan, K, Zhang, H & Wang, I 2006, "Materialism among Adolescents in Urban China",

Young Consumers, vol.7, no.2, p. 1-11, viewed 06/01/08, WARC.

Chavda, H, Haley, M & Dunn, C 2005, "Adolescents' Influence on Family Decision-

Making", Young Consumers, vol.6 no.3, p. 68-78, viewed 11/05/07, WARC. Business

Source Premier.

Chernin, M 2005, "The Relationship Between Children’s Television Viewing and

Academic Performance", Archives of Paediatrics and Adolescent Medicine, vol.159,

no.7, p. 686-689, viewed 07/01/08, ProQuest. ProQuest Social Science Journal.

Choi, SM & Ferle, CL 2004, "Convergence Across American And Korean Young

Adults: Socialisation Variables Indicate The Verdict Is Still Out", International Journal

of Advertising, vol.23, no.4, p. 479-506, viewed 16/05/07, WARC.

Churchill Jr, GA 1979, "A Paradigm for Developing Better Measures of Marketing

Construct", Journal of Marketing Research, vol.16, no. February, p.64-75.

247

Churchill Jr, GA & Moschis, GP 1979, "Television and Interpersonal Influences on

Adolescent Consumer Learning", Journal of Consumer Research, vol.6, no.1, p.23-37,

viewed 20/12/07, EBSCO Host Research Database. Business Source Premier.

Clark, L & Tiggemann, M 2006, "Appearance Culture in Nine- to 12-Year-Old Girls:

Media and Peer Influences on Body Dissatisfaction", Social Development, vol.15, no.4,

p. 628-643, viewed 05/02/08, EBSCO Host Research Database. Psychology and

Behavioural Science Collection.

Clark, S, Nabi, R & Moyer-Gusé 2007, Television Consumption and Young Women's

Expectations of Sexual Timing, Media Report to Women.

Clausen, JA 1968, Perspective on Childhood Socialisation Clausen, JA (eds),

Socialisation and Society, Little, Brown & Company, Canada.

Cogdon, K 2007, "Families say TV Educates Babies", Herald Sun, 9 May.

Colwell, J 1990, "Qualitative Market Research: A Conceptual Analysis and Review of

Practitioners Criteria", Journal of the Market Research Society, vol.32, no. October, p.

13.

Conger, RD & Dogan, SJ 2007, Social Class and Socialisation in Families, Conger, RD

& Dogan, SJ (eds), Handbook of Socialisation: Theory and research., Guilford Press,

New York, NY, US.

Conway, JM & Huffcutt, AL 2003, "A Review and Evaluation of Exploratory Factor

Analysis Practices in Organisational Research." Organisational Research Methods, vol.

6, no. 2, p.147.

Cram, F & Ng, SH 1999, "Consumer Socialisation", Applied Psychology: An

International Review, vol.48, no.3.

248

Dale, D 2007, "Stay in Touch", Sydney Morning Herald., 4 June.

Dammler, A 2005, "Life Frames: A New Approach to Researching Young Target

Groups", Age Matters Conference, London.

Denzin, NK & Lincoln, YS 1994, Entering the Field of Qualitative Research, Denzin,

NK & Lincoln, YS (eds), Hand book of Qualitative Research, SAGE, California.

Desvousges, WH & Frey, JH 1989, "Integrating Focus Groups and Survey: Examples

from Environmental Risk Surveys." Journal of Official Statistics, vol. 5, p. 349.

Deziel, S 2004, "Family Jewels. Cover story", Maclean's, vol.117, no.38, viewed

04/04/08, p. 51-52, EBSCO Host Research Databases. Business Source Premier.

Diekoff, G 1992, Statistics for the Social and Behavioural Sciences: Univariate,

Bivariate, Multivariate, Wm, C. Brown Publisher, Dubuque, USA.

Diamantopolous, A 1994. "Modelling with LISREL: A Guide for the Uninitiated."

Journal of Marketing Management, vol. 10, p. 105-136.

Dolliver, M 2006, "It Isn't So Unreal For Young Viewers", Adweek, vol.47, no.27,

viewed 040/04/08, p.22, EBSCO HosT Research Databases. Business Source Premier.

Donohue, TR 1975, "Effect of Commercials on Black Children", Journal of Advertising

Research, vol.15, no.6, p.41-47, viewed 20/12/07, EBSCO Host Research Database.

Business Source Premier.

Dotson, MJ & Hyatt, EM 2005, "Major Influence Factors in Children's Consumer

Socialisation", Journal of Consumer Marketing vol.22, no.1, p. 35-42, viewed 20/12/07,

Emerald Group Publishing Limited. Business Source Premier.

249

Driscoll, M 2007, "Taking a Flyer on American Eagle", Business Week 31 July.

Dubecki, L, Goodings, S & Mathieson, C 2008, "Green Guide's Critical View", The

Age, 4 March.

Dubow, EF, Huesmann, LR & Greenwood, D 2007, Media and Youth Socialisation:

Underlying Processes and Moderators of Effects, Dubow, EF, Huesmann, LR &

Greenwood, D (eds), Handbook of Socialisation-Theory and Research, The Guilford

Press United States of America.

Dunn, SC, Seaker, RT & Waller, MA 1994, Latent Variables in Business Logistics

Research: Scale Development and Validation, Journal of Business Logistics, vol. 18,

no. 2, p. 199-216.

Durham, MG 1999, "Girls, Media, and the Negotiation of Sexuality: A study of Race,

Class, and Gender in Adolescent Peer Groups", p. 193-216, Journalism and Mass

Communication, vol.76, no.2, viewed 09/02/08, ProQuest. ABI/INFORM Global.

Eagly, AH & Chaiken, S 2007, "The Advantages of an Inclusive Definition of an

Attitude", Social Cognition, vol.25, no.5, p. 582-602, viewed 04/01/08, EBSCO Host

Research Database. Psychology and Behavioural Science Collection.

Eastin, MS 2005, "Teen Internet Use: Relating Social Perceptions and Cognitive

Models to Behavior", Cyber Psychology & Behaviour, vol.8, no.1, p. 62-75, viewed

27/12/07, EBSCO Host Research Database. Psychology and Behavioural Science

Collection.

Edward, T & Grantham, S 2009, "Antecedent Influences on Children's Extrinsic

Motivation to Go Online." Journal of Applied Social Psychology, vol. 39, no. 3, p. 707-

733.

250

Elkin, F & Handel, G 1972, The Child and Society: The Process of Socialisation,

Random House, New York.

Englis, BG, Solomon, MR & Olofsson, A 1993, "Consumption Imagery in Music

Television: A bi-cultural Perspective", Journal of Advertising., vol.22, no.4, p. 21-35,

viewed 24/01/08, ProQuest. Social Science Journal.

Erametsa, HO 1990, "Media Distinction: An Empirical Study of Uses of Mass Media,

Social Classes, and Ways of Life", Helsinki.

Evans, M, Jamal, A & Foxall, G 2006, Consumer Behaviour, John Wiley and Sons

limited, England.

Fabrigar, LR, Wgener, DT, MacCallum, RC & Strahan, EJ 1999, Evaluating the Use of

Exploratory Factor Analysis in Psychological Research, Psychological Method, vol. 4,

no. 3, p. 272-299.

Fischer, C &.Wertz, F 1975, Empirical Phenomenological Analyses of Being

Criminally Victimized Phenomenology and Psychological Research Pittsburgh,

Duquesne University Press.

Fletcher, J 2006, "Social Interactions in Adolescent Television Viewing", Archives of

Paediatrics and Adolescent Medicine, vol.160, no.4, p. 383-386, viewed 07/01/08,

ProQuest Social Science Journals.

Fleishman, J & J. Benson 1987, "Using LISREL to Evaluate Measurement Models and

Scales Reliability” Educational and Psychological Measurement, vol. 47, p. 925-938.

Flick, U 1992, "Triangulation Revisited: Strategy of Validation or Alternative?" Journal

for the Theory of Social Behaviour, vol. 22.

251

Flouri, E 1999, "An Integrated Model of Consumer Materialism: Can Economic

Socialisation and Maternal Values Predict Materialistic Attitudes in Adolescents?"

Journal of Socio-Economics, vol.28, no.6, p. 707-724, viewed 20/02/08, EBSCO Host

Research Databases, Proquest Social Science Journals.

Fontana, A & Frey, JH 1994, Interviewing: The Art of Science, Fontana, A & Frey, JH

(eds), Handbook of Qualitative Research, SAGE.

Fornell, C & Yi, Y 1992, Assumptions of the Two-Step Approach to Latent Variable

Modelling. Sociological Methods and Research, vol. 20, p. 291-320.

Foxall, GR 1993, "Consumer Behaviour as an Evolutionary Process", European Journal

of Marketing vol.27, no.8, p. 46-57, viewed 12/12/07, Emerald.

French, DC, Pidada, S & Victor, A 2005, "Friendships of Indonesian and United States

youth", International Journal of Behavioural Development, vol.29, no.4, p. 304-313,

viewed 16/12/07, EBSCO Host Research Database. Psychology and Behavioural

Sciences Collection.

Frideres, JS 1973, "Advertising, Buying Patterns and Children", Journal of Advertising

Research, vol.13, no.1, p. 34-36, viewed 05/07/07, EBSCO Host Research Database.

PsycINFO.

Frutkin, AJ 2008, 'New Shows a Write-Off?' Mediaweek, 10 March.

Fujioka, Y & Austin 2002, "The Relationship of Family Communication Patterns to

Parental Mediation Styles", Communication Research, vol.29, no.6, p. 642-665, viewed

16/01/08, SAGE Full text Collection. ProQuest Social Science Journals.

Gabriel, C 1990, "The Validity of Qualitative Market Research", Journal of the

Marketing Research Society, vol.32, no.October, p. 507.

252

Gage, N 1989, "The Paradigm Wars and Their Aftermath: A Historical Sketch of

Research and Teaching Since 1989", Educational Research, vol.18.

Gainer, BP & Padanyl 2002, "Applying the Marketing Concept to Cultural

Organizations: An Empirical Study of the Relationship between Market Orientation and

Performance", International Journal of Non profit and Voluntary Sector Marketing,

vol.7, no.2, p. 182-193.

Garver, M S & Mentzer, JT 1999, "Logistics Research Methods: Employing Structural

Equation Modelling to Test for Construct Validity." Journal of Business Logistics, vol.

20, no. 1, p. 33-57.

Garson, D 2009, Structural Equation Modelling, North Carolina State University

Gerbener, G. 1990, Advancing on the path of righteousness, Signorielli & Morgan

(eds), Cultivation Analysis, Newbury Park, SAGE.

Geuens, M, Pelsmacker, PD & Mast, G 2002, "How Family Structure Affects Parent

Child Communication About Consumption", Young consumers, vol.4, no. 2, p. 57-62,

viewed 16/05/07, WARC.

Ghani, DNHA 2005, "Television Viewing and Consumer Behaviour." Young Consumers, vol. 6, no. 1, WARC.

Glaser, BG 1978, Theoretical Sensitivity: Advances in the Methodology of Grounded

Theory, Sociology Press, Mill Valley.

Gorn, GJ & Florsheim, Re 1985, "The Effects of Commercials for Adult Products on

Children", Journal of Consumer Research, vol.11, no.4, p. 962-967, viewed 05/07/07,

EBSCO Host Research Database. PsycINFO.

253

Goslin, DA 1971, Introduction, Goslin, DA (eds), Hand Book of Socialisation of

Theory and Research Rand McNally& Company U.S.A.

Gould, SJ & Gupta, PB 2006, "COME ON DOWN": How Consumers View Game

Shows and the Products Placed in Them", Journal of Advertising, vol. 35, no. 1, p. 65-

81, EBSCO Host Research Database, Business Source Premier.

Granhaug, K & Venkatesh, A 1986, "Products and Services in the Perspective of

Consumer Socialisation", European Journal of Marketing, vol.20, no.10, p. 55-65,

viewed 22/12/07, EBSCO Host Research Database. Business Source Premier.

Greenberg, BS 1974, "British Children and Televised Violence", Public Opinion

Quarterly, vol.38, no.4, p. 531, viewed 10/02/08, EBSCO Host Research Databases.

Business Source Premier.

Greenleaf, EA 1992, "Improving Rating Scale Measures by Detecting and Correcting

Bias Components in Some Response Styles", Journal of Marketing Research vol.29, no.

May, p. 176.

Grimes, T, Bergen, L, Nichols, K, Vernbergh, E & Fonagy, P 2004, "Is

Psychopathology the Key to Understanding Why Some Children Become Aggressive

When They Are Exposed To Violent Television Programming?" Human

Communication Research., vol.30, no.2, p. 153-181, viewed 24/10/8, ProQuest.

ABI/INFORM Global.

Griswold, W, McDonnell, T & Wright, N 2005, "Reading and the Reading Class in the

Twenty-First Century", Annual Review of Sociology, vol.31, p. 27-41, viewed

09/02/08, ProQuest Social Science Journals.

254

Groome, TH 2007, "Advice to Beginners - and to Myself", Religious Education,

vol.102, no.4, p. 362-366, viewed 04/01/08, EBSCO Host Research Database.

Psychology and Behavioural Science Collection.

Gruber, E & Thau, H 2003, "Sexually Related Content on Television and Adolescents

of Color: Media Theory, Physiological Development, and Psychological Impact", The

Journal of Negro Education, vol.72, no.4, p. 438-456, viewed 09/01/08, ProQuest Social

Science Journals.

Grusec, JE & Davidov, M 2007, Socialisation in the family: The Roles of Parents,

Grusec, JE & Davidov, M (eds), Hand Book of Socialisation-Theory and Research, The

Guilford Press, United states of America.

Guba, EG 1990, The Alternative Paradigm Dialog, Guba, EG (eds), The Paradigm

Dialog, SAGE, Newbury Park.

Guba, EG & Lincoln, YS 1994, Competing Paradigms in Qualitative Research, Guba,

EG & Lincoln, YS (eds), Handbook of Qualitative Research, SAGE, California.

Guilfoy, C 1998, "Value in Video Games Children's Obsession May Actually be

Teaching Important Skills" , Telegram & Gazette, 22 September.

Hair, JF, Anderson, RE, Tatham, RL & Black, WC. 1998. Multivariate Data Analysis.

New Jersey, Prentice Hall.

Hair, JF, Bush, RP & Ortinau, DJ 2006, Marketing Research-Within a Changing

Information Environment, New York, McGraw-Hill.

Hair, JF, Black, WC, Babin, BJ, Anderson, RE & Tatham, RL 2006, Multivariate Data

Analysis. New Jersey, Pearson.

255

Haron, SA, Paim, L & Yahaya, N 2005, "Towards Sustainable Consumption: an

Examination of Environmental Knowledge among Malaysians", International Journal of

Consumer Studies, vol.29, no. September, p. 426-436, viewed 10/01/08, Blackwell

Synergy. ProQuest Social Science Journals.

Harrison, K & Cantor, J 1997, "The Relationship Between Media Consumption and

Eating Disorders", Journal of Communication, vol.47, no.1, 40-67, viewed 04/01/08,

ProQuest. Business Source Premier.

Hasan, K 2002, "Violence in Film and Youth", Mass Line Media Centre, Dhaka.

Hastings, GB & Aitken, PP 1995, "Tobacco Advertising and Children's Smoking: A

Review of the Evidence", European Journal of Marketing, vol.29, no.11, p. 6-17,

viewed 04/01/08, EBSCO Host Research Database. Business Source Premier.

Hennessy, J 2006, "Today's Parent", vol.23, no.2, p. 77, viewed 06/01/08, ProQuest

Social Science Journal.

Hetsroni, A & Tukachinsky, RH 2006, "Television-World Estimates, Real-World

Estimates, and Television Viewing: A New Scheme for Cultivation", Journal of

Communication, vol.56, no.1, p. 133-156, viewed 04/01/08, Blackwell Synergy.

ProQuest Social Science Journal.

Hibberd, J 2005 TV, Violence Worse Than Nudity. Television Week.

Higie, RA & Feick, LF 1989. "Enduring Involvement: Conceptual and Measurement

Issues", Advances in Consumer Research, vol. 16, no. 1, p. 690-696.

Hill, A 2005, Reality TV: Audience and Popular Factual Television. London,

Routledge.

256

Hirschman, EC & Thompson, CG 1997. "Why Media Matter: Toward a Richer

Understanding of Consumers' Relationships with Advertising and Mass Media " Journal

of Advertising, vol. 26, no. 1, p. 43-60.

Holmes-Smith 2002, Applied Structural Equation Modelling, February, Canberra.

Holmes-Smith, P, & Rowe, K 1994. The Development and Use of Congeneric

Measurement Models in School Effective Research. The World Congress center.

Melbourne.

Homer, P. and L. R. Kahale 1988, "A Structural Equation Test of the Values-Attitude-

Behaviour Hierarchy." Journal of Personality and Social Psychology, vol. 54, no. April,

p. 638-646.

Hoyer, WD & Maclnnis, DJ 2007, Consumer Behavior,George T.Hoffman, United

states of America.

Hoyle, RH 1995, The Structural Equation Modelling Approach: Basic Concepts and

Fundamentals Issues, Hoyle, RH (eds), Structural Equation Modelling: Concepts,

Issues, and Applications, SAGE.

Hu, L & Bentler, PM 1998, Fit Indices indices in covariance structure modeling:

Sensitivity to under paramatarized method misspecification. Psychological Methods.

Hulland, J, Chow, YH & Lam, S 1996, Use of Casual Models in Marketing Research, A

review of International Journal of Research in Marketing, vol. 13, p. 181-197.

Hunt, SD 1991, Modern Marketing Theory, South Western Publishing Co, Cincinnati.

257

Hurley, AE, Scandara,TA, Scriesheim, CA, Brannick, MT, Seers, A, Vanderberg & RJ,

Williams, LJ 1997, Exploratory and Confirmatory Factor Analysis: Guidilines, Issues,

and Alternatives, Journal of Organisational Behaviour, vol. 18, p. 667

Idato, M 2007, "True Love's Dream Run Ends in The Poor: TV Review', Sydney

Morning Herald, July 14.

Islam, N 2005, Slums of Urban Bangladesh, Centre of Urban Studies, Dhaka.

Isler, L, Topper, ET & Ward, S 1987, " Children's Purchase Requests and Parental

Responses: Results from a Dairy Study", Journal of Advertising Research, vol.27, no.5,

p. 28-39, viewed 24/07/07, EBSCO Host Research Database. Business Source Premier.

Jacobs, E 2008, "What the Apprentices Did Next They Wanted a Career, Celebrity, or

Just Approval from a Father Figure". As the fourth series of the BBC's Reality Business

Show starts, Emma Jacobs Tracks Down Some of Those Who Were Fired - and One

Who Was Hired - to Talk about Life after 'Sir Alan'', Financial Times, 22 March.

James, S 2007, "Beyond Tomorrow", The Advertiser, 21 December.

Johnson, K, Cohen, P & Brook, JS 2007, "Extensive Television Viewing and the

Development of Attention and Learning Difficulties During Adolescence", Archives of

Pediatrics and Adolescent medicine, vol. 161, no.5, p. 480-486.

Johansson, JK & Nonaka, I 1987, "Market Research the Japanese Way", Harvard

Business Review, vol. May-June.

John, DR 1999, "Consumer Socialisation of Children: A Retrospective Look at Twenty-

Five Years of Research", Journal of Consumer Research, vol.26, no.3, p. 183-213,

viewed 05/07/07, EBSCO Host Research Database. PsycINFO.

258

Jordan, A 2004, "The Role of Media in Children's Development: An Ecological

Perspective", Journal of Developmental & Behavioural Paediatrics, vol.25, no.3, p. 196-

206, viewed 07/01/08, ProQuest Social Science Journal.

Joreskog, KG & Sorbom, D, 1993, LISREL8: User's Reference Guide, Scientific

Software, Chicago.

Joreskog, KG & Sorbom, D 1989, LISERAL 7: A Guide to the Program and

Applications (2nd ed), Chicago: SPSS.

Jung, J & Peterson, M 2007, "Body Dissatisfaction and Patterns of Media Use Among

Preadolescent Children", Family and Consumer Sciences Research Journal, vol.36,

no.September, p. 40-54, viewed 05/02/08, EBSCO Host Research Database.

Psychology and Behavioural Sciences Collection.

Kamaruddin, AR & Mokhlis, S 2003, "Consumer Socialisation, Social Structural

Factors and Decision-Making Styles: a Case Study of Adolescents in Malaysia",

International Journal of Consumer Studies,, vol.27, no.2, p. 145-156, viewed

16/05/2007, EBSCO Host Research Database. Business Source premier.

Kamineni, R 2005, "Influence of Materialism, Gender and Nationality on Consumer

Brand Perceptions", Journal of Targeting, Measurement and Analysis for Marketing,

vol.14, no.1, p. 25-33, viewed 12/10/08, ProQuest. ABI/INFORM Global.

Katz, H 2006, "The Media Hand Book", 24/01/08,

http://books.google.com.au/books?id=J2un6ihfiAC&dq=Types+of+Electronic+media&

pg=PA2&ots=Jfxum6ZoX&sig=haNc02dauBAvJmAzHVWVjMAEIO8&hl=en&prev=

http://www.google.com.au/search?hl=en&q=Types+of+Electronic+media&btnG=Searc

h&sa=X&oi=print&ct=title&cad=one-book-with-thumbnail.

259

Keenan, C 2008, "A Personal Approach to the Feminine Mystique", Sydney Morning

Herald, viewed, 21 March.

Kennedy, H 1996, Understanding Pay TV Markets in Africa. ESOMAR. Sanfransisco.

Keesling, JW 1972, Maximum Likelihood Approached to Causal Analysis. Department

of Education. Chicago. PhD Dissertation.

Keith, TZ 1986, "Parental Involvement, Homework, and TV Time: Direct and Indirect

Effects on High School Achievement", Journal of Educational Psychology, vol.78, no.5,

p. 373-380, viewed 24/01/08, EBSCO Host Research Databases. ABI/INFORM Global.

Kelloway, KE 1995 "Structural Equation Modeling in Perspective." Journal of Organisational Behaviour, vol. 16, p. 215.

Kim, J & Shin, H-J 1996, "Patterns of Television Co-viewing Behaviour", Broadcast

Audience Research, vol. April, p. 58-66, viewed 24/01/08, World ESOMAR Research.

WARC.

Kim, Y & Baylor, A 2006, "A Social-Cognitive Framework for Pedagogical Agents as

Learning Companions", Educational Technology Research & Development, vol.54,

no.6, p. 569-596, viewed 04/01/08, EBSCO Host Research Database. Psychology and

Behavioural Sciences Collection.

Kline, R1998, Principles and practices of Structural Equation Modelling. New York,

Guilford Press.

Knobloch-Westerwick, S 2007, "Gender Differences in Selective Media Use for Mood

Management and Mood Adjustment", Journal of Broadcasting & Electronic Media,

vol.51, no.1, p. 73-92, viewed 05/02/08, ProQuest. ABI/INFORM Global.

260

Kocher, B & Lalos, M 2008, "Let it Rock: The Effects of Brand Name Placement in

Songs on Attitudes toward the Artist and the Brand." Advances in Consumer Research -

North American Conference Proceedings, vol. 35, p. 769-770.

Kotler, P, Brown, L, Adam, S, Burton, S & Armstrong, G 2007, Marketing, Pearson-

Prentice Hall, Australia.

Krugman, HE 1962, "An Application of Learning Theory to TV Copy Testing".

Public opinion Quarterly, vol. 29, no. Fall, p. 349-356, viewed 04/10/08, EBSCO Host

Research Database, Business Source Premier.

Krugman, HE 1966, "The Measurement Of Advertising Involvement." Public Opinion

Quarterly, vol. 30, no. 4, p. 583-596.

Kuhl, C 2006 "Parental Control." Multi-Channel News, vol. 27, no. 43.

Lachance, MJ, Beaudoin, P & Robitaille, J 2003, "Adolescents Brand Sensitivity in

Apparel: Influence of Three Socialisation Agents", International Journal of Consumer

Studies, vol.27, no.1, p. 47-57, viewed 20/01/08, EBSCO Host Research Database.

Business Source Premier.

Lachance, MJ & Choquette-Bernier, N 2004, "College Students’ Consumer

Competence: a Qualitative Exploration", International Journal of Consumer Studies,

vol.28, no.5, p. 433-442, viewed 20/01/08, EBSCO Host Research Database Business

Source Premier.

Lam, K-C & Shi, G 2008, "Factors Affecting Ethical Attitudes in Mainland China and

Hong Kong", Journal of Business Ethics, vol.77, no.4, p. 463-479, viewed 20/01/08,

ProQuest. ABI/INFORM Global.

261

Larsen, JT, To, YM & Fireman, G 2007, "Children's Understanding and Experience of

Mixed Emotions", Psychological Science, vol.18, no.2, p. 186-191, viewed 16/12/07,

ProQuest Social Social Science Journal.

Lawrie, Z, Sullivan, EA, Davies, PSW & Hill, RJ 2006, "Media Influence on the Body

Image of Children and Adolescents", Eating Disorders, vol.14, no.5, p. 355-364, viewed

04/01/08, EBSCO Host Research Database.

Levy, S & Nebenzahl, ID 2006, "Programme Involvement and Interactive Behaviour in

Interactive Television", International Journal of Advertising & Marketing to Children,

vol.25, no.3, p. 1-13, viewed 06/01/08, WARC.

Levy, MR & Windahl, S 1984, The Concept of Audience Activity, K E , Rosengren,

Wenner & Palmgreen (eds), Media Gratification Research: Current Perspectives,

California, Beverly Hills.

Lin, CA 1993, "Modelling the Gratification-Seeking process of Television Viewing."

Human Communication Research, vol. 20, p. 224-244.

Lloyd, MA 1985, Adolescence, Harper & Row, Publishers, Inc., United States of

America.

Lofland, J & Lofland, L 1984, Analysing Social Settings: A Guide to Qualitative

Observations and Analysis, Wadsworth, California.

López-Sintas, J & García-Álvarez, E 2006, "Patterns of Audio-Visual Consumption:

The Reflection of Objective Divisions in Class Structure", European Sociological

Review, vol.22, no.4, p. 397-411, viewed 09/02/08, ProQuest. ABI/INFORM Global.

Lowry, B 2008, "Checks, but no Balance", Variety, 9 March.

262

Lueg, JE & Finney, RZ 2007, "Interpersonal Communication in the Consumer

Socialisation: Scale Development and Validation", Journal of Marketing Theory &

Practice, vol.15, no.1, p. 25-39, viewed 20/012/12, EBSCO Host Research Database.

Business Source Premier.

Lundy, S & Jacobson, S 2008, "Does You Tube Generation Lack Sense of Right,

Wrong?" McClatchy - Tribune Business News., 12 April.

Maccoby, EE 1992, "The Role of Parents in the Socialisation of Children: An Historical

Overview"', Developmental Psychology, vol.28, no.6, p. 1006-1017, viewed 24/07/07,

EBSCO Host Research Database. PsycINFO.

MacCullum, RM, Rosnowski, C & Reith, J 1994. Alternative Strategies for Cross-

Validation of Co-Variance Structure Models. Multivariate Behavioural Research, vol.

29, p. 1-32.

MacCullum, RC & Hong, S 1997, Power Analysis in Covariance Structure Modeling

GFI and AGFI, Multivariate Behavioral Research, vol. 32, p. 193-210.

Malhotra, NK 2003, Marketing Research: An Applied Orientation, Prentice Hall

International, Inc., New Jersey.

Malhotra, NK & Peterson, M 2006, Basic Marketing Research-A Decision-Making

Approach, New Jersey, Prentice Hall.

Malthouse, E & Calder, BJ 2007, Measuring Involvement with Editorial Content:

Conceptualization, Scale Development, and the Effects on Advertising. Worldwide

Multi Media Measurement (WM3). Dublin.

Malinowski, B 1989, A Diary in the Strict Sense of the Term, Stanford University

Press, California.

263

Mangleburg, TF & Grewal, D 1997, "Socialisation, Gender, and Adolescent's Self-

Reports of their", Journal of Consumer Affairs, vol.31, no.2, p. 255-280, viewed

20/120/07, EBSCO Host Research Database. Business Source Premier.

Mark, N 1998, "Birds of a Feather Sing Together", Social Forces, vol.77, no.2, p. 453,

viewed 10/02/08, EBSCO Host Research Database. Business Source Premier.

Marsh, HW & Hocevar, D 1988, "A New, More Powerful Approach to Multitrait-

Multi-Method Analysis: Application of Second-Order Confirmatory Factor Analysis",

vol.73, viewed 18/01/08, no.1, p. 107-117, Journal of Applied Psychology.

Maruyama, G. 1998. Basics of Structural Equation Modelling Thousand Oaks, SAGE

Publications.

McCarty, JA & Shrum, LJ 1993, "The Role of Personal Values and Demographic in

predicting Television Viewing Behaviour: Implications for Theory and Application",

Journal of Advertising, vol.22, no.4, p. 77-101, viewed ProQuest. ABI/INFORM

Global.

Mclaughlin, B 1971, Learning and Social Behaviour, Free Press, New York.

McClymont, H, Ogunmokun, G & Ali Akbari, J 2004, "An Exploratory Examination of

the Marketing Philosophy Utilization in the Health Industry: A Case of Preventative

Family Home Visiting Health Program in Australia", Problems & Perspectives in

Management, no.4, viewed 8/12/7, Buisness Source Premier. EBSCO Host Research

Database.

264

McDade, TW & Worthman, CM 2004, "Socialisation Ambiguity in Samoan

Adolescents: A Model for Human Development and Stress in the Context of Culture

Change", Journal of Research on Adolescence (Blackwell Publishing Limited), vol.14,

no.1, p. 49-72, viewed 14/012/07, ProQuest Social Science Journal. EBSCO Host

Research Database.

McDonald, RP & Marsh, HW 1990, Choosing a Multivariate Model: Non-centrality and

Goodness of fit. Psychological Bulletin.

Mcleod, JM & Keefe, GO 1972, The Socialisation Perspective and Communication

Behaviour: In Current Perspective of Mass Communication Research. Beverly Hills,

SAGE.

McNeal, JU & Ji, MF 1999, "Chinese Children as Consumers: An Analysis of Their

New Product Information Sources", Journal of Consumer Marketing, vol.16, no.4/5, p.

345-364, viewed 20/01/08, EBSCO Host Research Database. Business Source Premier.

McPhee, N 2004, "Reality TV and the Growth of Ethnographic/Qualitative Research –

Coincidence, or Context?" Consumer Insight Conference, Vienna, 23/01/08, WARC.

McQuil, D 1987, Mass Comminication Theory-An Introduction,SAGE Publication,

London.

McWhirter, E 2007, "An idol fascination", The Daily Telegraph., 30 june.

Melanie & Locke 2007, "The Socialisation of Adolescents Coping Behaviour:

Relationship with Families and Teachers", Journal of Adolescence, vol. 30, no. 1, p. 1-

16, viewed 15/05/2007, ABI/INFORM GLOBAL, ProQuest.

Mellor, S 2003, "Communicating Effectively to Millennium Children", Admap, March.

Middendorp, C 2008, "Down to the Last Chance", The Age, 1 March.

265

Miles, MB & Huberman, AM 1994, Qualitative Data Analysis: An Expanded

Sourcebook, SAGE, Newbury Park.

Ministry of Information 2001, "Media list", Government of Bangladesh.

Moore, ES & Bowman, GD 2006, "Of Friends and Family: How Do Peers Affect the

Development of Intergenerational Influences?" Advances in Consumer Research,

vol.33, no.1, p. 536-542, viewed 20/01/08, EBSCO Host Research Database. Business

Source Premier.

Moore, JN, Raymond, MA, Mittelstaedt, JD & Tanner Jr, JF 2002, "Age and Consumer

Socialisation Agent Influences on Adolescents' Sexual Knowledge, Attitudes, and

Behaviour: Implications for Social Marketing Initiatives and Public Policy", Journal of

Public Policy & Marketing, vol.21, no.1, p. 37-52, viewed 20/01/08, EBSCO Host

Research Databases. Business Source Premier.

Morse, JM1986, Qualitative Research: Issues in Sampling. Nursing Research

Methodology: Chinn (ed), Issues and Implementations. Rockville, Aspen.

Morse, JM 1991, Strategies for Sampling. J M Morse, JM (eds) Qualitative Nursing

Research: A Contemporary Dialogue. Newbury Park, SAGE.

Morse, JM 1994, Designing Funded Qualitative Research, Morse, JM (eds), Handbook

of Qualitative Research, SAGE, California.

Morgan, FW 1990, "Judicial Standards for Survey Research: An Update and

Guidelines", Journal of Marketing, vol.54, no. January.

Morgan, M & Shanahan, L 1998, Two Decades of Cultivation Research: An appraisal

and Meta-Analysis. Burleson Kunkel (eds) Communication yearbook. Thousands

Oaks, SAGE.

266

Moscardelli, D & Liston-Heyes, C 2005, "Consumer Socialisation in a Wired World:

The Effects of Internet Use and Parental Communication on the Development of

Skepticism of Advertising", Journal of Marketing Theory & Practice, vol.13, no.3, , p.

62-75, viewed 20/12/07, EBSCO Host Research Database. Business Source Premier.

Moschis, GP 1981, "Patterns of Consumer Learning", Academy of Marketing Science,

vol.9, no.1/2, p. 110-126, viewed 24/12/07, EBSCO Host Research Database.

ABI/INFORM Global.

Moschis, GP 1985, "The Role of Family Communication in Consumer Socialisation of

Children and Adolescents", Journal of Consumer Research, vol.11, no.4, viewed

23/07/07, EBSCO Host Research Databases. Business Source Premier.

Moschis, GP 1988, "Methodological Issues in Studying Intergenerational Influences on

Consumer Behavior", Advances in Consumer Research, vol.15, no.1, p. 898-914,

viewed 20/01/08, EBSCO Host Research Databases. Business Source Premier.

Moschis, GP & Churchill Jr, GA 1978, "Consumer Socialisation: A Theoretical and

Empirical Analysis", Journal of Marketing Research (JMR), vol.15, no.4, p. 599-609,

viewed 25/12/07, EBSCO Host Research Database. Business Source Premier.

Moschis, GP & Churchill Jr, GA 1979, "An Analysis of the Adolescent Consumer",

Journal of Marketing, vol.43, no.3, p. 40-51, viewed 12/12/07, Business Source

Premier. EBSCO Host Research Database.

Moschis, G P & Moore RL 1978, "An Analysis of the Acquisition of Consumer

Competencies Among Adolescents", Journal of Consumer Affairs, vol. 12, no. 2, p.

277-292.

267

Moschis, GP & Moore, RL 1979, "Decision Making Among the Young: A Socialisation

Perspective", Journal of Consumer Research, vol.6, no.2, p. 101-113, viewed 16/01/08,

ProQuest. ABI/INFOM Global.

Moschis, G P & Moore, RL 1982. "A Longitudinal Study of Television Advertising

Effects." Journal of Consumer Research, vol. 9, no. 3, p. 279-286.

Nacos, BL 2000, "Television and Its Viewers: Cultivation Theory and Research",

Political Science Quarterly, vol.115, no.2, p. 317-320, viewed 24/01/08, ProQuest.

ABI/INFORM Global.

Naert, P & Leeflang, P 1978, Building Implementable Marketing Models, International

Graphics Dordrecht, Netherland.

Nath, SR 2006, "Youth Access to Mass Media in Bangladesh", Adolescents and youth

in Bangladesh: Some selected issues.

Nathanson, AI 2001, "Parents versus Peers: Exploring the Significance of Peer

Mediation of Antisocial Television." Communication Research, vol. 28, no. 3, p. 251.

Neal, C, Quester, P & Hawkins, D 2004, Consumer Behaviour, MacGraw Hill,

Australia.

Neeley, S 2005, "Influences on Consumer Socialisation", Young Consumers, vol.6,

no.2, viewed 16/05/ 2007, World Advertising Research Centre Ltd. WARC.

Omura, GS 1983, "Correlations of Item Non-response", Journal of Market Research

Society, vol.October, p. 321.

268

Orlik, PB 2000, "Electronic Media Criticism-Applied Perspectives", 240/10/8,

http://books.google.com/books?id=v4NLBSmNGfoC&pg=PA60&lpg=PA60&dq=type

s+of+electronic+media&source=web&ots=rm19-Qh27W&sig=2J5fyWx-

fuBTTGftfh7CVfn9U1Q#PPP1,M1.

Parsons, T & Bales, RF 1955, Family, Socialisation and interaction process. Pearson.

Patton, MQ 1990, Qualitative Evaluation and Research Methods, SAGE, Newbury

Park.

Perreault, WD & Leigh, LE 1989, "Reliability of Nominal Data based on Qualitative

Judgements", Journal of Marketing Research, vol.25, no. May, p. 135.

Pecheux, C & Derbaix, C 2002, "Children's Reactions to Advertising Communication:

Multiple Methods, Moderating Variables and Construct Validity Issues." Advances in

Consumer Research, vol. 29, no. 1: 531-538.

Perry, C 1995, 'A Structured Approach to Presenting PhD Theses: Notes for Candidates

and Their Supervisors.' ANZMAC Doctoral Consortium, University of Sydney.

Perse E1990a "Audience Selectivity and Involvement in the Newer Media

Involvement." Communication Research, vol. 17. p. 675-697.

Perse, E. 1990b. "Involvement With the Local Television News: Cognitive and

Emotional Dimensions." Human Communication Research, vol. 16, p. 556-581.

Perse, E 1998 "Implications of Cognitive and Affective for Channel Changing." Journal

of Communication, vol. 48, no. 3, p. 49-68.

Peters, JW 2007, "When Reality TV Gets Too Real", New York Times, 08 October.

269

Peterson, SH, Wingood, GM, DiClemente, RJ, Harrington, K & Davies, S 2007,

"Images of Sexual Stereotypes in Rap Videos and the Health of African American

Female Adolescents", Journal of Women's Health (15409996), vol.16, no.8, p. 1157-

1164, viewed 04/01/08, EBSCO Host Research Database.

Pope, N, Brennan, L & Voges, K 2007, Consumer Behaviour, John Wiley & Sons, Ltd

Australia.

Population, RB 2006, The World's youth. Washington DC.

Potter, WJ 2005, Media Literacy, SAGE Publication, London.

Powell, C 2003, "Idol Worship", Marketing Magazine, 11 October.

Pratt, MW, Skoe, EE & Arnold, ML 2004, "Care Reasoning Development and Family

Socialisation Patterns in Later Adolescence: A longitudinal analysis", International

Journal of Behavioural Development, vol.28, no.2, p. 139-147.

Pursell, C & Hibberd, J 2007, "NBC Adopts 'Baby' for Fall Lineup. (Cover story)",

Television Week, 03 December.

Quraishi, SY, Bhuiya, I & Mohammad, N 2004, "Youth in Bangladesh- A Thematic

Review", UNFPA, Bangladesh.

Rahman, M 2008, "Family Entertainment and TV", Daily Manabjamin, March 3.

Rahman, MG 2006, Communication in Bangladesh: Media response and campaign

strategy, Shrabon Publication, Dhaka.

Rao, SL 2002, Phd Thesis, Faculty of Commerce and Management, Griffith University,

Australia.

270

Rao, S & Perry, C 2004, "Convergent Interviewing to Build a Theory in Under

Researched Area: Principles and an Example Investigation of the Role of the Internet in

Inter-Firm Relationships", Qualitative Market Research, An International Journal.

Reading, R 2008, "The effects of television advertisements for junk food versus

nutritious food on children's food attitudes and preferences", Child: Care, Health &

Development, vol.34, no.1, p. 134-140, viewed 04/01/008, EBSCO Host Research

Database. Psychology and Behavioural Science Collection.

http://www.realitytvworld.com 2008.

Robinson, P 2005, "Reality TV Trails Among Viewers 55+; film at 11", Marketing

Magazine, vol.110, no.12, p. 21-22, viewed 04/04/08, EBSCO HOST Research

Databases. Business Source Premier.

Robinston, SM, Crozier, SR, Borland, SE & J Hammond, ea 2004, "Impact of

Educational Attainment on the Quality of Young Women's Diets", European Journal of

Clinical Nutrition., vol.58, no.8, p.1174-1180, viewed 09/02/08, ProQuest Social

Science Journals.

Roopnarine, JL 1986, "Mothers' and Fathers' Behaviours Toward the Toy Play of their

Infant Sons and Daughters", Sex Roles, vol.14, no.1, p. 59-68.

Rose, RJ 2007. Peers, Parents, and Processes of Adolescent Socialization: A twin Study

Perspective. Engels, Kerr & Stattin (eds). Friends, Lovers and Groups-Key Relationship

in Adolescence. England, John Wiley & Sons, Ltd.

Rose, GM, Bush, VD & Kahle L 1998, "The Influence of Family Communication

Patterns on Parental Reactions toward Advertising: A Cross-National Examination."

Journal of Advertising, vol. 27. no. 4, p. 71-85, viewed 24/01/08, ProQuest Social

Science Journal

271

Rossiter, JR 1979, "Does TV Advertising Affect Children?", Journal of Advertising

Research, vol.19, no.1, p. 49-53, viewed 04/01/08, EBSCO Host Research Database.

ProQuest Social Science Journal.

Rubin, A & Perse, E 1987 "Audience Activity and Soap Opera Involvement: Human

Communication Research, vol. 14, p. 246-268, EBSCO Host Resaerch Database,

Business Source Premier.

Rubin, AM & Step, MM 1997. "Viewing Television Talk Shows." Communication

Research Report, vol. 14, no. 106-115, viewed 10/10/08, EBSCO Host Research

Database, Business Source Premier.

Russell, CA, Norman, AT & Heckler, SE 2004, "The Consumption of Television

Programming: Development and Validation of the Connectedness Scale", Journal of

Consumer Research, vol.31, no.1, p. 150-162, viewed 10/01/08, ProQuest.

ABI/INFORM Global.

Saint-Cyr,G 2003, "What Values Are Disseminated by Television", Carleton

University, viewed 12/01/08, http://chat.carleton.ca/~tneil/tv-values/a.html

Scardino, E 2005, "What to Wear? Teens look to reality TV", DSN Retailing Today,

vol.44, no.2, p. 10-12, viewed 04/04/08, EBSCO Host Research Databases. Business

Source Premier.

Schiffman, L, Bendall, D, O'cass, A, Paladino, A & Kanuk, L 2005, Consumer

Behavior, Pearson Education Australia, China.

Schmoll, NM, Hafer, J, Hilt, M & Reilly, H 2006, "Baby Boomers' Attitudes Towards

Product Placements", Journal of Current Issues & Research in Advertising, vol.28, no.2,

p. 33-54, viewed 10/01/08, EBSCO Host Research Database. ProQuest Social Science

Journal.

272

Schrobsdorff, S 2008, Reality's Believe It or Not. News week.

Schumacker, R & Lomax, R 1996, A Beginners Guide to Structural Equation Modlleing

New Jersey, Lawrence Erlbaum Associates.

Schwandt, TA 1994, Constructivist, Interpretivist Approaches to Human Inquiry

Schwandt, TA (eds), Handbook of Qualitative Research, SAGE, Thousand Oaks,

California.

Schwartz, L, Newsome, B & Mathieson, C 2007, "Green Guide's Critical View", The

Age, 11 July.

Schrest, L 1992, "Roots: Back to Our First Generations", Evaluation Practices.

Seedat, M 2006, "Wahiduddin Khan and Peace in Contemporary Islamic Thought",

South African Journal of Psychology, vol.36, no.4, p. 830-852, viewed 16/12/07,

EBSCO Host Research Database. Psychology and Behavioural Sciences Collection.

Senevirante, K 2007, Malaysia: Alternative media...Only on the Internet Senevirante, K

(eds), Media Pluralism in Asia: The Role and Impact of Alternative media, AMIC,

Singapore.

Shifrin 2006, "Effect of Media on Children and Adolescents", Archives of Paediatrics

and Adolescent Medicine, vol.160, no.4, p. 448-451, viewed 07/01/08, ProQuest Social

Science Journals.

Shim, S 1996, "Adolescent Consumer Decision-Making Styles: The Consumer

Socialisation Perspective; ABSTRACT", Psychology & Marketing, vol.13, no.6, p.

547-560, viewed 16/01/08, ProQuest. ABI/INFORM Global.

273

Shoham, A & Dalakas, V 2003, "Family Consumer Decision Making in Israel: The

Role of Teens and Parents", The Journal of Consumer Marketing., vol.20, no. 2/3, p.

238-250, viewed 12/05/07, ProQuest CSA LLC.

Shrum, LJ, Burroughs, JE & Gainesville, AR 2005, "Television's Cultivation if Material

values", Journal of Consumer Research., vol.32, no.3, p. 473-480, viewed 04/01/08,

ProQuest. Business Source Premier.

Solomon, MR 2004, Conusmer Behaviour-Buying, Having and Being, Pearson

education, Inc., New Jersey.

Solomon, MR, Dann, S, Dann, S & Russell-Bennett, R 2007, Consumer Behaviour,

Pearson Education Australia, China.

Sotirovic, M 2005, "Television and Child Development", Journalism and Mass

Communication vol.82, no.2, p. 477-488, viewed 06/01/08, ProQuest. ABI/INFORM

Global.

Stampfl, RW, Moschis, G & Lawton, JT 1978, "Consumer Education and the Preschool

Child", Journal of Consumer Affairs, vol.12, no.1, p. 12-30, viewed 23/12/07, EBSCO

Host Research Database. Business Source Premier.

Stanley, A 2003, ''Reality-Show Inbreeding: Some Mutations Have Survival Value'',

New York Times 5 March.

Stevens, J 2007, "PREVIEWS - THURSDAY JULY ", The Daily Telegraph, 18 July.

Strasburger, VC & Wilson, BJ 2002, Children, Adolescence, and the Media. Thousands

Oaks, SAGE, Canada.

274

Strauss, AL 1987, Qualitative Analysis for Social Scientists, Cambridge University

Press, Cambridge.

Strauss, A & Corbin, J 1994, Grounded Theory Methodology: An Overview', Strauss, A

& Corbin, J (eds), Hand Book of Qualitative Research, SAGE, California.

Strong, CA & Eftychia, S 2006 "The Influence of Family and Friends on Teenage

Smoking in Greece: Some Preliminary Findings", Marketing Intelligence & Planning,

vol.24, no.2, p. 119-127, viewed 17/05/07, Emerald Group Publishing Limited. Emerald

Full Text.

Sullivan, TN, Farrell, AD, Kliewer, W & Reynolds, MV 2007 "Exposure to Violence in

Early Adolescence: The Impact of Self-Restraint, Witnessing Violence, and

Victimization on Aggression and Drug Use." The Journal of Early Adolescence, vol. 27,

no. 3, p. 28, p. 296-323.

Suri, R & Phillip, MV 2004, "Impact of Gender Differences on the Evaluation of

Promotional Emails", Journal of Advertising Research, vol.44, no.460, p.1-9, viewed

05/02/08, WARC.

Sweeney, JC & Soutar, GN 2001, "Consumer Perceived Value: The Development of a

Multiple Item Scale", Journal of Retailing, vol.77, no.2, p. 203-220, viewed 12/01/08,

EBSCO Host Research Databases. ProQuest Social Scienece Journals.

Tabachnick, BG & Fidell, LS 2001, Clearing Up Your Act: Screening Data Prior to

Analysis, Tabachnick, BG & Fidell, LS (eds), Using Multivariate Statistics, Harper

Collins, New York.

Tanaka, JS.& Huba, GJ 1985, A Fit Index for Covariance Structure Models Under

Arbitrary GLS Estimation, British Journal of Mathematical and Statistical Psychology.

275

Tanaka, JS, 1993, "Multifaceted Conceptions of Fit in Structural Equations Models, in

Bollen, KA & Long, JS (Eds). Testing Structural Equation Models, p. 10-39, Newbury

Park, CA SAGE, Publication.

Tedesco, R. (2007). Cold Reality. Cableworld.

Tiggemann, M 2005, " Television and Adolescent Body Image: The Role of program

Content and Viewing Motivation", Journal of Social & Clinical Psychology, vol.24,

no.3, p. 361-381, viewed 05/02/08, EBSCO Host Research Database. Psychology and

Behavioural Sciences Collection.

Traeen, B, Nilsen, TS & Stigum, H 2006, "Use of Pornography in Traditional Media

and on the Internet in Norway", The Journal of Sex Research., vol.43, no.3, p.245-254,

viewed 09/02/08, ProQuest Social Science Journals.

Troseth, GL, Saylor, MM & Archer, AH 2006, "Young Children's Use of Video as a

Source of Socially Relevant Information", Child Development, vol.77, no.3, p.786-799,

viewed 27/12/07, EBSCO Host Research Database. Psychology and Behavioural

Science Collection.

Tufte, B 2003, "Children, Media and Consumption", Young Consumers, vol.5, no.1,

69-76, viewed 17/05/07, Emerald Group Publishing Ltd. WARC.

Tufte, B & Halling, J 2002, "The Gender Perspective: Children as Consumers in

Denmark", Young Consumers, vol.13, no.4, p. 1-12, viewed 05/20/08, WARC.

UNICEF 2005, Health, Education, Equality, Protection-Advance Humanity, UNICEF.

Uncles, M 2000, "Market Orientation", Australian Journal of Management, vol.25,

viewed 08/12/07, EBSCO Host Research Database, Business Source Premier.

276

Updegraff, KA, Debra A Madden-Derdich, Ana Ulloa Estrada, Lara J Sales & Leonard.,

SA 2002, "Young Adolescents' Experience with parents and Friends: Exploring the

connection", Academy Research Library vol.51, no. 1, p. 72-81 viewed 15/05/07,

ProQuest CSA LLC. ABI/NFORM Global.

Vakratsas, D & Ambler, T 1999, "How Advertising Works: What Do We Really

Know?", Journal of Marketing, vol.63, no. January, p. 26-43, viewed 05/07/07, EBSCO

Host Research Database. PsycINFO.

Van Eijck, K 2001, "Social Differentiation in Musical Taste Patterns", Social Forces,

vol.79, no.3, viewed 10/02/08, EBSCO Host Research Databases. Business Source

Premier.

Vandell, DL 2000, "Parents, Peer Groups, and Other Socialising Influences",

Developmental Psychology, vol.36, no.6, p.699-710.

Velleman, RDB, Templeton, LJ & Copello, AG 2005, "The Role of the Family in

Preventing and Intervening with Substance use and Misuse: A Comprehensive Review

of Family Interventions, with a Focus on Young People", Drug & Alcohol Review,

vol.24, no.2, p. 93-109.

Wagner, KV 2008, "What is Cognition", Aboutcom. Psychology

http://psychology.about.com/od/cindex/g/def_cognition.htm.

Wan, F & Faber, RJ 2003, "Perceived Impact of Thin Female Models in Advertising: A

Cross-Cultural Examination of Third Person Perception and its Impact on behaviours",

Asia Pacific Journal of Marketing and Logistics., vol.15, no.1/2, p. 51-74, viewed

09/01/08, ProQuest. Social Science Journals.

277

Wang, Q, Pomerantz, EM & Chen, H 2007, "The Role of Parents’ Control in Early

Adolescents" Psychological Functioning: A Longitudinal Investigation in the United

States and China', Child Development, vol.78, no.5, viewed 16/12/07, p. 1592-1610,

ProQuest Social Science Journal.

Ward, M & Rivadenyra, R 1999, "Contributions of Entertainment Television to

Adolescents" Sexual Attitudes and Expectations: The Role of Viewing Amount Versus

Viewer Involvement', The Journal of Sex Research., vol.36, no.3, p. 237-250, viewed

04/01/08, ProQuest Social Science Journal.

Ward, S 1974, "Consumer Socialisation", Journal of Consumer Research, vol. 1, no. 2,

p. 1-14.

Ward, S & Wackman, D 1974, "Consumer Socialisation: Initial Study Results",

Advances in Consumer Research, vol.1, no.1, p. 120-125, viewed 06/07/07.EBSCO

Host Research Database. Business Source Premier.

Wentzel, KR & Looney, L 2007, Socialisation in School Settings, Wentzel, KR &

Looney, L (eds), Handbook of Socialisation-Theory and Research, The Guilford Press,

New York.

Weber, R P 1990, Basic Content Analysis. California, SAGE Publication.

Werner, O & Schoepfle, GM 1987, Systematic Fieldwork, SAGE, California.

White, RC 1995, Social Class Differences in the Uses of Leisure, The University of

Chicago Press Chicago.

Wilder, G 2008, "Let's dance; Cover Story", Sydney Morning Herald., 5 April.

278

Wiley, DE 1973. The Identification Problem for Structural Equation Models in

Unmeasured variables, Goldberger & Duncan (eds)., Structural Equation Models in

Social Sciences. New York, Harcourt Brace.

Williams, LA & Burns, AC 2001, "Factors Affecting Children's Store Loyalty: An

Empirical Examination of Two Store Types", Journal of Applied Business Research,

vol.17, no.1, viewed 20/01/08, EBSCO Host Research Databases. Business Source

Premier.

Wiman, AR 1983, "Parental Influence and Children's Responses to Television

Advertising", Journal of Advertising, vol.12, no.1, p. 1-8, viewed 20/12/07, EBSCO

Host Research Database. Business Source Premier.

Windmiller, M, Lamber, N & Turie, E 1980, Moral Development and Socialisation

Allyn and Bacon Inc., United States of America

Wolin, LD 2003, "Gender Issues in Advertising - An Oversight Synthesis of Research:

1970-2002", Journal of Advertising Research, vol.43, no.1, p. 1-16, viewed 05/002/08,

WARC.

World Bank 2007, The Growth of Bangladesh, viewed 21/08/ 2007

http://go.worldbank.org/4HPB6A7P80

Wyllie, A, Zhang, JF & Casswell, S (1998). "Positive Responses to Televised Beer

Advertisement Associated with Drinking and Problems Reported by 18 to 29 years old."

vol. 93, no. 5.

Yavas, DU & Riecken, DG 1990, "Children's General, Product and Brand-Specific

Attitudes Towards Television Commercials", International Journal of Advertising &

Marketing to Children, vol.9, no.2, p.1-8, viewed 06/01/08, WARC.

279

Yoon, MT & Garma R 2006. "Reality Television Programs in Malaysia: A Dream Com

True." Sunway Academic Journal, vol. 3, p. 73-86.

Zaichkowsky, JL 1985, "Measuring the Involvement Construct", Journal of Consumer

Research, vol. 12, no. 3, p. 341-354, viewed 11/10/08, EBSCO Host Research

Database, Business Source Premier.

Zhang, YB & Harwood., J 2002, "Television Viewing Perception", Journal of

Broadcasting & Electronic Media vol.46, no.2, p. 245-265, viewed 09/01/08, ProQuest

Social Science Journals.

Zikmand, WG & Babin BJ 2007, Exploring Marketing Research, South Western

Cengage Learning.

Zikmund, W, 2000, Business Research Methods. Chicago, The Dydren Press.

Zimmerman, FJ & Christakis, DA 2005, "Children’s Television Viewing and Cognitive

Outcomes: A Longitudinal Analysis of National Data", Archives of Paediatrics and

Adolescent medicine, vol.159, no.7, p. 619-626, viewed 07/01/08, ProQuest, Social

Science Journal.

Zolo, P 2004, Getting wiser to Teens: More insights into Marketing to teenagers New

strategist publication, Inc., United states of America.

Zuckerman, PM, Ziegler, M & Stevenson HW 1978, "Children's Viewing of Television

and Recognition Memory of Commercials." Child Development, vol. 49, no. 1, p. 96-

104.

280

Appendix 1 Questionnaire

281

Role of reality TV as a consumer socialisation agent of

teenagers in a developing country

A Survey Questionnaire

Your completion of this questionnaire will be greatly appreciated.

All information will be treated in strict confidence.

Study Conducted By:

University of Western Sydney

SCHOOL OF MARKETING

On completion, Please return it to the researcher

If you have any questions regarding this questionnaire or the study please contact: Md. Ridhwanul Haq: (Phone) 61 2 98524127, Fax: 61 2 98524400, email: m.haq@uws.edu.au School of Marketing, Locked Bag 1797 Penrith South DC NSW 1797, University of Western Sydney, Australia. Any concerns about this research you have should be directed to: The Executive Officer, Ethics in Human Research Committee, University of Western Sydney, Rooms K.1.35 to K.1.53, First Floor, Building K, Second Avenue, Kingswood NSW 2747, Australia. Facsimile: 61 2 47360905.

Your completion of this questionnaire will be taken to indicate your consent to participate in the study.

282

Study of Reality TV (RTV) as Consumer Socialisation Agent

Explanatory Notes:

Coverage

This study will cover teenagers from urban middle and lower social classes in a developing country (Bangladesh).

Who should answer the questionnaire?

Teenagers, who watch reality TV (RTV), answer the questionnaire.

The questions

The questions require your considered opinion. Please answer the questions on the basis of your experiences of watching RTV, which results in consumer socialisation.

Some of the questions may appear similar, but your answer to each question will enable us to scientifically assess the responses and draw valid conclusions.

Confidentiality and use of data

All the data gathered through this survey will be held in the strictest confidence. The data gathered in this survey will be used in several reports, including a doctoral thesis. In all publications resulting from this survey only aggregated total figures will be published.

283

A. On a scale of 1 to 7, where 1 is strongly disagree and 7 is strongly agree, to what extent do you agree or disagree with the following statements? (Please circle the

number that best reflects your agreement and disagreement on the following

statements) Strongly Disagree Strongly Agree 1 2 3 4 5 6 7 Statement

Strongly Disagree Strongly Agree

I watch RTV 1 2 3 4 5 6 7 I pay close attention when I watch RTV 1 2 3 4 5 6 7 I read about RTV in news papers 1 2 3 4 5 6 7 I would attend a public appearance if my favorite RTV character from the show were there

1 2 3 4 5 6 7

I imagine how I would act in a similar situation shown on RTV

1 2 3 4 5 6 7

I watch other programs that include contestants from the RTV show

1 2 3 4 5 6 7

I wish I could be on RTV shows 1 2 3 4 5 6 7 RTV is just like real life 1 2 3 4 5 6 7 I wish I could watch RTV on web 1 2 3 4 5 6 7 RTV provides me a lot of information 1 2 3 4 5 6 7 RTV make me aware about particular brands

1 2 3 4 5 6 7

I understand product usage instruction from RTV

1 2 3 4 5 6 7

RTV helps me to recall about certain products

1 2 3 4 5 6 7

RTV is a good way to learn about what products/bands are available in market

1 2 3 4 5 6 7

The opinion of my favorite RTV celebrities influence me to say positive things about products or brands to other people

1 2 3 4 5 6 7

Advertisements shown in RTV gives me new product ideas

1 2 3 4 5 6 7

I prefer the brands that sponsor RTV to increase my brand awareness

1 2 3 4 5 6 7

RTV makes me creative to do some thing new (e.g., makeup, hair style)

1 2 3 4 5 6 7

284

A. On a scale of 1 to 7, where 1 is strongly disagree and 7 is strongly agree, to what extent do you agree or disagree with the following statements? (Please circle the

number that best reflects your agreement and disagreement on the following

statements) Strongly Disagree Strongly Agree 1 2 3 4 5 6 7 Statement Strongly Disagree Strongly Agree

I get fashion tips from RTV 1 2 3 4 5 6 7 I like the product shown on RTV 1 2 3 4 5 6 7 I often buy clothing styles that I have seen in RTV

1 2 3 4 5 6 7

My favorite RTV celebrities make me fashion conscious

1 2 3 4 5 6 7

I want to get good figure like RTV celebrities (e.g. slimmer, muscular)

1 2 3 4 5 6 7

I see my favorite RTV celebrity as my idol 1 2 3 4 5 6 7 I would like to have lunch with my favorite RTV celebrity

1 2 3 4 5 6 7

I feel very strongly about banning unethical products (e.g., tobacco, beer, alcohol) shown in RTV

1 2 3 4 5 6 7

I prefer to buy multinational brands if is shown by RTV

1 2 3 4 5 6 7

I like to own the newest things shown on RTV 1 2 3 4 5 6 7 I like to own the best things shown on RTV 1 2 3 4 5 6 7 It is better to have more allowance from parents to buy the brand shown on RTV

1 2 3 4 5 6 7

I often request my parents to get the products shown on RTV

1 2 3 4 5 6 7

I often request my parents to get the products shown on advertisements in RTV shows

1 2 3 4 5 6 7

Some RTV shows promote use of prohibited items (e.g., alcohol, cigarette) among teenagers

1 2 3 4 5 6 7

I talk to my parents about RTV content 1 2 3 4 5 6 7 My parents and I discuss about RTV show selection

1 2 3 4 5 6 7

My parents and I watch RTV on weekdays 1 2 3 4 5 6 7 My parents and I watch RTV during weekends 1 2 3 4 5 6 7 My parents know what I watch on RTV 1 2 3 4 5 6 7 My parents prohibit me to watch certain RTV shows

1 2 3 4 5 6 7

285

A. On a scale of 1 to 7, where 1 is strongly disagree and 7 is strongly agree, to what extent do you agree or disagree with the following statements? (Please circle the

number that best reflects your agreement and disagreement on the following

statements) Strongly Disagree Strongly Agree 1 2 3 4 5 6 7 Statement Strongly Disagree Strongly Agree

I talk to my friends about the RTV shows 1 2 3 4 5 6 7 My friends and I talk about content of RTV shows (e.g., entertainment, sex, information)

1 2 3 4 5 6 7

My friends and I watch often RTV together 1 2 3 4 5 6 7 My friends think the content of RTV shows can provide useful information

1 2 3 4 5 6 7

My friends think content of RTV are entertaining

1 2 3 4 5 6 7

My peers’ opinions on RTV shows are important to me

1 2 3 4 5 6 7

286

B. For each of the following questions, please tick (� ) the answers that apply to

you from the options provided,: a. In which of these age groups you belong to? 13-15 years 16-19 years b. What is your gender? Male Female c. In which class you are currently in? _____________ d. What is the highest educational level your father completed? No Formal Education Bellow SSC SSC HSC Graduate and above e. What is the highest educational level your mother completed? No Formal Education Bellow SSC SSC HSC Graduate and above f. What is your father’s occupation? House duties Blue collar job White collar job Others (please specify) --- g. What is your mother’s occupation? House duties Blue collar job White collar job Others (please specify) --- h. In which of the following groups your total annual household income falls into (Taka)? 30,000-50,000 51,000-70,000 71,000-90,000 90,000 Plus I. In which suburb you live in?--------------------------------------

THANK YOU VERY MUCH FOR YOUR ASSISTANCE AND CO-OPERATION

287

Appendix 2 Flyer for Volunteers

288

Volunteers Needed for Focus Group Discussion If you are male or female

1) from middle and lower social class background 2) age between 13-19 years 3) can speak and read Bengali

---then you may wish to participate in a PhD research conducted by University of Western Sydney titled “Role of reality TV as a consumer socialisation agent of teenagers in a developing

country”

This research is the first of its kind done in Bangladesh which will explore the knowledge, attitude and values about consumption behaviors of teenagers who are involved in reality TV (RTV). Also effect of RTV will be measured considering the impact of social class differences, gender differences, and parental control on teenagers’ RTV watching behaviour in Dhaka city, capital of Bangladesh.

The study is conducted by the School of Marketing and approved by the Human Ethics Committee at University of Western Sydney (Ref no. H6890). You will be invited to participate in one group discussion held in Dhanmandi, Dhaka which will take about 2 hours and will be conducted by highly trained people. Participation in this research is totally voluntary. Your privacy is our priority! So you do not have to provide your name, contact details or any personal information in the discussion as it will be audio recorded. You can withdraw your participation at any time during the discussion. Should you need more information and/or to express your interest in participation, please contact Md. Ridhwanul Haq on 01710895896/0423640048 or email m.haq@uws.edu.au. The date and time of the group discussion will be negotiated once you have expressed your interest in participating.

289

Appendix 3

Research Protocol

290

University of Western Sydney

SCHOOL OF MARKETING

Dr. Syed H. Rahman Parramatta Campus

Senior Lecturer Room: EI.G.17 PO Box: 2116 NSW, Australia Phone: 61 2 9685 9162 Fax: 61 2 98524400 Email: s.rahman@uws.edu.au PARTICIPANT INFORMATION STATEMENT

Research Project Title: Role of reality TV as a consumer socialisation agent of teenagers in a

developing country

(1) What is the study about?

This research aims to gain a better understanding of teenagers consumption related cognition, attitude, and values through their involvement with reality television. The study is done involving teenagers, both boys and girls, who are13-19 years old in Dhaka city.

(2) Who is carrying out the study?

The study is being conducted by Md. Ridhwanul Haq and will form the basis for the degree of Doctor of Philosophy at University of Western Sydney under the supervision of Dr. Syed H. Rahman and Associate professor Felicitas Evangelista.

(3) What does the study involve?

The study involves you participating in discussion among a focus group involving up to 10 (5 boys and 5 girls) teenagers from Dhanmandi area to be facilitated by highly a trained person. The topics and questions discussed will be relevant to the research aims and will not be pointed at any one participant. You do not have to identify yourself by name and you will be requested not to identify others by name at any time during the discussion. You can respond only to the questions that you want to. The discussion will be audio recorded for transcribing purposes only and will not identify any individual participant. These focus group discussions will be organised around Dhanmondi area.

291

As your participation is voluntary, you will not be paid for this participation. However, at the end of the discussion session light refreshments will be served. (4) How much time will the study take? The focus group discussion may take up to maximum of 2 hours. (5) Can I withdraw from the study?

Being in this study is completely voluntary - you are not under any obligation to consent and - if you do consent - you can withdraw at any time without prejudice or penalty. If you withdraw from participation during the discussion, any audio information that you have provided during the discussion, till the point of withdrawal, can not be separated and destroyed as your responses are not individually identifiable. Any other information you may have given up to that point will be destroyed. (6) Will anyone else know the results?

All aspects of the study, including results, will be strictly confidential and only the researchers will have access to information on participants. A report of the study may be submitted for publication, but individual participants will not be identifiable in such a report

(7) Will the study benefit me?

Your participation in this study may have benefits for the entire teenaged population of Dhaka city and in other similar teenaged population in all around Bangladesh.

(8) Can I tell other people about the study?

Yes.

(9) What If I require further information?

When you have read this information, Md. Ridhwanul Haq will discuss it with you further and answer any questions you may have. If you do like to know more at any stage, please feel free to contact Md. Ridhwanul Haq, PhD candidate on 01711895896/0423640048 or email m.haq@uws.edu.au or Dr. Syed H.Rahman on 61 2 9685 9162 or email s.rahman@uws.edu.au. (10) What if I have a complaint or concern? Any concerns about this research you have should be directed to: The Executive Officer, Ethics in Human Research Committee, University of Western Sydney, Rooms K.1.35 to K.1.53, First Floor, Building K, Second Avenue, Kingswood NSW 2747, Australia. Facsimile: 61 2 47360905. This information sheet is for you to keep

292

Appendix 4 Participant Consent Form

293

University of Western Sydney

SCHOOL OF MARKETING

Parramatta Campus Room: EI.G.17 PO Box: 2116 NSW, Australia Phone: 61 2 9685 9162 Fax: 61 2 98524400 Email: s.rahman@uws.edu.au PARTICIPANT CONSENT FORM

I,--------------------------------------, give consent to my participation in the research project Title: Role of reality TV as a consumer socialisation agent of teenagers in a

developing country

In giving my consent I acknowledge that:

1. The procedures required for the project and time involved have been explained to me, and any questions I have about the project have been answered to my satisfaction.

2. I have read participant Information Statement and have been given the opportunity to discuss the information and my involvement in the project with the researcher/s.

3. I understand that my involvement in this study is strictly confidential and no information about me will be used in any way that reveals my identity.

4. I understand that being in this study is completely voluntary - I am not under any obligation to consent.

5. I consent to the audio recording of the interview. 6. I understand that I can withdraw from the study at any time if I do not wish to

continue and any recorded information provided by me during the discussion, till the point of withdrawal, cannot be separated and destroyed as my responses are not individually identifiable. Any other information that I have given to the interviewer up to this point will be destroyed.

Signed:-------------------------------------- First Name:--------------------------------- Surname:----------------------------------- Date:---------------------------------

294

Appendix 5

Focus Group Discussion Script

295

Focus Group Discussion Script: Consumer socialisation by reality TV

(CSRTV)

A) PRE-AMBLE (10 minutes) � Thanks & welcome by researcher � Nature of a focus group (informal, multi-way, all views acceptable, disagree) � There are no right or wrong answers – all about finding out what people think � Participants’ viewing � Going to be talking about the involvement of teenagers with TV particularly RTV

� Questions or concerns?

B) INTROS & WARM-UP (20 minutes) I’d like to go round the room and have you introduce yourselves … � First name � Do you understand the concept of RTV? � Best thing of watching RTV � Worst thing about watching RTV � How many days/hours you watch RTV? � What are some of the most common types of RTV you watch? BRIEFLY EXPLORE C) WATCHING OF RTV FEATURES (30 minutes) � Thinking now about RTV features, I’d like to start by making a list of all the

features you watch, learn and follow that influence your buying behaviors- anything that RTV affect your buying decisions, etc

� Are there any features RTV doesn’t have that might help you to take proper buying decisions but you wish to have that features?

EXPLORE

D) MOTIVATIONS FOR WATCHING RTV (30 minutes) � You’ve all been invited here because you watch TV, RTV … � What motivated you to watch RTV that helps you to think critically about your

buying decisions? EXPLORE � What do you think are some of the reasons that teenagers would consider RTV as

source of information for teenagers’ buying decisions? � What would be the single biggest reason? And some mention some others. EXPLORE

296

E) PARENTAL INVOLVEMENT AND ROLE OF RTV SELECTION (20 minutes) � Do you watch RTV with your parents? � What is the attitude of your parents about RTV? � How you parents restrict and share with you about RTV? CLOSING EXERCISE (10 minutes) � How you watch RTV and what are the nature of this involvement with RTV? � Finally, I’d like your creativity for a few minutes – to come up with ideas … � Don’t worry about whether it’s a good idea or a bad idea. � Supposing RTV sponsors want to encourage you to buy their product tomorrow

…What could they do? � Just call out anything at all that occurs to you – obvious, profound, serious, silly,

whatever … EXPLORE & REFINE � Thank the respondents and close the session.

297

Appendix 6 Parental Consent Form

298

Consent Form

Name of Research Project: Role of reality TV as a consumer socialisation agent of teenagers in a developing country Name, Address and Phone No. of Investigator: Md. Ridhwanul Haq

School of Marketing University of Western Sydney Lockd Bag Penrith South DC NSW 1797

1. I, ................................................................... consent to my son/daughter’s participation in the research project titled Role of reality TV as a consumer socialisation agent of teenagers in a developing country. 2. I understand that my son/daughter is free to withdraw their participation in the research at any time. 3. The purpose of the research has been explained to me and I have read and understood the written explanation given to me. I have been given the opportunity to ask questions about the research and received satisfactory answers. 4. I understand that any information or personal details gathered in the course of this research about my son/daughter is confidential and that neither their name nor any other identifying information will be used or published without my written permission. 5. I understand that if I have any complaints about this research I can contact: The Executive Officer Ethics in Human Research Committee University of Western Sydney Kingswood NSW 2747 Phone: 61 2 47360905 Signed by: .............................................................................................. Date.......................................................................................

299

Appendix 7

Permission from Schools in Dhaka

300

301

302

Appendix 8

A Proposed Model for this Research

303

Peer

Group

inf luences

pi6e39

1

1pi5e38

1pi4e37

1pi3e36

1pi2e35

1pi1e34

1

Parental

control

pc5e44

11

pc4e43

1pc3e42

1pc2e41

1pc1e40

1

pc6e45

1

RTV

Involvement

inv1 e25

1

1

inv2 e26

1

inv3 e27

1

inv4 e28

1

inv5 e29

1

inv6 e30

1

inv7 e31

1

inv8 e32

1

inv9 e33

1

cognition

cog1 e1

1

1

cog2 e2

1

cog3 e3

1

cog4 e4

1

cog5 e5

1

cog6 e6

1

cog7 e7

1

cog8 e8

1

cog9 e9

1

Attitude

att1 e10

1

1

att2 e11

1

att3 e12

1

att4 e13

1

att5 e14

1

att6 e15

1

att7 e16

1

att8 e17

1

att9 e18

1

Values

val1 e19

1

1

val2 e20

1

val3 e21

1

val4 e22

1

val5 e23

1

val6 e24

1

res 1

1

res 21

res 31

res 4

1

304

Cognition

cog1

1.77

e1

1.00

1

cog2

1.78

e2

1.04

1

cog3

1.46

e31.13

1

cog4

1.82

e41.181

cog5

1.92

e5

.96 1

cog6

2.89

e6

.841

cog7

1.94

e7

1.03

1

cog8

2.30

e8

1.00

1

cog9

2.90

e9

.91

1

Attitude

att1

2.27

e10

1.00

1

att2

1.63

e11

1.05

1

att3

2.84

e121.06

1

att4

1.96

e131.081

att5

3.32

e14

.91 1

att6

2.67

e15

.921

att7

3.96

e16

.92

1

att8

6.04

e17

.28

1

att9

2.43

e18

.85

1

Values

val1

2.53

e19

1.00

1

val2

2.65

e20.831

val3

3.83

e21.93 1

val4

1.54

e22

1.411

val5

1.37

e23

1.401

val6

5.71

e24

.17

1

RTV

Inv

inv1

2.74

e25

1.00

1

inv2

2.56

e26

1.05

1

inv3

3.66

e271.15

1

inv4

3.93

e281.161

inv5

2.87

e29

1.14 1

inv6

2.60

e30

1.211

inv7

2.85

e31

1.321

inv8

2.51

e32

.87

1

inv9

3.89

e33

1.05

1.70

.93

.26

.37

.471.51

Peers'

Inf luences

pi6

2.43

e39

1.00

1pi5

1.81

e38

.901

pi4

1.71

e37

1.021pi3

3.58

e36.921

pi2

3.76

e35 .881

pi1

2.13

e34

.99

1

.19

Parental

control

pc6

5.08

e45

1.00

1pc5

3.14

e44

2.031

pc4

1.67

e43

3.561pc3

2.09

e423.441

pc2

1.44

e41 3.971

pc1

2.45

e40

3.25

1

.50

.53

.40

res 11

.53

res 2

1

.45

res 31

1.11

res 4

1

305

Appendix 9

Computer Output of the SEM Analysis of the Empirically

Tested Model (Main Model)

306

Cognition

cog 1 e1

1

1

cog 2 e2

1

cog 3 e3

1

cog 4 e4

1

cog 5 e5

1

cog 6 e6

1

Attitude

att 1 e7

1

1

att 2 e8

1

att3 e9

1

att 4 e10

1

att5 e11

1

val 1 e12

1

values

val 1 e 1311

val 2 e14

1

val 3 e 15

1

val 4 e 16

1

RTV

Inv

inv 1 e 17

1

1

inv 2 e 18

1

inv 3 e 19

1

inv 4 e 20

1

inv 5 e 21

1

inv 6 e 22

1

inv7 e 23

1

inv 8 e 24

1

inv 9 e 25

1Parental

control

pc 5e 36

1

1pc 4e 35

1pc 3e34

1pc 2e 33

1pc 1e 32

1

Peers

Inf luences

pi 6e 31

1

1

pi 5e 30

1

pi 4e 29

1

pi 3e 28

1

pi 2e 27

1

pi 1e 26

1

res 1

1

res 2

1

res 3

1

res 4

1

307

.54

Cognition

.62

cog 1 e1

.78.53

cog 2 e2.73 .45

cog 3 e3.67

.45

cog 4 e4

.67

.36

cog 5 e5

.60

.37

cog 6 e6

.61

.60

Attitude

.49

att 1 e7

.70.46

att 2 e8.68 .41

att3 e9.64

.36

att 4 e10

.60

.26

att5 e11

.51

.50

val 1 e12

.70

.35

values

.38

val 1 e 13.62 .65

val 2 e14.81

.52

val 3 e 15

.72

.25

val 4 e 16

.50

.50

RTV

Inv

.27

inv 1 e 17

.52

.23

inv 2 e 18

.48

.24

inv 3 e 19

.49.32

inv 4 e 20.56 .19

inv 5 e 21.44

.21

inv 6 e 22

.45

.32

inv7 e 23

.57

.22

inv 8 e 24

.47

.27

inv 9 e 25

.52

.74

.60

.30

.39

.21

Parental

control

.20

pc 5e 36

.45.45

pc 4e 35

.67

.52

pc 3e34

.72

.60

pc 2e 33 .77

.67

pc 1e 32

.82

Peers

Inf luences

.38

pi 6e 31

.62.26

pi 5e 30

.51

.40

pi 4e 29

.64

.41

pi 3e 28.64

.48

pi 2e 27 .70

.24

pi 1e 26

.49

.25

.67

res 1

res 2

res 3

res 4

308

Appendix 10

Multi-Group Analysis Output of Gender Differences

309

Standardised Regression Weights (Male)

Estimate Involvement <--- Parental_control .269 Involvement <--- Peer_group_influences .630 Cognition <--- Involvement .756 Values <--- Involvement .609 Attitude <--- Involvement .425 Attitude <--- Values .420 Attitude <--- Cognition .058 closeatten <--- Involvement .586 readrtvinpapers <--- Involvement .508 watchrtv <--- Involvement .476 otherprogrtv <--- Involvement .515 watchrtvweb <--- Involvement .432 rtvreallife <--- Involvement .524 couldbeonrtv <--- Involvement .580 publicapparancertvcharacter <--- Involvement .506 imaginertv <--- Involvement .525 instructrtv <--- Cognition .769 rtvecallpro <--- Cognition .719 rtvawarebr <--- Cognition .654 rtvlotinfo <--- Cognition .668 rtvlearnpro <--- Cognition .611 tvconrtv <--- Cognition .609 rtvcelebmakesfashionconscious <--- Attitude .715 fashiontipsrtv <--- Attitude .661 clothingstylesrtv <--- Attitude .591 rtvmakescreative <--- Attitude .559 figurelikertvceleb <--- Attitude .535 likeprortv <--- Attitude .704 multibrandsbyrtv <--- Values .548 newestthingsonrtv <--- Values .854 bestthingsonrtv <--- Values .720 allowancefromparents <--- Values .483 parentsknowrtv <--- Parental_control .396 talktoparents <--- Parental_control .681 rtvonweekdays <--- Parental_control .814 rtvonweekends <--- Parental_control .804 parentsnidiscuss <--- Parental_control .827 peersopiniononrtv <--- Peer_group_influences .609 friendsnirtvtogether <--- Peer_group_influences .656 talktofriends <--- Peer_group_influences .653 friendsthinkrtventert <--- Peer_group_influences .687 friendsthinkrtvuseful <--- Peer_group_influences .701 talktofriendsrtvcontent <--- Peer_group_influences .437

310

Goodness of Fit (Male)

Goodness of Fit Values Chi-Square 900.660 Degree of Freedom 587 P .000 Normed Chi-square (CMIN/df) 1.534

Root mean square of error of estimation (RMSEA) .054

Goodness of fit index (GFI) .789

Adjusted of goodness-of-fit index (AGFI) .760

Normed fit index (NFI) .698

Tucker-Lewis Index (TLI) .857

Comparative fir index (CFI) .866

Empirically Tested Model for Multi-Group Analysis (Male)

cognition

cog3 e1

1

1

cog4 e2

1

cog2 e3

1

cog1 e4

1

cog5 e5

1

cog7 e6

1

values

att9 e1311

val1 e14

1

val2 e15

1

val3 e16

1

involvement

inv2 e17

1

1

inv3 e18

1

inv1 e19

1

inv6 e20

1

inv9 e21

1

inv8 e22

1

inv7 e23

1

inv4 e24

1

inv5 e25

1

attitude

att4 e7

1

1

att1 e8

1

att3 e9

1

cog9 e10

1

att5 e11

1

att2 e12

1

peers'

inf luences

pi6e31

1

1pi3e30

1pi1e29

1pi5e28

1pi4e27

1pi2e26

1

parental

control

pc5e36

1

1pc1e35

1pc3e34

1pc4e33

1pc2e32

1

res 1

1

res 2

1

res 3

1

res 4

1

311

Standardised Regression Weights (Female)

Estimate

Involvement <--- Peer_group_influences .754

Involvement <--- Parental_control .133

Cognition <--- Involvement .729

values <--- Involvement .598

Attitude <--- Involvement .186

Attitude <--- values .374

Attitude <--- Cognition .339

closeatten <--- Involvement .454

readrtvinpapers <--- Involvement .443

watchrtv <--- Involvement .499

otherprogrtv <--- Involvement .599

watchrtvweb <--- Involvement .470

rtvreallife <--- Involvement .359

couldbeonrtv <--- Involvement .551

publicapparancertvcharacter <--- Involvement .421

imaginertv <--- Involvement .508

instructrtv <--- Cognition .797

rtvecallpro <--- Cognition .726

rtvawarebr <--- Cognition .692

rtvlotinfo <--- Cognition .691

rtvlearnpro <--- Cognition .609

tvconrtv <--- Cognition .613

rtvcelebmakesfashionconscious <--- Attitude .680

fashiontipsrtv <--- Attitude .690

clothingstylesrtv <--- Attitude .673

rtvmakescreative <--- Attitude .642

figurelikertvceleb <--- Attitude .538

likeprortv <--- Attitude .710

multibrandsbyrtv <--- values .684

newestthingsonrtv <--- values .770

bestthingsonrtv <--- values .731

allowancefromparents <--- values .511

parentsknowrtv <--- Parental_control .486

talktoparents <--- Parental_control .625

rtvonweekdays <--- Parental_control .621

rtvonweekends <--- Parental_control .716

parentsnidiscuss <--- Parental_control .807

peersopiniononrtv <--- Peer_group_influences .626

friendsnirtvtogether <--- Peer_group_influences .371

talktofriends <--- Peer_group_influences .614

312

Estimate

friendsthinkrtventert <--- Peer_group_influences .592

friendsthinkrtvuseful <--- Peer_group_influences .695

talktofriendsrtvcontent <--- Peer_group_influences .564

Goodness of Fit (Female)

Goodness of Fit Values Chi-Square 1088.097 Degree of Freedom 587 P .000 Normed Chi-square (CMIN/df) 1.854

Root mean square of error of estimation (RMSEA) .068

Goodness of fit index (GFI) .763

Adjusted of goodness-of-fit index (AGFI) .732

Normed fit index (NFI) .642

Tucker-Lewis Index (TLI) .776

Comparative fir index (CFI) .792

Empirically Tested Model for Multi-Group Analysis (Female)

cognition

cog3 e1

1

1

cog4 e2

1

cog2 e3

1

cog1 e4

1

cog5 e5

1

cog7 e6

1

attitude

att4 e7

1

1

att1 e8

1

att3 e9

1

cog9 e10

1

att5 e11

1

att2 e12

1

values

att9 e131

1

val1 e14

1

val2 e15

1

val3 e16

1

involvement

inv2 e17

1

1

inv3 e18

1

inv1 e19

1

inv6 e20

1

inv9 e21

1

inv8 e22

1

inv7 e23

1

inv4 e24

1

inv5 e25

1

peers'

inf luences

pi6e31

1

1pi3e30

1pi1e29

1pi5e28

1pi4e27

1pi2e26

1

parental

control

pc5e36

1

1pc1e35

1pc3e34

1pc4e33

1pc2e32

1

res 1

1

res 2

1

res 3

1

res 4

1

313

Appendix 11

Multi-Group Analysis Output of Social Class Differences

314

Standardised Regression Weights (Middle Social Class)

Estimate

Involvement <--- Parental_control .231

Involvement <--- Peer_group_influences .715

Cognition <--- Involvement .740

Values <--- Involvement .554

Attitude <--- Involvement .274

Attitude <--- Values .410

Attitude <--- Cognition .251

closeatten <--- Involvement .646

readrtvinpapers <--- Involvement .578

watchrtv <--- Involvement .581

otherprogrtv <--- Involvement .644

watchrtvweb <--- Involvement .459

rtvreallife <--- Involvement .465

couldbeonrtv <--- Involvement .581

publicapparancertvcharacter <--- Involvement .484

imaginertv <--- Involvement .563

instructrtv <--- Cognition .809

rtvecallpro <--- Cognition .751

rtvawarebr <--- Cognition .797

rtvlotinfo <--- Cognition .754

rtvlearnpro <--- Cognition .636

tvconrtv <--- Cognition .695

rtvcelebmakesfashionconscious <--- Attitude .781

fashiontipsrtv <--- Attitude .780

clothingstylesrtv <--- Attitude .724

rtvmakescreative <--- Attitude .676

figurelikertvceleb <--- Attitude .546

likeprortv <--- Attitude .751

multibrandsbyrtv <--- Values .677

newestthingsonrtv <--- Values .858

bestthingsonrtv <--- Values .740

allowancefromparents <--- Values .595

parentsknowrtv <--- Parental_control .401

talktoparents <--- Parental_control .669

rtvonweekdays <--- Parental_control .767

rtvonweekends <--- Parental_control .840

parentsnidiscuss <--- Parental_control .741

peersopiniononrtv <--- Peer_group_influences .600

friendsnirtvtogether <--- Peer_group_influences .518

talktofriends <--- Peer_group_influences .686

315

Estimate

friendsthinkrtventert <--- Peer_group_influences .706

friendsthinkrtvuseful <--- Peer_group_influences .714

talktofriendsrtvcontent <--- Peer_group_influences .682

Goodness of Fit (Middle Social Class)

Goodness of Fit Values Chi-Square 996.690 Degree of Freedom 587 P .000 Normed Chi-square (CMIN/df) 1.698

Root mean square of error of estimation (RMSEA) .062

Goodness of fit index (GFI) .776

Adjusted of goodness-of-fit index (AGFI) .746

Normed fit index (NFI) .724

Tucker-Lewis Index (TLI) .852

Comparative fir index (CFI) .862

Empirically Tested Model for Multi-Group Analysis (Middle Social Class)

cognition

cog3 e1

1

1

cog4 e2

1

cog2 e3

1

cog1 e4

1

cog5 e5

1

cog7 e6

1

attitude

att4 e7

1

1

att1 e8

1

att3 e9

1

cog9 e10

1

att5 e11

1

att2 e12

1

values

att9 e131

1

val1 e14

1

val2 e15

1

val3 e16

1

involvement

inv2 e17

1

1

inv3 e18

1

inv1 e19

1

inv6 e20

1

inv9 e21

1

inv8 e22

1

inv7 e23

1

inv4 e24

1

inv5 e25

1

peers'

influences

pi6e31

1

1

pi3e30

1

pi1e29

1

pi5e28

1

pi4e27

1

pi2e26

1

parental

control

pc5e36

1

1

pc1e35

1

pc3e34

1

pc4e33

1

pc2e32

1

res 1

1

res 2

1

res 3

1

res 4

1

316

Standardised Regression Weights (Lower Social Class)

Estimate

Involvement <--- Parental control .230

Involvement <--- Peer group influences .578

Cognition <--- Involvement .697

values <--- Involvement .563

Attitude <--- Involvement .336

Attitude <--- values .363

Attitude <--- Cognition .170

rtvcelebmakesfashionconscious <--- Attitude .648

fashiontipsrtv <--- Attitude .532

clothingstylesrtv <--- Attitude .531

rtvmakescreative <--- Attitude .452

figurelikertvceleb <--- Attitude .465

likeprortv <--- Attitude .635

closeatten <--- Involvement .392

readrtvinpapers <--- Involvement .436

watchrtv <--- Involvement .399

otherprogrtv <--- Involvement .394

watchrtvweb <--- Involvement .364

rtvreallife <--- Involvement .373

couldbeonrtv <--- Involvement .500

publicapparancertvcharacter <--- Involvement .368

imaginertv <--- Involvement .517

multibrandsbyrtv <--- values .547

newestthingsonrtv <--- values .745

bestthingsonrtv <--- values .660

allowancefromparents <--- values .360

instructrtv <--- Cognition .755

rtvecallpro <--- Cognition .683

rtvawarebr <--- Cognition .552

rtvlotinfo <--- Cognition .559

rtvlearnpro <--- Cognition .524

tvconrtv <--- Cognition .477

parentsknowrtv <--- Parental_control .567

talktoparents <--- Parental_control .604

rtvonweekdays <--- Parental_control .634

rtvonweekends <--- Parental_control .700

parentsnidiscuss <--- Parental_control .855

peersopiniononrtv <--- Peer group influences .620

friendsnirtvtogether <--- Peer group influences .470

talktofriends <--- Peer group influences .616

317

Estimate

friendsthinkrtventert <--- Peer group influences .585

friendsthinkrtvuseful <--- Peer group influences .658

talktofriendsrtvcontent <--- Peer group influences .284

Goodness of Fit (Lower Social Class)

Goodness of Fit Values Chi-Square 991.583 Degree of Freedom 587 P .000 Normed Chi-square (CMIN/df) 1.689

Root mean square of error of estimation (RMSEA) .061

Goodness of fit index (GFI) .774

Adjusted of goodness-of-fit index (AGFI) .744

Normed fit index (NFI) .576

Tucker-Lewis Index (TLI) .746

Comparative fir index (CFI) .763

Empirically Tested Model for Multi Group Analysis (Lower Social Class)

cognition

cog3 e1

1

1

cog4 e2

1

cog2 e3

1

cog1 e4

1

cog5 e5

1

cog7 e6

1

attitude

att4 e7

1

1

att1 e8

1

att3 e9

1

cog9 e10

1

att5 e11

1

att2 e12

1

involvement

inv2 e17

1

1

inv3 e18

1

inv1 e19

1

inv6 e20

1

inv9 e21

1

inv8 e22

1

inv7 e23

1

inv4 e24

1

inv5 e25

1

peers'

inf luences

pi6e31

1

1

pi3e30

1

pi1e29

1

pi5e28

1

pi4e27

1

pi2e26

1

parental

control

pc5e36

1

1

pc1e35

1

pc3e34

1

pc4e33

1

pc2e32

1

res 1

1

res 2

1

res 3

1

values

att9 e1311

val1 e14

1

val2 e15

1

val3 e16

1

res 4

1

318

Appendix 12

List of Constructs and Variables

319

Constructs Variables RTV Involvement I watch RTV

I pay close attention when I watch RTV I read about RTV in news papers I would attend a public appearance if my favorite RTV character from the show were there I imagine how I would act in a similar situation shown on RTV I watch other programs that include contestants from the RTV show I wish I could be on RTV shows RTV is just like real life I wish I could watch RTV on web

Consumption Related Cognition RTV provides me a lot of information RTV make me aware about particular brands I understand product usage instruction from RTV RTV helps me to recall about certain products RTV is a good way to learn about what products/bands are available in market The opinion of my favorite RTV celebrities influence me to say positive things about products or brands to other people Advertisements shown in RTV gives me new product ideas I prefer the brands that sponsor RTV to increase my brand awareness RTV makes me creative to do some thing new (e.g., makeup, hair style)

Consumption Related Attitude I get fashion tips from RTV I like the product shown on RTV I often buy clothing styles that I have seen in RTV My favorite RTV celebrities make me fashion conscious I want to get good figure like RTV celebrities (e.g. slimmer, muscular)

320

I see my favorite RTV celebrity as my idol I would like to have lunch with my favorite RTV celebrity I feel very strongly about banning unethical products (e.g., tobacco, beer, alcohol) shown in RTV I prefer to buy multinational brands if is shown by RTV

Consumption Related Values I like to own the newest things shown on RTV I like to own the best things shown on RTV It is better to have more allowance from parents to buy the brand shown on RTV I often request my parents to get the products shown on RTV I often request my parents to get the products shown on advertisements in RTV shows Some RTV shows promote use of prohibited items (e.g., alcohol, cigarette) among teenagers

Parental Control

I talk to my parents about RTV content My parents and I discuss about RTV show selection My parents and I watch RTV on weekdays My parents and I watch RTV during weekends My parents know what I watch on RTV My parents prohibit me to watch certain RTV shows

Peer Group Influences I talk to my friends about the RTV shows My friends and I talk about content of RTV shows (e.g., entertainment, sex, information) My friends and I watch often RTV together My friends think the content of RTV shows can provide useful information My friends think content of RTV are entertaining My peers’ opinions on RTV shows are important to me

321

Appendix 13 Questionnaire (Bengali Version)

322

������������� �����������������������������������

���������������(RTV)�������������� �����������������!�

��

������������ ������������������������������������������ �

���������!�����"��#$��%&�"�#$%���������������#��'�(��&�� �

�

�

��� ��&�

University of Western Sydney

SCHOOL OF MARKETING

�

�

�����������������%�&��)"������������!������*��+���������&,-�.-�������'���� ����#������������������(��#�)�������������*������+����+�������"�,��������+��-.&����/���0!���1�&�2��3��4�56�7�89:7;67<2�3$�=4�56�7�89:7;;>>2��?���14�m.haq@uws.edu.au�,@A1�B3�����C'2� 1�DE� �$�*� 6<8<� �(F� ����#� EG�G� ��G��GE�H)2� 6<8<2� �)�I��C� B�� 0�!��FC� �E�2�B�JK1!� �����*��+F����$������������������������#���1������L�M����!����*����*������B����������������4���C�&N�B3���2����� ���$�!��,O� �I�*���2��)�I��C�B��0�!��FC� �E�2��P�?���?6GQ:��#�����?6G:Q2� R�N!� �1�2� �S'?� ��2� �I�)?72� �'�)E2� ��G��GE�H)� 7<;<2� B��K1!�2� 3$�=?� 56� 7�;<Q5>8>: �������������� �����B'�*�&F���������$������*��+F�!����������T��*U&��&�1� �

323

������������� �������������������������������(RTV)��/0�1��

��� ��

�

������!������2���

�

��� ����3!&�

• ���*��+F����C����)��!��N1�������V��'1����W��X�����$�"�0�����"������������N�������Y�

����1��&�� �

�

�����'�������!���4������5�

• �����������N2��������!1�I�(RTV)����Z�#��� �

�

���������

• ���*��+F���������%&�����������.-�����������$������� RTV ��Z���BI[����B��1�����!������

�����L�����\�1���)"����02������������I�L��������N��F���]!�!�I%�������Z �

• ��� *��+F��� ��������� ��A� ������%&� ������� ����� �����N!� ���� &��� ����2� �.̂� ������� ���"� ����

�������)"�����*��+F���31�3�1���(����#��#�I%������Z����&�!��&�� �

�

��)#������������

���*��+F�!����_��#$��1N�B�$.-��*���N!�������#��'�(��&�� �����#$�)��"��%&��I���*��+F�!��#�`�

�G���GE�#�����$�a��&�� �����N(���#�����0!����1��#$���.b��31�31�&�����*��+F�!��������

&�� �

324

�W� �����L��,@�1�6?<2���Z����6�c�B�$.-�B��T����'�<�c�B�$.-���T��%���B#C���������� ������L��.-�$\�1������#��A��������������R������+F�������)�1HZ����6����������4��/7�8������������29��#���������:���������������������;4�

"�#$%�"�/<�������������������������������������������������������������������������������������������"�#$%��/<�

6� 7� Q� ;� :� 5� <�

� � � � � � �

�$%�#� "�#$%�"�/<���"�#$%��/<�

B���RTV ��Z � 6���7���Q���;���:���5���<�RTV ��Z�����!�B���*IN���������*���!�#�� � 6���7���Q���;���:���5���<�B�������!�RTV �']�.-�Z����d � 6���7���Q���;���:���5���<�B�������!�RTV �$L�e���������B��f����)�,g��#������&�1�B��0���Z����)�,g��#�������h������� �

6���7���Q���;���:���5���<�

B������!��RTV-����+!�,�̂��!�I�����2�i����+���,g����#���1�B���������� �

6���7���Q���;���:���5���<�

RTV ��������B�$������B��f����B'�*�&F����1�����B�����Z � 6���7���Q���;���:���5���<�B����������&!����B���RTV?��B��f����B'�*�&F������������ � 6���7���Q���;���:���5���<�RTV �����$����,-��������Z�! � 6���7���Q���;���:���5���<�B����������&!����B����h������RTV ��Z���������j� 6���7���Q���;���:���5���<�RTV B�������X����#$���! � 6���7���Q���;���:���5���<�RTV B��������C���F$��%&��� ��C����F����! � 6���7���Q���;���:���5���<�B���RTV����Z���F$���$�&����� ��C���Y������ � 6���7���Q���;���:���5���<�RTV B��������C���F$��� ��C������������&��$���� � 6���7���Q���;���:���5���<����������C�����F$��� ��C���������$�RTV ���I��1�����$� � 6���7���Q���;���:���5���<�RTV ���������������B������B�$������������K$��k���$��������������.-�$��������I������� �

6���7���Q���;���:���5���<�

RTV �������C���[����B��������A���F$��� ��C����F����! 6���7���Q���;���:���5���<�B������K$�k��+!�����F���Ul���(���B���RTV �����m��$���N����f������K$�k���B�������h���� �

6���7���Q���;���:���5���<�

RTV B��������F$���$�&����� ��C��U���N1�&�����&��$�����V�#�`����B��2��A�1����'W �

6���7���Q���;���:���5���<�

B���RTV �#���3$�����+!�������C���� � 6���7���Q���;���:���5���<�B���RTV �������C���F$��%&���h���� � 6���7���Q���;���:���5���<�B������Y����Y RTV �������C���I���J���1�����d��� � 6���7���Q���;���:���5���<�B��������h���RTV ��������B������3$������������� � 6���7���Q���;���:���5���<�B���RTV ��������������&���mh��C$�V�#�`������2����N�X1W��������h���� �

6���7���Q���;���:���5���<�

B���B��������h���RTV ����������!��$L�e�&����������� � 6���7���Q���;���:���5���<�B���B��������h���RTV ����������#���������Z������Z�����h������� � 6���7���Q���;���:���5���<�

325

�W� �����L��,@�1�6?<2���Z����6�c�B�$.-�B��T����'�<�c�B�$.-���T��%���B#C���������� ������L��.-�$\�1������#��A��������������R������+F�������)�1HZ����6����������4��/7�8������������29��#���������:���������������������;4�

"�#$%�"�/<� "�#$%��/<��6� 7� Q� ;� :� 5� <�� � � � � � �

�$%�#� "�#$%�"�/<��������"�#$%��/<�

B�� RTV �������C��Bn������F$��V�#�`���2��!��W�������������1�I����B��#C���� �

6���7���Q���;���:���5���<�

B���RTV �������C���X��������f������K$�k������B�������h���� �

6���7���Q���;���:���5���<�

B���RTV �������C�������!���A���F$��������h���� � 6���7���Q���;���:���5���<�B���RTV �������C�������!�I��1���F$��������h���� � 6���7���Q���;���:���5���<�RTV �������C����A���K$��k���F$���������$�BII�����������#�������&��Z��������B�����h���� �

6���7���Q���;���:���5���<�

RTV�� ����C���F$���������$�B���B�����BII��������������Y����Y���!�����

����6���7���Q���;���:���5���<

RTV ����[���������C���F$���������$�B���B�����BII��������������Y����Y���!����� �

6���7���Q���;���:���5���<�

�����������N�������Y���A�RTV B��f���Bn������F$��V�#�`���2��!��W�������o�! �

6���7���Q���;���:���5���<�

B���B�����BII���������#�RTV�����+!�,�̂��!�B��1������� �

6���7���Q���;���:���5���<�

B�����BII������'�B���M�����RTV?��������B��f�����Z��� � 6���7���Q���;���:���5���<�B�����BII������'�B������C������RTV���Z � 6���7���Q���;���:���5���<�B�����BII������'�B����A������RTV ��Z � 6���7���Q���;���:���5���<�B�����BII���������B���RTV �����B��f�����Z � 6���7���Q���;���:���5���<�B�����BII����B����������������RTV B��f�����Z�����+����� �

6���7���Q���;���:���5���<�

B���B������pA�������#�RTV ���B��f�����!��#���1 � 6���7���Q���;���:���5���<�B������pA�����'�B���RTV ����+!�,�̂��!�B�1������V�#�`�������2���=2��#$��+!�W�

6���7���Q���;���:���5���<�

B������pA�����'�B������Y����Y�������#�RTV ��Z � 6���7���Q���;���:���5���<�B������pA�����������RTV ����+!�,�̂�%&����!���N!��#$���! � 6���7���Q���;���:���5���<�B������pA�����������RTV ����+!�,�̂�%&��������%1� � 6���7���Q���;���:���5���<�RTV ���B��f����� ��C�B���B������pA������������\�P�e���� � 6���7���Q���;���:���5���<��

326

=;�������������8�������*��4���������=����>���6����;�����29���?@

A@ �����������������B ������C��:���"�DE����5�

��6Q?6:����� ��65?68�����

F@ ���������G5�

� �����1� �����!�

H@ �:�������I������J�5�___________________�

K@ ����������������1�-2���3�����+��#������L=��M�

� �����������f�����(������ ��B���C����$���

� �����$��� ��)������$���

� ����������'�����C�

N@ ����������������1�-2���3�����+��#������L=��M�

� �����������f�����(������ ��B���C����$���

� �����$��� ��)������$���

� ����������'�����C�

O@ ���������������������L=��M�

� ��*U&��C� �����������N��V�#�`�(A�K��$���!N2����Z�����K��2�,bq�B��!�����A���

��$��W�

� ��)��������N��V�#�`�E�L��2��(�2��$���!N���$��W���B�$��$�

P@ ���������������������L=��M�

� ��*U&��C� �����������N��V�#�`�(A�K��$���!N2����Z�����K��2�,bq�B��!�����A�����

��$��W�

� ��)��������N��V�#�`�E�L��2��(�2��$���!N���$��W���B�$��$�

Q@ ����������� 1�����"�����������������"������"�DE������6���;5�

� ��Q>2>>>?:>2>>>� ��:62>>>?<>2>>>�

� ��<62>>>?8>2>>>� ��8>2>>>���'�����C�

R@ �:����������������������5�_____________________________�

�������������������������'������!����/0��������+��������#��������"�� ���#����

327

Appendix 14 List of Middle and Lower Social Class Suburbs

328

List of Middle and Lower social classes Suburbs and Respective Schools in Dhaka

#) Suburbs are Highlighted as Selected from Two Social Classes A) Middle Social Class Suburb 1) Mohammad poor 2) Shamoly 3) Dhanmondi 4) Kalabagan 5) Hatirpool 6) Shah bagh 7) Palashi (Selected for survey)

8) Azimpoor 9) Lalmatia 10) Firm Gate 11) Majgbazar 12) Malibagh 13) Khilgaon 14) Rampoora 15) Zikatalla B) Lower Social Class Suburb

1) Adabar 2) Hazaribagh 3) Rayer Bazar 4) Nilkhet (Selected for Survey) 5) Nawabpoor 6) Badda