Thesis_Muhammad Asif KHAN

Transcription

AIX-MARSEILLE UNIVERSITE
INSTITUT D’ADIMINISTRATION DES ENTREPRISES
Ecole Doctorale de Sciences Economiques et de Gestion d’Aix-Marseille – ED 372
Centre d’Etudes et de Recherche en Gestion d’Aix-Marseille (CERGAM)
Pre-Purchase Determinants of Brand
Avoidance: A Study of Negative
Attitudes and Intentions
Dissertation defended on 21 May, 2012 in partial fulfillment of the requirements
for the degree of Doctorate in Management Sciences by
Muhammad Asif KHAN
Jury
Directeur de recherche:
M. Dwight MERUNKA
Professeur, IAE d’Aix en Provence
Aix-Marseille Université
Co-directeur de la recherche:
Mme. Cécile BOZZO
Maître de Conférences, IAE d’Aix en Provence
Aix-Marseille Université
Rapporteurs:
M. Pierre VALLETTE-FLORENCE
Professeur, IAE de Grenoble
Université Pierre Mendès France, Grenoble
Mme. Géraldine MICHEL
Professeur, IAE de Paris
Université PARIS 1 – Panthéon Sorbonne
Dedication
This thesis is dedicated to my parents (late) and family, especially to my mother, who gave
me unconditional love, encouragement, and support. My dream came true due to her love,
prayers, and sacrifices. She was anxiously waiting for me to complete this thesis and rejoin
her but she passed away on 27 November 2010. I always miss her.
I love you Mom!!!
Muhammad Asif KHAN
“If you have a mom, there is nowhere you are likely
to go where a prayer has not already been”
– Robert Brault
ii
Acknowledgements
First of all, thanks to my creator who made my work easy by blessing me with all potentials,
talents and capabilities to accomplish the task of writing my thesis.
This thesis was an extremely challenging academic voyage and there are several people who
helped me in successfully completing it. First of all, I would like to thank my supervisor
Professor Dwight MERUNKA from IAE, Graduate School of Management, Aix-Marseille
Université, who provided valuable guidance, useful suggestions and motivated me with
positive and enthusiastic spirits all along the way. His insightful comments, suggestions and
specialized knowledge guided me to refine my research tremendously. I am of equal gratitude
to my co-supervisor Professor Cécile BOZZO, IAE, Graduate School of Management, AixMarseille Université, for being available at the time of need, for reading through multiple
refinements of this thesis. She has been a source of encouragement for the completion of this
hard work.
I would like to express my gratitude to all the jury members for being a part of the jury;
especially to both the ‘rapporteurs’ Professor Pierre VALLETTE-FLORENCE and Professor
Géraldine MICHEL for sharing their valuable comments.
Special thanks go to Professor Elyette ROUX, Director CERGAM, and Professor Jean-Louis
CHANDON, for their boundless support and encouragement during my PhD program,
especially for feedback and comments during quarterly marketing workshops. I am also
thankful to the administration staff of IAE especially to Ms. Brigitte MARCELLIN, Ms.
Leonie GADES, Ms. Marie LAVIRON, Ms. Michelle COMB and M. Said BOUDJEMAI for
logistics support throughout the research program.
iii
I also want to acknowledge the financial support from Higher Education Commission (HEC)
of Pakistan, and providing me all the necessary arrangements to pursue my studies in France.
I would like to thank my family members for their support, trust and confidence in my
capabilities and for always praying for me. Besides this I am also very thankful to my friends
in PAKISTAN who have been appreciating and supporting me throughout my research, and
especial thanks to Akram, Amir, Ramiz, Rizwan, and Yasir for helping me in data collection
from Pakistan at different phases during my research period.
It is a great honor to have a liaison with Professor Michael Shyue Wai Lee from ‘The
University of Auckland’, New Zealand. I am thankful to him for motivating and developing
interest in learning theories about my domain of interest and help me in data collection from
New Zealand.
I would like to extend my sincere appreciation to my colleagues and friends Abdul-Karim,
Ali, Aneela, Asad, Aziz, Basma, Charlène, Daniel, Elena, GM Kundi, Haroon, Houda, Imran,
Ishtiaq, Jamal, Julie, Junaid, Magalie, Marie-aude, Naeem, Niaz, Omer, Richard, Rizwan,
Rohail, Samina, Sana, Sergio, Shakaib, Skander, Tareq, Thuy VO-THI, Tingting, Zeeshan
Bhatti, Zeeshan Qader, and Zulqarnain, who commented on my research work and shared
their knowledge during meetings and for all the great moments we enjoyed during the time of
my stay in France.
Finally, I would like to acknowledge my beautiful soul mate Dalila, who has progressed from
friend, to girlfriend, to partner, to fiancé, during this challenging time. Thank you for your
ongoing love, support, and understanding, and yes, it is finally finished!
iv
Abbreviations
Abbreviation
What They Stand For?
A
Animosity
BA
Brand Avoidance
BAA
Brand Avoidance Attitude
BAI
Brand Avoidance Intention
CFI
Comparative Fit Index
COO
Country –of-origin
DVs
Dependent Variables
ETH
Ethnocentrism
FCO
Familiarity with country-of-origin
IVs
Independent Variables
MFI
Monthly Family Income
MM
Measurement Model
NSI
Negative Social Influence
NWoM
Negative word-of-mouth
NZ
New Zealand
PK
Pakistan
PR
Perceived Risk
PUCC
Perceived Unethical Corporate Citizenship
RMSEA
Root Mean-Square Error of Approximation
SEM
Structural Equation Modeling
SME
Small to Medium Enterprise
SR
Structural Regression
SRMR
Standardized Root Mean-Square Residual
TLI
Tucker-Lowis Index
US
Undesired self
v
Abstract
Brands are considered as asset, therefore, are of considerable importance in both marketing
academia and practice. Although most research has focused on the positive attitudes and
behaviors that consumers have towards brands, there is a growing interest in anticonsumption. This thesis contributes to anti-consumption research by exploring the
phenomenon of pre-purchase brand avoidance from a consumers’ perspective.
The current research investigates the reasons for pre-purchase brand avoidance behavior.
Because of the exploratory nature of the problem, after reviewing literature about the brand
avoidance phenomenon, empirical investigation has been done in this thesis. This thesis is a
combination of qualitative (in-depth interviews) and quantitative (field-survey) research
efforts. The themes emerged from 13 in-depth interviews, supported with literature, proposed
a research model. Consumers’ undesired self congruence (US), negative social influence
(NSI), animosity (A), perceived unethical corporate citizenship (PUCC) and perceived risk
(PR) have been identified as the predictors of consumers’ attitude toward pre-purchase brand
avoidance attitude. Further, consumers’ familiarity with country-of-origin plays a moderating
role on the relationship between animosity and brand avoidance attitude; similarly,
consumers’ ethnocentrism moderates the relationship between animosity and brand avoidance
attitude.
In empirical testing of proposed research model, data was collected from two countries
(Pakistan = 278 and New Zealand = 198) through online web-survey. Structural equation
modeling (SEM) has been used to measure the psychometric properties of the scales and for
hypotheses testing. Data was pooled (n = 476) and sampling weights has been used in pooled
vi
data to make the samples representative of target populations. Measurement invariance has
been assessed to establish configural and metric invariance between the two samples.
The results show that undesired self congruence, animosity and perceived risk positively
influence consumers’ brand avoidance attitude, in pooled data whereas, negative social
influence has mixed results in Pakistan and New Zealand. Consumers’ perceived unethical
corporate citizenship has no significant effect on brand avoidance attitude in all three data
sets.
Concerning moderating variables, it is found in all three data sets that the relationship
between animosity and consumers’ brand avoidance attitude is stronger with high familiarity
with country-of-origin and vice versa. Similarly, the relationship between animosity and
consumers’ brand avoidance attitude was stronger for high ethnocentric consumers for New
Zealand and pooled data, whereas for Pakistan, this interaction was not significant.
Combined, the findings of this thesis provide a comprehensive appreciation of why
consumers avoid certain brands before experience. Overall, this thesis contributed knowledge
to the growing field of anti-consumption research by providing an overview and an integrative
understanding based on empirically tested conceptual model on an under-explored domain of
brand avoidance.
Key Words: Brand avoidance, consumer behavior, undesired self congruence, negative
social influence, animosity, perceived risk, familiarity with country-of-origin, and consumer
ethnocentrism.
vii
Résumé
Les marques sont considérées comme un atout, donc, sont d'une importance considérable dans
le milieu universitaire du marketing ainsi que dans la pratique. Bien que la plupart des
recherches ont porté sur le comportement et l’attitude positif des consommateurs envers la
marque, il y a un intérêt croissant pour l’anti-consommation .Cette thèse contribue à la
recherche de l’anti-consommation en explorant le phénomène de l’évitement de la marque au
pré-achat du point de vue du consommateur.
La recherche actuelle étudie les raisons du comportement d’évitement de marque au préachat. En raison de la nature exploratoire du problème, et après avoir consulté la revue
littéraire sur le phénomène de l’évitement de la marque, une investigation empirique a été
appliquée sur cette thèse. Cette thèse est une combinaison
de méthodes qualitatives
(entretiens en profondeurs) et quantitatives (enquêtes dans le terrain). Les thèmes qui ont
émergé des 13 entretiens en profondeur, ont constitué avec l’appui de la littérature le model
de recherche. La Congruence Indésirable du Soi du Consommateur (CIS), l’Influence Sociale
Négative (ISN), l’Animosité (A), la Citoyenneté Inéthique Perçue de l’Entreprise (CIPE) et
le Risque Perçu (RP) ont été identifiés comme des facteurs prédictifs ‘de l’attitude du
consommateur envers le phénomène de l’évitement de la marque au pré-achat. En outre la
familiarité du consommateur avec le pays d’origine joue un rôle modérateur sur la relation
entre l’animosité et
l’attitude de l’évitement de la marque, de la même manière
l’ethnocentrisme du consommateur modère la relation
entre l’animosité et l’attitude de
l’évitement de la marque.
Dans les tests empiriques du modèle de recherche proposé, les données ont été recueillies en
ligne de deux pays différents (le Pakistan = 278 et la Nouvelle-Zélande = 198). La
viii
modélisation par équations structurelles (SEM) a été utilisée pour mesurer les propriétés
psychométriques des échelles et pour tester des hypothèses. Les données ont été collectées (n
= 476) et les poids d'échantillonnage ont été utilisés dans les données regroupées pour créer
des échantillons représentatifs de la population cible. L’invariance de mesure a été évaluée
pour établir l'invariance configurale et métrique entre les deux échantillons.
Les résultats montrent que la congruence indésirable du soi, l’animosité et le risque perçu
influencent positivement l’attitude de l’évitement de la marque, tandis que dans les données
regroupées, l’influence sociale négative a des résultats différents au Pakistan et NouvelleZélande. La citoyenneté inéthique perçue de l’entreprise
n'a pas d'effet significatif sur
l'attitude d'évitement de marque dans les trois ensembles de données
Concernant les variables modératrices, les résultats ont démontré dans les groupes de données
que la relation entre l’animosité et l’attitude de l’évitement de la marque est plus forte avec
une grande familiarité avec le pays d’origine et vice versa. Pareil, la relation entre l’animosité
et l’attitude de l’évitement de la marque du consommateur est plus forte avec les
consommateurs les plus ethnocentriques pour Nouvelle-Zélande et les données regroupées,
tandis que pour le Pakistan, cette interaction n'est pas significative.
En conclusion, les résultats de cette thèse fournissent une appréciation globale des raisons
pour lesquelles les consommateurs évitent certaines marques avant l'expérience. Dans
l'ensemble, cette thèse a contribué à la connaissance du domaine en plein essor de l'anticonsommation en donnant un aperçu et une compréhension intégrée, basée sur le modèle
conceptuel testé empiriquement sur un domaine sous-exploré de l'évitement de la marque.
Mots clés: Marque d'évitement, le comportement des consommateurs, la congruence
indésirable du soi, l'influence sociale négative, l'animosité, le risque perçu, la familiarité avec
le pays d'origine et de l'ethnocentrisme des consommateurs
ix
Table of Contents
DEDICATION .................................................................................................................................................... II
ACKNOWLEDGEMENTS ................................................................................................................................... III
ABBREVIATIONS .............................................................................................................................................. V
ABSTRACT ....................................................................................................................................................... VI
RESUME ........................................................................................................................................................ VIII
TABLE OF CONTENTS........................................................................................................................................ X
LIST OF TABLES .............................................................................................................................................. XV
LIST OF FIGURES........................................................................................................................................... XVII
INTRODUCTION ..............................................................................................................................................18
RESEARCH BACKGROUND ........................................................................................................................................ 18
PROBLEM IDENTIFICATION ....................................................................................................................................... 20
SIGNIFICANCE OF THE STUDY .................................................................................................................................... 21
ACADEMIC AND PRACTICAL AIMS AND OBJECTIVES....................................................................................................... 22
ORGANIZATION OF THE THESIS ................................................................................................................................. 27
PART-1: THEORETICAL FRAMEWORK & MODEL BUILDING…………………………………………………………...30
CHAPTER 1: LITERATURE REVIEW....................................................................................................................32
1.1 OVERVIEW OF THE CHAPTER ........................................................................................................................... 32
1.2 DEFINITION OF BRAND .................................................................................................................................. 32
1.3 DEFINITION OF BRAND AVOIDANCE ................................................................................................................. 37
1.4 ANTI CONSUMPTION AND CONSUMER RESISTANCE............................................................................................. 41
1.5 CHARACTERISTICS OF ANTI-CONSUMERS ........................................................................................................... 43
1.5.1
General-societal: Global impact consumer.................................................................................... 44
1.5.2
General-personal: Simplifier .......................................................................................................... 44
1.5.3
Brand-societal: Market activist...................................................................................................... 45
1.5.4
Brand-personal: Anti-loyal consumers........................................................................................... 45
1.6 CONCLUSION ............................................................................................................................................... 48
CHAPTER 2: QUALITATIVE STUDY....................................................................................................................49
2.2 QUALITATIVE APPROACH ............................................................................................................................... 49
2.3 ANALYSIS METHOD ...................................................................................................................................... 50
2.4 EMERGING THEMES ...................................................................................................................................... 54
2.4.1
Undesired Self Congruence ............................................................................................................ 54
2.4.2
Negative Social Influence............................................................................................................... 55
2.4.3
Animosity ....................................................................................................................................... 56
2.4.4
Country-of-Origin........................................................................................................................... 57
2.4.5
Ethnocentrism................................................................................................................................ 58
2.4.6
Perceived Unethical Corporate Citizenship .................................................................................... 58
2.4.7
Perceived Risk ................................................................................................................................ 60
x
2.5
2.6
DISCUSSION AND CONCLUSION OF THE QUALITATIVE STUDY ................................................................................. 61
CONCLUSION ............................................................................................................................................... 63
CHAPTER 3: THEORETICAL FRAMEWORK & HYPOTHESES ...............................................................................64
3.2 DETERMINANTS OF BRAND AVOIDANCE ............................................................................................................ 64
3.3 PRE-PURCHASE DETERMINANTS OF BRAND AVOIDANCE ...................................................................................... 66
3.3.1
Undesired Self Congruence ............................................................................................................ 67
3.3.2
Negative Social Influence............................................................................................................... 74
3.3.3
Animosity ....................................................................................................................................... 78
3.3.4
Perceived Unethical Corporate Citizenship .................................................................................... 82
3.3.5
Perceived Risk ................................................................................................................................ 85
3.4 BRAND AVOIDANCE ATTITUDE AND BRAND AVOIDANCE INTENTION ......................................................................... 89
3.5 MODERATING VARIABLES ............................................................................................................................... 91
3.5.1
Familiarity with COO...................................................................................................................... 91
3.5.2
Consumer Ethnocentrism............................................................................................................... 94
3.6 SOCIO-DEMOGRAPHIC CHARACTERISTICS .......................................................................................................... 97
3.7 RESEARCH MODEL ....................................................................................................................................... 98
3.8 CONCLUSION ............................................................................................................................................... 98
PART-II: RESEARCH METHODOLGY, EMPIRICAL ANALYSIS & RESULTS…………………………..………….…100
CHAPTER 4: RESEARCH METHODOLOGY ......................................................................................................103
4.1 OVERVIEW OF THE CHAPTER ......................................................................................................................... 103
4.2 RESEARCH CONTEXT ................................................................................................................................... 103
4.3 RESEARCH DESIGN...................................................................................................................................... 105
4.4 POPULATION OF THE STUDY ......................................................................................................................... 105
4.4.1
Target Population ........................................................................................................................ 105
4.4.2
Accessible Population .................................................................................................................. 105
4.5 SAMPLE.................................................................................................................................................... 106
4.6 VARIABLES OF THE STUDY ............................................................................................................................ 107
4.7 MEASUREMENT OF CONSTRUCTS .................................................................................................................. 108
4.7.1
Brand Avoidance.......................................................................................................................... 108
4.7.2
Undesired Self Congruence .......................................................................................................... 109
4.7.3
Negative Social Influence............................................................................................................. 110
4.7.4
Animosity ..................................................................................................................................... 111
4.7.5
Perceived Unethical Corporate Citizenship .................................................................................. 111
4.7.6
Perceived Risk .............................................................................................................................. 112
4.7.7
Familiarity with Country-of-origin ............................................................................................... 113
4.7.8
Consumer Ethnocentrism............................................................................................................. 113
4.8 PILOT STUDY ............................................................................................................................................. 117
4.9 DATA COLLECTION ..................................................................................................................................... 118
4.9.1
Through Emails ............................................................................................................................ 119
4.9.2
Through Social Media Networks.................................................................................................. 119
4.10
CONCLUSION ........................................................................................................................................ 120
CHAPTER 5: DATA ANALYSIS .........................................................................................................................121
5.2 DATA ANALYSIS: PAKISTAN SAMPLE ............................................................................................................... 122
5.2.1
Preliminary Data Analysis ............................................................................................................ 122
xi
5.2.2
Analytical Methods and Procedure.............................................................................................. 125
5.2.3
Confirmatory Factor Analysis....................................................................................................... 127
5.2.4
CFA: Independent Variables......................................................................................................... 127
5.2.5
CFA: Dependent Variables ........................................................................................................... 132
5.2.6
CFA: Full Measurement Model .................................................................................................... 135
5.2.7
Common Method Bias ................................................................................................................. 137
5.3 DATA ANALYSIS: NEW ZEALAND SAMPLE ........................................................................................................ 139
5.3.1
Preliminary Analysis..................................................................................................................... 139
5.3.2
Analysis Technique and procedure .............................................................................................. 141
5.3.3
Confirmatory Factor Analysis....................................................................................................... 141
5.3.4
CFA: Independent Variables......................................................................................................... 141
5.3.5
CFA: Dependent Variables ........................................................................................................... 144
5.3.6
CFA: Full Measurement Model .................................................................................................... 147
5.3.7
Common Method Bias ................................................................................................................. 149
5.4 HYPOTHESES TESTING ................................................................................................................................. 153
5.5 MODERATION EFFECT ................................................................................................................................. 156
5.5.1
Moderating effect of familiarity with country-of-origin .............................................................. 157
5.5.2
Moderating effect of consumer ethnocentrism ........................................................................... 159
5.5.3
Socio-Demographic Effects .......................................................................................................... 161
5.6 CONCLUSION ............................................................................................................................................. 163
CHAPTER 6: DATA ANALYSIS & RESULTS .......................................................................................................164
6.2 SAMPLING WEIGHTS................................................................................................................................... 164
6.3 CFA WITH SAMPLE WEIGHTS ....................................................................................................................... 166
6.4 MEASUREMENT INVARIANCE ........................................................................................................................ 167
6.4.1
CFA: Measurement Model for Pooled Data ................................................................................. 170
6.5 HYPOTHESES TESTING ................................................................................................................................. 174
6.6 MODERATION EFFECT ................................................................................................................................. 176
6.6.1
Moderating effect of familiarity with country-of-origin .............................................................. 176
6.6.2
Moderating effect of consumer ethnocentrism ........................................................................... 177
6.7 RESULTS OF THE STUDY ............................................................................................................................... 179
6.7.1
Undesired self congruence and brand avoidance attitude .......................................................... 179
6.7.2
Negative social influence and brand avoidance attitude ............................................................ 181
6.7.3
Perceived animosity and brand avoidance attitude .................................................................... 184
6.7.4
Perceived unethical corporate citizenship and brand avoidance attitude ................................... 187
6.7.5
Perceived risk and brand avoidance attitude .............................................................................. 189
6.7.6
Brand avoidance attitude and brand avoidance intention .......................................................... 190
6.8 CONCLUSION ............................................................................................................................................. 191
CHAPTER 7: GENERAL DISCUSSIONS & CONCLUSION ....................................................................................193
7.1
7.2
7.3
7.4
7.5
7.6
7.7
OVERVIEW OF THE CHAPTER ......................................................................................................................... 193
SUMMARY OF THE RESEARCH FINDINGS .......................................................................................................... 193
CONTRIBUTIONS ........................................................................................................................................ 195
IMPLICATIONS ........................................................................................................................................... 197
LIMITATIONS ............................................................................................................................................. 200
FUTURE RESEARCH...................................................................................................................................... 202
CONCLUSION ............................................................................................................................................. 204
REFERENCES..................................................................................................................................................206
xii
APPENDICES..................................................................................................................................................236
APPENDIX I: PARTICIPANTS’ DETAIL-PRE-TESTING IN QUALITATIVE STUDY ....................................................................... 236
APPENDIX II: PRELIMINARY ANALYSIS OF SAMPLE 1.................................................................................................... 237
APPENDIX III: COMMON METHOD BIAS FOR SAMPLE 1............................................................................................... 240
APPENDIX IV: PRELIMINARY ANALYSIS OF SAMPLE 2 .................................................................................................. 241
APPENDIX V: COMMON METHOD BIAS FOR SAMPLE 2 ............................................................................................... 243
APPENDIX VI: QUESTIONNAIRE FOR PAKISTAN........................................................................................................... 244
APPENDIX VII: QUESTIONNAIRE FOR NEW ZEALAND................................................................................................... 248
APPENDIX VIII: MODERATION EFFECT...................................................................................................................... 249
APPENDIX IX: SAMPLE WEIGHTS ............................................................................................................................ 251
APPENDIX X: DISCRIMINANT VALIDITY WITH WEIGHTED SAMPLE .................................................................................. 264
APPENDIX XI: LIST OF BRANDS ............................................................................................................................... 265
APPENDIX XII: DISCRIMINANT VALIDITY-SEM BASED APPROACH .................................................................................. 267
APPENDIX XIII: RESUME DE THESE EN FRANÇAIS ........................................................................................................ 278
9.1 INTRODUCTION .......................................................................................................................................... 278
9.1.1
Contexte de la recherche ............................................................................................................. 278
9.1.2
Identification des problèmes ....................................................................................................... 279
9.1.3
Importance de l'étude.................................................................................................................. 280
9.1.4
Objectifs académiques et pratiques ............................................................................................ 281
9.2 CADRE THEORIQUE ET HYPOTHESES ................................................................................................................ 283
9.2.1
Définition de l’Evitement de la marque ....................................................................................... 283
9.2.2
Approche qualitative ................................................................................................................... 285
9.2.3
Méthode d'analyse ...................................................................................................................... 285
9.2.4
Thèmes émergents ...................................................................................................................... 285
9.2.5
Les déterminants pré-achats de l’évitement de la marque ......................................................... 286
9.2.6
La congruence du soi non désirée ................................................................................................ 286
9.2.7
L’influence sociale négative ......................................................................................................... 288
9.2.8
Animosité..................................................................................................................................... 289
9.2.9
Citoyenneté non éthique perçue de l’entreprise ......................................................................... 290
9.2.10
Risque perçu............................................................................................................................ 292
9.2.11
Attitude d’évitement de marque et Intention d’évitement de marque................................... 293
9.2.12
Variables modératrices ........................................................................................................... 293
9.2.13
Familiarité avec le pays d’origine............................................................................................ 294
9.2.14
Ethnocentrisme du consommateur ......................................................................................... 295
9.2.15
Caractéristiques sociodémographiques .................................................................................. 296
9.2.16
Modèle de recherche............................................................................................................... 297
9.3 METHODOLOGIE DE RECHERCHE.................................................................................................................... 298
9.3.1
Contexte de recherche ................................................................................................................. 298
9.3.2
Plan de recherche ........................................................................................................................ 298
9.3.3
Collection des données ................................................................................................................ 298
9.3.4
Dimension de construction .......................................................................................................... 299
9.4 ANALYSE DES DONNEES ET LES RESULTATS ....................................................................................................... 302
9.4.1
Procédure et méthodes analytiques ............................................................................................ 302
9.4.2
L’analyse factorielle confirmatoire .............................................................................................. 303
9.4.3
Hypothèses testées et résultats de l’étude .................................................................................. 303
9.4.4
Effets sociodémographiques........................................................................................................ 303
9.4.5
Congruence du soi indésirable et comportement d’évitement de marque.................................. 304
9.4.6
Influence social négative et le comportement d’évitement de la marque................................... 304
9.4.7
Animosité perçue et le comportement d’évitement de marque .................................................. 305
xiii
9.4.8
Citoyenneté non éthique perçue de l’entreprise et l'attitude d'évitement de marque ............... 307
9.4.9
Le risque perçu et de l'attitude d'évitement de marque .............................................................. 307
9.4.10
L’attitude d’évitement de la marque et l’intention d’évitement de la marque....................... 308
9.5 DISCUSSIONS GENERALES ET CONCLUSION....................................................................................................... 309
9.5.1
Contributions ............................................................................................................................... 309
9.5.2
Implications ................................................................................................................................. 310
9.5.3
Limites.......................................................................................................................................... 311
9.5.4
Les recherches futures ................................................................................................................. 312
9.5.5
Conclusion.................................................................................................................................... 313
xiv
List of Tables
TABLE 1.1: BRAND DEFINITIONS ................................................................................................................................... 34
TABLE 1.2: FOUR TYPES OF ANTI-CONSUMERS................................................................................................................. 43
TABLE 2.1 : PARTICIPANTS DETAIL (INTERVIEWED) ........................................................................................................... 52
TABLE 2.2: INTERVIEW GUIDE ...................................................................................................................................... 53
TABLE 3.1: MAJOR ANIMOSITY STUDIES ......................................................................................................................... 80
TABLE 3.2: DESCRIPTION AND DEFINITION OF PERCEIVED RISK FACETS ................................................................................. 86
TABLE 4.1: DEMOGRAPHIC, ECONOMIC AND LIFESTYLE INDICATORS .................................................................................. 104
TABLE 4.2: VARIABLES OF THE STUDY........................................................................................................................... 107
TABLE 4.3: QUESTIONNAIRE ITEMS WITH REFERENCES .................................................................................................... 115
TABLE 5.1: SOCIO-DEMOGRAPHIC CHARACTERISTICS (PK)............................................................................................... 124
TABLE 5.2: FIT INDICES CRITERIA ................................................................................................................................. 127
TABLE 5.3: FIT INDICES OF CFA FOR IVS (PK)................................................................................................................ 129
TABLE 5.4 : RELIABILITY OF SCALE FOR IVS (PK)............................................................................................................. 130
TABLE 5.5: DISCRIMINANT VALIDITY OF IVS (PK) ........................................................................................................... 131
TABLE 5.6: FIT INDICES OF DVS (PK) ........................................................................................................................... 132
TABLE 5.7: RELIABILITY OF SCALE OF DVS (PK).............................................................................................................. 133
TABLE 5.8: FIT INDICES FOR FULL MEASUREMENT MODEL (PK)........................................................................................ 135
TABLE 5.9: DISCRIMINANT VALIDITY OF FULL MEASUREMENT MODEL (PK) ........................................................................ 137
TABLE 5.10: DEMOGRAPHIC CHARACTERISTICS (NZ) ...................................................................................................... 140
TABLE 5.11: FIT INDICES OF CFA FOR IVS (NZ) ............................................................................................................. 142
TABLE 5.12: RELIABILITY OF SCALES FOR IVS (NZ) .......................................................................................................... 144
TABLE 5.13: DISCRIMINANT VALIDITY OF IVS (NZ)......................................................................................................... 144
TABLE 5.14 : FIT INDICES FOR DVS (NZ) ...................................................................................................................... 145
TABLE 5.15: RELIABILITY OF SCALES FOR DVS (NZ)......................................................................................................... 146
TABLE 5.16: FIT INDICES OF FULL MEASUREMENT MODEL (NZ) ....................................................................................... 147
TABLE 5.17 : DISCRIMINANT VALIDITY OF FULL MEASUREMENT MODEL (NZ)...................................................................... 148
TABLE 5.18: MEANS, STANDARD DEVIATIONS, AND CORRELATIONS AMONG STUDY VARIABLES (PK)....................................... 151
TABLE 5.19: MEANS, STANDARD DEVIATIONS, AND CORRELATIONS AMONG STUDY VARIABLES (NZ)....................................... 152
TABLE 5.20: STRUCTURAL MODEL FIT INDICES .............................................................................................................. 153
TABLE 5.21: HYPOTHESES TESTING ............................................................................................................................. 155
TABLE 5.22: UNSTANDARDIZED ESTIMATES FOR INTERACTION OF A AND FCO ..................................................................... 157
TABLE 5.23: UNSTANDARDIZED ESTIMATES FOR INTERACTION OF A AND ETH...................................................................... 160
xv
TABLE 5.24: SUMMARY OF HYPOTHESIS TESTING ............................................................................................................ 163
TABLE 6.1: SAMPLING WEIGHTS .................................................................................................................................. 166
TABLE 6.2: MODEL COMPARISONS FOR MEASUREMENT INVARIANCE................................................................................. 170
TABLE 6.3: FIT INDICES OF MEASUREMENT MODEL OF POOLED DATA ................................................................................ 170
TABLE 6.4: DISCRIMINANT VALIDITY OF MEASUREMENT MODEL OF POOLED DATA .............................................................. 171
TABLE 6.5: MEANS, STANDARD DEVIATIONS, AND CORRELATIONS AMONG STUDY VARIABLES OF POOLED DATA........................ 173
TABLE 6.6: STRUCTURAL MODEL FIT INDICES ................................................................................................................ 174
TABLE 6.7: HYPOTHESES TESTING ............................................................................................................................... 175
TABLE 6.8: UNSTANDARDIZED ESTIMATES FOR INTERACTION OF A AND FCO ....................................................................... 176
TABLE 6.9: UNSTANDARDIZED ESTIMATES FOR INTERACTION OF A AND ETH........................................................................ 178
TABLE 6.10: RESULTS OF HYPTHESIS 1 ......................................................................................................................... 179
TABLE 6.11: RESULTS OF HYPOTHESIS 2 ....................................................................................................................... 182
TABLE 6.12: SOCIO-DEMOGRAPHIC EFFECTS ................................................................................................................. 182
TABLE 6.19: SUMMARY OF RESULTS ............................................................................................................................ 192
xvi
List of Figures
FIGURE 0.1: STRUCTURE OF THE THESIS.......................................................................................................................... 29
FIGURE 0.3: FLOW CHART OF PART-I ............................................................................................................................. 31
FIGURE 1.1: CONCEPTUALIZATION OF CONSUMER BEHAVIOR ............................................................................................. 40
FIGURE 1.2: CONSUMPTION ANTI-CONSTELLATION........................................................................................................... 46
FIGURE 2.1: PRE-TEST RESULT OF QUESTION 1................................................................................................................ 51
FIGURE 2.2: PRE-TEST RESULT OF QUESTION 2................................................................................................................ 51
FIGURE 3.1: DETERMINANTS OF BRAND AVOIDANCE......................................................................................................... 65
FIGURE 3.2: PRE-PURCHASE (FOCUSED) DETERMINANTS OF BRAND AVOIDANCE ................................................................... 65
FIGURE 3.3: LEVELS AND FACTORS OF BRAND DISLIKE ....................................................................................................... 73
FIGURE 3.4: CONCEPTUAL RESEARCH MODEL .................................................................................................................. 98
FIGURE 3.5: FLOW CHART OF PART-II .......................................................................................................................... 102
FIGURE 5.1: SEQUENCE OF DATA ANALYSIS ................................................................................................................... 122
FIGURE 5.2: FLOW CHART FOR MODEL SPECIFICATION .................................................................................................... 126
FIGURE 5.3: FINAL CFA MODEL FOR IVS (PK)............................................................................................................... 129
FIGURE 5.4: CFA MODEL FOR DVS (PK) ...................................................................................................................... 133
FIGURE 5.5: ONE FACTOR MODEL OF DVS (PK)............................................................................................................. 134
FIGURE 5.6: CFA: FULL MEASUREMENT MODEL (PK) .................................................................................................... 136
FIGURE 5.7: FINAL CFA MODEL FOR IVS (NZ) .............................................................................................................. 143
FIGURE 5.8: CFA MODEL FOR DVS (NZ)...................................................................................................................... 145
FIGURE 5.9: ONE FACTOR MODEL FOR DVS (NZ) .......................................................................................................... 146
FIGURE 5.10: CFA OF FULL MEASUREMENT MODEL (NZ)............................................................................................... 149
FIGURE 5.11: STRUCTURAL MODEL RESULTS ON PAKISTAN SAMPLE .................................................................................. 154
FIGURE 5.12: STRUCTURAL MODEL RESULTS ON NEW ZEALAND SAMPLE ........................................................................... 154
FIGURE 5.13: MODERATION EFFECT OF FAMILIARITY WITH COO (PK) ............................................................................... 158
FIGURE 5.14: MODERATION EFFECT OF FAMILIARITY WITH COO (NZ) ............................................................................... 159
FIGURE 5.15: MODERATING EFFECT ON ETHNOCENTRISM (NZ) ........................................................................................ 161
FIGURE 6.1: MEASUREMENT MODEL OF POOLED DATA .................................................................................................. 172
FIGURE 6.2: STRUCTURAL MODEL RESULTS ON POOLED DATA.......................................................................................... 175
FIGURE 6.3: MODERATION EFFECT OF FAMILIARITY WITH COO......................................................................................... 177
FIGURE 6.4: MODERATION EFFECT OF ETHNOCENTRISM .................................................................................................. 178
xvii
Introduction
Introduction
“Write to be understood, speak to be heard, read to grow…”
---Lawrence Clark Powell
Research Background
Consumers are independent in their choices for products or brands. They could love some
brands, be indifferent or express negative opinions towards other brands. A considerable work
has been done by researchers in positive responses to brands. For example, brand love
(Albert, Merunka, & Valette-Florence, 2008; Carroll & Ahuvia, 2006; Whang, Allen,
Sahoury, & Zhang, 2004), brand attachment (Thompson, Rindfleisch, & Arsel, 2006;
Thomson, MacInnis, & Whan Park, 2005), brand passion (Fournier, 1998), brand satisfaction
(Fournier & Mick, 1999), and brand delight (Rust, Varki, & Oliver, 1997). Consumers have
the tendency to buy those brands with images congruent to their self-concepts, or those that
will give desired meaning to their lives (Aaker, 1999; Grubb & Harrison L. Grathwohl, 1967;
Levy, 1959; Sirgy, 1982; Solomon, 1983).
Whereas the literature on the negative side that studies why people avoid certain brands is
limited or narrowly defined (Dalli, Romani, & Gistri, 2006). This inequality is understandable
because companies and institutions are very much interested in practical results of the positive
form of the knowledge. They are interested in knowing what consumers want and willing to
buy (Dalli et al., 2006). At the same time, it is also difficult to justify this inequality on the
18
Introduction
theoretical level; in order to better understand the purchasing and consumption behaviors, it is
essential to take both positive and negative aspects into the same frame (Dalli et al., 2006).
When we pay close attention to the existing literature, we can see that major studies in Anticonsumption area are focusing on dissatisfaction with products and services (Banister &
Hogg, 2004; Oliver, 1980), or on counter-cultural phenomenon such as voluntary
simplification (Craig-Lees & Hill, 2002; Piacentini & Banister, 2009), boycotts (Klein, Smith,
& John, 2004; Kozinets & Handelman, 1998) and consumer resistance (Cherrier, 2009a;
Hogg, 1998; Kozinets, 2002; Zavestoski, 2002a).
In general Anti-consumption domain and particularly, brand avoidance topic has
progressively gain the attention and become more relevant to scholars, managers and
consumers. In the Journal of Business Research’s special issue on anti-consumption, anticonsumption was defined as ‘against consumption’ (Lee, Fernandez, & Hyman, 2009), they
posit that it is not synonymous with alternative, conscientious, or green consumption and also
that anti-consumption studies do not merely deal with ethics, sustainability or public policy.
Anti-consumption research focuses on the reasons against consumption instead of pro-social
movements. Generally, in consumer research the focus is on approach aspects of consumers’
attitudes and behaviors; for example, why people choose a product or brand. In contrast, the
focus of anti-consumption research is on the reasons, for which consumer avoid a product or
brand (Lee, Fernandez, et al., 2009). Most of the studies in anti-consumption area have been
on general products and services instead of specifically on brands; and also the research on
negative attitudes and their relative behaviors towards brands is fragmented into many
perspectives (Banister & Hogg, 2001, 2004; Englis & Solomon, 1997; Klein, 2000; Kozinets
& Handelman, 2004; Rumbo, 2002; Thompson & Arsel, 2004). So in our study, we will
provide a structure based on pre-purchase determinants of brand avoidance, and contribute to
an integrative understanding of the area.
19
Introduction
Problem Identification
Consumer is independent; s/he may or may not choose a brand. The basic question is what
does a consumer need from the product or brand? Of course, he wants to fulfill his need by
consuming the product or brand. If the product is able to satisfy his basic need with an
acceptable level of quality then the consumer is satisfied with it; he will continue to consume
that product or brand (Jacoby & Kyner, 1973). But now the world has become a global village
and consumers can find a range of different brands in almost every category fulfilling the
same desire with the required level of quality. Research emphasizes that apart from satisfying
the basic need, consumer also want to represent him/her self by consuming a certain product
or brand (Aaker, 1999; Grubb & Harrison L. Grathwohl, 1967; Levy, 1959; Sirgy, 1982;
Solomon, 1983). For that purpose marketing managers create an acceptable image of their
product or brand with which consumer can feel affiliation and can give a symbolic meaning to
public.
Marketing managers are always interested in the acceptance of their brand in the market and
to increase their market share and to maximize the profit is their ultimate goal or target. To
get their product/brand accepted by consumer they have to offer their product/brand with an
acceptable level of quality to meet the need of the consumer. Since the world has become a
global village, technology is at its peak, and in this technological era, almost every
firm/organization has the capability to fulfill the basic need of the consumer with an
acceptable quality level. Then the question that arises then is that what should a company do
for the acceptance of their product or brand? They should differentiate their product/brand
from others available. For that they have to know the self of the consumer and to create an
acceptable image. A lot of research has been done on the positive factors which help
marketing managers to achieve their goals. At the same time, it is equally important to know
about which negative factors/reasons are creating a problem for the acceptance of their
20
Introduction
product. In our study, our focus is to explore those reasons/factors which restrict consumer to
use the product or brand beyond product-related attributes. More specifically, what are the
pre-purchase determinants of brand avoidance?
Significance of the Study
By explicitly exploring the reasons why consumers avoid brands before consumption, this
thesis will contribute to brand avoidance area in different ways. First, from a managerial
point of view, it is very important for the managers to know why consumers are not willing to
buy their product for the first time; although they have fulfilled the pre-requisites of the
purchase regarding the price, quality level, after sales services, etc. Second, from an academic
point of view, a significant amount of literature is available on positive attitude towards brand
but, on the negative side, literature is still not enough to answer managers’ question. So in the
form of this thesis, a conceptual framework is developed by joining together the existing
literature available in the anti-consumption area. Our study offers, for the first time, a
comprehensive framework based on conceptual and empirical findings of pre-purchase brand
avoidance phenomenon.
The knowledge of pre-purchase brand avoidance attitudes and behaviors contributes
theoretically as well as practically to an interesting and under-researched area. The traditional
view of marketing managers is concentrating on convincing consumers to choose their brands.
Whereas, this research will exhibit that consumers do avoid certain brands even before
consuming them and it is essential to have the knowledge about negative side of brands. The
limited efforts have been exerted on this side that is evident from the past research. Thompson
et al. (2006) emphasized the importance of brand avoidance that the result of ignoring or not
paying attention to brand avoidance attitudes and behaviors may lead to a ‘brand crises’. This
21
Introduction
may lead to unidentified loss in profits, a reduction in company or brand market share, or an
overall fall in the reputation of the company (Lee, 2007).
Companies may take advantage over its competitors by having this knowledge and translating
it into appropriate strategies to cope with the consumers who have pre-purchase negative
attitudes and behaviors towards their brand. The focus of relational marketing is to understand
consumer and build customer relationship, so companies in general and marketing managers
should take it as a long term objective to work on those consumers and to mitigate prevailing
negative feelings about their brand or company.
In existing literature, none of the above perspectives have been attempted before. Although a
few studies have discussed and emphasized individually on the constructs such as undesired
self (Banister & Hogg, 2001; Ogilvie, 1987), animosity (Klein, Ettenson, & Morris, 1998) and
consumer resistance (Kozinets & Handelman, 2004). Our study not only confirmed
empirically the findings of past studies, but it also recognizes and integrates some additional
factors, on the basis of cues obtained from qualitative interviews, that play influential role in
consumers’ decision making.
Academic and Practical Aims and Objectives
Anti-consumption research area is growing and mainly focuses on the side of ‘against’
consumption. Our study will provide an applicable and novel approach in understanding
consumer behavior and in general it will help in the improvement of the understanding of
consumption pattern. Brand avoidance is the phenomenon that study avoidance behavior of
people towards certain brands, even they have the financial capability to consumer them (cf.
Definition of Brand Avoidance). Therefore, brand avoidance could be a link between
branding research and the anti-consumption. In symbolic consumption literature, many
studies have concluded that cognitively consumer has a clearer understanding of what he does
22
Introduction
not want, contrary to what he wants or desires (Banister & Hogg, 2001, 2004; Hogg, 1998;
Hogg, Cox, & Keeling, 2000; Hogg & Michell, 1996; Ogilvie, 1987). In a similar way, some
researchers argued that what a consumer dislikes is equally important for his decision making
as that of his liking, and negative information about brands has the same effect as that of
positive information about brands (Aaker, 1996; Bourdieu, 1984; Herr, Kardes, & Kim, 1991;
Lutz, 1975; Wilk, 1994, 1997). Whereas, impression-formation literature tells that negative
information has a stronger impact on impressions than positive information (Kanar, Collins,
& Bell, 2010). People rely more on negative information because it is more diagnostic in
nature or valuable for discriminating between alternative judgments than positive information
(Skowronski & Carlston, 1989). These perspectives suggest the importance of brand
avoidance, but the motivation behind the study of brand avoidance is to emphasize that people
may define themselves with their consumption tastes but they can also achieve the desired self
image by staying away from the things that they do not want to consumer or avoid consuming
(Wilk, 1996).
According to Bourdieu (1984) preference and tastes may be defined by the refusal of other
tastes. Therefore, brand avoidance could be an output of dislikes and distastes and exploring
the reasons/factors behind the avoidance of certain brands, instead of focusing only on the
positive side, will contribute knowledge in negative symbolic consumption areas and it also
helps in understanding general consumption behavior. For marketers, the study of negative
preferences is also essential to fully understand the positive preferences of consumers.
In the global view, brands are made to target international markets, and global companies
want to expand their businesses to a larger and more diverse group of customers in the world.
In addition, with the inception of internet now consumers can contact to any company almost
anywhere in the world, and this freedom results in more competition among international
companies (Holt, Quelch, & Taylor, 2004). The more heterogeneous or dispersed market
23
Introduction
means the higher the chances of misinterpreting the marketing messages. Because of the
diverse target market, the firm cannot be certain that the message has been conveyed as they
wish to convey it or if it is interpreted as they want it to be interpreted. There are chances that
the brand values or position which the companies wish to create in the minds of consumers
may lost during the communication process from sender to receiver (Aaker, 1996). It is
argued that consumers have doubts in company communication and they more likely hold the
firms responsible for all kind of the messages ‘positive or negative’(Holt et al., 2004; Klein,
2000). Beyond these doubts the most harmful thing for the companies is the additional
associations that consumer may relate to the brand, for instance, consumer may have negative
corporate reputation through media exposure (Holt et al., 2004; Klein, 2000).
Apart from the misinterpretation of the message; even if the message is received successfully
and understood as intended some of the messages could result in the opposite of the desired
effect. Many studies have concluded that consumers may avoid those brands that are
perceived ideologically or symbolically unwanted or inconsistence with their own self
concepts (Banister & Hogg, 2001, 2004; Bhattacharya & Elsbach, 2002; Hogg & Michell,
1996). Sometimes it is not just because of the communication errors but the problem is that
companies are targeting same markets/audience with similar messages and as a result
consumer cannot differentiate among different brands competing in the same markets.
Due to increased competing advertising material and clutter the efforts for creating clear
message are not as required (Kent & Allen, 1993; Rumbo, 2002). There could be chances that
due to excessive advertising, consumers might not be willing to pay attention towards ads and
also he may perceive the messages from different brands as similar (Aaker, 1996; Keller,
1993). Consequently, it is complicated to establish and predict consumer’s loyalty towards
specific brands (Oliver, 1999).
24
Introduction
Nowadays, the markets reached to saturation point in all categories of product especially in
developed countries and majority of the consumers are in a position to satisfy their need with
more than one brand or by using multiple brand in the same category. Hence, consumers can
be loyal to multiply brands; it means he can select as many alternative brands as he wants
depending upon the circumstances (Abougomaah, Schlacter, & Gaidis, 1987). But on the
other hand, there might be some specific brands that consumers choose to avoid consistently
(Lee, 2007).
Practically, although it is fundamentally necessary to be one of the many chosen brands, and
at the same time, it is also important to not to be among the avoided brands. Academically, a
significant number of research has been done on consumer selection of brands, but perhaps
now is the time to understand that what consumer think while rejecting of a brand.
In recent times, marketing practices are shifting from a transitional to a relationship paradigm
and it is also influencing academic research priorities. This shift suggests marketing managers
to think about building a profitable long-term relationship with customers, rather than just
customer acquisition. They should focus on the long term retention and to build successful
networks and think out of the box instead of traditional marketing practices (Lusch & Vargo,
2004; Morgan & Shelby D. Hunt, 1994). In addition, some researchers have also argued on
the importance of negative aspects of popular themes like market based assets, business
relationships, brand loyalty, and brand meanings. (Morgan & Shelby D. Hunt, 1994;
Srivastava, Fahey, & Christensen, 2001; Stern, 2006).
Hence, we emphasized that as much as the concern of marketing for favorable attitudes and
loyalty towards a brand are important, it is equally important to focus on investigating the
anti-consumption and negative attitudes that consumers may have towards brands.
25
Introduction
In today’s marketing environment, consumers’ concerns in brand selection have been
increased. As the number of brands increases in the market the consumer is becoming more
skeptical and hard to satify (Firat & Venkatesh, 1995; Fournier, Dobscha, & Mick, 1998;
Zavestoski, 2002a). Massive advertising has challenged the true brand loyalty and result in
frequent brand switching (Aaker, 1996; Banister & Hogg, 2001; Rumbo, 2002). Consumers’
attitudes and behaviors are now counter to marketers’ desire. We have seen these counter
behaviors in the past, in the form of protests (Kozinets & Handelman, 2004), boycotts (Klein
et al., 2004; Kozinets & Handelman, 1998), ‘culture jamming’(Rumbo, 2002; Thompson &
Arsel, 2004), and ‘BurningMan’ (Kozinets, 2002). Mostly, regular consumers reject those
products/brands that do not show congruence with their own sense of self (Fournier et al.,
1998; Hogg, 1998) or those products/brands that fail to meet their expectations (Halstead,
1989; Oliver, 1980). Even though customers are showing negative attitudes and behaviors
towards brands, but branding research have a few strategies to cope with these negative
attitudes and behaviors such as sub branding, value augmentation, image adaption and
positive WOM (Lee, 2007; Thompson et al., 2006).
This thesis entitled, “Pre-purchase determinants of brand avoidance” is aiming to explore the
possible negative attitudes and Intentions existing prior to experience, which lead to brand
avoidance. Literature about negative attitudes and intentions provide a fragmented picture but
in this study our aim is to put them all under one umbrella.
Our aim is also to find any robust determinants of brand avoidance among the various factors
identified by earlier studies and further their relative importance among others determinants in
predicting such behavior.
26
Introduction
Organization of the Thesis
To address the outlined objectives of the study, our research consists of 7 chapters mainly
divided into two parts ‘Theoretical Framework and Model Building’, and ‘Research
Methodology, Empirical Analysis and Results. The summary of the structure of whole thesis
is presented in Figure 0.1, but a brief introduction of each chapter is as follows:
Chapter 1 Problem identification, justification and objectives of the study have been
explained in the introduction section of the study. First chapter reviews and examines the
literature regarding many aspects of brand avoidance, such that the definition of brand and
brand avoidance on which our thesis will be based. Types and characteristics of anti
consumers are discussed as well.
Chapter 2 gives a detailed explanation about preliminary data collection based on qualitative
approach. In depth interviews technique is used to explore the concept. The responses of the
interviewees (verbatim) are discussed in detail and analyzed by using software to fully
understand the emerging themes which help us in developing our theoretical framework.
Chapter 3 provides the theoretical framework which comprised of key pre-purchase
determinants that are expected to influence on brand avoidance attitude and intention.
Hypotheses are formulated and supported with literature, leading us to conceptual model of
the study. The conceptual model shows the relationships between the pre-purchase
determinants of brand avoidance and the brand avoidance attitude, further leading to brand
avoidance intention. Role of moderating variables is also discussed.
Chapter 4 represents the research methodology applied in conducting this study. The data is
collected into two phases, preliminary data (through interviews) as explained in chapter 2 and
the quantitative data through online survey methods. In this chapter, the research
27
Introduction
methodology related issues such as research process, design, sampling, questionnaire
formulation and its pre-testing, and data collection methods are explained in detail.
Chapter 5 & 6 present the data analysis consists of three data sets (Pakistan sample, New
Zealand sample and Pooled sample). Preliminary data analysis for purification of the data and
inferential data analysis related to testing of proposed research model by utilizing the
Structural Equation Modeling technique is explained. Construct and discriminant validity
among the variables is explained and established. The moderation effects are also explained
in detail.
Finally, Chapter 7 discusses the main research findings, the contributions and implications.
The limitations are acknowledged, and future avenues opened through this thesis are
described in this chapter.
28
Introduction
General Introduction
 Research background
 Research Problem
 Significance of the Study
 Academic and Practical Aims and Objectives

Chapter 1: Literature Review
 Literature review on Brand avoidance
 Concept positioning
 Types of anti-consumers
Part I
Theoretical Framework and
Model Building
Chapter 2: Qualitative Study
 Qualitative Approach
 Interviews
 Themes identification
Chapter 3: Theoretical Framework & Hypotheses
 Pre-purchase determinants of brand avoidance
 Hypotheses generation
 Conceptual model
Part II
Research Methodology,
Empirical Analysis & Results
Chapter 4: Research Methodology
 Research Approach
 Field of study
 Sample selection & instrument design
 Data collection

Chapter 5 & 6: Data Analysis
 Data preparation
 Confirmation of scales
 Sampling weights & Measurement Invariance
for pooled data
 Hypotheses testing & Moderation testing
Chapter 7: Discussion & Conclusion
 Summary of the results
 Contribution & Implication
 Limitation & future research

Figure 0.1: Structure of the Thesis
29
Part-I
Part-I: Theoretical Framework & Model
Building
Chapter 1: Literature Review
 Literature review on Brand avoidance
 Concept positioning
 Types of anti-consumers
Part I
Theoretical Framework and
Model Building
 Qualitative Approach
 Interviews
 Themes identification
Chapter 3: Theoretical Framework & Hypotheses
 Pre-purchase determinants of brand avoidance
 Hypotheses generation
 Conceptual model
30
Part-I
Introduction to Part-I
In the introduction chapter the general research background has been discussed. This part of
the study provides the theoretical basis of the thesis. We need to understand what is meant by
brand avoidance, its definitions, its theoretical position in literature, and the characteristics
of consumers having anti-consumption attitudes and behaviors. This is explained in chapter 1.
After establishing the theoretical grounds for brand avoidance, in chapter 2, we look for the
pre-purchase determinants of brand avoidance and identify its drivers from qualitative study
through in-depth interviews. The identified factors are supported by literature and hypotheses
are formulated to test the proposed relationship among the factors. This is explained in
chapter 3.
What is brand avoidance? Where it stands in literature?
Chapter 1: literature Review
How do consumers perceive about pre-
What are the pre-purchase determinants of
purchase brand avoidance?
brand avoidance?
Chapter 3: Theoretical Framework and
Hypotheses
Figure 0.2: Flow Chart of Part-I
31
Literature Review
Chapter 1
Chapter 1:
1
Literature Review
“To achieve lasting literature, fictional or factual, a writer need perceptive
vision, absorptive capacity, and creative strength”
--- Lawrence Clark Powell
1.1
Overview of the chapter
The previous introductory chapter outlined the overall context of this research. In this chapter
we build the basis of our thesis by first defining brand and its related functions. Definition of
brand avoidance and its positioning in the existing literature is discussed. A brief introduction
of anti-consumption and consumer resistance domain is provided and characteristics of
different types of anti-consumers are also discussed.
1.2
Definition of Brand
If we see the extensive literature in the area of anti-consumption, it becomes evident that a
significant amount of literature consist of studies exploring consumers’ dissatisfaction with
products and services (Oliver, 1980) voluntary simplification phenomenon (Craig-Lees &
Hill, 2002), culture jamming or boycotts (Kozinets, 2002; Kozinets & Handelman, 1998,
2004) and consumer resistance (Banister & Hogg, 2004; Cherrier, 2009b; Penaloza & Price,
1993). Most of the studies conducted are based on general product or service categories
except (Thompson & Arsel, 2004; Thompson et al., 2006). Our research focuses negative
attitudes and behaviors about brands so it is important to define the concept of brand upon
which this thesis will be focused.
32
Literature Review
Chapter 1
Brands perform plenty of valuable functions for consumers. For example, a brand can make
choices easy, promise a specific level of quality, can reduce risk, and /or engender trust
(Keller & Lehmann, 2006). It can also help managers in determining the effectiveness of
marketing mix efforts, such as advertising and channel placement (Keller & Lehmann, 2006).
Brands are considered as an asset in financial sense and should be treated carefully (Keller &
Lehmann, 2006; Kotler & Armstrong, 2007).
In contemporary world, the term ‘brand’ is taken for granted; and it has been experienced that
most of the consumers are able to name many brands when asked; which is a result of
company’s branding efforts. Albeit, a lot of literature available on brand and branding
research, but a single definition of brand is still not available (de Chernatony & Dall’Olmo
Riley, 1998; Stern, 2006).
A brand can be defined as “A name, term, sign, symbol, or design, or combination of them
which is intended to identify the goods and services of one seller or group of sellers and to
differentiate them from those of competitors” (Kotler, 1991)
According to Keller (1993) individual components of a brand are known as ‘brand identities’
and their total is known as ‘brand’. Brand knowledge is also important. ‘Brand awareness’
and ‘brand image’ are the two main dimensions of brand knowledge (Keller, 1993). Whereas
‘brand awareness’ is the strength of brand trace in memory, as reflected by the consumers’
ability to identify the brand under different conditions (Rossiter & Percy, 1987), ‘brand
image’ is the perception about a brand as reflected by the brand associations held in consumer
memory (Keller, 1993). Further, he linked brands with different types of associations on the
basis of brand attributes, benefits and attitudes. The ‘multi-dimensional value constellation’
perception of brand proposed by de Chematony and Dall’ Olmo Riley (1998) seems realistic
33
Literature Review
Chapter 1
as compared to ‘product centric’ traditional view of the brand (Ambler & Styles, 1996) (Table
1.1).
Table 1.1: Brand Definitions
Perspective
Brand definition
Function of Brand
1.Legal Instrument The brand is used by the firm as a mark of ownership and as legal
protection from imitation by its competitors.
The brand is viewed from the
2.Logo
firm’s perspective; for
instance, what does a brand do
3.Company
for its company? The
development of the brand is
also seen as relying on the
‘input’ of the firm.
Communication regarding the 4.Identity System
brand is ‘one-way’ from the
firm to the consumer.
5.Shorthand
6.Risk reducer
7.Image
8.Value System
The brand is viewed from the
consumer’s perspective.
Development of the brand is
also seen as being influenced by
both the organization and the 9.Personality
consumers that use the brand.
10.Relationship
11.Added Value
Brand development is driven
by the consumers’ activities.
The brand is used as a name or symbol to differentiate one
company’s products from their competitors
The brand is used as a unifying force for the entire corporation. A
consistent message is conveyed to all stakeholders so that the
products, culture, employees and CEO all work towards the same
goals
The brand is holistically made up of culture, personality, selfprojection, physique, reflection, and relationship. The resulting
unique identity communicates a consistent and integrated vision to
all stakeholders. This allows for better strategic positioning and
protection from imitation.
The brand represents a number of functional and symbolic
characteristics. The brand acts as a mental shortcut and memory cue
to make information processing easier for the consumer.
The brand acts as an informal contract between the consumer and
the company ensuring consistent quality or reliable service.
The brand is what consumers perceive it to be, and is made up of
the consumer’s perceptions of functional and psychological
attributes.
Organizations imbue their brands with certain values; consumers are
driven by their own values. Consumers can also assign values to
brands. Thus, the success of the brand depends not only on the
functional aspects of the brand, but also on the match between the
values that the brand represents and the values of its target market.
Each brand symbolizes a combination of psychological traits. The
match between the consumer’s and the brand’s personality
influences his or her attitudes towards the brand.
The brand is personified as an entity that interacts with its
consumers. A successful brand is one that has successful
relationships with its customers.
In addition to its functional benefits, customers also use a product
for its non-functional benefits. When a brand is construed to satisfy
the consumer’s symbolic needs and wants more closely, the brand
adds unique value above and beyond functional attributes.
12.Evolving Entity The brand starts as a product of the organization and gradually
becomes a product of its consumers.
Source: de Chematony & Dall’Olmo Riley (1998)
34
Literature Review
Chapter 1
According to this ‘holistic’ perspective of brand’s value, a consumer could have many
different things in his mind towards brands, for example “a legal instrument, a logo, a
promise/contract, a risk reducer, an identity, a value system, an evolving entity, or a
corporation” (Lee, 2007 p. 2). Eventually, as stated by Banister and Hogg (2001) that
consumer makes a network of association in his mind based on the ‘multi-dimensional values
constellation’ and brand ultimately becomes a set of information in the memory of consumers.
To our understanding, a brand must satisfy the following three criteria:
1. “A brand must have the ability to convey the message that it is distinct from another
entity”. As AMA ‘American Marketing Association’ (2011) defines a brand as “A
name, term, design, symbol, or any other feature that identifies one seller's good or
service as distinct from those of other sellers. A brand may identify one item, a family
of items, or all items of that seller”. It is the least required level for the brand that it
must be able to distinguish itself from the others. Otherwise brand cannot achieve the
benefits for which it is created.
2. “A brand must have the ability to give more meaning about the entity to which it
belongs, beyond the simple fact that it is not the same as other brands”. After
achieving the basic level, a brand is required to give more meaning to consumer it
must show (symbolize) that these products or services are different from any other
product or service. Only the difference in names is not enough to attract consumer
towards brands. The brand must be able to create value/benefits. The objective of
marketing is to convey great amount of information with one brand name or logo.
Brand intangibles – features and characteristics of brand image that do not involve
physical, tangible, or concrete attributes or benefits (Levy, 1999) are common means
with which marketers used to differentiate the perceptions of their brands with
35
Literature Review
Chapter 1
consumers (Park, Jaworski, & Maclnnis, 1986) and transcend physical products
(Kotler, Keller, Brady, Goodman, & Hansen, 2009). Generally, a company creates a
brand with a bundle of positive meaning and once consumer has perceived the brand
as reliable a performer or a has a positive consumption experience then his or her
behavior towards that brand, and other similarly branded product/services, will be
based on those positive meanings. But it does not always happen, sometimes the
meanings that brand represents could be negative. In that case, consequently,
behaviors towards those brands may be based on negative meanings. Apart from
positive or negative meaning, the important thing is that brand must not be considered
just as a tool for identification. Instead it should be considered as a tool for
communication with extra meaning and values (Lee, 2007).
3. “A brand must be used with commercial intention or the perception of altered
worth/value”. The third criterion is related to the financial benefit attached with the
brand. For example, celebrities are considered as brands, and it is true in a sense that
their names are used to create and increase profitability. It is only possible when their
name represent a constellation of values which convey ‘extra meaning’ to many
people. It is not possible for every ordinary person because they do not have such level
of fame. For ordinary people their names are just for identification only. That is why
they cannot be considered as brands, unless they develop their name with distinct
values and extra meaning attached to them. Similar is the case of countries as a brand.
Not every country can be considered as brand unless it has been created with extra
meaning to other people or for foreigners. For example, Japan has highlighted its
name as a superior quality, and a vacation to France will be clean and memorable.
These examples show that these countries have undergone the process of branding.
36
Literature Review
1.3
Chapter 1
Definition of Brand Avoidance
In the preceding section we established the definition of brand, in this section we will define
what we meant by brand avoidance and in the following section we will give a brief
introduction to the field of anti-consumption and consumer resistance. The term ‘brand
avoidance’ will be discussed here to elucidate the parameters of the study. In psychology
literature, approach and avoidance motivation were studied in detail. Elliot (1999) said that in
approach motivation, the behavior is “instigated or directed by a positive/desirable event or
possibility”, on the other hand in avoidance motivation; the behavior is “instigated or
directed by negative/undesirable event or possibility”. The term brand avoidance was not
properly defined before Lee (2007) defined it as follows:
“The conscious, deliberate, and active rejection of a brand that the consumer can afford,
owning to the negative meaning associated with that brand.”
So by taking into account the psychological definition of avoidance, the characteristics of
brand avoidance may be defined as “The attitudes and behaviors consumers have towards a
brand, as motivated by the negative meaning/consequences related to the brand.”
If we go into extent literature we can find limited studies such as Thompson et al. (2006) and
Oliva et al. (1992) who explicitly used the term ‘brand avoidance’ in their study. Brand
avoidance has been described as anti-thesis of brand loyalty (Oliva et al., 1992). He further
proposed that satisfaction leads to brand loyalty whereas dissatisfaction leads to brand
avoidance or brand switching. Whereas brand avoidance and brand switching are not the same
phenomenon.
In marketing literature, brand switching has received substantial attention (Oliva et al., 1992;
Raju, 1984; Zahorik, 1994). Broadly speaking, one can identify two types of brand switching.
37
Literature Review
Chapter 1
One is due to dissatisfaction with the previous or existing brand. And the other is due to some
intrinsic desire for novelty, change or variety (Raju, 1984). In his study Raju (1984) used the
term exploratory brand switching for the brand switching due to intrinsic desire. And also he
divided it in active and passive brand switching. In active brand switching, the consumer
would be highly motivated to look for change or variety, whereas in passive brand switching
the consumer might accept a change with little or no objection, but not pursue it actively.
Desire for variety plays a more important role in brand switching than changes in the attitudes
towards the brand (Raju, 1984).
There could be many reasons that why consumers switch brands, even when the intended
usage occasions are quite similar. According to Zahorik (1994), a consumer may switch
because he or she:

Consider the two brands highly substitutable for the intended use

Finds the favorite brand out of stock

Switches to a brand which is being offered on deal

Is bored or satiated with some aspect of the last brand purchased

Experiences a change in needs or circumstances

Tries an unfamiliar brand to learn about it

Has undergone a change of perceptions of the relative merits of brands
Dissatisfaction leads to brand avoidance and there are many other determinants that also
contribute to brand avoidance (cf. Chapter 2 & 3). Brand switching is a behavior that involves
a consumer observing negative experience with one brand and making a relationship with
another brand. In contrast, he posits that brand avoidance may include brand switching and
also involves the ongoing rejection of brand.
38
Literature Review
Chapter 1
The term brand avoidance was also discussed by Thompson et al (2006). They used the term
‘Doppelgänger brand image’ which they defined as “A family of disparaging images and
stories about a brand that are circulated in popular culture by a loosely organized network of
consumers, anti-brand activists, bloggers, and opinion leaders in the news and entertainment
media”. Their study suggests that when consumer’s emotional-branding promises are viewed
as inauthentic then consumers start avoiding brand and this is the time when marketing
managers investigate the reasons and resolve them to make their brand healthy.
Existing boycotting literature (Bodur & Day, 1978; Hirschman, 1970; Klein & Dawar, 2004;
Kozinets & Handelman, 1998) may arise a question of similarity of the concept but brand
avoidance and boycotting are not the same in nature. According to Hirschman (1970), boycott
may be defined as a “temporary act of exiting a relationship from an organization, due to
some form of dissatisfaction, accompanied by an assurance of re-entering the relationship
once certain conditions have been met, such as a change of policy by the offending party”
(Lee, 2007 p. 136). It is quite possible that the boycotts towards a specific brand or company
involve cases of brand avoidance, but the guarantee that consumption relationship will resume
in future is not a prerequisite of brand avoidance, and brand avoidance may continue even if
the party has fulfilled the lacking commitment (Lee, Motion, & Conroy, 2009).
The concepts Narayana and Markin (1975) proposed are relevant to our study. They proposed
three sets under the conceptualization of consumer behavior. They initially divided the
decision made by consumers towards total set of brands available in the market into two sets:
awareness set and unawareness set. If the brand falls into unawareness set then the chances of
being considered for that brand does not exist at that particular time. On the other hand if the
brand falls into awareness set, then there are chances to be considered. Narayana and Markin
39
Literature Review
Chapter 1
(1975) further divided choice decision into three sets: evoked set, inert set and inept set as
shown in the Figure 1.1.
Evoked
Set
Awareness
Set
t’
New
brands
t’ (+)
t’
Total Set at
any given
time t
Inert
Set
(O)
t’
t’
Unawareness
Set
t’
Inept
Set
t’
(-)
Figure 1.1: Conceptualization of Consumer Behavior
Source: Narayan & Markin (1975)
Consumer is likely to reduce his choice options in awareness set by narrowing down the
category further and want to make his purchase selection from a smaller group of brands,
known as evoked set (Howard & Sheth, 1969). Campbell (1969) defined the evoke set as “the
set of brands of a product which the buyer actually considers when making a specific brand
choice”. According to Howard and Sheth (1969), “the brands that become alternatives to the
buyer’s choice decision are generally a small number, collectively called his ‘evoked set’”.
Consumer positively evaluates the brands in their evoked set for purchase and consumption.
In the awareness set group, second set is ‘inert set’ which is based on the alternatives which
consumers know of but they do not consider them as a better option as compared to already
selected options. They evaluate these brands neither positive nor negative (Narayana &
Markin, 1975). They are aware of them but they may not have sufficient information to
40
Literature Review
evaluate them properly.
Chapter 1
In other words, for this set consumers’ attitudes and intentions to
purchase remain ‘inert’.
Lastly, the ‘inept set’, this is most relevant set to our study of brand avoidance. It consists of
those brands which the consumer has rejected from his purchase consideration, due to his
unpleasant experience with those brands or negative feedback from various sources. Narayana
and Markin (1975), defined this set as the group of brands that consumers have resolved not
to purchase at all in their present form. They negatively evaluate these brands while making a
purchase decision. As our study is focused on pre-purchase factors, so we proposed that
avoidance set (inept) may be equal or less than in features and quality to their evoked set but
they do not want to buy that brands because of negative meaning associated with that brands.
Negative side of consumption is also relevant to our domain of study, which we will discuss
in the following section.
1.4
Anti Consumption and Consumer Resistance
Brand avoidance positioning in literature may fall in the area of anti-consumption. Literature
about anti-consumption attitudes and behaviors are now more diversified in nature and
widespread. The term anti-consumption includes many forms like relatively harm beliefs,
such as negative perceptions of fast food, also violent and illegal behaviors, such that the act
of destruction of targeted companies for instance Mcdonald’s, Coca Cola and Nike (SandIkci
& Ekici, 2009).
Zavestoski (2002b) suggests that anti-consumption attitudes in simple terms can be defined as
“function of a preference to consume one object over another” whereas in a deeper sense
anti-consumption attitudes include “a resistance to, distaste of, or even resentment or
rejection of, consumption more generally”.
41
Literature Review
Chapter 1
Anti-consumption literature tells that there could be many reasons associated with negative
brand attitudes. Sometimes they were instigated due to unethical business practices or
disrespect for human rights or environment and people start boycotting those brands
(Friedman, 1985). Sometimes it is the perceived animosity toward the country from which the
brand originates and to use that brand is considered to be as unpatriotic (Klein et al., 1998;
Shimp & Sharma, 1987). Another possible reason is that the brand represents an undesired
self which the consumer does not want to be. Or that brand is associated with the group of
people with which the consumer does not want to be identified (Banister & Hogg, 2001;
Englis & Solomon, 1995). Also people refuse to use those brands which are not compatible
with the products they are already using (Hogg, 1998).
The rejection of brand might be politically motivated (SandIkci & Ekici, 2009). Political
consumerism involves the market actions and consumer choice as a political tool (Micheletti,
2003; Micheletti, Follesdal, & Stolle, 2003). Micheletti (2003) proposes that political
consumerism “Represents actions by people who make choices among producers and
products with the goal of changing objectionable institutional or market practices; their
choices are based on attitudes and values regarding issues of justice, fairness, or noneconomic issues that concern personal and family well-being and ethical or political
assessment of favorable and unfavorable business and government practice”.
There involve both individual and collective actions in political consumerism, such that
negative actions (boycott) or positive actions (buycott). Boycott is a special form of politically
motivated rejection or avoidance of brands or companies. It is different from an individual’s
decision to not to consume a particular product or service; instead it is an organized and
collective effort to refuse consuming a particular good (Friedman, 1985). Sandlkci and Ekici
(2009) suggest that boycotts can be triggers by economical, social or ethical reasons. It may
42
Literature Review
Chapter 1
aim to change the unfair business practices of the target company such as lowering the prices
or increasing the quality; or the purpose of boycotting is to force the target company to take a
socially responsible and ethical position in the society such as labor union laws or green
practices. Boycotts are relatively a short-term reactions and the act of boycotting normally
ends up immediately once the target company meets the demands of the group.
In the following section we will elaborate the characteristics of the consumers involving in
anti-consumption.
1.5
Characteristics of Anti-consumers
To understand the phenomenon of brand avoidance, understanding the characteristics of anticonsumers is essential. In their work Iyer & Muncy (2009) divided anti-consumers into four
categories in vertical and horizontal dimensions as presented in the Table 1.2.
Table 1.2: Four Types of Anti-consumers
Purpose of Anti-Consumption
Object of
Anti-Consumption
General (All
Consumption)
Specific (Individual
Brand or Products)
Societal Concerns
Personal Concerns
Global Impact
Consumers
Simplifiers
Market Activists
Anti-Loyal
Consumers
Source: Iyer & Muncy (2009)
The vertical dimension is the comparison between the consumers who want to decrease their
overall consumption versus those who are interested in decreasing the consumption of
specific brands or products. Whereas, in horizontal dimension they presented the comparison
between those who are more concerned for societal issues like environmentalism versus those
who are interested on personal issues like life-simplification. The important point is that these
43
Literature Review
Chapter 1
categories were not mutually exclusive, i.e. there could be one or possibly all reasons to
reduce the consumption. Following are the characteristics of each sub-group of the anticonsumers.
1.5.1
General-societal: Global impact consumer
In this group, people are more concerned with the environment and material inequity. In her
study Dobscha (1998), she found that anti-consumers did not follow the common view that in
nation’s prosperity the consumption is a major indicator (Borgmann, 2000). Instead it was the
over consumption of some people which created many of society’s problems. “Buy Nothing
Day” is the event which many socially concerned consumers look forward each year. The
purpose of that day is to encourage consumers to think that why they buy and how their
consumption affects society and the environment (Carty, 2002).
1.5.2
General-personal: Simplifier
In this group, people are concerned in reducing their overall consumption. They are not the
‘frugal materialists’ identified by Lastovica (2006), who are reducing their consumption in
one area and increase it in other areas. This need is not due to changing economic
circumstances to reduce consumption. Instead they believe that increasing the consumption
has undesirable results, such that stress and distraction from higher pursuits (Shaw &
Newholm, 2002; Zavestoski, 2002a). There could be a spiritual or ethical belief to reduce
consumption. Simplifiers believe that it is morally disgusting to focus so much on self-serving
consumption activities (Shaw & Newholm, 2002). Research tells that a significant portion of
the population believes that over-consumption can be a reason of stress, fatigue, unhappiness
or disillusionment (Zavestoski, 2002a).
44
Literature Review
1.5.3
Chapter 1
Brand-societal: Market activist
This is a group of people called market activist who try to use the power of consumer dollars
to impact societal issues (Friedman, 1985). “Market activist consumer might avoid using
product and brand because they feel that a specific brand or product causes a specific societal
problem” (Iyer & Muncy, 2009). Market activists have publications as the source of
information which keeps them informed about the brands and companies to avoid. For
instance, The Media Foundation, a very visible anti-consumption organization, circulates a
number of anti-consumption “advertisements” in its Adbusters magazine (Penaloza & Price,
1993). Similarly, other publications force consumers to avoid products which are potentially
harmful for them such that consuming alcohol or smoking cigarettes (Iyer & Muncy, 2009).
1.5.4
Brand-personal: Anti-loyal consumers
This group of people behaves opposite to the loyal consumers. Since brand loyalty is the
reflection of a commitment to repurchase a brand because of its real or imagined superiority
(Jacoby, Chestnut, & Fisher, 1978); so the anti-loyalty is the reflection of a commitment to
avoid purchasing a product or brand because of its perceived inferiority or due to a negative
experience associated with it (Lee, Motion, et al., 2009). According to Englis and Solomon
(1997), products or brands that consumers actively avoid are most of the times as personally
and socially important to them as products or brands that they actively seek to purchase. Also
they argued that to avoid undesired products is equally important to shape individual
consumer’s ideal self.
In a relevant context, Muniz & Hamer (2001) explained the phenomenon of ‘oppositional
brand loyalty’ in such way that loyal users of a specific brand may develop an important
connection between a given brand and their self on the basis of their perceptions of competing
brands and may communicate their brand loyalty by opposing those competing brands.
45
Literature Review
Chapter 1
Anti-constellation is the key concept in the study of brand avoidance. According to Hogg
(1998) anti-constellation involves two aspects of consumers’ negative choices: Non choice
and anti-choice. Products and services which are beyond the means of the consumers were
considered as non choice, they are not bought because of availability, accessibility and
affordability. Whereas those products and services, which are within the means of the
consumer, but were not seen as compatible or consistent with consumers’ consumption
choices or preferences, were considered as anti-choice.
Figure 1.2, explains the division of consumption anti-constellation. Non choice includes
Affordability, Accessibility and Availability whereas Anti-choice includes Avoidance,
Abandonment and Aversion.
Consumption
Anti-constellation
Non Choice
Anti choice
Affordability/Accessibility/Availability
Avoidance/Abandonment/Aversion
Figure 1.2: Consumption Anti-constellation
Source: Hogg (1998)
As already mentioned in the definition of brand avoidance, it occurs when the rejection of
brand by consumer is motivated by the negative meanings/consequences linked with that
brand. Therefore, it is important to specify that our research is primarily focused on those
brands that consumers actively choose to avoid, not because of that they have no choice. In
other words, our study does not include those brands which consumers do not purchase
because of issues concerning affordability (price), availability, or accessibility. The exclusion
of these elements is understandable, because rejection due to shortage of money, accessibility,
46
Literature Review
Chapter 1
or availability, is inevitable and therefore it does not give any sense to our concept of brand
avoidance. Therefore, our study deals with Hogg’s concept of ‘anti-choice’ instead of ‘nonchoice’ (Banister & Hogg, 2001; Hogg, 1998; Hogg & Michell, 1996).
In anti-choice, the distinction between aversion, avoidance and abandonment associated to the
strength of feelings and behaviors they express (Hogg, Banister, & Stephenson, 2009),
whereas earlier research suggested the overlaps in elements (Hogg, 1998). Hogg et al. (2009)
precisely defined the three terms as “‘Aversion’ involves the psychological or physical action
of turning away from something, ‘Avoidance’ involves the act of staying away from or
moving away from something, and ‘Abandonment’ involves the action of giving up
something previously consumed.”
Aversion links strongly to the affective part of attitudes, whereas avoidance and abandonment
have stronger associations to the behavioral elements of attitudes; also aversion (expressed as
dislike, disgust, revulsion) seem to stimulate behavioral responses of either avoidance or
abandonment (Hogg et al., 2009).
The inter-relationship among aversion, avoidance and abandonment is complex and important
in articulation of anti-consumption. Although the three components of anti-choice (aversion,
avoidance and abandonment) overlap to some extent
(Hogg, 1998), in our study we
established that aversion and abandonment are the actions related to after consumption and
our study is mainly focus on avoidance i.e. the act of staying away from something even
before using it.
Our research focuses on brand avoidance attitudes and behavior and contributes to the
existing literature by investigating why consumers may intentionally choose not to consume a
specific brand beyond product-related reasons.
47
Literature Review
1.6
Chapter 1
Conclusion
In this chapter we established the basis of our thesis and provided the definition of the main
dependent variable and its positioning. The brief introduction to our broader domain of anticonsumption and consumer resistance has been provided. Characteristics of different types of
anti-consumers are also discussed. After establishing the basis of this thesis, in next chapter
we will identify the key drivers of the pre-purchase brand avoidance behavior through indepth interviews.
48
Qualitative Study
Chapter 2
Chapter 2:
2
Qualitative Study
“The numbers themselves tell a pretty grim story, so if
you’re going to hang your hat on this, you have to
believe the qualitative commentary rather
than the quantitative information”
--- Drew Peck
2.1
This chapter provides detail about how the themes emerged from interviews, on the basis of
which the conceptual model is formed (cf. chapter 3). It starts with dealing the issues related
to the methodology of qualitative study, why this approach was selected, sample selection,
research guide to be used, and the role of the interviewer in the whole process of conducting
in-depth interviews. Data transcription and its analyses with software are also explained.
2.2
Qualitative Approach
In Patton’s view (1990), all types of sampling in qualitative research may fall under the broad
term of ‘purposeful sampling’. He states that “qualitative inquiry typically focuses in depth on
relatively small samples, even single cases, selected purposefully” (p.169). He described 15
different strategies for purposefully selecting information.
Sandelowski (1995) in agreement with Patton’s view suggests three different kinds of
purposeful sampling: maximum variation, phenomenal variation and theoretical variation.
Sandelowski suggests that maximum variation in one of the most frequently employed kinds
of purposeful sampling and in which “researchers wanting maximum variation in their sample
49
Qualitative Study
Chapter 2
must decide what kind(s) of variation they want to maximize and when to maximize each
kind” (p.181). Examples may be race, class, gender or other personal-related characteristics.
Phenomenal variation is variation of the target phenomenon under study and the decision to
seek phenomenal variation is “often made a priori in order to have representative coverage of
variables likely to be important in understanding how diverse factors configure as a whole”
(p. 182).
Since our purpose of the study is to explore the reasons of consumers’ behavior for brand
avoidance. So consistent with previous research by using a phenomenological (Patterson, Hill,
& Maloy, 1995) approach, we have used personal interviews as the primary data collection
technique (Hirschman, 1992; Thompson, Locander, & Pollio, 1989) to understand the
consumer’s thought process and to explore the reasons behind his decision-making for not
buying certain brands.
2.3
Analysis Method
To ensure that the concept of brand avoidance is relevant or not in Pakistani context, the pretest was performed by sending emails to 50 people in his social networks and 27 responses
were obtained what gives us a rate of 54%. The details of the respondents are provided in
appendix I.
Initially, following question was asked in the pre-test:

Do you think that in Pakistani context brand avoidance exists or not?
50
Qualitative Study
Chapter 2
Figure 2.1: Pre-Test Result of Question 1
Further, if the answer to above question was ‘Yes’ then in order to have an idea about the
context in which this type of study was suitable we asked the following question:

Do you think that brand avoidance exists in a Business-to-Business (B2B)
environment, in a Business-to-Consumer (B2C) environment or in both?
Figure 2.2: Pre-Test Result of Question 2
The result of this exercise is that 22 (81%) of the participants do agree on the fact that this
phenomenon exists in the Pakistani context (Figure 2.1). Whereas the result of the second
51
Qualitative Study
Chapter 2
question is that 21 (78%) have the opinion that this behavior is observed in B2C environment
and 22% of the respondents have the opinion that it exists in both B2B and B2C environment
(Figure 2.2). The results of the pre-test encouraged researcher to further explore this concept
in B2C environment.
Because of the exploratory nature of the research in-depth interviews are used to explore the
reasons behind consumers’ brand avoidance behaviors. A self-selected group (Table 2.1) of
thirteen (8 males, 5 females) participants are interviewed until saturation point where the
theme start repeating (Taylor & Bogdan, 1998). The interviews are conducted in university
group rooms, and personal offices. All interviews lasted for approximately 60 to 75 minutes
and are audio-taped and transcribed.
Table 2.1 : Participants Detail (Interviewed)
No.
Initial
Gender
Age
1
2
3
4
5
6
7
8
9
10
11
12
13
AK
AM
SQ
HH
AA
QJ
AG
ZA
IH
GA
SH
XB
AJ
F
F
F
M
F
F
M
M
M
M
M
M
M
23
38
32
32
27
30
50
31
29
27
29
27
38
Education
Bachelor
PhD
PhD
Master
PhD
Master
Engineer
Bachelor
PhD
PhD
PhD
PhD
PhD
Professional
Status
Job
Academic
Academic
Job
Student
Job
Job
Job
Student
Student
Academic
Student
Academic
Income
Rs.* 40,001-60,000
Rs. 80,001-100,000
Rs. 80,001-100,000
Rs. 60,001-80,000
Rs. 80,001-100,000
Rs. 40,001-60,000
Rs. 80,001-100,000
Rs. 40,001-60,000
Rs. 80,001-100,000
Rs. 80,001-100,000
Rs. 80,001-100,000
Rs. 80,001-100,000
Rs. 80,001-100,000
Status
Married
Married
Single
Married
Single
Married
Married
Married
Married
Married
Single
Married
Married
*Pakistan currency rupee (Rs.)
The data is analyzed by using QSE NVivo 8. It is a powerful analyzing tool which helps
researchers in managing and analyzing qualitative data such as transcripts of the interviews.
The interviews are conducted in English as the official language of Pakistan is English and
where it is required the respondents’ mother tongue (Urdu) is used for further explanation, so
52
Qualitative Study
Chapter 2
that they may express their response conveniently. Later on the transcripts were translated
into English with the help of two PhD colleagues. Open-ended questions are asked which let
the interviewee share his/her views openly as there are no prescribed answer to be selected by
participants.
In accordance with conventional in-depth interview design, a guide (Table 2.2) is developed
to gain response on main research questions.
Table 2.2: Interview Guide
Statements and questions
Research Agenda
Project Introduction
Developing mutual
understanding of brand
Which brands are avoided
by the cunsumer?
Before we start I want to make it clear that there are no right or wrong
answers. I’m interested in knowing about what normal consumers think
about brands, and perticularly about certain brands that they might not like,
or the brands they might avoid even if they had the money to buy.
Before starting I would like to make sure that what brands are between the
two of us, so that we will be sure that we are talking about the same thing.
By brand I mean a name of a product, service or company that is a form of
communication that conveys any number of meanings between people. For
example, Nokia is a brand; Sony is a brand. They are names that give you
some sort of information about the product. This information may come
from the company, or from other independent parties like your friends or
family. So if you could share some of the names of brands, then we can go
further.
Do you avoid any brand by choice?
Are there any brands that you could afford but actively choose not to
purchase?
What are your reasons for avoiding brand X?
When did you first start avoiding this brand?
What are the possible
reasons that cause people
to avoid certain brands?
What happened? Tell me the story of what happened.
Think of the process you go through to make your purchase decisions.
What is the process you go through to decide which brands you will
purchase?
How do you decide which brands you like and which brands you choose to
avoid?
53
Qualitative Study
2.4
Chapter 2
Emerging Themes
The data analyzed is consistent with Gentry et al. (2006) and Patterson et al. (1995) which
consist of several stages of analysis. In the first stage of our analysis the interview
transcription is read till the “empathetic knowing” of the interviewee is achieved. In the
second stage interviewee’s experience and perception is summarized. In the third stage the
common themes are searched across interviews. For that purpose the responses of each
interview are coded and analyzed by using software, which facilitated thematic content
analysis and lastly emergent themes are identified.
The responses are categorized electronically using NVivo and then evaluated manually by the
authors. The final stage is to explore the “relationships among the themes” to develop a
comprehensive understanding of the interviewee’s experiences. Further each interview
summary is reviewed again to detect the ‘interconnections among themes” and to look for
negative and qualifying evidence that failed to support this pattern. Following are the themes
emerged from the analysis of the interviews.
2.4.1
Undesired Self Congruence
Apart from experiential reasons to avoid brands, there are other factors which play role in
avoiding brands. Hogg and Michell (1996) found that individual’s self concept is valuable to
him. Self esteem is a driver that compels people to enhance their self-concept and self
consistency forces them to be consistent with their perception of themselves (Sirgy, 1982).
“I don’t think that those who use IPhone are like me.” AA (Female, 27)
Most of the studies investigated this concept in self-image congruence area (Englis &
Solomon, 1997; Sirgy, 1982). The idea behind those studies is that people are equally
54
Qualitative Study
Chapter 2
motivated to avoid their undesired self as they want to approach the ideal self or maintain
their actual self.
“I may avoid buying Ferrari because it is a sports car and I rather buy a Mercedes
because it gives me social/professional look as I want to give…” AG (Male, 50)
“…every one like Nokia and wants to buy but I think Nokia is not for me, and second
thing is that everyone has Nokia and I want to be different from others. I want to have
something different that’s why I don’t use Nokia mobile. My father has gifted me a
Nokia mobile which I am using, just only because I got it free... Otherwise I would
never buy Nokia mobile.” XB (Male, 27)
People do not want to lose their sense of individuality, if they perceive that the consumption
of specific brand will damage their self-identity or take away their sense of individuality, this
will motivate them to avoid such brands. Such as depicted in the following statements:
“I don’t buy things which are very common. If you say … I am kind of a different
person… I never do things which others do.” QJ (Female, 30)
“… I don’t want to buy certain perfumes because most of the people are buying it.
Even it is a brand and everyone is using then it loses its uniqueness and its unique
position in the market. If the brand is not differentiating you then it is useless.
Although, it is differentiating you from those who cannot buy but who can afford you
are not different from them.” SQ (Female, 32)
“It is no more a brand in my opinion. It loses its uniqueness and when it loses its
uniqueness then there is no utility to buy that brand.” SH (Male, 29)
2.4.2
Social influence is very important in collectivist countries. It has been emphasized that
consumers who are dissatisfied with a purchase are telling friends about their experience and
55
Qualitative Study
Chapter 2
influence them to avoid it (Bodur & Day, 1978; Leonard-Barton, 1985). Many researchers
suggested that consumers have tendency to give more weight to negative information in
product or brand’s evaluation (Lutz, 1975; Mizerski, 1982). As following participants explain:
“You always trust people who have experience of that brand. Direct reference is
necessary to buy or avoid brands.” IH (Male, 29)
“…peers’ pressure is a very important factor in selection or avoidance.”AA
(Female, 27)
“… I will surely avoid this brand… This is the first thing and second thing I will do, I
will recommend others to not to buy this brand.” SH (Male, 29)
“… From friends or relatives you come to know that this brand is not of that worth.”
GA (Male, 27)
2.4.3
Animosity
Klein et al. (1998) in their study of testing animosity model on Chinese consumers proposed
that consumers might not willing to buy the products from the offending country not because
of the quality of the products, but because the country has been engaged in economic, political
or military acts that the consumer find difficult to forgive. Following statements of the
participants clearly shows:
“I avoid Indian products. I don’t want my money goes to my enemy’s pocket. There
are a lot of enemies but some enemies are just immediate to you.” AJ (Male, 38)
“Normally I avoided Indian product, because I don’t want to give them favor.”QJ
(Female, 30)
“I would avoid Indian product as it is our geographical rivals.” XB (Male, 27)
India-Pakistan relationships are always bad since their separation and establishment as
independent countries (1947). People from both countries show animosity towards each other.
56
Qualitative Study
2.4.4
Chapter 2
Country-of-Origin
Most of the time consumer does not know the country of origin of the brand. Familiarity with
COO may increase the chances of avoidance if the consumer has negative information about
COO, as following statement explains:
“Normally we don’t know the country of origin but if someone tells that it belongs to
X country (negative meaning associated) then I will not buy those brands.” AA
(Female, 27)
Literature tells that consumers from developing countries perceive foreign brands as being of
higher quality than domestic brands (Raju, 1995).
“I had an experience of using yogurt. I tried both local and Nestlé but I preferred
Nestlé; no matter, where it is being produced.” AA (Female, 27)
Past studies has explored the impact of COO on product evaluation (Batra, Boush, ChungHyun, & Kahle, 2000; Janda & Rao, 1997). They found that brand may take functional
attributes of the country in case of strong perceived link between brand and country of origin.
The results might be negative if consumer linked the brand with the country of producing
substandard products/goods.
“If a brand claims that it is a French brand and when you open the box and see
made in Bangladesh or China then it is useless… you are buying a French or Italian
brand but when you find that actually it is made in China then you get hurt. If it is
made in china then why should I buy an Italian brand why not a Chinese brand.”SH
(Male, 29)
Clear depiction that respondent perceive the brand produced in china or Bangladesh as lower
quality as compared to branded as originating from European countries like France or Italy.
57
Qualitative Study
2.4.5
Chapter 2
Ethnocentrism
Elliot and Cameron (1994) found that instead of buying foreign imported goods, people are
willing to buy locally made products unless the product quality and the price remain same.
“According to me, when we don’t have any national product/brand in our local
market we use the foreign brand, but when we see that a national product is available
now, with a good quality then why should not we use our national product/brand, it is
for me and I believe in it.” AJ (Male, 38)
“…and the second thing is, it included many things for example design, color, tastes
and many more. If all these things are good in our national brand then I will go for
our national brand. I would prefer to purchase our national product without
compromising on these quality related things.”AJ (Male, 38)
It is clear in above statement that people preferred to buy their local brands (if available) as
compared to foreign brands, unless they have not to make compromise on quality and esthetic
attributes.
2.4.6
Corporate social responsibility (CSR) is an old concept and well studied in literature under
different names such as business ethics, corporate citizenship (Carroll, 1998, 1999). Esrock
and Leichty (1998) emphasized CSR has four central elements such as “morally fair and
honest practices, product safety and reliability, treating employees well, and improving the
environment”. Any unethical action towards society perceived by consumer may lead to brand
avoidance. It might be related to corporate advertising message or based on the tactics used to
gain share in the market.
58
Qualitative Study
Chapter 2
“… I tell you that ARIEL (washing brand) they are emotionally black mailing people
with the name of mother. Also in the case of Dalda (Oil brand) they said « where,
there is mother there is Dalda». It means you are attaching Dalda with most beloved
persons in the family. I think it is unethical so I will not prefer these brands instead I
will buy other ones.” SH (Male, 29)
“I will never buy ‘fcuk’ brand because of its unethical branding/marketing. Initially it
was French Connection UK but later on they promote it like f**k and their sale have
been increased by 10 to 20 percent. It is totally unethical brand.” HH (Male, 32)
If the organization is not taking care of society in general and employee in particular, people
perceive it as unethical and have negative attitude towards their product.
“… Nike does not care about human rights and most of the NGO’s define its logo as
it is a cut on human to exploit it and divide it in two parts. Not providing facilities to
its employees, so I do not want to buy its brands.”AG (Male, 50)
The act of getting products manufactured in third world countries is also perceived an
unethical act as the companies promote their brands as European or US brands but get
manufactured in china or other countries.
“… It should be manufactured in equally comparable country, for example, an Italian
or French product must be manufactured in Europe. At least the company has to give
me some value as they are charging me a good value for their own brand name but
getting manufactured from a third world country.” SH (Male, 29)
Heavy investments in advertising are also perceived as unethical as explained by following
participants.
“I don’t feel that a well advertised brand is of good quality than the non advertised
brand. Secondly I think intellectual capital is being used by these companies in
making consumer goods and selling to the people.” SQ (Female, 32)
59
Qualitative Study
Chapter 2
“Lux soap, after some time they advertise on a large scale. You can see their bill
boards everywhere like they have launched a new product but actually it is not. In
actual it is just a change of packaging. So excess of advertising force me to think that
why should not they spend money on poor people. In my opinion, they just want to
make us habitual of these unimportant things.” AG (Male, 50)
2.4.7
Perceived Risk
When participants were asked about what kind of risks they perceive in buying that brand,
they responded in such a way that sometime they feel that they may lose money in buying the
product as in their opinion that product /brand is not worth of it.
“I want to buy IPhone but it is very much costly. It is not worthy of this price. It is just
that it is earning its name ‘Apple’.” GA (Male, 27)
Sometimes people do not want to buy an expensive brand because they have intuition that
perhaps they will be robbed and they might lose the expensive product and lose money.
“I think it is a costly brand and I know the conditions what is happening in Pakistan.
Mobile snatching is very common in Pakistan as you know. And if I lost IPhone then I
think I will lose much… At least, one week I would be absent minded.” HH (Male, 32)
Some brands are being avoided because of the health consciousness. People feel that the
product is not good for their health and they might get hurt with it.
“In particular if I speak about Coca Cola; personally I am avoiding Pepsi and Coca
Cola after knowing that these are not good for stomach. My health consciousness
stops me to drink Pepsi or Coca Cola. Secondly, medically I know that it destroy
tooth enamel.” AM (Female, 38)
60
Qualitative Study
2.5
Chapter 2
Discussion and Conclusion of the Qualitative Study
From the above mentioned statements (verbatim) of the participants, we would like to discuss
certain points to further explain our research problem. We have observed during our
interviews that the main reason of avoidance is based on unmet expectation which we can call
performance based avoidance. It is quite logical that if the consumer is not satisfied with the
performance of the product as expected, he will be disappointed and this disappointment will
turn in dissatisfaction and dissatisfaction lead to brand avoidance (Halstead, 1989; Hirschman,
1970; Oliva et al., 1992; Oliver, 1980).
The focus of the available research on negative responses is mainly on brand dissatisfaction
(Bodur & Day, 1978; Lee, 2007). As observed by Bagozzi et al. (1999), the focus on
dissatisfaction was due to its position as primary emotion to receive attention. These
researchers suggested to include other possible (apart from dissatisfaction) emotions and
behavior in the negative side but the research done is still limited (Bougie, Pieters, &
Zeelenberg, 2003).
We find that undesired self restricts consumers to buy certain brands. People want to make
and maintain their self concept through selection of product/brands. They have a positive
attitude towards the products/brands which enhance their self concept and have negative
attitude towards the products/brands which are inconsistence with their self concept.
In collectivist culture, people have no free will and choice. Their opinion is not independent
of others. Instead their self is perceived as collective self. Pakistan culture is collectivist in
nature, and people like to do things in groups. They go to eat and shop things in groups.
People affect on the decision making. They trust in others’ opinions for selection and
avoidance of brands.
61
Qualitative Study
Chapter 2
Animosity is well explored concept in literature. It is assumed that people avoid
products/brands originated from the country with which they feel animosity. Sometimes
people do not know brand’s country of origin and they evaluate brands on the basis of their
projected image and other features. Country-of-origin effect has been interesting topic in
research. For brands, it is important to highlight the COO because it helps consumer to select
a brand; because country’s perceived attributes also get linked with the product and brands.
On the other hand, if the country is perceived to produce substandard products. People avoid
products/brands associated with a country producing substandard products. Ethnocentrism
also plays a role in consumer’s choice decision. The higher the ethnocentrism, the higher the
inclination of the people to buy their national products without compromising the quality
related features of the product/brand (Elliott & Cameron, 1994).
Since last decade, the focus on CSR practices has been increased. Consumers are more
concerned with these practices. If they find that these companies are practicing illegal or
unethical act, then the ethical action they can take is to avoid their products/brands.
There are many perceived risks inherent in consumers purchase decision. We asked questions
concerning performance risk, social risk and overall risk before buying a product/brand.
Because in buying a brand, consumer are more concerned with the product’s performance and
the social risk of losing their self in using that brand.
In our study, the reasons that are based on experience (e.g. after using the product people feel
that it is not serving as per expectation and start avoiding that product) we consider them as
post-purchase determinants of brand avoidance. The reasons that are not based on
experience, instead of based on consumer’s personal, ideological, societal or environmental
beliefs are called pre-purchase determinants of brand avoidance.
62
Qualitative Study
Chapter 2
To establish the link between above mentioned concepts a survey study is conducted
including questions on consumer’ undesired self congruence, negative social influence,
Animosity, familiarity with country-of-origin, ethnocentrism, perceived unethical corporate
citizenship and perceived risk along with demographic items.
2.6
Conclusion
This chapter provided the detail of the qualitative approach used for initial explorative study
and the method opted to capture the emerging themes from the interviews. Interview guide is
made according to which the responses are obtained. How the themes are identified and
emerged from data is also discussed. The emerged themes have been supported in the
literature (cf. chapter 3).
63
Theoretical Framework & Hypotheses
Chapter 3
Chapter 3:
3
“A fact is a simple statement that everyone believes. It is innocent,
unless found guilty. A hypothesis is a novel
suggestion that no one wants to believe.
It is guilty, until found effective”
– Edward Teller
3.1
In previous chapters, we have defined the research domain and basis of our thesis. This
chapter provides the theoretical background of the variables (determinants) emerged from
qualitative interviews (cf. chapter 2) that may have significant effect on pre-purchase brand
avoidance. It provides a bird’s eye view of the research conducted in our domain. It also
identifies the lead researchers and discusses their work. Constructs are conceptualized to
develop hypotheses on the basis of which a hypothetical research model is proposed.
3.2
Determinants of Brand Avoidance
On the basis of our qualitative study which is based on interviews (cf. Chapter 2); we have
found diverse factors predicting the brand avoidance attitude. In our study, we divided the
determinants into two broad categories (Figure 3.1): Pre-purchase determinants and Postpurchase determinants. The former is based on the personal, social and societal needs;
whereas, the later one is related personal experience that consumer has with the product or
brand (Lee, Motion, et al., 2009).
64
Chapter 3
Post-Purchase Determinants
Brand Avoidance
Pre-Purchase Determinants
Figure 3.1: Determinants of Brand Avoidance
A significant literature is available on post-purchase determinants such as unmet expectation,
poor performance of the product or brand, and dissatisfaction with the product or brand
(Fournier & Mick, 1999; Lee, Motion, et al., 2009; Oliva et al., 1992; Oliver, 1980).
The focus of the available research on post purchase negative responses is mainly on brand
dissatisfaction (Bodur & Day, 1978; Lee, 2007). As observed by Bagozzi et al. (1999), the
focus on dissatisfaction is due to the fact that it is primary emotion to receive attention. Some
people avoid certain products or brand because of negative associations and meanings that
they do not want to represent (Banister & Hogg, 2004; Englis & Solomon, 1997; Levy, 1959;
Thompson & Arsel, 2004). These researchers suggested to include other possible (apart from
dissatisfaction) emotions and behavior on the negative side but the research done on this issue
is still limited (Bougie et al., 2003). So in our study our focus is on the pre-purchase
determinants of the brand avoidance as highlighted in the Figure 3.2.
Post-Purchase Determinants
Brand Avoidance
Pre-Purchase Determinants
Figure 3.2: Pre-Purchase (Focused) Determinants of Brand Avoidance
65
Chapter 3
In the next section, we will elaborate the literature particularly related to the pre-purchase
determinants of brand avoidance.
3.3
Pre-Purchase Determinants of Brand Avoidance
Our study focuses on the reasons that motivate consumers not to consume a particular brand
beyond product-related attributes. In literature we can find that a variety of reasons can form
negative brand attitudes which motivate people to refrain from using a brand. Consumers may
boycott a brand because of its explicit commercialism, violation of human rights or the
environment, or involvement in unethical business practices (Friedman, 1985; SandIkci &
Ekici, 2009). Animosity towards the country from which the brand originates may be the
cause of avoiding a brand as consumer may believe that buying a foreign-made product/brand
is unpatriotic (Klein et al., 1998; Shimp & Sharma, 1987). Furthermore, Consumers may
choose not to buy a brand as the brand symbolizes an undesired self or unwanted reference
group that they do not want to identify with (Banister & Hogg, 2001; Englis & Solomon,
1997). Also consumers may reject brands that are not well-matched with their social roles or
with other products or brands that they already use (Hogg, 1998). Whereas, Garretson and
Clow (1999) noticed that during the process of purchasing the products/brands, the consumers
perceive different types of risks related to product itself or associated with the usage of that
product. In the following section, we will discuss these themes in more detail.
66
3.3.1
Chapter 3
"Without a tangible, undesired self, the real self would lose its navigational cues"
(Ogilvie, 1987)
Consumers form their self-concepts and identify their social reference groups through what
they choose not to consume as well as what they consume (Banister & Hogg, 2001; Englis &
Solomon, 1997; Hogg, 1998; Hogg et al., 2000). As Englis and Solomon (1995) posit,
“consumers may eschew purchase, ownership, and use of such products and activities owing
to their reluctance to be identified with an avoidance group”. In the same way, consumers
define, “not me” by refusing “anti-constellation” to stay away from an association with the
related stereotype (Hogg, 1998; Hogg & Michell, 1996). Anti-constellation “represent the
complementarity of negative choices across multi-category products” (Hogg, 1998).
Undesired self concept is introduced by Ogilvie (1987), which he described as a least desired
identity, consist of the sum of negatively valenced traits, memories of unhappy experiences,
embarrassing situations, fearsome events, and unwanted emotions, the individual is
consistently motivated to avoid (Bosnjak & Rudolph, 2008). Undesired self should more
strongly predict the consumer’s well being relative to idealized selves as predicted by Ogilvie
(1987). He gave the reason that undesired selves are relatively more concrete self –images
based on the past experiences of failure, humiliation, and guilt. In contrast, idealized selves
are mostly based on abstract, perfectionistic and unattainable goals, so people may have little
direct experience with actualizing their ideal selves (Phillips, Silvia, & Paradise, 2007). As a
consequence, undesired selves should have an emotional strength that ideal selves lack. In the
first ever study of the undesired self, Ogilvie (1987) evaluated aspects of the real self ("how I
am most of the time"), the ideal self ("how I would like to be"), and the undesired self ("how I
hope to never be"). In his study he used actual-ideal discrepancies and actual-undesired
67
Chapter 3
discrepancies to predict life satisfaction and both of the discrepancies predicted the wellbeing. But actual-undesired discrepancies had stronger relations with well-being (r = -0.72)
relative to actual-ideal discrepancies (r = 0.37). Regression analyses found that only the
undesired self exclusively predicted life satisfaction (Phillips et al., 2007).
Undesired self is a psychological construct and most relevant to brand avoidance. Ogilvie
(1987) suggests that undesired self is a set of associations and values with which people do
not want to be linked and afraid of giving a negative self. Other researchers argue that in
comparison of what people want to be, they may have clear idea about what they do not want
to be, and this ‘push’ away from undesired self might be more effective than the ‘pull’
towards the ideal self (Lee, 2007; Ogilvie, 1987). Previous study mentioned the response of a
participant, “you can get away with not having a complete positive image by just completely
avoiding the negative image” (Banister & Hogg, 2001).
Most of the studies investigated this concept in self-image congruence area (Englis &
Solomon, 1997; Sirgy, 1982). The idea behind those studies is that people are equally
motivated to avoid their undesired self as they want to approach the ideal self or maintain
their actual self. Lee (2007) proposed that undesired self motivate people to avoid behavior,
such as an affluent women may avoid the stores who use the label ‘urgent sale’ to attract and
the teenagers who avoid product symbolic of ‘family fun’ (Levy, 1959). As in first example
affluent women may think that only poor or inferior people purchase on ‘urgent sales’ so this
could be her undesired self ‘to be inferior’ which motivate her to not shop from those stores.
And similarly in later example, teenager’s undesired self of ‘not being protected by family’
motivate them to avoid such stores or products.
Since research in anti-consumption is rare and only specific studies have been done in
developing the link between specific brand and undesired self. In this domain Banister &
68
Chapter 3
Hogg (2004) investigated the role of undesired self and the avoidance of some products. The
notion from the literature can be taken as “brand selection is based on construction of an ideal
self or the maintenance of the actual-self, through consumption of desired congruent brands,
and brand-avoidance involves consumers distancing themselves from their undesired selves
by rejection symbolic incongruent brands” (Lee, 2007). A relevant concept in the literature is
organizational disidentification which helps in understanding undesired self.
3.3.1.1 Organizational Disidentification
It is evident in literature that not only products and brands help people to define and enhance
their self concept but also they formulate their self-concept through the organization that they
identify or disidentify with (Elsbach & Bhattacharya, 2001; SandIkci & Ekici, 2009).
Organizational identification defined as “ the degree to which a person define him or herself
as having the same attributes that he or she believes defines the organization” (Dutton,
Dukerich, & Harquail, 1994). By developing a connection with the organization a person can
preserve his or her self-concept. However, Elsbach and Bhattacharya (2001) suggest that by
separating oneself from the organization can also help maintain self-concept. Disidentification
is a helpful concept in learning undesired self and brand avoidance. They defined
organizational disidentification “as a self-perception based on (1) a cognitive separation
between one’s identity and one’s perception of the identity of an organization, and (2) a
negative relational categorization of one-self and the organization” (Elsbach & Bhattacharya,
2001). In their study Elsbach and Bhattacharya (2001) gave a framework that outlines the
indicators, antecedents, and consequences of organizational disidentification. In some cases it
happens that companies fail in portraying the real image of the brand and the message they
wish to covey and this discrepancy result in undesired self (Aaker, 1996; de Chernatony &
Dall’Olmo Riley, 1998).
69
Chapter 3
A second relevant concept under the head of undesired self congruence that exists in the
literature is negative reference group. People do not want to be associated with unwanted
reference groups.
3.3.1.2 Negative Reference Group
Social identity theory tells that it is inherent in a person’s personality to be associated with a
group that represents a meaning to him (Abrams & Hogg, 1988; Tajfel, 1978). So People try
to have and maintain relationship with the groups evaluated as being positive and stay away
from the groups evaluated as negative (Lee, 2007). It is also supported in academic literature
that consumers develop their self concept by identifying with positive reference groups, while
disidentifying with negative reference groups (Bhattacharya & Elsbach, 2002; Elsbach &
Bhattacharya, 2001; Englis & Solomon, 1995, 1997). Marketers are very much interested in
creating a link of their brand with positive reference groups so that consumer may choose to
consume their brand to be associated with their aspiration reference groups (Solomon, 2002).
In their study Escalas & Bettman (2005) found that consumers form high self brand
connections with the brands that have the images congruent with the images of an in-group
‘one’s own group’ in comparison with the brands that have inconsistent images in-group. On
the other hand, they have also found that self-brand connections are lower for the brands that
have images consistent with the images of an out-group “group to which one does not
belong”. Consumers will most probably accept meaning from brands linked with an in-group
and reject meanings associated or consistent with an out-group (Escalas & Bettman, 2005).
Consumers’ likes, dislikes and distastes also play an important role in making his/her choices
for products or brands.
70
Chapter 3
3.3.1.3 Dislike and Distastes
“Tastes (i.e. manifested preferences) are the practical affirmation of an inevitable
difference. It is no accident that when they have to be justified, they are asserted
purely negatively, by the refusal of other tastes”
(Bourdieu, 1984)
A significant number of publications (Ogilvie, 1987; Wilk, 1995, 1996, 1997) suggests that
what we choose not to consume is an important aspect of both individual and group identity.
Our distastes or the ‘refusal of tastes’ may say as much about us personally and socially as
that which we choose to consume. Tastes normally keep positive connotations, but it is also
noticed that tastes are mainly emphasized in negative terms through the ‘refusal’ of other
tastes (Bourdieu, 1984). Hence, we can define ‘good taste’ only through a complete
knowledge of what makes ‘bad taste’, and this varies among individuals. As Wilk (1997) said
that consumers feel less difficulty in articulating their distastes and dislikes than they do in
their desires and likes, and also in communicating the negative product user stereotypes ‘the
undesired self’ and negative implication that can be linked with product cues.
Rozin and Fallon (1987) specifically focus on the concept of ‘disgust’ as an extreme type of
rejection. People are able to define their self through contrasting their tastes with those of
others, and differentiate themselves through the dislikes of something with others. They posit
that these notions of disgust and rejection can be linked with the undesired self and the themes
of avoidance and aversion (Hogg, 1998).
In reality, we cannot separate consumption from non-consumption. There are two groups of
choices: choices based on want and need, and choices based on aversion and distastes, but
they are not alternatives or opposites. Both sets can be seen alternatively depending upon
socially or individually positive or negative valences, as being materialistic or spiritual, active
71
Chapter 3
or passive (Wilk, 1996). In a same way, not having, not wearing, not eating, can be a form of
conformity or social distinction as having, wearing or eating.
In his work, Wilk (1996) found that both desires and distastes can work as relations of
inclusion and exclusion. People can define themselves as members of a group because they
share taste with others, or because they share distastes with others. Our dislikes and aversions
are our personal things and mostly known to friends and relatives, where as our likes are
publicly stated in our conspicuous choices of clothes, cars, houses and other goods.
In his interviews, he exclusively stated that when he asked questions about people that why
they like certain TV shoes or music, they gave statements like “Well, I just like it” or “It just
tastes good to me, that’s all”. But he asked about what they did not like, people have a lot to
say and he often could not stop them (Wilk, 1996). So people are very clear about what they
do not want to choose rather what they want. Sometimes people overtly define themselves by
contrasting their tastes with others, and they emphasize that for example, they were not the
kind of person who wears those kinds of clothes, drink that kind of liquor, or eat those sorts of
dishes.
In a recent work on brand dislike, Dalli et al. (2006), defined brand dislike as “the negative
judgment expressed by the consumer and/or implied in the choice not to buy a brand”. They
divided dislike into three levels: product brand level, user brand level and corporate brand
level (Figure 3.3). ‘Product Level’ respondents criticize those brands which shows an unfair
price/quality ratio, or whose products perform poorly, and that do not provide effective
customer care services. In other words, consumers felt dissatisfied with some products and
service characteristics and they end up disliking the brand. They found a mismatch between
their expectation and the performance of the product. ‘User Level’ respondents criticize those
brands which are associated with negative stereotypes users and they do not want to be linked
72
Chapter 3
with. Some time brands are not liked for its commonness: consumers are convinced that
particular brand is not capable of portraying them as they really want to be. ‘Corporate
Level’ respondents criticize those brands which belong to those companies whose behaviors
and activities they judge illegal, immoral and unethical. No matter what the product
characteristics are, consumer dislikes the brand because of their companies’ perceived unfair
behaviors and abuses.
There are a number of reasons upon which consumers can develop negative brand attitudes,
some of which are individualistic nature (functional, egotistic) and others may have
collectivistic an ideological nature i.e. social, cultural, ethical (Dalli et al., 2006).
LEVELS AND FACTORS OF BRAND DISLIKE
Collectivistic
Exploitation and transgression
Fake and persuasive
communication
Dislike as resistance
practice
Negative stereotypes
Dislike as social
communication
Individualistic
(materialistic, snobbish, elitist)
Poor self expression properties
Price/quality relation
Dislike as refusal
of inadequate
partner-brands
Product performance
Customer services
Corporate
brand
level
User
brand
level
Product
brand
level
Figure 3.3: Levels and Factors of Brand Dislike
Source: Dalli et al. (2006)
The literature mentioned suggests that undesired self in all dimensions is an important aspect
in consumer’s set of selves to choose or not to choose a specific product or brand. So in this
study we have following hypothesis to see the relationship between undesired self and brand
avoidance:
H1 :
Undesired self congruence positively influences consumers’ brand avoidance
attitude.
73
3.3.2
Chapter 3
Consumers believe that they are capable of making their independent opinions but what other
people think can also influence them, especially people with negative attitudes can make a big
change in their decisions (Duhacheck, Zhang, & Krishnan, 2007). Consumer attitudes towards
products/brands are influenced by others around them. In current social networking era, the
society is becoming more inter-connected through online social networks such as Facebook,
twitter, and Myspace. This influence is even becoming a significant driver of individual
consumer attitudes (Duhacheck et al., 2007).
Theory of reasoned action (TRA) suggests that one’s behavior is based on one’s intention to
engage in the behavior and in turn, the behavioral intention is best predicted by both
attitudinal and normative influences (Fishbein & Ajzen, 1975). Whereas, subjective norm is
defined as an individual’s perceptions of what significant others think about the individual
performing a specific behavior (Fishbein & Ajzen, 1975). This theory emphasized that
subjective norms are influenced by other people’s expectations and an individual’s motivation
to comply with these expectations. People may choose to perform a behavior which they do
not want to perform with their opinion but they do it just to comply with reference group’s
desirability (Kulviwat, Bruner II, & Al-Shuridah, 2009).
Social influence is defined as, the degree to which members of a reference group influence
one another’s behavior and experience social pressure to do a specific behaviors (Kelman,
1958; Kulviwat et al., 2009). In a similar way, in this study the term negative social influence
may be taken as the degree to which members of a reference group influence one another’s
behavior and experience social pressure to not to do a specific behavior. In their study,
Duhacheck et al. (2007) found that negative opinions had a particularly strong influence on
74
Chapter 3
the attitudes of others. They further, found that people with negative opinions about products
have tendency to become more negative while participating in group discussions.
It is a common phenomenon that when people confronted a dissatisfaction situation with a
product/brand or service, they may or may not speak up with others depending upon their
personality characteristics (Lau & Ng, 2009). For example; a quite person may not speak up
and a social type person may willingly engage in negative WOM. Sociable people tend to be
in contact with more people and discuss their negative product experiences. Especially,
consumers who are dissatisfied with a purchase or rejected or discontinued using a product
are telling friends about the experience and influence them to avoid it (Bodur & Day, 1978;
Lau & Ng, 2009; Leonard-Barton, 1985). Three types of personality construct namely self
confidence, sociability, and social responsibility, which motivate them to influence others.
Self-confidence: It is defined as the extent to which an individual believes himself to be
capable, significant, successful, and worthy (Coopersmith, 1967). Past research in this area
has revealed the effects of self-confidence on persuasibility (Cox & Bauer, 1964), and
consumers who complain tend to be more self confident and assertive (Day, 1978). Higher
social class also play vital role in individual’s personality and people for higher social class
found to be more self confident and tend to perceive less risk of embarrassment in
complaining (Bearden & Teel, 1983).
Sociability: A sociable person is one who likes outgoing, enjoys being with others, and has a
participative temperament (Lau & Ng, 2009). It is a characteristic of a sociable person that he
will tend to be in contact with more people, try to increase his chances of discussing negative
product experiences with others.
Social responsibility: Socially responsible people are inclined to help other people even when
there is nothing to be obtained from them. They expend their efforts due to strong standard of
75
Chapter 3
right and wrong (Berkowitz & Lutterman, 1968). From the literature of word-of-mouth
(WOM), it is suggested that if someone has the experience of dissatisfaction with product,
brand or service, his desire to prevent others from experiencing a similar fate may motivate
him to engage in negative WOM. Since socially responsible people are concerned about other
people’s welfare, they influence by spreading negative WOM to warn others about the
unsatisfactory product, brand, or service (Lau & Ng, 2009).
According to Harrison-Walker (2001), WOM may be described as “informal, person-toperson communication between a perceived noncommercial communicator and a receiver
regarding a brand, a product, an organization, or a service”. Past studies say that WOM is an
essential channel of interpersonal influence in consumer decision making for purchasing
(Grewal, Cline, & Davies, 2003). WOM communications can be positive, negative or both
(Brown, Barry, Dacin, & Gunst, 2005; Laczniak, DeCarlo, & Ramaswami, 2001; Nyer &
Gopinath, 2005), and they have a stronger influence than printed information on product
judgment (Herr et al., 1991). Negative WOM is one form of consumer reaction to
dissatisfaction with a product/brand which has received little attention from business firms
(Lau & Ng, 2009). These days mass media is effectively generating awareness of consumer
products, but consumers mostly rely on word-of-mouth (positive or negative) when making
their purchase decisions for products/services (Davis, Guiltinan, & Wesley, 1979). Many
researchers have suggested that consumers have tendency to give more weight to negative
information in product or brand’s evaluation (Lutz, 1975; Mizerski, 1982). In other ways,
marketing scholars and managers have the opinion that non-marketing sources of information
such that word-of-mouth (positive or negative), are given substantial weight by consumers
while evaluating the product or brand (Richins, 1983).
76
Chapter 3
Richins (1983, 1984) discussed the importance and impact of NWOM, and she defined
NWOM as “interpersonal communication among consumers concerning a marketing
organization or product which denigrates the object of the communication” (1984, p.697). She
emphasized that continuous and organized NWOM by a large group of consumers who are
not satisfied with product/brand or services will have substantial impact on consumer choice.
The strength of WOM (positive or negative) communications is a result of its high rate of
diffusion. In their empirical study, Engel et al. (1969) found that the people who tried a new
service , 90% of those has told something (positive or negative) about their experience to at
least one person and about 40% had told two or more people within a few days after their
trial. Blodgett et al. (1993) mentioned in their study that 75% of all their respondents were
engaged in NWOM.
In our study, we would like to know the effect of negative social influence towards the
avoidance of certain brands. So we have following hypothesis:
H2 :
Negative Social Influence positively influences consumers’ brand avoidance
attitude.
77
3.3.3
Chapter 3
Animosity
Literature about animosity has established that people avoid products/brands from specific
countries, not because of the quality of the product, but because of the perceived animosity
towards the country of origin (Charles, 2012; Klein et al., 1998). There could be many
reasons, such as history of conflict, political positions, religious, ethnic or cultural difference,
overt or covert acts of violence or terrorism, of why people of one country develop animosity
feelings towards another country and as a result restrict the consumption of products or
services originating from that country.
Klein et al. (1998) introduced the concept ‘Animosity’ into literature by developing and
presenting the Animosity Model of Foreign Product Purchase, and tested and validated it on
Chinese consumers’ attitude towards Japanese-made products. They defined the construct of
animosity as “the remnants of antipathy related to previous or ongoing military, political, or
economic events”, affect consumers’ purchase behavior in the international marketplace. As
the world has become global village and companies have more access to deliver their product
to other countries and at the same time consumers from different countries have the liberty to
choose products and brands. So it is evident from literature that products’ origin can affect
the consumer buying decision irrespective of the product judgment. i.e. consumer might avoid
products or brands from offending country not concerning to the quality of goods, but because
the exporting country has engaged in military, political, or economic acts (Klein et al., 1998).
There could be many sources of having animosity towards another country, from a minor
reason that the two countries are sharing a contiguous border (like the US and Canada), or due
to some serious manifestation stemming from previous military events or recent economic or
diplomatic disputes (like Pakistan and US, US and Iran).
78
Chapter 3
Recent international tensions over the war in Iraq and Afghanistan have increased antiAmerican sentiments and consumers’ boycotting of U.S. products throughout the world
(CNN, 2003; Economist, 2005). History is full of such incidents of the dramatic and
damaging effects of hostility between nations. If international tension can lead to armed
conflict and atrocities, it seems reasonable that animosity toward a current or former enemy
also will affect willingness to buy products produced in or by firms from that country. To
date, the marketing and consumer behavior literature largely has ignored the construct of
animosity between nations and its potential impact on foreign product purchase (Russell &
Russell, 2006).
Majority of the studies has explored animosity against American or Japanese products/brands
(Table 3.1). Klein and Ettenson (1999) have explored US consumers of Japanese products to
find out antecedents for the animosity and ethnocentrism constructs. They found that
animosity and ethnocentrism are two separate and distinct factors and the profile of the
ethnocentric consumer is different from the consumer having animosity towards a particular
exporting country.
Further, in an additional study on US consumers, Klein (2002) validated the notion that
international animosity and consumer ethnocentrism are distinct constructs which play
differently according to the products available to them. The results proposed that the construct
animosity towards foreign country is related to the choices among foreign products, whereas
the construct ethnocentrism is related to the choices between domestic and foreign products.
In an empirical study, Shin (2001) concluded that animosity model does work in Korea, by
testing and validating the animosity model on Korean consumers using Japanese-made
products.
79
Chapter 3
Table 3.1: Major Animosity Studies
Klein et al.
1998
Home
Country
China
Klein and Ettenson
1999
USA
Japan
Shin
2001
Korea
Japan
Klein
2002
USA
Animosity: Japan
Neutral: Korea
Nijssen and
Douglas
2004
The
Netherlands
Germany
Convenience Cars and TVs
Sampling
Shimp et al.
2004
USA South
US North
Hinck
2004
Eastern
Germany
Western Germany
Convenience Computers,
Sampling
electricity, film,
ISP, cellular
phone service
Convenience No specific
Sampling
Product
Klein et al.
2005
Croatia
Cicic et al.
2005
BosniaHerzegovina
Ettenson and Klein
2005
Australia
BosniaHerzegovina,
Serbia, USA, W.
Europe
BosniaHerzegovina,
Serbia, Croatia
France
Russell and Russell
2006
USA, France
USA, France
Shoham et al.
2006
Jewish
Israelis
Arab Israelis
Riefler and
Diamantopoulos
2007
Austria
Not Specific
Rose et al.
2008
Israel
UK, Italy
Sampling
Product
Russell and Russell
2010
France
USA
Sampling
Product
Charles
2012
USA
Japan
Author (s)
Year
Foreign Country
Japan
80
Sampling
Product (s)
Convenience TVS, stereos,
Sampling
radios,
cameras,
refrigerators
Random
No specific
Probability
Product
Sampling
Sampling
Product
Judgment
Sampling
Cars
Quota
Sampling
No specific
Product
Not
Mentioned
No specific
Product
Area
Probabilities
No specific
Product
Simple
Film Industry
Random
Sampling
Convenience Bread and
Sampling
pastry, olives
and olive oil
Sampling
Product
Not
Mentioned
No specific
Product
Chapter 3
The animosity model has been extended and confirmed by Nijssen and Douglass (2004) in
small countries where domestic products or brands in many product categories are not
available in local markets.
They explored the Dutch consumers’ response for German
products and concluded that even in those countries ethnocentrism and feelings of animosity
have an impact on the evaluation of foreign products. Further to extend this model at regional
level, Shimp, Dunn and Klien (2004) found the same results of animosity model on southern
US consumers with products originating from the northern US.
In the original animosity model (Klein et al., 1998) consumer judgments and attitudes’ effect
were not examined, although other studies empirically proved that animosity also related to
consumers evaluative response (Russell & Russell, 2010). Shoham et al. (2006) in assessing
Israeli Jewish consumers’ opinion about Israeli-Arab products found that animosity equally
impacted product quality judgments and willingness to buy. Similarly, Yang and TSO (2007)
showed that animosity towards China was significantly related to Taiwanese consumers’
attitude toward TV program imported from China.
In reference to above discussion about animosity towards country of origin, we have proposed
following hypothesis:
H3 :
Consumers’ perceived animosity towards the Country-of-origin positively
influences consumers’ brand avoidance attitude.
81
3.3.4
Chapter 3
The role of business in society has been a subject of discussion since the middle of the last
century (Turker, 2009). Concerns towards humanity and the natural environment have now
been increased among people all around the world considerably. Today, stakeholder in
national and international communities anticipates more balanced and responsible use of
increased business power. In literature some researchers call it corporate social responsibility
(CSR), others refer to it as corporate ethics, and in recent times businesses’ social
performance has been framed as ‘corporate citizenship’ (Carroll, 1998).
Practically, a stakeholder is any entity, typically (but not essentially) outside of the firm, that
impacts the organization and that the organization seeks to influence (Murray & Vogel, 1997).
Generally, stakeholders include: customers, competitors, financial community, government
and regulatory agencies and /or political-activists groups. As it has been acknowledged long
ago that these groups can considerably change the viability of the firm, so knowledge of
relevant stakeholder groups is of significant interest to business firms (Dill, 1958). In this
regard, CSR contributions represent an important “exchange” between the firm and its
stakeholders (Murray & Vogel, 1997). Apart from the fact that CSR is most widely studied
concept in the literature, yet it is difficult to give a precise and commonly accepted definition.
According to Bowen (1953) CSR is the obligations of businessmen, “to pursue those policies,
to make those decisions, or to follow those lines of action which are desirable in terms of the
objectives and values of our society”. Whereas, Carroll’s (1999) elaborative review traces the
evaluation of the CSR construct since 1950s, and he clearly indicated that one of the main
problems in the literature of CSR is to sketch out a conceptual framework for it.
Carroll (1998) has divided corporate citizenship into four faces: economic face, a legal face,
an ethical face and a philanthropic face. Similarly, Esrock and Leichty (1998) emphasized
82
Chapter 3
CSR has four central elements such as “morally fair and honest practices, product safety and
reliability, treating employees well, and improving the environment”.
Davis (1973) made the distinction between social responsibilities of a business and its
economic, technical and legal obligations. Further, Carroll (1999) distinguished between the
economic and non economic components of CSR; the former one is what the business does
for itself, whereas the later component is what the business does for others. In a recent study
Turker (2009) excluded the economic component from the definition of CSR and define it
exclusively a corporate behaviors that aims to affect stakeholders positively and that go far
away its economic interest.
Consumers’ evaluation towards corporate citizenship has been explored through industry
surveys (Jones, 1997). The results show that consumers are willing to make an effort to
support proactive corporate citizens. For instance, a 1997 Cone/Roper study showed that 76%
of consumers are ready to switch to brands or stores that show concern about the community
(Jones, 1997). Brown and Dacin (1997) in a series of laboratory experiments showed that
negative CSR associations may have a negative effect on overall product evaluation, whereas
positive associations may enhance product evaluations. Further, Maignan et al. (1999) by
conducting a survey on managers established a positive relationship between proactive
citizenship and customer loyalty.
In behaviors evaluation or impression processing psychology affirms two different types of
attributional biases or tendencies that is negativity effect or positivity effect. A group of social
science researchers have studied the negativity effect, and argued that people put more weight
on negative information as compared to positive information when evaluating an object or
target (Fiske, 1980; Hamilton & Huffman, 1971; Klein, 1996). Some scholars expanded this
negativity bias to a product level by conducting studies into the context of word-of-mouth
83
Chapter 3
communication (East, Hammond, & Lomax, 2008; Granbois, Walters, & Blodgett, 1993; Herr
et al., 1991; Laczniak et al., 2001). Similarly for CSR, if it is considered as information
related to morality, then negative CSR may be given more weights (McWilliams & Siegel,
2001).
Frooman’s (1997) meta analysis argued that the stock market reacts negatively to a firm if
they do socially irresponsible or illicit activities, whereas he could not find significant
evidence for socially responsible activities’ impact on firm value. In a similar way, Johnson
(2003) build up a CSR continuum framework which emphasized that a firm’s socially
irresponsible or illegal activities hurt financial performance, but socially responsible activities
cannot be linked to any fiscal advantages.
How the firm is being viewed and evaluated by stakeholders is the main concern for the
subsequent interaction. Although the concept of corporate citizenship is commonly employed
by business practitioners but academically it is not conceptually formulated. Research in the
area of CSR suggests that corporate citizenship may be defined as “the extent to which
businesses assume the economic, legal, ethical, and discretionary responsibilities imposed on
them by their stakeholders” (Maignan & Ferrell, 2001). Most importantly, whether the
firm/organization is perceived to be ethical and socially responsible is a matter of interest for
managers (Freeman, 1994; Langtry, 1994; Schlossberger, 1994). We have the supposition that
when consumer perceives corporate citizenship unethical, he may be motivated to avoid that
firm or its products/brands. Hence, our hypothesis posits as:
H4 :
Perceived Unethical Corporate Citizenship (PUCC) positively influences
consumers’ brand avoidance attitude.
84
3.3.5
Chapter 3
Perceived Risk
As the results of an exchange are uncertain, so consumers want to reduce their risk involved
in their purchasing. Since 1960s (Bauer, 1960), perceived risk theory has been used to explain
consumers’ behavior. In literature we can find different definitions for perceived risk. For
instance, Peter and Ryan(1976) defined perceived risk as a kind of subjective expected loss
and Featherman and Pavlou (2003) defined perceived risk as the possible loss when pursuing
a desired result (Lee, 2009). Many researchers have examined the impact of risk on traditional
consumer decision making (Lin, 2008; Mitra, Reiss, & Capella, 1999). Some researchers
argued that perceived risk is a two-dimensional construct which consist of uncertainty related
to purchasing decision and the consequences of taking an adverse action (Bettman, 1973;
Cunningham, 1967; Mitra et al., 1999; Schiffman, 1972).
Although perceived risk expose a mixture of dimensions and meanings based on different
research targets and purposes, but most of the scholars claimed that the consumers’ perceived
risk is a kind of multi-dimensional structure. In marketing literature, six components or types
of perceived risk have been identified (Jacoby & Kaplan, 1972; Kaplan, Szybillo, & Jacoby,
1974). Jacoby and Kaplan (1972) identified five dimensions as follows:
1. Financial Risk
2. Performance Risk
3. Physical Risk
4. Social Risk
5. Psychological Risk
Whereas the sixth risk parameter, ‘time risk’ was identified by Roselius (1971). In their
research Peter and Tarpey (1975) studied all the six dimensions for the automobile purchase.
However, it is not obligatory to study all the dimensions of perceived risk; it may vary
85
Chapter 3
according to the product (or service) class (Featherman & Pavlou, 2003) (Table 3.2). From
company’s perspective, exploring those factors which influence consumer’s perceived risk is
one of the challenges for managers, to reduce deficits and have efficient response (Lin, 2008).
Table 3.2: Description and Definition of Perceived Risk Facets
Source: Featherman & Pavlou (2003)
Originally perceived risk was studied in psychology but now in recent years, it is considerably
applied in decision-making and in explaining consumer’ behavior (Chaudhuri, 1997; Dowling
& Staelin, 1994; Folkes, 1988). In other disciplines such as psychology, economics, statistical
decision theory and game theory, the risk concept is associated to choice situation concerning
86
Chapter 3
both potentially positive and potentially negative outcomes (Stone & Grønhaug, 1993). A
typical example of this situation is a lottery involving possible chances for gain and loss.
While studying perceived risk in consumer behavior, the focus is primarily on potentially
negative outcomes only. This is the significant distinction about risk, as understood in
marketing, versus as it understood in other disciplines.
In the past studies, the methods were explored that affect the perception of product quality
evaluation or reduce the perceived risk upon internal or external clues of the products or
services. Among the external clues, brand equity is the critical factor. Before purchasing the
products/brands, when consumers cannot completely understand and assess the properties of
the products, at that time, clues of brand image and awareness can reduce the risk perceived
by the consumer while purchasing the products/brands (Lin, 2008).
According to Lin (2008), the perceived risk concept must include following four elements:
First element is the stimulation of the external motivation that produce the visible drive in the
consumers’ mind. Second element refers to uncertainty that implies the meaning of possible
happening of certain events intuitively believed by the consumers. Third element is the result
that clarifies the risk of the result after specific indent happens, and the fourth element is the
disparity of different individuals towards risk perception and the influence on the evaluation
of risks.
Our research followed the statement of Stone and Grønhaug (1993) and divided perceived risk
into financial, performance, physical, social and psychological risks:

Financial risk: The probability that a purchase results in loss of money as well as the
subsequent maintenance cost of the product

Performance risk: The possibility of the product malfunctioning and not performing
as it was designed and advertised and therefore failing to deliver the desired benefits
87

Chapter 3
Physical risk: The probability that a purchased product results in a threat to human
life

Overall risk: A general measure of perceived risk when all criteria are evaluated
together
Zeithaml (1988) and Dodds et al. (1991) argued that the consumers’ wish to purchase depend
upon and would be influenced by objective price, perceived quality, perceived value and
properties of the products. Garretson and Clow (1999) noticed that during the process of
purchasing the products/brands, the consumers would perceive different types of the risks.
Especially, when these perceived risks were considerably significant, they would influence
the consumers to not to purchase that product or brands. Taylor (1974) in his study proposed
the theory of consumers’ perceived risk and suggested that during the purchasing decision
making, consumers’ choice would be affected because of different levels of perceptions. Thus
perceived risks had significant influence on the consumers’ purchasing or not to purchase
specific brands. So our research believes that when the level of perceived risk increases the
consumers is more likely to have a negative attitude towards those brands. In reference to
above discussion we have following proposition:
H5 :
Consumers’ Perceived Risk positively influence consumers’ brand avoidance
attitude.
88
3.4
Chapter 3
Brand avoidance attitude and brand avoidance intention
The consumer behavior literature (Fishbein & Ajzen, 1975; Lutz, 1975) suggests that
consumer’s attitude influence his choice. According to Fishbein & Ajzen (1975) attitude is an
individual’s positive or negative feelings about performing a particular behavior. It is also
identified as one of the main determinants of behavior (Shepherd & Raats, 1996), and
strongest predictor of behavioral intentions (Honkanen, Verplanken, & Olsen, 2010; Olsen,
2004). Some studies argue that attitudes alone are often poor predictors of market place
behaviors (Iyer & Muncy, 2009; Shaw & Clarke, 1999).
Intention is defined as indication of how much effort people are planning to exert in order to
perform a behavior (Ajzen, 1991). He further, emphasized that the most immediate
determinant of any behavior is the intention to perform that behavior. Intention is also
considered as a cognitive component that mediates the relationship between attitude and
behavior (Ajzen & Fishbein, 1980; Debbabi, Daassi, & Baile, 2010). Past studies have
provided the evidence of positive relationship between attitude and intention (Povey, Wellens,
& Conner, 2001; Verbeke & Vackier, 2005). Bagozzi and Yi (1989) suggested that the
formation of intentions is initiated with attitudes and the will to perform a particular action
will be a function of the strength of one’s attitude.
engage in the behavior and in turn, the behavioral intention is best predict by both attitude
(one’s positive or negative evaluation of performing the behavior) and normative influences
(Bagozzi, 1992; Fishbein & Ajzen, 1975).
Consumers’ attitude towards brands plays a significant role in influencing their purchase
intention (Phau, Teah, & Lee, 2009). The theory of planned behavior (TPB) also emphasized
that the purchase behavior is determined by the purchase intention, which is further predicted
89
Chapter 3
by attitude (Fishbein & Ajzen, 1975). Purchasing of brands is not only based on the price
factor but also some non-price factors influence the consumption decision such as ‘attitude’
(Kwong, Yau, Lee, Sin, & Tse, 2003).
Unethical decision making, including the purchase of brands performing unethical
advertising, is mainly explained by attitude, irrespective of product class (Phau et al., 2009;
Wee, Ta, & Cheok, 1995). If consumers have a favorable or positive attitude towards certain
brands they will prefer to purchase those brands; on the other hand, if consumers have a
negative or unfavorable attitude towards’ them, they will be less prone to purchase them (Wee
et al., 1995). In the light of above discussion we propose our hypothesis:
H6 :
Consumers’ brand avoidance attitude is positively linked to their brand
avoidance intention.
90
3.5
Chapter 3
Moderating variables
In the previous section, we have framed literature about pre-purchase factors (determinants)
of brand avoidance and developed hypotheses accordingly. In this section we will discuss the
moderator role of the variables (Familiarity with country-of-origin and Consumer
Ethnocentrism) on the relationship between animosity and brand avoidance attitude.
3.5.1
Familiarity with COO
In the era of globalization, companies have the opportunities to distribute their goods to
consumers all over the world. At the consumer side, consumers have more options to take
their decision as they have a broad range of products and services in almost any category. So,
country of Origin (COO) has become an essential factor to consider while studying the
consumer assessment of foreign products (Jiménez & Martín, 2010). Researchers interested in
exploring COO effect have diverse findings due to the wide-range of backgrounds, conception
and contexts involved in the analysis of this variable. In broader sense, COO is an image
element that consists of product characteristics (e.g. innovation, technology, reliability, price,
overall quality, typical products) and the country associations about the characteristics of a
country (Jiménez & Martín, 2010). Whereas other studies describe COO as one aspect of a
brand which give an idea to consumers to identify the organization with the original domicile
of country rather that the country where the product is being manufactured (Ahmed et al.,
2004; Shankarmahesh, 2006; Thakor, 1996).
Diverse literature is available on COO, and majority perceived it as a cue that is able to
summarize the information regarding products, brands and firms from different countries
(Hamzaoui & Merunka, 2006; Jiménez & Martín, 2010; Klein et al., 1998). COO is a well
researched area and found conceptually relevant to the study of brands.
91
Chapter 3
Lee (2007) argued that countries become brands when we associate the country name with
commercial use. COO is perceived as an image variable or an extrinsic cue attached with
brand information, and consumer may use these cues during product/brand evaluation (Elliott
& Cameron, 1994; Zeithaml, 1988). Elliot and Cameron (1994) emphasized that when
consumers have minimum product related information or less motivated to process
information, COO knowledge plays a robust effect in product evaluation. Many studies has
been conduct to know the impact of COO on consumer’s assessments of products (Batra et
al., 2000; Janda & Rao, 1997).
In their study Hamzaoui & Merunka (2006) decomposed COO into country of design (COD)
and country of manufacture (COM) and found their influence on consumer evaluation of binational products. They have found that in product evaluation the concept of fit between
product category and country image (both COD and COM) is an important determinant.
Hence the product or brand which is strongly linked to a certain country , there is a possibility
that it might take attribute of that country, and on the contrary, consumers may avoid brands
linked with those countries that they believe to produce inferior goods (Lee, 2007).
Consumers build up direct and indirect product-related experience from advertising
exposures, interaction with salespersons, word of mouth communications, trial and
consumption (Alba & Hutchinson, 1987; Tam, 2008). Therefore, consumers develop country
images through familiarity with foreign products (Roth & Romeo, 1992). Familiarity reflects
the capability of consumers to recognize a particular brand and to relate it to a certain product
category with which he/she has direct or indirect experience (Matthiesen & Phau, 2005).
Luce, Barer, & Hillman (2001) have defined familiarity as “a general overall level of
acquaintance with the firm, most likely without reference to a specific, identifiable source of
92
Chapter 3
information”. Based on an extensive literature review, familiarity can be an important factor
in explaining the propensity to use COO information and its effects on other variables.
Previous studies have shown that familiarity influences consumers’ decision-making process,
above all consumers’ evaluations of COO products (Moorman, Diehl, Brinberg, & Kidwell,
2004). Jiménez and Martin (2010) in their study of automobiles in Spain found significant
moderating effect of familiarity with country’s product and brand on the relationship between
animosity and trust towards firms. Consumers may consider not buying an unfamiliar foreign
brand because they may make unfavorable inferences about the quality of this product (Han,
1990). We have found no prior evidence in the literature on the moderating role of familiarity
with COO in the Relationships between animosity and brand avoidance. Keeping in view the
above discussion we hypothesize:
H7 :
Familiarity with COO moderates the relationship between perceived animosity
and brand avoidance attitude; the higher the familiarity with COO the stronger
the relationship and vice versa.
93
3.5.2
Chapter 3
Consumer Ethnocentrism
Ethnocentrism is a very old concept (Sumner, 1906) and consumers are considered to be
‘ethnocentric’ when they show a tendency to buy locally manufactured products (Sharma,
Shimp, & Shin, 1995). The product is purchased, not because of its merit, but because
consumers think that it would be better for their in-group/local economy. In a similar way,
foreign products are avoided because consumer’ beliefs that it is ‘wrong’ to help an outgroup/foreign competition above their own nation (Shimp & Sharma, 1987).
Shimp & Sharma (1987) introduced consumer ethnocentric tendencies scale (CETSCALE).
They apply ethnocentrism to the study of marketing and consumer behavior and have coined
the term "consumer ethnocentric tendencies" (CET) to represent beliefs held by consumers
regarding the appropriateness and morality of purchasing foreign-made products. Studies of
consumer ethnocentrism generally have found that scores on the scale are inversely related to
willingness to purchase imports, perceptions of the quality of imported goods, cultural
openness, education, and income (Sharma et al., 1995; Shimp & Sharma, 1987), Consumer
ethnocentric tendencies play a significant role when products are perceived to be unnecessary
and when consumers believe that either their personal or national well-being is threatened by
imports (Sharma et al., 1995; Shimp & Sharma, 1987).
It is also observed that consumers preferred local made products against imported products, as
far as the perception of quality and price were equivalent (Elliott & Cameron, 1994). Many
studies resulted in negative attitude against foreign products (Sharma et al., 1995). Also a
positive relationship between CET and purchase intention of domestic/local made products
(Han, 1988).
Sumner (Sumner, 1906) was the first to provide a formal definition of ethnocentrism (Shimp
& Sharma, 1987). He defined ethnocentrism as: “. . . the view of things in which one’s own
94
Chapter 3
group is the center of everything, and all others are scaled and rated with reference to it. . .
Each group nourishes its own pride and vanity, boasts itself superior, exalts its own divinities
and looks with contempt on outsiders”.
Lewis (1985) contended that ethnocentrism is “a universal phenomenon that is rooted deeply
in most areas of inter-group relations” (Sharma et al., 1995). CET was regarded as a “unique
economic form of ethnocentrism that captures the beliefs held by consumers about the
appropriateness and indeed morality of purchasing foreign-made products” (Shimp & Sharma,
1987).
Consumer ethnocentrism has been studied widely as moderator or mediator in many contexts
such as US (Han, 1988; Klein & Ettenson, 1999; Shimp & Sharma, 1987), China (Klein et al.,
1998), Korea (Sharma et al., 1995), France, Australia and Mexico (Clarke, Shankarmahesh, &
Ford, 2000) and Turkey (Balabanis, Mueller, & Melewar, 2002). The concepts of
ethnocentrism and animosity are variables associated with the origin of a product. They have
been introduced into marketing literature from others disciplines such as psychology and
sociology (Balabanis et al., 2002). Marketing literature describes animosity and ethnocentrism
as concepts that imply psychological and behavioural reactions to specific countries of origin
(Nijssen & Douglas, 2004). We could describe ethnocentrism as a link between social and
moral norms and consumer behavior, while animosity is a variable that emphasizes a
consumer’s emotional attachment to the geographic origin of a product. Ethnocentric
consumers believe that buying foreign products hurts the domestic economy and national
employment (Balabanis et al., 2002; Shimp & Sharma, 1987). Animosity refers to remnants
of antipathy, or hostility towards a country (Klein et al., 1998). This emotion can act as a
protective instinct and increase in-group solidarity, domestic defensive behaviors and
ethnocentrism (Nijssen & Douglas, 2004).
95
Chapter 3
Moreover, consumer ethnocentrism and animosity may have different implications for
perceptions of product quality. Consumers who hold strong ethnocentric beliefs are more
likely to evaluate foreign products negatively than are those who do not hold such beliefs.
Those who believe that it is wrong to buy foreign goods also tend to perceive those goods as
lower in quality than domestic goods; ethnocentric consumers prefer domestic goods not only
because of economic or moral beliefs, but also because they believe that their own country
produces the best products. In contrast, it is possible that a consumer can harbor animosity
toward a specific country without denigrating the quality of goods produced by that country.
Thus consumer might be unwilling to buy these goods but might still believe that foreign
products- including the products produced by that country-are of high quality.
In the light of above discussion, we hypothesize:
H8 :
Consumer’s Ethnocentrism moderates the relationship between perceived
animosity and brand avoidance attitude; the higher the consumer’s
ethnocentrism, the stronger the relationship and vice versa.
96
3.6
Chapter 3
Socio-Demographic Characteristics
Socio-demographic characteristics of consumers have always been interested and investigated
in a number of studies to determine their potential effects on variables of interest (Charles,
2012; El-Adly, 2010; Heeter & Greenberg, 1985; Speck & Elliott, 1997). In their study,
Heeter and Greenberg (1985) found that gender and age are the most influential demographic
variables on TV ads avoidance, whereas, Education found to have no significant effect on the
avoidance behavior. Moreover, El-Adly (2010) found that young adults as compared to older
people have more tendencies to avoid TV ads.
Charles (2012) specifically conducted his study to explore the relationship of demographic
variables with Animosity and Ethnocentrism and found a significant relationship between age
and animosity, whereas gender has no significant relationship with animosity and
ethnocentrism.
Speak and Elliot (1997) found a significant relationship between household income and the
avoidance behavior. They concluded that people belongs to high income social class are more
likely to avoid TV ads.
Keeping in view the effects of demographic characteristics, current study has included four
socio-demographic characteristics of consumers (gender, age, education and monthly family
income) as potential predictors of brand avoidance attitude and behaviors.
97
3.7
Chapter 3
Research Model
On the basis of above discussion, we propose following conceptual research model (Figure
3.4) to be tested empirically by using online survey method:
Pre-Purchase Factors
Familiarity with
COO
Undesired Self
Congruence
H1
Negative Social
Influence
H2
Perceived
Animosity
Perceived Unethical
Corporate
Citizenship
H7
H3
Brand Avoidance
Attitude
H6
Brand Avoidance
Intention
H4
H5
H8
Perceived
Risk
Ethnocentrism
Figure 3.4: Conceptual Research Model
3.8
Conclusion
Based on extended literature review, this chapter has provided a model of pre-purchase
determinants of brand avoidance. Hypotheses are developed with the support of existing
literature for each determinant in the model. The effects of moderating variables are also
discussed and hypotheses are proposed accordingly. Following hypotheses are proposed in
this study:
98
Chapter 3
H1:
Undesired self congruence positively influences consumers’ brand avoidance attitude.
H2:
Negative Social Influence positively influences consumers’ brand avoidance attitude.
H3:
Consumers’ perceived animosity towards the Country-of-origin positively influences
consumers’ brand avoidance attitude.
H4:
Perceived Unethical Corporate Citizenship (PUCC) positively influences consumers’
brand avoidance attitude.
H5:
Consumers’ Perceived Risk positively influence consumers’ brand avoidance attitude.
H6:
Consumers’ brand avoidance attitude is positively linked to their brand avoidance
intention.
H7:
Familiarity with COO moderates the relationship between perceived animosity and
brand avoidance attitude; the higher the familiarity with COO the stronger the
relationship and vice versa.
H8:
Consumer’s Ethnocentrism moderates the relationship between perceived animosity
and brand avoidance attitude; the higher the consumer’s ethnocentrism, the stronger
the relationship and vice versa.
There may be other factors that have affects on pre-purchase brand avoidance behavior, but
our purpose is to identify the most influential factors predicting brand avoidance attitude and
behaviors.
In this part-I of the thesis, we made the basis of our research and identified the influential
factors that may have influence on pre-purchase brand avoidance. In the next part-II of the
thesis, the empirical issues are to be discussed in detail
99
Part-II
Part-II: Research Methodology, Empirical
Analysis & Results
Part II
Research Methodology,
Empirical Analysis & Results
 Research Approach
 Field of study
 Sample selection & instrument design
 Data collection

Chapter 5 & 6: Data Analysis
 Data preparation
 Confirmation of scales
 Sampling weights & Measurement Invariance
for pooled data
 Hypotheses testing & Moderation testing
Chapter 7: Discussion & Conclusion
 Summary of the results
 Contribution & Implication
 Limitation & future research

100
Part-II
Introduction to Part-II
In previous part of the study (part-I), research domain, preliminary data collection through
qualitative interviews and subsequently, on the basis of literature review, hypotheses are
formulated and a research model is proposed.
This part of the study deals with empirical issues of testing of hypothetical research model
proposed in part-I and provides the conclusion of the study. Chapter 4 addresses the issues
such as context of study, the target population, the sampling techniques, the formulation of
the questionnaire, its pre-testing, and the data collection method applied in this study.
In chapter 5, we present the data analysis of two countries (Pakistan and New Zealand)
separately. Data analysis includes the preliminary data cleaning, scale confirmation and
hypotheses testing. After analyzing separately, we pooled the data of two samples by
assigning sampling weights to make the samples representative of the target populations and
rerun the analysis for hypotheses testing. This is explained in chapter 6. The discussion and
interpretation of the results of all three data sets is provided at the end of analysis.
After detailed analyses and extensive discussion on results, the contribution and managerial
implications are provided in chapter 7. This chapter also talks about the limitation identified
and highlights the future research avenues.
101
Part-II
Where the study is conducted? How the constructs are
measured? How the data is collected?
What factors are significant in each
What factors are significant in pooled
country data?
data?
Chapter 5: Data Analysis
Chapter 6: Data Analysis (Pooled)
What are the theoretical and managerial implications of
the study?
Chapter 7: General Discussions and Conclusion
Figure 3.5: Flow Chart of Part-II
102
Research Methodology
Chapter 4
Chapter 4:
4
“Art and science have their meeting point in method”
---Edward G. Bulwer
4.1
Where and how the research is conducted? This chapter provides the information about the
context of the study and methodological aspects of the research. The objective is to elaborate
the research contexts of the studies. Moreover, the methodological topics such as research
design, populations, samples, variables of the study and research instrument used to measure
the constructs are discussed in detail. The formulation of questionnaire, it’s pre-testing and
data collection method is also explained.
4.2
Research Context
The primary context of our study is Pakistan based on the convenience of the researcher in
data collection. To test the proposed model on multiple markets and for the validity of the
results, the study is conducted in New Zealand. Why to choose New Zealand? In selecting
New Zealand, the first objective is to get data from a country where English is native
language because to translate questionnaire into another language may raise many other
concerns. The second reason is to have diversity in results by including an emerging country
(Pakistan) and a developed country (New Zealand).
Many researchers suggested consumers’ perception and evaluative judgments are based on
their cultural, social, experiential and technological factors (Bloch, 1995; Seva & Helander,
103
Chapter 4
2009). According to Hofsted (1980, 2011) , Pakistani society is a collectivistic society having
low score on individualism scale (around 15) as compared to New Zealand having high score
on this scale (around 80). But the purpose of this study was not to compare the findings of
two contexts, that’s why no cultural variables e.g. individualism/collectivism were included in
the study.
Following Table 4.1 shows some demographics and economic indicators of Pakistan. It has
the population of around 190 million (6th in world).
Table 4.1: Demographic, Economic and Lifestyle Indicators
Indices
Real GDP Growth (% growth)
Inflation (% growth)
Population Aged 65+: January 1st ('000)
Population Density (persons per sq km)
GDP Measured at Purchasing Power Parity
(international $ million)
Consumer Expenditure (US$ million)
2007
2008
2009
2010
2011
6.8
3.7
1.7
3.8
2.6
7.6
10.2
13.9
13.9
13.9
6,790.4
7,037.2
7,288.3
7,540.2
7,791.7
224.6
229.5
234.5
239.7
244.9
408,583.9
433,044.4
445,679.4
461,895.0
483,886.9
106,442.7
109,598.5
124,223.6
140,792.2
147,707.9
Annual Gross Income (US$ million)
126,724.8
129,832.6
146,402.4
166,387.1
174,391.8
Annual Disposable Income (US$ million)
114,671.4
117,941.1
133,292.6
150,760.7
158,123.6
Internet Users ('000)
17,500.0
18,500.0
20,431.3
22,885.7
25,552.6
Consumer Expenditure on Food (US$ million)
42,309.9
44,229.4
50,388.2
57,074.9
59,481.4
506
547
735
841
1,105
63,160
88,020
94,342
99,185
108,894
Workers’ Remittances (US$ million)
Mobile phone Users ('000)
Source: http://www.euromonitor.com/pakistan/country-factfile
http://www.sbp.org.pk/ecodata/homeremmit/remittance.pdf
http://www.pta.gov.pk/index.php?Itemid=1
According to economic indicators, Pakistan has the potential for foreign direct investment
(FDI) in almost every sector of the economy. With the globalization, Pakistani market has
become an attraction for the foreign brands.
Pakistan was one of the fastest growing economies of the world and had touched a GDP
growth rate of 8.4% in 2005 (Economic Survey of Pakistan 2005-2006). However, in recent
years the economy of Pakistan has suffered a lot due to many factors like war against terror,
104
Chapter 4
political instability, dictatorship crisis, flood disaster and law and order situation in areas near
Afghanistan border.
4.3
Research Design
The current study is conducted in two phases.

First phase: qualitative study (in-depth interviews) has been done to understand the
concept in Pakistani context. (cf. chapter 2)

Second Phase: quantitative study has been conducted in Pakistan and New Zealand by
using cross-sectional design with an online web-based survey. (cf. chapter 5 &6)
4.4
Population of the Study
The population of this study is described as follows in accordance with the guidelines
provided by Fraenkel and Wallen (2006).
4.4.1
Target Population
According to Fraenkel and Wallen (2006) target population of a study is the actual population
(consists of total number of the individuals/groups/organizations) for whom a researcher
would really interested to generalize the findings and results of his/her study. As per this
definition, all the consumers in Pakistan and New Zealand are the target population of this
study.
4.4.2
Accessible Population
According to Fraenkel and Wallen (2006), the population (consisting of all individuals or
groups) from which the sample of a research study is directly drawn out is termed as the
accessible population of that study. So according to this definition, in our study, the
accessible population is the individuals having internet access all over Pakistan because our
105
Chapter 4
method of data collection is an online web-based survey. Similarly, from New Zealand, our
accessible population is all the students of The University of Auckland.
We conducted an online survey study to test the hypothesized model (Figure 3.4). Data was
collected from general consumers of brands. There were a number of reasons for choosing
online survey method of data collection as argued by Wright (2005). Firstly, we are interested
to collect data from all parts of the country so it is much convenient and cost effective way to
collect data from the respondents at geographical distances. Secondly, the main reason behind
that is the amount of cost associated in duplication and sending the questionnaire for each
respondent. In electronic based surveys, there is no need to duplicate or create a separate link
for each respondent. The third major reason is the time saving, since the time taken in sending
and receiving back the postal questionnaire is quite lengthy process. Also the follow up is
convenient and there is an easy possibility of sending a quick reminder to the nonrespondents.
4.5
Sample
A ‘sample’ in a research study refers to any group (usually of individuals) from which the
information is obtained whereas ‘sampling’ refers to the process of selecting these individuals
(Fraenkel & Wallen, 2006). Convenience sampling technique, referred as the collection of
information from members of the population who are conveniently available (Sekaran, 2002)
has been used to get the responses from the consumers, through online survey method, who
have experienced brand avoidance phenomenon. Data collection through email or web-based
survey has certain disadvantages along with the advantages mentioned in last section.
Particularly, sampling frame (complete list of individuals to be contacted) in online sampling
survey is a difficult task. Although when you have a list of emails and you can send emails to
all of these members, again there is problem of having multiple email addresses for the same
106
Chapter 4
persons. There are other disadvantages argued by Wright (2005), as the online survey is a
voluntary act of filling up the questionnaire, so it has a low response rate. Market researchers
use different types of financial incentive techniques to increase response rates such as lottery
or coupon; but again it will undermine the credibility of the survey. Dealing with all these
problems we have a sample of 286 consumers collected from Pakistan and 199 consumers
from New Zealand. Detail of data collection is provided in later section (cf. data collection
method).
4.6
Variables of the Study
Table 4.2 enlists all the variables of the study including independent and dependent variables
along with moderating variables. The demographic information was also collected from the
respondents.
Table 4.2: Variables of the Study
Independent/Predictor Variables
Undesired Self
US
NSI
Animosity towards Country-of-Origin
Perceived Risk
A
PUCC
PR
Dependent/Criterion Variables
BAA
BAI
Moderator Variables
Familiarity with Country-of-Origin
FCO
Ethnocentrism
ETH
Demographic variables
Gender
GENDER
Age
AGE
Education Level (Highest Level Completed)
EDU
Monthly Family Income
MFI
107
4.7
Chapter 4
Measurement of Constructs
In this section, we have provided the details of the research instruments used to measure the
variables involved in the study:
4.7.1
Brand Avoidance
Currently, there is no scale available to measure brand avoidance behavior directly, so in
order to measure it indirectly, we measured attitude and intention towards brand avoidance.
engage in the behavior and in turn, the behavioral intention is best predicted by both attitude
and normative influences (Fishbein & Ajzen, 1975). Moreover, Iyer and Muncy (2009)
argued that attitude alone is often a poor predictor of marketplace behavior.
4.7.1.1 Brand Avoidance Attitude
According to Fishbein & Ajzen (1975) attitude is an individual’s positive or negative feelings
about performing a particular behavior. We measured brand avoidance attitude (BAA) from 4
items. Three items were adapted from Cheng, Lam, & Yeung (2006) and one item added by
the researcher (in consultation with 2 professors and 2 doctoral colleagues) to measure
negative attitude towards the avoided brand. The scale has been widely used to measure the
attitude (Lai, Chau, & Cui, 2010; Lee, 2009). Lee (2009) has validated this scale in his study.
All the factor loadings were greater than 0.8 and Cronbach’s alpha and composite reliability
were more than 0.90. The participants were asked to mention their agreement/disagreement
with the following statements on a Likert scale of 7 points. Where (1) shows strongly disagree
and (7) means strongly agree.

Using that brand is not a good idea

In my opinion, it is not desirable to use that brand
108
Chapter 4

Using that brand is not pleasant

I have a negative attitude towards that brand
4.7.1.2 Brand Avoidance Intention
The concept intention is referred as a cognitive component that mediates the relationship
between attitude and behavior (Ajzen & Fishbein, 1980; Debbabi et al., 2010). We used three
items to measure brand avoidance intention (BAI) adapted from Baker & Churchill’s (1977)
model, which has already been implemented and validated by other researchers (Debbabi et
al., 2010; Griffith & Chen, 2004). In their study in France, Debbabi et al (2010), used this
scale to measure attitude for two different products (watches & coats) and found a reliability
of more than 0.85 for both products. The participants’ agreements/disagreements were asked
on Likert scale of 7 points. Ranging from (1) strongly disagree to (7) strongly agree.

I would not buy that brand

I would not buy that brand even if I see it in a store

I would rather buy any other brand than this brand
4.7.2
In past studies, an attribute-based scale has been used to measure undesired self (Bosnjak &
Rudolph, 2008; Ogilvie, 1987; Phillips et al., 2007). Bosnjak and Rudolph (2008), they used
one statement in their study to measure undesired self “they never want to be or become” and
asked respondents to rate themselves on 20 attributes on seven point scale. Similarly, Phillips
et al (2007), they used one measure developed by Carver et al (1999), and people rated
themselves how similar they were to the trait by using 7-point scale. The same method has
been used for measuring actual/ideal self congruence (Sirgy, 1985; Sirgy et al., 1997). After a
long discussion on the problems in this type of measurement, Sirgy et al (1997) proposed, on
the basis of six studies, a new method of measuring self-image congruence directly rather than
109
Chapter 4
indirectly through the use of product image and the self image. Their findings have provided
support for high predictiveness for new method over traditional method. In a similar way, we
used 5 items to measure undesired self directly by following the method suggested by
Churchill (1979) for developing marketing constructs. Items were generated with the help of
PhD colleagues and approved by two research professors. The response was measured on a
Likert scale of 7 points from (1) strongly disagree to (7) strongly agree.

The brand does not represent what I wish to become

I associate that brand with something unappealing, which I do not want to become

I do not want to be associated with the type of people using that brand

The brand is trying to be something that it is not

That brand is not capable of portraying my unique individuality
4.7.3
To measure negative social influence, we adapted the items from social influence scale
frequently measured in information technology adoption studies (Gupta, Dasgupta, & Gupta,
2008; Kulviwat et al., 2009). We adapted 3 items from Kulviwat et al. (2009) and Gupta et al.
(2008), in which one item is commonly used in both the studies. The other items used in those
studies were particularly related to the organizational management support. The response was
measured on a likert scale of 7 points from (1) strongly disagree to (7) strongly agree.

People important to me think I should not use that brand

Others do not expect people like me to use that brand

People who influence my behavior think that I should not use that brand
110
4.7.4
Chapter 4
Animosity
Consumer animosity first entered into literature in 1998 by Klein et al. (1998). The author
studied the level of Chinese consumer animosity toward the Japanese. We adapted the scale
provided by Klein et al. (1998). It is a very well studied construct and previous authors
consistently used and validated this scale (Bahaee & Pisani, 2009; Charles, 2012; Nijssen &
Douglas, 2004; Riefler & Diamantopoulos, 2007; Russell & Russell, 2010, 2006; Shimp et
al., 2004). The response has been measured on a 7-point Likert scale from (1) strongly
disagree to (7) strongly agree.

I feel angry towards that country

That country is not reliable trading partner

That country wants to gain economic power over my country

That country is taking advantage of my country

That country has too much economic influence in my country
4.7.5
Corporate Social Responsibility (CSR) is a multidimensional concept (Carroll, 1999) and has
been measured through different scales (Maignan & Ferrell, 2001; Maignan et al., 1999;
Narver & Slater, 1990; Turker, 2009). Mainly the scales were made to measure the dimension
from internal employees’ perspectives. Turker (2009) in his study in Turkey, proposed and
validated a new scale consisting of four factors to measure corporate social responsibility. We
adapted 5 items from his scale according to the need of our study. The respondents were
asked to mention their agreement or disagreement on 7-point Likert scale.

The company does not participate in activities which aim to protect and improve the
quality of the natural environment
111

Chapter 4
The company does not contribute to campaigns and projects that promote the wellbeing of the society

The company does not emphasize the importance of its social responsibilities to the
society

The company does not respect consumer rights

The company does not provide full and accurate information about its products to its
customers
4.7.6
Perceived Risk
Perceived risk has been studied in many dimensions in past studies e.g. financial, social,
performance, psychological, security and physical risk. Since our study is focused on general
brands including all product categories so to have a general orientation of perceived risk we
prefer to use a composite scale to measure perceived risk. We used 4 items in our study to
measure perceived risk. Out of which three items were adapted from the scale of Peter and
Tarpey (1975) according to the need of our study, scale has already been validated by Evans
(1982) and one item was self generated by the researcher and approved by two research
professors. The overall perceived risk was measured on a Likert scale of 7-points. The
respondents were asked to mention their agreement or disagreement for the following items:
To buy this brand…

Is a financial risk for me because of things such as its poor warranty, high
maintenance costs, or high monthly payments

Is a performance risk for me because it would run extremely poorly

Is a physical risk for me because it would hurt me

Is too risky in general for my liking
112
Chapter 4
Familiarity with Country-of-origin
4.7.7
Following single item has been taken to measure the familiarity with country-of-origin
adapted from Laroche et al. (2005). The scale has already been used by other authors
(Nagashima, 1977; Papadopoulos, Marshall, & Heslop, 1988). Familiarity with the country of
origin of that brand was measured on a 7-point Likert scale with (1) not familiar and (7) very
familiar.

4.7.8
I know the country of that brand
Shimp & Sharma (1987) developed a 17 items Consumer Ethnocentric Tendency Scale
(CETSCALE) to measure consumer ethnocentrism. Out of 17 items, we adapted following
four items to measure the concept of ethnocentrism as these values are more relevant to the
context and also in their original scale these items have the higher factor loadings. The scale
has been validated in various countries such as France, Japan, and West Germany
(Netemeyer, Durvasula, & Lichtenstein, 1991), New Zealand (Watson & Wright, 2000),
Russia (Durvasula, Andrews, & Netemeyer, 1997), South Korea (Sharma et al., 1995) and US
(Durvasula et al., 1997; Shimp & Sharma, 1987). The respondents were asked to rate their
agreement or disagreement on the Likert scale of 7-points.

It is not right to buy foreign products because it puts our people out of job

We should buy products manufactured in our country instead of letting other countries
get rich off us

We should not buy foreign products because this hurts our business and causes
unemployment
113

Chapter 4
We should buy from foreign countries only those products that we cannot obtain
within our own country
It is emphasized here that all the scales used were 7-point Likert scale measuring either
agreement-disagreement (where, 1 = strongly disagree and 7 = strongly agree) or familiar-not
familiar (where, 1 = not familiar and 7 = very familiar) of the participants with the given
statements in the questionnaire. The reason to use seven point scales throughout the
questionnaire is to have uniformity in the measurements and in order to achieve measurement
and analysis benefits due to increased variability in the data.
For simplicity, Table 4.3 presents all the items with their references.
114
Chapter 4
Table 4.3: Questionnaire Items with References
S.No.
Reference
S.No.
Cheng et al. 2006
1
Familiarity with Country-of-Origin (COO)
Laroche et al. 2005
1
I think that using that brand is not a good idea
2
3
Cheng et al. 2006
1
Klien et al. 1998
4
I think that using that brand is not pleasant
Cheng et al. 2006
2
Klien et al. 1998
3
Klien et al. 1998
Self made
Reference
Animosity
1
Baker and
Churchill 1977
4
2
I would not buy that brand if I saw that brand in a store
Baker and
Churchill 1977
5
3
I would not prefer to buy that brand over any other brand
Baker and
Churchill 1977
Self made
1
2
I associate that brand with something unappealing, which I do not
want to become
Self made
3
I do not want to be associated with the type of people using that
brand
Self made
4
5
Self made
Klien et al. 1998
Klien et al. 1998
1
The company does not participate in activities which aim to protect and improve
the quality of the natural environment
2
The company does not contribute to campaigns and projects that promote the
well-being of the society
Turker 2009
3
The company does not emphasize the importance of its social responsibilities to
the society
Turker 2009
4
The company does not respects consumer right
5
The company does not provide full and accurate information about its products
and its customers
Self made
115
Turker 2009
Turker 2009
Turker 2009
Chapter 4
Perceived Risk
1
Kulviwat et al 2009
1
Buying of that brand is a financial risk for me because of such things as its poor
warranty, high maintenance costs, or highly monthly payments.
2
It is expected that people like me do not use that brand
Kulviwat et al 2009
2
Buying of that brand is a performance risk for me because it would run
extremely poorly.
3
People who influence my behavior think that I should not use that
brand
Gupta et al. 2008
3
Buying of that brand is a physical risk for me because it would hurt me
4
Buying of that brand is too risky in general for my liking
1
It is not right to purchase foreign products because it puts our
people out of job
Shimp and Sharma
1987
2
We should purchase products manufactured in our country instead
of letting other countries get rich off us
Shimp and Sharma
1987
3
We should not buy foreign products because this hurts our
business and cause unemployment
Shimp and Sharma
1987
4
We should buy from foreign countries only those products that we
cannot obtain within our own country
Shimp and Sharma
1987
116
Peter & Tarpey
1975
Peter & Tarpey
1975
Peter & Tarpey
1975
Self made
4.8
Chapter 4
Pilot Study
Evrard et al. (2003) emphasized that, it is always recommended to check the quality of the
questionnaire in a pretest with a smaller sample size representing the diversity of the
population that will be studied in the main sample. The purpose of a small number of
respondents is to find whether the questions are apt and easy to understand. It gives the
researcher a chance to make up any deficiency in time, if exists. Churchill (1999)
recommended that data collection should not be done without the pre-testing of the
instrument. So after finalizing the questionnaire, the first question was how many respondents
should be used for pilot study? A small sample size varying from 15 to 30 respondents should
be used (Kinnear & Taylor, 1996; Malhotra, 1999). So we distributed to 15 people (mostly
research colleagues/friends of the researcher including two professors) for conducting pilot
study.
Some of them were interrogated regarding the difficulties in the language or format of the
questionnaire. This main purpose of this step is to detect any possible difficulties, errors and
omissions in the language or legibility, or format of the questionnaires. In result of this
exercise some typographical mistakes were corrected, some difficult words in the scales were
simplified by using their easy or common alternatives for better understanding. After
removing all discrepancies the final questionnaire was prepared.
At the time of pilot study, an alternative questionnaire in native language ‘Urdu’ was prepared
and distributed among friends/colleagues. All the participants in this exercise unanimously
emphasized to use questionnaire in ‘English’ language rather than in ‘Urdu’ language. The
main reason behind this choice, they feel comfortable in English as compared to its literal
translation in Urdu language. Since the official language of Pakistan is English and the mode
117
Chapter 4
of education after 10th grade in Pakistan is mostly English1. So it was decided to use the
questionnaire in English language.
4.9
Data Collection
There are a number of methods to administer the questionnaire: the researcher may opt for a
self-administered option in which the respondent answers the questions himself and all the
instructions and guidelines are provided in the form or assisted administration in which the
respondents answers the questions with a text expressed by the investigator. After making
this choice the next step is to consider the possible methods for launching of questionnaire for
instance, as a postal mail questionnaire, electronic questionnaire (via internet/email),
telephonic surveys or face to face surveys.
While opting for self-administered questionnaire, one has to opt for by post or electronic
survey. In the first choice such as by post, the questionnaire is a paper based document sent
directly to the respondents by post and a request is being made to fill it. Whereas, in the later
choice, the questionnaire is a file or a web link sent to respondents by email or social medial
web pages.
As argued by Wright (2005), that the online data collection has advantages such as an easy
access to respondents that are difficult to contact and also to minimize the data collection cost
and time. The final questionnaire also includes socio-demographic variables: Gender, Age,
Education Level (Highest Level Completed), and Monthly Family Income. The complete
online questionnaire along with the cover letter explaining the purpose of the research, and
assuring respondents’ complete anonymity, was built by using online free services provided
1
Punjabi 48%, Sindhi 12%, Saraiki 10%, Pashtu 8%, Urdu (official) 8%, Balochi 3%, Hindko 2%,
Brahui 1%, Burushaski, and other 8%, English (official; lingua franca of Pakistani elite and most
government ministries [source: https://www.cia.gov/library/publications/the-world-factbook/geos/pk.html]
118
Chapter 4
by Kwiksurveys (http://www.kwiksurveys.com/). The link of the survey was sent to the
potential respondents by 2 ways:

Through emails

Through social networks by using social media web pages
4.9.1
Through Emails
Typically, in this method a survey invitation e-mail is being sent to potential respondents
including the web site address, or URL (uniform resource locator), that the respondent can
click on or cut and paste into a browser address box to access online survey (Grover &
Vriens, 2006). The researcher sent email invitations to personal email addresses and also used
different student’s group email addresses to contact them. The privacy was kept by not
allowing anyone to see the personal email address of any respondent. To deal with potential
problem during launching the survey, the researcher has provided personal email and phone
number to be contacted. Only two reminder emails were sent to increase the response rate of
the survey after every 10 days because multiple reminders may annoy the respondent or he
may consider it as ‘spam’ (Grover & Vriens, 2006). A special thanks note was being
mentioned in the reminders for those who have already filled the survey.
To collect data from New Zealand, the link of online survey questionnaire was sent to
institutional email addresses of the students through university professors and request was
made to voluntarily fill in the questionnaire.
4.9.2
Through Social Media Networks
Snowball sampling technique has been used while collecting data from social media
networks. Snowball sampling also known as chain referral sampling, is a non-probability
sampling technique. This is a method in which, the respondents or informants who have
119
Chapter 4
already been contacted are requested to use their social networks to further refer the
researcher to other people who could potentially take part in the study. To collect data by
using social media networks, the researcher put the online survey link to his social web pages
including:

Facebook (http://www.facebook.com)

Linkedin (http://www.linkedin.com)

Twitter (http://twitter.com)
The researcher requested the respondents to respond to the online survey. Furthermore, they
were requested to spread this link among their networks by putting the online survey link to
their wall pages. Snowball technique is a simple and cost efficient method to collect data but
the sample representativeness is questionable issue that the researcher has to compromise.
4.10 Conclusion
This chapter provided answers to methodological issues regarding, research design,
populations, samples and sampling technique. How the questionnaire is built and the source
of the instruments is also provided. Pre-testing of questionnaire and data collection methods
are explained in detail.
Next chapter will provide the empirical analysis of the data.
120
Data Analysis
Chapter 5
Chapter 5:
5
Data Analysis
“Although we often hear that data speak for themselves,
their voices can be soft and sly”
--- Frederick Mosteller
5.1
Chapter 5 and chapter 6, present data analyses and the findings of the study. The data analyses
of two different samples (sample 1: Pakistan & sample 2: New Zealand) are presented
separately for clarity in chapter 5. Data analysis is divided into three parts: first part will
discuss the preliminary data cleaning. Second part of the analysis will discuss the
confirmatory factor analysis (CFA), using structural equation modeling (SEM), to analyze the
measurement properties and convergent and discriminant validity of the scales. Subsequently
structural regression model is analyzed to test the proposed hypotheses. Third part of the
chapter will explain the moderation effects of consumer’s familiarity with country-of-origin
and consumer ethnocentrism. In chapter 6, we pooled the data of two countries after
establishing measurement invariance between samples and analyze convergent and
discriminant validity among variables.
We follow the following sequence (Figure 5.1) for data analysis in Pakistan and New Zealand
samples in chapter 5 and pooled data in Chapter 6:
121
Data Analysis
Chapter 5
• Missing Values
• Single and Multivariate outliers
• Socio-demographic characteristics
Preliminary
Analysis
Confirmatory
Factor
Analaysis (CFA)
Pooled Data
• Default Model
• Optimum Model
• Reliability and Validity
• Sampling weights
• Measurement invaraicne (MI)
(Chapter 6)
Hypothesis
Testing
• Direct Reletionships
• Indirect relationships (moderation effects)
Figure 5.1: Sequence of Data Analysis
5.2
Data Analysis: Pakistan Sample
First, we perform analysis on sample 1 (i.e. Pakistan sample); the sequence of analysis is,
initially we perform preliminary analysis for data cleaning, subsequently we perform
confirmatory factor analysis and hypotheses testing of the study.
5.2.1
Preliminary Data Analysis
In preliminary data analysis, we analyze missing values, aberrant values (single and
multivariate outliers) and socio-demographic characteristics of the respondents.
122
Data Analysis
Chapter 5
5.2.1.1 Missing Values Analysis
Missing values in the data set can seriously affect the results and certain analyses cannot be
performed in the presence of missing values. For that reasons, we put the check in our online
survey that survey cannot be submitted until it is complete i.e. free from missing values. So in
our data there are no missing values.
5.2.1.2 Outliers Detection
The basic assumption in multivariate data analysis techniques, e.g. exploratory factor analysis
(EFA) and confirmatory factory analysis (CFA), is that the data is free from outliers
especially the multivariate outliers. There are many perspectives to identify the outliers, for
example, univariate, bivariate, and multivariate perspectives. A good researcher should take
into account as many of these perspectives as possible to detect any outliers present in data
(Hair, Anderson, Tatham, & Black, 1998; Tabachnick & Fidell, 2007). In our study, we use
Z-scores and box plot graph to detect univariate outliers (see Appendix II: Table (a) & (b))
and for multivariate outliers we use Mahalanobis distances, Cook’s distance and Leverage
distance values (see Appendix II: Table (c)). As a result, we find 7 influential multivariate
outliers whose cases numbers are marked [31, 65, 183, 231, 245, 261, and 265]. One case
[157] is also identified as outlier because the respondent used the same response modality for
all questions. These cases are deleted from the data set because of their potential influence on
the parameters of interest as compared to other cases and we have 278 (=286-8) observation
for further analysis.
5.2.1.3 Socio-Demographic Characteristics of the Respondents
The socio-demographic characteristics (gender, age, education, and monthly family income)
of all the respondents are presented in Table 5.1.
123
Data Analysis
Chapter 5
Table 5.1: Socio-Demographic Characteristics (PK)
Characteristics
Gender
Age
Education
Monthly Family
Income
Frequency
Percentage
Male
203
73.0
Female
75
27.0
Less than 20 years
8
2.9
20-25 years
66
23.7
26-30 years
117
42.1
31-35 years
64
23.0
36-40 years
19
6.8
More than 40 years
4
1.4
Intermediate
19
6.8
Bachelor
57
20.5
Master or Above
202
72.7
Below or Rs. 60,000
68
24.5
Between Rs. 61,000 to Rs. 80,000
54
19.4
Between Rs. 81,000 to Rs. 100,000
61
21.9
Above or Rs. 100,000
95
34.2
In the sample of 278 valid respondents, 203 (73.0%) are male and 75(27.0%) are female.
Male2 dominant sample may be the reason of using convenience non-probability sampling
(snowball sampling) technique in online data collection. Another reason of low rate of
response from female might be the inaccessibility to internet or reluctant to respond to online
surveys. The sample consists of respondents having education master or above (72.7%) could
be the reason that majority of the people who use online social media networks are higher in
education. Also 75.5% of the sample belongs to middle to upper social class (MFI more than
Rs. 60,000) of the population
2
Pakistan Population: Male = 0.512% and Female= 0.488% Source : http://www.pbs.gov.pk/content/labourforce-survey-2010-11
124
Data Analysis
5.2.2
Chapter 5
Analytical Methods and Procedure
Structural equation modeling (SEM) technique is mainly used to test the hypothetical
relationships of proposed research model. The decision of using SEM is based on the fact that
the researcher can test the complete model of theoretical relationship simultaneously rather
than in separate small segments; Secondly, it is easy to apply and useful to observe the effect
of several independent variables on the dependent variable; and thirdly, it provides a reliable
analysis by allowing to correlate the measurement errors and decreases the chance of making
a type II error.
There are two main steps involved in structural equation modeling process: examining the
latent structure of measurement model and testing of the structural regression model. This
two-steps approach is employed in order to evaluate the proposed regression model
(Anderson & Gerbing, 1988). In order to examine the latent structure of the concepts we
perform confirmatory factor analysis (CFA), by using AMOS and following the guidelines
provided by Tabachnick and Fidell (2007), Hair et al (2009) and Kline (2010) as described in
the following section.
Model specification is the process in which hypotheses are represented in the form of a
structural equation model (Kline, 2010). Figure 5.2 shows the flowchart of the basic steps of
model specification in structural equation modeling.
125
Data Analysis
Chapter 5
Figure 5.2: Flow Chart for Model Specification
Source: Kline (2010)
The assessment of the adequacy of model is to assess the quality of “fit” of the empirical data
to the theoretical model. The adequacy of the structural equation models is assessed in terms
of absolute, incremental and parsimonious fit indices. Following are the fit indices used for
assessment of the model adequacy:

The chi-square (χ2/df) test statistics with corresponding degrees of freedom and level
of significance (Kline, 2010)

Normed chi-square (χ2/df)

Comparative Fit Index [CFI;(Bentler, 1990)]
126
Data Analysis
Chapter 5

Tucker-Lowis Index [TLI; (Bentler & Bonett, 1980)]

Standardized Root Mean-Square Residual [SRMR; (Kline, 2010)]

Root Mean-Square Error of Approximation [RMSEA;(Steiger, 1990)]
Table 5.2 represents the model fit criteria to be used in our study:
Table 5.2: Fit Indices Criteria
Fit Indices
Normed Chi Square
CFI
TLI
RMSEA
SRMR
5.2.3
Recommended /
Threshold Value
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
Reference
Kline (2010)
Bentler (1990)
Bentler & Bonett (1980)
Steiger (1990)
Kline (2010)
Confirmatory Factor Analysis
For simplicity, a separate CFA for independent and dependent variables is performed to test
the measurement properties of the underlying latent structure of measurement. We also
analyze the convergent and discriminant validities of all the measures.
5.2.4
CFA: Independent Variables
First, we run the CFA for five independent variables (US, NSI, A, PUCC and PR) of our
model. All the independent variables are allowed to correlate with each other; this helps us to
test the convergent and discriminant validities of the scale. The fit indices show a poor fit to
the data (χ2= 610.301, df= 199, χ2/df= 3.067, CFI= 0.887, TLI= 0.869, RMSEA= 0.086,
SRMR= 0.063).
Also we found that some of the items have very low standardized regression estimates (factor
loadings). As a general rule, loadings having value above 0.71 are excellent, 0.63 very good,
0.55 good, 0.45 fair, and 0.32 poor (Tabachnick & Fidell, 2007). Since we have multiple
127
Data Analysis
Chapter 5
items for each construct, we decided to remove all those items having standardized regression
estimate less than 0.55 (good). This criterion suggests to remove one item from perceived risk
(PR1).
With the removal of this item, the regression estimate of one item (PR2) fell below 0.55 the
minimum required criterion set for our study; so we remove this item. We also remove one
item form animosity construct (A2), two items from perceived unethical corporate citizenship
construct (UC4 & UC5), on the basis of having high standardized residual covariances (Kline,
2010).
After removing all the problematic items, our re-specified model shows an acceptable fit yet
we explore the opportunity for any improvement in the model. We look for modification
indices reported by the AMOS for possible improvements. AMOS provides modification
indices on the basis of regression weights, variances and covariances of the parameters
involved. Since our regression weights are above 0.60, modification on the basis of regression
weights and variances will not improve the model fit. So we look for modification indices on
the basis of covariances and found no value higher than 9 for modification indices. So there is
no higher value (magnitude) which can improve the model significantly. Hence, the respecified measurement model (Figure 5.3) shows a better fit (χ2= 165.546, df= 109, χ2/df=
1.519, CFI= 0.979, TLI= 0.974, RMSEA= 0.043, SRMR= 0.041). All the minimum required
criteria are met (Table 5.3). To verify the improvement in the re-specified model, a chisquare difference test is applied. The results of chi-square difference test show that the respecified model is better than the basic or initial model, at a risk level of 5% (with Δχ2=
444.755, df= 90, at p<0.00%). Hence our re-specified model (Final Model) has a better fit to
the data and we proceed to test convergent and discriminant validities of the constructs.
128
Data Analysis
Chapter 5
Table 5.3: Fit Indices of CFA for IVs (PK)
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
Initial Model
Obtained Values
610.301
199
3.067
0.887
0.869
0.086
0.063
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
e5
US1
e4
US2
e3
US3
e2
US4
e1
US5
e8
NSI1
e7
NSI2
e6
NSI3
e12
A1
e11
A3
e10
A4
e9
A5
e15
UC1
e14
UC2
e13
UC3
e17
PR3
e16
PR4
.65
.64
Final Model
Obtained Value
165.546
.79
.79
.68
109
1.519
0.979
0.974
0.043
0.042
US
.47
.89
.86
.80
-.10
NSI
.31
-.21
.91
.83
.74
.87
.41
.40
A
.43
-.47
.93
.85
.88
-.28
PUCC
.52
.71
.83
PR
Figure 5.3: Final CFA Model for IVs (PK)
5.2.4.1 Convergent and discriminant validity
To assess whether the items measure hypothesized constructs as proposed by the researcher is
referred to as establishing construct validity. To establish construct validity, convergent
129
Data Analysis
Chapter 5
validity and discriminant validity of the underlying constructs are important because usually
constructs are measured indirectly (being latent in nature) through multiple observed scores or
indicators (Kline, 2010). Convergent validity of a construct suggests that items/indicators are
at least moderately inter-correlated (Kline, 2010). To establish convergent validity, factor
loadings, average variance extracted (AVE), internal consistence (Cronbach’s alpha) and
composite reliability (CR) are good measures to use. Fornell and Larcker (1981) suggest that
in order to establish convergent validity the AVE>0.5 for each construct. Hair et al. (2009)
recommend that for reliability CR>0.70 and CR>AVE for convergent validity.
Table 5.4 shows that all standardized regression estimates (factor loadings) are significant at
p< 0.001 and the corresponding criteria for CR and AVE are also fulfilled. Internal
consistency (Cronbach’s alpha) values are well above the minimum criteria of 0.70
(Nunnally, 1978), Hence the required conditions are met to establish convergent validity.
Table 5.4 : Reliability of Scale for IVs (PK)
Constructs
Undesired
Self
Congruence
Negative
Social
Influence
Animosity
Perceived
Unethical
Corporate
Citizenship
Perceived
Risk
Items
US1
US2
US3
US4
US5
NSI1
NSI2
NSI3
A1
A3
A4
A5
UC1
UC2
UC3
PR3
PR4
Factor
Loading
0.790
0.790
0.677
0.649
0.642
0.858
0.801
0.888
0.738
0.867
0.909
0.830
0.881
0.932
0.853
0.831
0.709
Cronbach's
Alpha
Composite
Reliability
0.837
0.508
0.885
0.886
0.722
0.900
0.904
0.703
0.917
0.919
0.791
0.741
0.746
0.597
0.833
130
AVE
Data Analysis
Chapter 5
Discriminant validity refers to the fact that inter-correlation among different constructs is not
too high (Kline, 2010). Fornell and Larcker (1981) suggested that for establishing
discriminant validity among different constructs the AVE for each construct should be greater
than the shared variance. i.e. squared correlation.
Table 5.5: Discriminant Validity of IVs (PK)
PUCC
US
NSI
A
PUCC
0.791
US
0.093
0.508
NSI
0.159
0.225
0.722
A
0.220
0.010
0.043
0.703
PR
0.275
0.171
0.187
0.079
PR
0.597
*Diagonal elements are the AVE and off diagonal are share variances
The AVE for each construct is greater than the shared variance, which confirms discriminant
validity among constructs (Table 5.5).
We also check the discriminant validity among variables with SEM based alternative
approach, widely accepted by many researchers. In this method, discriminant validity is
established by running CFA on unconstrained (the correlation between constructs is free) and
constraining the correlation between constructs to 1.0. For discriminant validity the chisquare difference test should be significant (Bagozzi, Yi, & Phillips, 1991). In other words,
for discriminant constructs, each constrained model should result in significantly improved
chi-square from the less constrained models (Anderson & Gerbing, 1988). We run the models
and find chi-square difference test significant for each pair of constructs (cf. Appendix XII).
Hence, the convergent and discriminant validity is established among independent variables.
Next, we look for dependent variables.
131
Data Analysis
5.2.5
Chapter 5
CFA: Dependent Variables
We follow the same procedure for confirmatory factor analysis on the dependent variables
(BAA and BAI). Both the variables are allowed to correlate to assess the discriminant validity
of the scales. The fit indices (χ2= 28.926, df= 13, χ2/df= 2.225, CFI= 0.990, TLI= 0.984,
RMSEA= 0.067, SRMR= 0.020) show a good fit to the data, although RMSEA is high but
falls in the acceptable region. All the standardized regression estimates (factor loadings) are
higher than 0.70 and significant at p< 0.001. Hence the model provides good fit (Table 5.6).
We check the convergent and discriminant validities of the constructs.
Table 5.6: Fit Indices of DVs (PK)
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
132
Obtained Value
28.926
13
2.225
0.990
0.984
0.067
0.020
Data Analysis
Chapter 5
e4
BA1
e3
BA2
e2
BA3
e1
BA4
.83
.83
BAA
.80
.81
.83
e7
BA5
e6
BA6
e5
BA7
.94
.97
.76
BAI
Figure 5.4: CFA Model for DVs (PK)
To establish the convergent and discriminant validity of the scales, the standardized
regression estimate, internal consistency (Cronbach’s alpha), CR and AVE are computed and
presented in the Table 5.7. Required criteria are fulfilled to establish the convergent validity
of the scales.
Table 5.7: Reliability of Scale of DVs (PK)
Construct
Brand Avoidance
Attitude
Brand Avoidance
Intention
Items
Factor
Loading
Cronbach's
Alpha
Composite
Reliability
AVE
BA1
BA2
BA3
BA4
BA5
BA6
BA7
0.830
0.829
0.802
0.814
0.936
0.969
0.756
0.890
0.891
0.670
0.911
0.920
0.796
*Criteria: Cronbach’s alpha >0.7; CR>0.7 ; AVE>0.5 ; CR>AVE
Similarly, we follow the same principle for discriminant validity, that AVE should exceed the
shared variance between the variables.
133
Data Analysis
Chapter 5
The shared variance (0.690) between BAA and BAI is slightly greater than AVE for BAA
(0.670) but less than the AVE for BAI (0.796), which is the sign of absence of discriminant
validity. For that reason we merged the two factors into one factor and rerun the analysis for
one factor model (Figure 5.5).
e4
BA1
e3
BA2
e2
BA3
e1
BA4
.74
.75
.72
.74
.93
e7
BA5
e6
BA6
e5
BA7
BA
.94
.75
Figure 5.5: One factor Model of DVs (PK)
Apparently, the one factor model shows a poor fit (χ2= 178.572, df= 14, χ2/df= 12.755, CFI=
0.895, TLI= 0.843, RMSEA= 0.206, SRMR= 0.064) as compared to the two factor model. A
chi-square difference test is conducted to assess which model shows better fit. The results of
chi-square difference test shows that two factor model is better than the one factor model, at
the risk level of 5% (with Δχ2= 149.646, df= 1, at p<0.00%).
Also, with SEM alternative approach the chi-square difference test between unconstrained
and constrained model is significant, which shows the discriminant validity between BAA
and BAI (cf. Appendix XII). We, therefore, decide to continue our analysis with the two
factor model (Figure 5.4).
134
Data Analysis
5.2.6
Chapter 5
CFA: Full Measurement Model
After testing convergent and discriminant validity separately for independent and dependent
variables, we perform a combined CFA including IVs and DVs of the model to test their
discriminant validity before running the structural regression (SR) model. Sometimes, it
happens that independent variables share variances with dependent variables and when we put
them together in structural model they influence the results. For that reasons we perform CFA
together to resolve in advance their discriminant validities issue. The fit indices of full
measurement model (χ2= 343.889, df= 231, χ2/df= 1.489, CFI= 0.974, TLI= 0.969, RMSEA=
0.042, SRMR= 0.044) shows a better fit, meeting all the required conditions (Table 5.8). Also
all the factor loadings are above 0.60 and significant at p< 0.001.
Table 5.8: Fit Indices for Full Measurement Model (PK)
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
135
Obtained Value
343.889
231
1.489
0.974
0.969
0.042
0.044
Data Analysis
Chapter 5
e5
US1
e4
US2
e3
US3
e2
US4
e1
US5
e8
NSI1
e7
NSI2
e6
NSI3
e12
A1
e11
A3
e10
A4
e9
A5
e15
UC1
e14
UC2
e13
UC3
e17
PR3
e16
PR4
e21
BA1
e20
BA2
e19
BA3
e18
BA4
e24
BA5
e23
BA6
e22
BA7
.65
.66
.63
.80
.80
US
.47
.80
.89
.91
.83
.93
.85
.71
.81
.82
.86
-.10
NSI
.30
-.21
.74
.87
.42
.40
A
-.47
.88
.82
PR
.08
.24
.83
.82
.26
.10
.52
BAA
.49
.30
-.28
PUCC
.59
.44
.10
.03
.14
.83
.97
.76
.93
BAI
Figure 5.6: CFA: Full Measurement Model (PK)
For discriminant validity, AVE for each construct is greater than shared variance (Table 5.9),
which shows discriminant validity among the constructs.
136
Data Analysis
Chapter 5
Table 5.9: Discriminant Validity of Full Measurement Model (PK)
BAA
US
NSI
A
PUCC
PR
BAI
BAA
0.671
0.353
0.092
0.010
0.006
0.059
0.691
US
NSI
A
PUCC
PR
BAI
0.507
0.222
0.010
0.093
0.172
0.244
0.722
0.042
0.159
0.189
0.069
0.703
0.220
0.079
0.009
0.790
0.275
0.001
0.595
0.020
0.796
*Diagonal elements are the AVE and off diagonal is shared variance
The shared variance between BAA and BAI is slightly greater than AVE for BAA, which
shows lack of discriminant validity between these two constructs, which we have already
discussed (cf. CFA for DVs).
Since our data collection method is self-reported response with single source measurement
method, common method bias could be involved. The presence of common method bias could
make our findings biased and less generalizable.
5.2.7
Common Method Bias
Data collection from single source such as the same respondents were used for independent
and dependent variables may cause to have a common method bias (Podsakoff, MacKenzie,
Lee, & Podsakoff, 2003). Most commonly used method to test the common method bias is
Harman’s single factor test which is based on the hypothesis that data may suffer from
common method bias if a single factor accounts for more than 50% of the total variance
(Podsakoff et al., 2003). We performed exploratory factor analysis (EFA) with all the items in
the study and found that no single factor accounted for majority of variance. All of the 9
factors account for 70% of variance with the first factor which accounts for 26%. This implies
that common method bias is not a serious threat in our study (cf. Appendix III).
137
Data Analysis
Chapter 5
Moreover, Spector (2006) suggests to look for the highly significant correlation between the
variables of the study. We have found that the variables were not highly correlated and none
of the correlations with dependent variables exceeds 0.50. Hence, statistically, common
method bias is not detected in our study.
In next section, we will analyze New Zealand sample.
138
Data Analysis
5.3
Chapter 5
Data Analysis: New Zealand Sample
In this section, we perform preliminary analysis and then CFA on the data collected from
New Zealand. We follow the same sequence and procedure as in sample1.
5.3.1
Preliminary Analysis
In preliminary data analysis, we analyze missing values, aberrant values (single and
multivariate outliers) and socio-demographic characteristics of the variables.
5.3.1.1 Missing Values Analysis
Similar to the sample 1 (Pakistan sample), we control the responses by putting check that
survey cannot be submitted until it is complete i.e. free from missing values. So in our data
there are no missing values at all.
5.3.1.2 Outliers Detection
We used Z-scores and box plot to detect univariate outliers (cf. Appendix IV Table (a) & (b))
and for multivariate outliers we use Mahalanobis distances, Cook’s distance and Leverage
distance values (cf. Appendix IV Table (c)).
As a result we find only one influential
multivariate outlier whose case number is marked [85]. This case is deleted from the data set
and we are left with 198 (=199-1) observation for further analysis.
5.3.1.3 Socio-Demographic Characteristics of the Respondents
The demographic characteristics (gender, age, education, and monthly family income) of all
the respondents are presented in able 5.10.
139
Data Analysis
Chapter 5
Table 5.10: Demographic Characteristics (NZ)
Gender
Age
Education
Monthly Family
Income
Characteristics
Male
Female
Less than 20 years
20-25 years
26-30 years
31-35 years
36-40 years
More than 40 years
Secondary School Certificate
Under Graduate Tertiary Degree/Diploma
Post Graduate Tertiary Degree/Diploma
Less or $ 30.000
$31.000 to $ 50.000
$51.000 to $ 70.000
$ 71.000 to $ 90.000
$ 91.000 and Above
Frequency
77
121
52
108
10
14
2
12
63
104
31
101
39
18
7
33
Percentage
38.9
61.1
26.3
54.5
5.1
7.1
1.0
6.1
31.8
52.5
15.7
51.0
19.7
9.1
3.5
16.7
Although the gender ratio3 in New Zealand is almost m/f= 1/1; this sample is female
dominant as opposite to our Pakistani sample which is male dominant.
3
New Zealand Population: Male= 0.49% and Female= 0.51% Source: http://www.stats.govt.nz/
140
Data Analysis
5.3.2
Chapter 5
Analysis Technique and procedure
After dealing with data cleaning, we discuss the method used for inferential data analysis. We
perform structural equation modeling technique as in previous sample. After establishing
convergent and discriminant validity through confirmatory factor analysis we will test
hypotheses through structural regression model by using AMOS.
5.3.3
Confirmatory Factor Analysis
For simplicity, we perform separate CFA for independent and dependent variables same as in
Pakistan sample.
5.3.4
CFA: Independent Variables
We perform confirmatory factor analysis on all the independent variables (US, NSI, A,
PUCC, and PR) together. All the variables are allowed to correlate. The fit indices (χ2=
(581.886, df= 199, χ2/df= 2.924, CFI= 0.860, TLI= 0.837, RMSEA= 0.099, SRMR= 0.092)
show a poor fit to the data.
All the items have standardized regression estimates (factor loadings) above 0.55, the
acceptable criteria set for our study. We remove two items from undesired self (US4 &US5),
two items from perceived risk (PR1 & PR2), two items from perceived unethical corporate
citizenship (UC4 & UC5) and two items from animosity (A1 & A2) on the basis of high
standardized residual covariances (Kline, 2010).
After removing all these items, we look for the modification indices. There are no high
modification indices so the fit indices of re-specified model (χ2= 139.496, df= 67, χ2/df=
2.082, CFI= 0.962, TLI= 0.948, RMSEA= 0.074, SRMR= 0.052) are better as compared to
the initial model (Table5.11). We further perform chi-square difference test to verify the
improvement in the re-specified model. The results of chi-square difference test show that the
141
Data Analysis
Chapter 5
re-specified model is better than initial model, at the risk level of 5% (with Δχ2= 442.390, df=
132, at p<0.00%). Hence our re-specified model (final model) shows better fit to the data. We
proceed to study convergent and discriminant validity of the constructs.
Table 5.11: Fit Indices of CFA for IVs (NZ)
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
Initial Model
Obtained Value
581.886
199
2.924
0.860
0.837
0.099
0.092
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
142
Final Model
Obtained Value
139.496
67
2.082
0.962
0.948
0.074
0.052
Data Analysis
Chapter 5
e5
US1
e4
US2
e3
US3
.69
.87
.64
US
.35
e8
NSI1
e7
NSI2
e6
NSI3
.67
.97
.77
NSI
-.27
.52
-.30
e11
A3
e10
A4
e9
A5
e15
UC1
e14
UC2
e13
UC3
e17
PR3
e16
PR4
.94
.88
.98
A
.56
-.19
.96
.91
.06
.16
.88
-.21
PUCC
.12
.77
.76
PR
Figure 5.7: Final CFA Model for IVs (NZ)
To assess the convergent validity among the constructs (for detail see Sample 1: CFA for
IVs), we computed and presented standardized regression estimates, internal consistency
(Cronbach’s alpha), composite reliability (CR) and average variance extracted (AVE).
All the factor loadings are above 0.60 and significant at p<0.001 (Table 5.12). Also the
required conditions for Cronbach’s alpha, CR, and for AVE are met, hence convergent
validity is established.
143
Data Analysis
Chapter 5
Table 5.12: Reliability of scales for IVs (NZ)
Construct
Items
Undesired
Self
Congruence
US1
US2
US3
NSI1
NSI2
NSI3
A3
A4
A5
UC1
UC2
UC3
PR3
PR4
Negative
Social
Influence
Animosity
Perceived
Unethical
Corporate
Citizenship
Perceived
Risk
Factor
Loading
0.868
0.644
0.686
0.771
0.675
0.969
0.979
0.938
0.885
0.882
0.959
0.907
0.763
0.774
Cronbach's
Alpha
Composite
Reliability
0.780
0.546
0.830
0.852
0.663
0.952
0.954
0.874
0.940
0.940
0.840
0.743
0.743
0.591
0.757
AVE
*Criteria: Cronbach’s alpha >0.7; CR>0.7; AVE>0.5; CR>AVE
AVE and shared variances are presented in the following Table 5.13.
Table 5.13: Discriminant Validity of IVs (NZ)
PUCC
US
NSI
PUCC
0.840
US
0.275
0.546
NSI
0.025
0.125
0.663
A
0.034
0.071
0.090
A
PR
0.874
PR
0.015
0.003
0.312
0.046
0.591
AVE for each construct is greater than the shared variance, which confirms discriminant
validity among constructs. Also, the chi-square difference tests between unconstrained and
constrained models are significant for each pair of constructs (cf. Appendix XII).
5.3.5
CFA: Dependent Variables
We run a separate CFA for the dependent variables (BAA and BAI) of the model. The fit
indices (χ2= 35.547, df= 13, χ2/df= 2.504, CFI= 0.971, TLI= 0.954, RMSEA= 0.087, SRMR=
144
Data Analysis
Chapter 5
0.042) show a good fit to the data for all the indices except RMSEA which is slightly higher
than the recommended value. All standardized regression estimates (factor loadings) are
greater than 0.60 and significant at p<0.001. Next we check the convergent and discriminant
validity of the constructs.
Table 5.14 : Fit Indices for DVs (NZ)
Recommended
/Threshold Value
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Obtained Value
35.547
13
2.504
0.971
0.954
0.087
0.042
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
e4
BA1
e3
BA2
e2
BA3
e1
BA4
.65
.84
BAA
.84
.62
.81
e7
BA5
e6
BA6
e5
BA7
.78
.87
.75
BAI
Figure 5.8: CFA Model for DVs (NZ)
To establish the convergent and discriminant validity of the scales of dependent variables, the
standardized regression estimate, internal consistency (Cronbach’s alpha), CR and AVE are
computed (Table 5.15).
145
Data Analysis
Chapter 5
Table 5.15: Reliability of scales for DVs (NZ)
Construct
Brand Avoidance
Attitude
Brand Avoidance
Intention
Items
Factor
Loading
Cronbach's
Alpha
Composite
Reliability
AVE
BA1
BA2
BA3
BA4
BA5
BA6
BA7
0.651
0.838
0.843
0.619
0.777
0.868
0.746
0.825
0.830
0.555
0.828
0.840
0.638
*Criteria: Cronbach’s alpha >0.7; CR>0.7; AVE>0.5; CR>AVE
All factor loadings are above 0.60 and significant at p< 0.001, and required criteria for other
indices are met, so convergent validity is established. For discriminant validity, AVE for each
construct should be greater than the shared variance. i.e. squared correlation.
The shared variance (0.656) between BAA and BAI is greater than AVE for both (BAA=
0.555 & BAI= 0.638) of the dependent variables, which is the sign of absence of discriminant
validity between them. For that reason we merge the two factors into one factor. The results
are presented in Figure 5.9.
e4
BA1
e3
BA2
e2
BA3
.8 1
e1
BA4
.5 9
.6 2
.8 0
.7 4
e7
BA5
e6
BA6
e5
BA7
BA
.7 9
.6 8
Figure 5.9: One Factor Model for DVs (NZ)
Apparently, the one factor model shows poor fit (χ2= 82.160, df= 14, χ2/df= 5.869, CFI=
0.900, TLI= 0.849, RMSEA= 0.157, SRMR= 0.066) as compared to the two factor model.
146
Data Analysis
Chapter 5
The results of a chi-square difference test show that the two factor model is better than the one
factor model, at risk level of 5% (with Δχ2= 46.613, df= 1, at p<0.00%).
Also, the chi-square difference test between unconstrained and constrained model is
significant (cf. Appendix XII). We, therefore, continue our analysis with the two factor model
(Figure 5.8).
5.3.6
CFA: Full Measurement Model
After running separate CFAs, we perform a combined CFA including IVs and DVs. The fit
indices of the combined measurement model are presented in the Table 5.16.
Table 5.16: Fit Indices of Full Measurement Model (NZ)
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
Obtained Value
376.764
168
2.243
0.925
0.906
0.079
0.062
The fit indices show an adequate fit of the data. All factor loadings are higher than 0.60 and
significant at p< 0.001. AVE for each construct is greater than shared variance which ensures
discriminant validity (Table 5.17). Only BAA and BAI lack discriminant validity which we
have already discussed (see Sample 2: CFA for DVs).
147
Data Analysis
Chapter 5
Table 5.17 : Discriminant validity of Full Measurement Model (NZ)
BAA
US
NSI
A
PUCC
PR
BAA
0.561
US
0.249
0.542
NSI
0.117
0.134
0.662
A
0.014
0.078
0.091
0.873
PUCC
0.120
0.281
0.024
0.034
0.841
PR
0.301
0.003
0.307
0.038
0.017
0.594
BAI
0.654
0.198
0.040
0.013
0.123
0.123
148
BAI
0.638
Data Analysis
Chapter 5
e5
US1
e4
US2
e3
US3
e8
NSI1
e7
NSI2
e6
NSI3
e11
A3
e10
A4
e9
A5
e15
UC1
e14
UC2
e13
UC3
e17
PR3
e16
PR4
e21
BA1
e20
BA2
e19
BA3
e18
BA4
e24
BA5
e23
BA6
e22
BA7
.68
.68
.97
.94
.88
.84
.68
US
.37
.77
NSI
-.28
-.30
.98
A
.53
.55
-.19
.96
.91
.82
.65
.50
.88
-.20
PUCC
.34
-.12
.13
.81
.06
.16
.20
.73
PR
.35
.55
.69
.81
BAA
.44
-.11
.35
.33
.81
.87
.75
.77
BAI
Figure 5.10: CFA of Full Measurement Model (NZ)
5.3.7
Common Method Bias
We use the same method as for sample 1 to look for Common method bias in our data. The
result of the EFA shows that all of the 9 factors account for 69% of variance with first factor
149
Data Analysis
Chapter 5
only account for 27%, which implies that common method bias is not a serious threat in our
study (see Appendix V).
Table 5.18 and Table 5.19, present the mean, standard deviation (SD) and inter-correlation
among all variables including socio-demographic variables, for Pakistan sample and New
Zealand sample respectively.
150
Data Analysis
Chapter 5
Table 5.18: Means, Standard Deviations, and Correlations among Study Variables (PK)
Mean
SD
Gender
Age
EDU
MFI
US
NSI
A
PUCC
PR
BAA
BAI
Gender
1.27
0.44
Age
3.12
0.99
-.145*
EDU
3.66
0.60
-.194**
.552**
MFI
2.66
1.18
-.003
.123*
.170**
US
4.21
1.34
.010
.101
.022
-.067
NSI
3.91
1.55
.008
.112
.090
.087
.434**
A
5.00
1.48
.104
.145*
.114
.131*
-.098
-.197**
PUCC
3.66
1.30
-.061
.080
-.055
-.089
.281**
.369**
-.442**
PR
3.51
1.51
.005
.060
-.097
.023
.332**
.369**
-.244**
.436**
BAA
4.66
1.57
.024
.010
.075
.075
.499**
.260**
.085
.072
.206**
BAI
5.32
1.59
-.036
-.028
.097
.029
.448**
.233**
.079
.042
.127*
.769**
FCO
4.53
1.45
.026
.139
.057
.000
-.108
-.119*
.300**
-.094
-.070
-.095
-.126*
ET
3.64
1.53
.039
-.017
-.019
.052
-.140*
-.300**
.398**
-.404**
-.247**
-.044
-.006
FCO
.107
US=Undesired Self, NSI=Negative Social Influence, A=Animosity, PUCC= Perceived Unethical Corporate Citizenship, PR= Perceived Risk, BAA= Brand Avoidance
Attitude, BAI= Brand Avoidance Intention, FCO= Familiarity with Country-of-origin, ET= Ethnocentrism, EDU= Education, MFI= Monthly Family Income.
*p< 0.05
**p< 0.01
151
Data Analysis
Chapter 5
Table 5.19: Means, Standard Deviations, and Correlations among Study Variables (NZ)
Mean
SD
Gender
Age
EDU
MFI
US
NSI
A
PUCC
PR
BAA
BAI
Gender
1.61
0.49
Age
2.20
1.27
.135
EDU
1.84
0.68
-.001
.590**
MFI
2.15
1.49
.262**
-.035
-.131
US
5.53
1.17
.052
.056
.066
.061
NSI
3.73
1.39
-.025
.023
-.002
.094
.325**
A
5.07
1.73
.066
-.047
-.103
.088
-.262**
-.271**
PUCC
4.38
1.45
-.027
-.015
-.005
-.034
.429**
.099
-.180*
PR
3.49
1.59
-.065
.142*
.032
.018
.059
.483**
-.207**
.099
BAA
5.32
1.14
.039
.106
.053
.092
.412**
.353**
-.127
.320**
.484**
BAI
6.04
0.97
.109
.036
-.009
.084
.369**
.177*
-.129
.317**
.277**
.653**
FCO
4.38
1.62
-.016
-.048
-.106
.056
-.085
-.169*
.361**
-.055
-.186**
-.167*
-.116
ET
4.52
1.53
.033
-.064
.018
.025
-.129
-.271**
.234**
-.182*
-.348**
-.238**
-.205**
FCO
.096
*p< 0.05
**p< 0.01
152
Data Analysis
5.4
Chapter 5
Hypotheses Testing
After developing an acceptable measurement model and establishing the convergent and
discriminant validity, we validate the structural model (Figure 2.4) according to our proposed
hypotheses. We check the validation of the conceptual model on both sets of data (sample 1,
sample 2).
Fit indices of the structural model for both sets of data are presented in Table 5.20. All fit
indices are above the minimum criteria, showing an adequate fit of the data with the structural
models (see Figure 5.11 and Figure 5.12).
Table 5.20: Structural Model Fit Indices
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
153
Pakistan
New Zealand
346.600
236
1.469
0.975
0.970
0.041
0.044
382.540
173
2.211
0.924
0.908
0.078
0.065
Data Analysis
e5
e4
e3
e2
e1
e8
e7
e6
e12
e11
e10
e9
e15
e14
e13
e17
e16
US1
US2
US3
US4
US5
NSI1
NSI2
NSI3
A1
A3
A4
A5
UC1
UC2
UC3
PR3
PR4
Chapter 5
.65
.67
.63
.80
.80
US
.47
.80
.89
.91
.83
.93
.85
.71
.86
NSI
-.10
e19
e20
e21
e22
e23
e24
BA4
BA3
BA2
BA1
BA7
BA6
BA5
A
.42.15
.40
-.47.43
.88
PUCC
.82 .83
.82 .81
.08
.30
-.21
.74
.87
.83
e18
.58
.76 .94
.97
.83
BAA
-.09
BAI
e26
e25
.04
-.28
.52
PR
Figure 5.11: Structural Model Results on Pakistan Sample
e5
e4
e3
e8
e7
e6
e11
e10
e9
e15
e14
e13
e17
e16
US1
US2
US3
.68
NSI1
NSI2
NSI3
.68
.97
A3
A4
A5
UC1
UC2
UC3
PR3
PR4
.94
.88
.84
.68
US
.37
.77
NSI
.98
A
.80
.88
.16.06
PUCC
.74
-.20
e19
e20
e21
e22
e23
e24
BA4
BA3
BA2
BA1
BA7
BA6
BA5
-.19
-.30
.53
.55
-.19
.96
.91
.55
-.28
e18
.10
.04
.82 .68
.64 .83
BAA
.87
.75 .78
.81
e25
.61
.13
PR
Figure 5.12: Structural Model Results on New Zealand Sample
154
BAI
e26
Data Analysis
Chapter 5
Table 5.21 shows standardized regression estimates along with t-values for all the paths of the
structural model for both data sets (Pakistan sample & New Zealand sample).
Table 5.21: Hypotheses Testing
Hypotheses
Direct effect
Std. estimate
t-value
Result
Pakistan Sample
H1
US  BAA
0.581***
7.306
Supported
(n = 278)
H2
NSI  BAA
0.082
1.158
Not Supported
H3
A  BAA
0.145*
2.241
Supported
H4
PUCC  BAA
-0.095
-1.236
Not Supported
H5
PR  BAA
0.041
0.501
Not Supported
H6
BAA  BAI
0.832***
12.204
Supported
New Zealand Sample
H1
US  BAA
0.548***
4.426
Supported
(n = 198)
H2
NSI  BAA
-0.192
-1.811
Not Supported
H3
A  BAA
0.104
1.461
Not Supported
H4
PUCC  BAA
0.042
0.488
Not Supported
H5
PR  BAA
0.610***
4.750
Supported
H6
BAA  BAI
0.806***
7.638
Supported
US=Undesired Self, NSI=Negative Social Influence, A=Animosity, PUCC= Perceived Unethical Corporate
Citizenship, PR= Perceived Risk, BAA= Brand Avoidance Attitude BAI= Brand Avoidance Intention.
*p<0.05, **p<0.01, *** p<0.001
155
Data Analysis
5.5
Chapter 5
Moderation Effect
Moderation occurs when the strength or direction of the relationship between the independent
and dependent variable depends upon a second independent variable, known as a moderator.
Statistically, moderation can be seen as an interaction of the independent and moderator
variables (Baron and Kenny 1986, Hayes and Matthes 2009).
Following the recommendations of Aiken and West (1991), the predictor and moderator
variables should be mean centered. The advantage of having mean centered independent
variables include:

Reduction of multicollinearity among predictor variables, and

Meaningful interpretation of regression coefficients
According to Baron and Kenny (1986), and James and Brett (1984), the test of moderation
should include three terms: a term for the direct effect of the predictor, a term for the direct
effect of the moderator, and the interaction term. The moderation hypothesis is supported if
the interaction term is significant. There may also be significant effects for the predictor
and/or the moderator, but conceptually, these are not directly relevant in testing the
moderation hypothesis (Baron & Kenny, 1986).
Interactions between variables can be probed through different approaches, one of which is
known as subgroup analysis or separate regression approach. In this method, the data file is
split into various subsets based on the values of the moderator (high and low or high, medium
and low) and the whole analysis is repeated on these subsets of the data. The difference in the
results is then evaluated for significance. Hayes and Matthes (2009) propose to estimate the
effect of the focal predictor variables at different values of the moderator variable. As in our
study, we have two moderators, familiarity with country-of-origin and ethnocentrism. Based
156
Data Analysis
Chapter 5
on their nature of measurement (familiarity is a single item observed construct and
ethnocentrism is a latent construct measured by four items), we choose two different methods
to look for interaction effect.
5.5.1
Moderating effect of familiarity with country-of-origin
In hypothesis H7, we proposed that consumers familiarity with country-of-origin moderates
the relationship between perceived animosity and brand avoidance attitude; the higher the
familiarity with country-of-origin the stronger the relationship and vice versa. We used Hayes
and Matthes’s (2009) approach and used their SPSS macro to find the interaction effect of
familiarity. The interaction effect of an independent variable (animosity) and a moderating
variable (familiarity with COO) on a dependent variable (brand avoidance attitude) is
considered. The average scores for all three variables (IV, Moderator, and DV) are calculated.
The independent variable and moderating variable are mean centered for both the data sets as
recommended by Aiken and West (1991). The unstandardized regression estimates for the
effects of independent, moderator and interaction between them, are presented in the Table
5.22.
Table 5.22: Unstandardized estimates for interaction of A and FCO
Relation
Pakistan Sample
New Zealand Sample
Estimate
S.E.
P
A  BAA
0.128
0.066
0.052
FCO  BAA
-0.106
0.070
0.128
A * FCO  BAA
0.086
0.042
0.045
A  BAA
-0.053
0.049
0.280
FCO  BAA
-0.062
0.055
0.265
A * FCO  BAA
0.057
0.028
0.046
A=Animosity, FCO= Familiarity with COO, BAA= Brand avoidance attitude
157
Data Analysis
Chapter 5
The interaction term between animosity and familiarity with COO is significant at p< 0.05 in
both data sets. Hence, familiarity with country-of-origin moderates the relationship between
perceived animosity and brand avoidance attitude. This supports hypothesis H7.
Figure 5.13: Moderation effect of Familiarity with COO (PK)
158
Data Analysis
Chapter 5
Figure 5.14: Moderation effect of Familiarity with COO (NZ)
Graphically Figure 5.13 and 5.14 show the interaction effect in both data sets (Pakistan
sample and New Zealand sample respectively). For Pakistan data, in case of low familiarity
with COO, the attitude for brand avoidance remains constant. But in case of high familiarity
with COO, brand avoidance attitude increases with perceived animosity. It supports our
hypothesis. For New Zealand, brand avoidance attitude is constant when familiarity with
COO is high whereas with low animosity, the avoidance attitude varies downward. It is
contrary to our proposed hypothesis. Hence, our hypothesis partially supported in New
Zealand.
5.5.2
Moderating effect of consumer ethnocentrism
In hypothesis H8, we proposed that consumer ethnocentrism moderates the relationship
between perceived animosity and brand avoidance attitude; the higher the consumer
ethnocentrism the stronger the relationship.
159
Data Analysis
Chapter 5
To probe the moderation effect, the interaction effect of independent variable (animosity) and
moderator variable (ethnocentrism) on the dependent variables (brand avoidance attitude) is
calculated. As the moderator is a latent construct measured by multiple items, we follow Little
et al (2006)’s approach to find out the interaction effect among latent variables (cf. Appendix
VIII). The fit indices of the estimated model for Pakistan (χ2= 350.079, df= 298, χ2/df=
1.175, CFI= 0.992, TLI= 0.990, RMSEA= 0.025, SRMR= 0.043) show an adequate fit to the
data. Similarly, for New Zealand sample fit indices (χ2= 220.715, df= 196, χ2/df= 1.126,
CFI= 0.996, TLI= 0.995, RMSEA= 0.025, SRMR= 0.035) show a better fit to the data. The
unstandardized regression estimates are presented in the Table 5.23.
Table 5.23: Unstandardized estimates for interaction of A and ETH
Relation
Pakistan Sample
New Zealand Sample
Estimate
S.E.
P
A  BAA
0.145
0.074
0.050
ETH  BAA
-0.139
0.096
0.145
A * ETH  BAA
0.143
0.189
0.448
A  BAA
-0.031
0.048
0.518
ETH  BAA
-0.215
0.066
0.001
A * ETH  BAA
0.178
0.110
0.100
A=Animosity, ETH= Ethnocentrism, BAA= Brand avoidance attitude
The interaction term is significant only for New Zealand at p< 0.100. This implies that
ethnocentrism plays a significant role on the relationship between animosity and brand
avoidance attitude, supporting hypothesis H8.
160
Data Analysis
Chapter 5
Figure 5.15: Moderating effect on Ethnocentrism (NZ)
Graphically Figure 5.15 shows that being low ethnocentric consumer, the relationship
between animosity and brand avoidance attitude is constant, whereas being high ethnocentric
consumer, the relationship between animosity and brand avoidance attitude is upward i.e.
higher the ethnocentrism stronger the relationship. It supports H8, that consumer
ethnocentrism moderates the relationship between perceived animosity and brand avoidance
attitude.
5.5.3
Socio-Demographic Effects
We check socio-demographic variables’ (gender, age, education, and monthly family income)
effect on the dependent variable. We check the effect in two samples separately. In Pakistan
sample; we find age, education, and monthly family income have significant effect on the
brand avoidance attitude and whereas gender has no significant effect on brand avoidance
attitude. In New Zealand sample, we find no significant effect of demographic variables on
brand avoidance attitude. Although we find significant effect of gender on the brand
161
Data Analysis
Chapter 5
avoidance intention (standardized regression estimate= 0.105; p< 0.10), this shows that
women as compared to men have more avoidance intention.
In Pakistan sample, education with standardized regression estimate (0.121) significant at
p<0.10, shows that it has a positive relationship with brand avoidance attitude. As the level of
education increases, the brand avoidance attitude also increases. We can say that the more
educated person is, the more information he has about the brands that affect his decision
making. Our findings are contradictory to Heeter and Greenberg’s (1985) finding that
education has no significant effect on TV ads avoidance behavior.
In Pakistan sample, age with standardized regression estimate (-0.156) significant at p<0.05,
shows that it has negative relationship with the brand avoidance attitude. The negative
relationship emphasized that the young adults have more avoidance attitude towards brands as
compared to older people. We find support from Heeter and Greenberg’s (1985) study of TV
ads avoider. They found that gender and age are the most influential demographic variables
on TV ads avoidance. Another study proposed that young adults tend more to avoid TV ads as
compared to older people (El-Adly, 2010). Speck and Elliott (1997) concluded that people
have more tendencies of avoiding TV ads if they are young and have high income.
Concerning monthly family income, in Pakistan sample, it shows positive relationship with
brand avoidance attitude (standardized regression estimate= 0.094; p<0.10). This shows that
as the income increases, people have more avoidance attitude towards brands. In other words,
we can say that they have more choices to make selection among the brands and exercise their
avoidance behavior. Our findings are similar to Speck and Elliott’s (1997) study. They found
that people are more likely to avoid TV ads if they have high income.
162
Data Analysis
5.6
Chapter 5
Conclusion
This chapter provided the analysis on separate samples and the results are summarized in
Tables 5.24. Next chapter will provide pooled data analysis. The discussion and interpretation
of the results will also be provided in next chapter.
Table 5.24: Summary of hypothesis testing
Hypotheses
Direct effect
Std. estimate
t-value
Result
Pakistan Sample
H1
US  BAA
0.581***
7.306
Supported
(n = 278)
H2
NSI  BAA
0.082
1.158
Not Supported
H3
A  BAA
0.145*
2.241
Supported
H4
PUCC  BAA
-0.095
-1.236
Not Supported
H5
PR  BAA
0.041
0.501
Not Supported
H6
BAA  BAI
0.832***
12.204
Supported
H7
FCO Interaction
0.086 *
a
2.015
Supported
H8
ETH Interaction
0.143
a
0.758
Not Supported
New Zealand Sample
H1
US  BAA
0.548***
4.426
Supported
(n = 198)
H2
NSI  BAA
-0.192
-1.811
Not Supported
H3
A  BAA
0.104
1.461
Not Supported
H4
PUCC  BAA
0.042
0.488
Not Supported
H5
PR  BAA
0.610***
4.750
Supported
H6
BAA  BAI
0.806***
7.638
Supported
H7
FCO Interaction
0.057 *
a
2.011
Partially Supported
H8
ETH Interaction
0.178 +
a
1.622
Supported
a
= unstandardized regression estimate
+p<0.10, *p<0.05, **p<0.01, *** p<0.001
163
Data Analysis & Results
Chapter 6
Chapter 6:
6
“There is no such thing as failure, there are only results”
---Anthony Robin
6.1
This chapter is an extension of data analysis. We pooled the data of two samples (Pakistan
and New Zealand) to test the model as a whole irrespective of the country differences.
Sampling weights are used to reduce the effect of demographic skewness of the samples and
covariances matrices are being used as input in AMOS after applying weights in SPSS. We
assess the measurement invariance between the two samples before pooling them. We follow
the sequence of analysis such as confirmatory factor analysis on weighted data to establish
convergent and discriminant validity and subsequently, hypothesis testing. Moderation effects
of familiarity with country-of-origin and consumer ethnocentrism are tested in the last part of
the analysis. Findings and interpretation of the results are discussed in the end of this chapter.
6.2
Sampling Weights
After analyzing the model at country level, to have an overall validity of our model, we
pooled the data of two countries. Because of the convenience sampling technique used in our
data collection method, our samples are not true representative of the actual population in
some demographics. For example, Pakistan sample is male4 dominant (73%) and the
4
Pakistan Population: Male = 0.512% and Female= 0.488% Source : http://www.pbs.gov.pk/content/labourforce-survey-2010-11
164
Chapter 6
respondents having education level ‘master or above’ (72.7%), whereas New Zealand sample
is female5 dominant (61.1%) and majority of the respondents are less than 25 years (80.8%).
Sampling weights is a technique used to weigh the sample data to make the sample
representative with respect to the target population of interest. These weights reflect the
probabilities of the unequal sample selection and compensate for differential non-response or
under coverage framework. It has been verified that when the sample is selected with unequal
selection probabilities that are related to the values of the response variables can yield
misleading results (Pfeffermann, 1996). He further, claims that probability weighting of the
sample observations give consistent estimators of the model parameters and protects against
model misspecification. Pfeffermann (1993) suggest that usage of sampling weights can play
an important role in two different aspects of modeling process.

To test and protect against non-ignorable sampling design which could cause selection
bias.

To protect against misspecification of the model holding in the population.
In other words, Yansaneh (2005) emphasized to use sample weights:

To compensate for unequal probabilities of selection.

To compensate for (unit) non-response.

To adjust the weighted sample distribution for key variables of interest (e.g. age,
education, and sex) to make it representative of known population distribution.
Although the usage of weights in statistical analysis of survey data is very common, yet the
role of sampling weights is still a subject of controversy among theorists (Pfeffermann, 1993).
Majority of the researchers are in agreement of using weights for descriptive inferences such
5
New Zealand Population: Male= 0.49% and Female= 0.51% Source: http://www.stats.govt.nz/
165
Chapter 6
as mean, variances and proportions but for analytical inference about model parameters, there
is a wide spectrum of opinion on the role of the sampling weights (Pfeffermann, 1993).
In the case of survey design of sampling individuals with unequal probabilities of selection,
the sample weights effectively represent the number of individuals in the population that each
sampled individual represents. The sample weights attached with an individual is the inverse
of that individual’s probability of being included in the sample for non response (Korn &
Graubard, 1995).
In our study, we calculate the sample weights for Pakistan and New Zealand at two levels;
such that in Pakistan we calculate the weight for gender and education, whereas in New
Zealand we calculate the weights for gender and age (cf. Appendix IX).
Table 6.1: Sampling weights
Gender
Male
Female
Pakistan
Education
Less than Master
Master or Above
Less than Master
Master or Above
Weights
3.576
0.328
1.744
0.383
New Zealand
Age
Weights
Less than 25 years
0.572
More than 25 years
3.570
Less than 25 years
0.631
More than 25 years
2.285
We generate a variable ‘weight’ in SPSS data file by using weights presented in Table 6.1, to
further use the weighted data for pooled sample analysis.
6.3
CFA with Sample Weights
Since direct sample weights cannot be used in AMOS. So we use weighted covariances
matrices as input in AMOS (Kline, 2010). The weighted covariances matrices are calculated
by applying sample weights in SPSS data file (Data → weight cases → weight cases by). We
run CFA with weighted covariance matrices and estimate the models separately for both
166
Chapter 6
samples. The convergent and discriminatory validity has been assessed and established for
both the samples (cf. Appendix X).
6.4
Measurement Invariance
In consumer behavior research, it is recommended to validate the models developed in one
country, in other countries as well (Bagozzi & Baumgartner, 1994). If the constructs are not
understood in similar way across the countries, then the results based on these scales may be
ambiguous. There are many approaches recommended to assess the measurement invariance
(MI) of data collected (Steenkamp & Baumgartner, 1998). Horn (1991 p. 119), precisely
states “without evidence of measurement invariance, the conclusion of a study must be weak”.
We have collected data from two different countries, along with separate analysis for each
country; we want to understand the relationships irrespective of the country differences. For
that reasons we pooled the data and check for measurement invariance between two samples.
There are a variety of techniques available to assess measurement equivalence (Hui &
Triandis, 1985), but multi-group confirmatory analysis model (Jöreskog, 1971) has the most
powerful and versatile approach to test for cross-national measurement invariance
(Steenkamp & Baumgartner, 1998).
Myers et al. (2000) argue that consumers from different cultures may not perceive and
interpret the given construct in a similar way, or their tendency to respond to certain scale
items may vary cross-culturally (Hui & Triandis, 1985). To deal with these problems, scholars
have suggested examining, the contract, metric, and scalar equivalence before analysis across
cultural or national boundaries (Mullen, 1995; Steenkamp & Baumgartner, 1998).
Steenkamp and Baumgartner (1998), proposed to measure Configural Invariance, i.e. the
basic meanings and structure of research constructs are well understood and conceptualized in
167
Chapter 6
a similar way across difference groups (countries); Metric Invariance, i.e. whether the
intervals of scales used are similarly perceived across groups (countries). They also proposed
three more steps with stricter requirements to assess MI.
Configural invariance is supported if the specified model fits the data well in all groups
(countries); all the factor loadings are significantly and substantially different from zero. Also
the constructs have the discriminant validity among them (Steenkamp & Baumgartner, 1998).
Metric invariance can be tested by constraining the factor loadings to be the same across
countries (Steenkamp & Baumgartner, 1998).
According to He et al. (2008), Configural invariance is sufficient, if the research purpose is to
explore cross-nationally the basic structure of the constructs. Steenkamp and Baumgartner
(1998) and Myers et al. (2000), highlight the importance of conducting MI assessment in
cross-national empirical studies in marketing research field and have provided a step-by-step
approach to conduct MI assessment.
We check the measurement invariance by following Steenkamp and Baumgartner’s (1998)
step-by-step approach. The Configural invariance model is estimated first. This is a basic
model with which other models can be compared. The configural invariance model shows a
satisfactory fit indices (χ2= 965.675, df= 298, CFI= 0.901, TLI= 0.874, RMSEA= 0.069,
SRMR= 0.068). All the standardized regression estimates (factor loadings) exceed 0.60 and
are significant in both the countries. The constructs are discriminated as well (cf. Appendix X)
Hence, we can conclude that our model exhibits configural invariance across two countries.
Full metric invariance, is analyzed by constraining all factor loading, for both countries. By
applying constraint and running the analysis, the fit indices (χ2= 1094.207, df= 318, CFI=
0.885, TLI= 0.863, RMSEA= 0.072, SRMR= 0.075) remain in the acceptable range with a
168
Chapter 6
significant increase in χ2-value as compared to configural model. A chi-square difference test
is conducted to study whether the difference between configural and full metric model is
significant. The result of chi-square difference test shows that the two models are not
invariant (with Δχ2= 128.532, df= 20 at p<0.00%). Hence, the full metric invariance condition
is not fulfilled.
Many cross-cultural researchers including Steenkamp and Baumgartner (1998) recognized
that few scales achieve full cross-cultural metric invariance. Consequently, they recommend
that researchers should conduct a test of partial measurement invariance, and if the
measurement instruments are at least partially invariant, the cross-cultural (country) study can
be conducted even when the ideal condition of full invariance is not realized. So to examine
partial metric invariance, we look on the modification indices, we find that there are four
items (NSI3, BA3, NSI2, BA7) for which the modification indices are 93.983, 44.257,
43.358, 38.929 respectively in full item-level metric invariance model. It should be noted that
these values might change in the model modification process. We sequentially relax the
constraints on these parameters, starting with the loading having high modification index.
After setting free all four parameters the fit indices (χ2= 988.464, df= 314, CFI= 0.885, TLI=
0.868, RMSEA= 0.070, SRMR= 0.073), are satisfactory whereas, the χ2-value decreased with
degree of freedom in comparison to the full metric invariance model (Table 6.2). Again a chisquare difference test is applied to look for invariance between configural model and partial
metric model. The chi-square difference test shows that the two models are invariant (with
Δχ2= 22.789,
df= 16 at p>0.10). Hence, partial metric invariance is fulfilled.
Now we check the convergent and discriminant validity for pooled data.
169
Chapter 6
Table 6.2: Model Comparisons for Measurement Invariance
χ2 value
df
RMSEA
SRMR
CFI
TLI
Configural Invariance
965.675
298
0.069
0.068
0.901
0.874
Full Metric Invariance
1094.207
318
0.072
0.075
0.885
0.863
Partial Metric Invariance
988.464
314
0.070
0.073
0.885
0.8618
Calibration Data
6.4.1
CFA: Measurement Model for Pooled Data
The results of measurement invariance show that our two samples fulfill the configural and
partial metric invariance, so we pool up the two samples and run the analysis on the pooled
data. We initially check the model fit of the pooled data (476= 278 + 198), and also look for
the discriminant validity among the constructs.
The fit indices are satisfactory (χ2= 400.379, df= 149, χ2/df= 2.687, CFI= 0.962, TLI= 0.952,
RMSEA= 0.060, SRMR= 0.042), all the indices meet the required conditions (Table 6.3). All
factor loadings are significant and exceed the value 0.60 (Figure 6.1).
Table 6.3: Fit Indices of Measurement Model of pooled data
Fit Indices
χ2
df
χ2/df
CFI
TLI
RMSEA
SRMR
Recommended
/Threshold Value
≤ 3.00
≥ 0.90
≥ 0.90
≤ 0.08
≤ 0.08
Obtained Value
400.379
149
2.687
0.962
0.952
0.060
0.042
The condition of discriminant validity among the variables is fulfilled. Average variance
extracted (AVE) for each construct is above the shared variance among the variables (Table
6.4). The shared variance among BAA and BAI is slightly greater than AVE of BAA. We
170
Chapter 6
decide to take two factors in further analysis because the two factor model is better than one
factor model (cf. CFA for DVs).
Table 6.4: Discriminant Validity of Measurement Model of Pooled Data
BAA
US
NSI
A
PUCC
PR
BAA
0.655
US
0.433
0.632
NSI
0.063
0.126
0.730
A
0.001
0.019
0.108
0.770
PUCC
0.060
0.166
0.102
0.125
0.809
PR
0.134
0.064
0.319
0.132
0.129
0.603
0.726
0.391
0.048
0.001
0.043
BAI
0.048
BAI
0.791
Table 6.5 presents the mean, standard deviation (SD) and inter-correlation among all variables
including socio-demographic variables for Pooled data.
171
e5
US1
e4
US2
e3
US4
e8
NSI1
e7
NSI2
e6
NSI3
e11
A3
e10
A4
e9
A5
e15
UC1
e14
UC2
e13
UC3
Chapter 6
.73
.76
.94
.90
.86
.83
.82
US
.35
.86
NSI
-.14
-.33
.87
A
.41
.25
.32
.56
-.35
e17
PR3
e16
PR4
e21
BA1
e20
BA2
e19
BA3
.95
.88
.87
-.36
PUCC
.81
PR
.78
.84
.24
.37
BAA
BA5
e23
BA6
e22
BA7
.95
.82
.22
.89
BAI
Figure 6.1: Measurement Model of Pooled Data
172
-.03
.21
.63
.22
.78
.85
e24
.25
-.04
.36
.78
.66
Data Analysis
Chapter 6
Table 6.5: Means, Standard Deviations, and Correlations among Study Variables of Pooled Data
Mean
SD
Gender
Age
EDU
MFI
US
NSI
A
PUCC
PR
BAA
BAI
Gender
1.46
0.50
Age
2.76
1.17
-.130
EDU
2.90
1.08
-.241
.647**
MFI
2.43
1.35
-.048
.067
.192**
US
4.89
1.44
.108*
-.047
-.263**
-.138**
NSI
3.92
1.55
-.046
.211**
.183**
.083
.301**
A
4.84
1.59
.166**
-.062
-.143**
.003
-.118**
-.299**
PUCC
4.07
1.34
.016
.014
-.146**
-.064
.354**
.285**
-.334**
PR
3.77
1.54
-.006
.144**
.101*
-.066
.205**
.471**
-.298**
.309**
BAA
4.93
1.55
.125**
-.091
-.191**
-.093
.566**
.211**
-.035
.226**
.304**
BAI
5.55
1.51
.089
-.135**
-.207**
-.071
.568**
.190**
-.041
.207**
.195**
.768**
FCO
4.41
1.46
.022
.092*
.022
-.087
-.129**
-.104*
.317**
-.033
-.076
-.153**
-.184**
ET
3.82
1.56
.034
-.207**
-.298**
-.025
-.003
-.344**
.298**
-.249**
-.311**
.017
.055
FCO
.000
*p< 0.05
**p< 0.01
173
Data Analysis
6.5
Chapter 6
Hypotheses Testing
To check the hypothetical relationship among the variables, the structural regression model is
estimated (Figure 6.2). Fit indices of the structural regression model for pooled data shows a
good fit to the data (Table 6.6). All fit indices are above the minimum criteria.
Table 6.6: Structural Model Fit Indices
Fit Indices
Recommended
/Threshold Value
Pooled data
χ2
414.794
df
154
χ2/df
≤ 3.00
2.693
CFI
≥ 0.90
0.961
TLI
≥ 0.90
0.951
RMSEA
≤ 0.08
0.060
SRMR
≤ 0.08
0.044
174
Data Analysis
e5
e4
e3
e8
e7
e6
e11
e10
e9
e15
e14
e13
e17
e16
US1
US2
US4
NSI1
NSI2
NSI3
A3
A4
A5
UC1
UC2
UC3
PR3
PR4
Chapter 6
.73
.76
.94
.90
.86
.83
.82
US
.35
.86
NSI
.87
A
.78
PUCC
.77
e21
e22
e23
e24
BA3
BA2
BA1
BA7
BA6
BA5
BAA
-.07
-.36
.95
.82 .89
.81 .77
.84
.10
.32.25
.87
e20
-.09
-.33
.41
.56
-.35
.95
.88
.68
-.14
e19
.86
BAI
e26
e25
.28
.36
PR
Figure 6.2: Structural Model Results on Pooled Data
Table 6.7 shows standardized regression estimates along with t-values for all the paths of the
structural regression model.
Table 6.7: Hypotheses Testing
Hypotheses
Direct effect
Std. estimate
t-value
Result
Pooled Data
H1
US  BAA
0.684***
11.332
Supported
(n = 476)
H2
NSI  BAA
-0.092
-1.625
Not Supported
H3
A  BAA
0.101*
2.136
Supported
H4
PUCC  BAA
-0.069
-1.388
Not Supported
H5
PR  BAA
0.275***
4.254
Supported
H6
BAA  BAI
0.858***
17.430
Supported
*p<0.05, **p<0.01, *** p<0.001
175
Data Analysis
6.6
Chapter 6
Moderation Effect
We check the moderating effect of familiarity with country-of-origin and consumer
ethnocentrism for the pooled data with weighted samples.
6.6.1
Moderating effect of familiarity with country-of-origin
We use Hayes and Matthes’s (2009) approach and their SPSS macro to find the interaction
effect of familiarity. The average score is computed for independent, dependent variables. The
weighted independent variable and moderating variable are mean centered for all the data sets
as recommended by Aiken and West (1991). The unstandardized regression estimates for the
6.8.
Table 6.8: Unstandardized estimates for interaction of A and FCO
Relation
Pooled Data
Estimate
S.E.
P
A  BAA
0.684
0.183
0.000
FCO  BAA
0.569
0.060
0.000
A * FCO  BAA
-0.039
0.005
0.000
The interaction term between animosity and familiarity with COO is significant at p< 0.01.
Hence, familiarity with country-of-origin moderates the relationship between perceived
animosity and brand avoidance attitude. This supports hypothesis H7.
176
Data Analysis
Chapter 6
Figure 6.3: Moderation effect of Familiarity with COO
Graphically Figure 6.3 shows the interaction effect in pooled data. It shows that with high
familiarity with country-of-origin, the brand avoidance attitude is higher as compared to low
familiarity with COO.
6.6.2
Moderating effect of consumer ethnocentrism
Similarly, we use Hayes and Matthes’s (2009) approach to find the interaction effect of
ethnocentrism. The average scores for independent, moderator and dependent variables are
computed. The weighted independent variable and moderating variable are mean centered as
recommended by Aiken and West (1991). The unstandardized regression estimates for the
6.9.
177
Data Analysis
Chapter 6
Table 6.9: Unstandardized estimates for interaction of A and ETH
Relation
Pooled Data
Estimate
S.E.
P
A  BAA
0.609
0.047
0.000
ETH  BAA
0.653
0.068
0.000
A * ETH  BAA
-0.032
0.005
0.000
The interaction term is significant at p< 0.05. This implies that ethnocentrism plays a
moderating role on the relationship between animosity and brand avoidance attitude, it
supports hypothesis H8.
Figure 6.4: Moderation effect of Ethnocentrism
Graphically, Figure 6.4 shows the interaction effect of ethnocentrism in pooled data. It shows
that high ethnocentric consumers have high brand avoidance attitude as compared to low
ethnocentric consumers.
178
Data Analysis
6.7
Chapter 6
Results of the Study
We discuss and interpret findings of our study for all three sets of data i.e. Pakistan, New
Zealand and Pooled sample.
6.7.1
Undesired self congruence and brand avoidance attitude
We proposed in our first hypothesis that undesired self congruence positively influences
consumers’ brand avoidance attitude. The significance of standardized estimates at p<0.001,
in all three data sets show that undesired self is a positive predictor of brand avoidance. It
supports H1. Results confirm the phenomenon of staying away from undesired self image or
group of people. The higher the congruence of one’s undesired self with brand’s image, the
greater the avoidance attitude for that brand.
Since our core objective was to find out the influential pre-purchase determinants of the brand
avoidance, and among them which factor is relatively more important in predicting the brand
avoidance attitude and behavior. In all three data sets we have found that undesired self
congruence is the essential factor in predicting brand avoidance attitude. As depicted from the
standardized regression estimates (Table 6.10), it has relatively high standardized estimate as
compared to other factors in the study, showing that it is relatively more concrete and
important factor in making the avoidance attitude towards brands.
Table 6.10: Results of Hypthesis 1
Hypotheses
Direct effect
Std. estimate
Result
Pooled Data
H1
US  BAA
0.684***
Supported
Pakistan Sample
H1
US  BAA
0.581***
Supported
New Zealand Sample
H1
US  BAA
0.548***
Supported
*** p< 0.001
179
Data Analysis
Chapter 6
The results obtained strongly prove our hypothesis that undesired self congruence positively
influences consumers’ brand avoidance attitude. The findings confirm earlier research (Englis
& Solomon, 1995; Ogilvie, 1987; Phillips et al., 2007) which indicate that consumers may
abstain from purchasing or use of product/brands owning to their reluctance to be identified
with an avoidance image/group. Clear support is provided to Banister & Hogg’s (2004)
findings that respondents have negative feeling towards unwanted image of self.
Many studies concluded that purchasing or approaching towards brands is motivated by their
actual or ideal self but they are also motivated to avoid undesired self of the consumer
(Banister & Hogg, 2004; Bhattacharya & Elsbach, 2002; Elsbach & Bhattacharya, 2001;
Englis & Solomon, 1995; Grubb & Harrison L. Grathwohl, 1967; Hogg, 1998; Hogg et al.,
2000; Hogg & Michell, 1996; Sirgy, 1982; Wilk, 1996). We have found that undesired self is
the strongest motivator in making avoidance attitude.
Ogilvie (1987) proposed that undesired self is based on those associations and values which
the consumer does not want to be linked with and abstain from giving the negative self image.
He also gave the reason that undesired self is relatively more concrete self image based on the
past experience of failure, humiliation and guilt. This study further extends his findings that
prior experience of failure or unmet expectations are not the only reason for staying away
from undesired self image. Our results emphasized that undesired self congruence is a strong
predictor of brand avoidance attitude even the consumer did not use that brand before. In
other words, it is a strong pre-purchase determinant of brand avoidance.
In his study based on qualitative interviews in New Zealand, Lee (2007), argued that as the
brand selection is the way of construction of ideal self or maintenance of the actual self, by
consuming the desired brands, similarly, the avoidance of brand involves staying away from
the undesired selves by rejection or avoiding the incongruent brands. Our findings empirically
180
Data Analysis
Chapter 6
proved that the avoidance attitude is strongly predicted by the undesired self congruence with
brands.
Escalas & Bettman (2005) in making self brand connection, they found that consumers make
strong connections with the brands that were being used by their own groups, and contrarily,
they make weak connections with those brands which were being used by outer group (the
group of people with which the consumer is not willing to be associated). Similarly, we have
found that people have high avoidance attitude towards those brands which have
inconsistence image congruence with their self image.
6.7.2
Negative social influence and brand avoidance attitude
The second hypothesis proposed that negative social influence positively influences
consumers’ brand avoidance attitude. We obtained mixed results in all three data sets. In
Pakistan sample the standardized estimate is positive but not significant at p< 0.05, which
does not support H2. In Pooled and New Zealand data sets, we found significant regression
estimates at p< 0.10 but in opposite direction to our proposed relationship. Here too, H2 is not
supported.
According to Hofstede (1980), Pakistan has a collectivistic culture whereas New Zealand has
an individualistic culture. In collectivistic culture, people take decisions under the influence of
others important to them. An important characteristic of the people having collectivistic value
orientation is that they will subordinate their personal interest to the common goals of their
collective entity with whom they work and identify (Triandis & Albert, 1987). Individualistic
cultural orientation, on the other hand focuses on the self-sufficiency, control and fulfillment
of individual goals which may or may not remain consistent with in-group goals and
expectations. Collectivistic, as compared to individualistic, have self concept derived from ingroup identity and relationship with others (Parkes, Schneider, & Bochner, 1999). We
181
Data Analysis
Chapter 6
proposed that negative social influence positively influences consumers’ brand avoidance
attitude. Logically, this relationship should be expected to exist in collectivistic culture
(Pakistan). As expected, we have found this relationship positive in Pakistan but not
significant at p<0.05 (Table 6.11).
Table 6.11: Results of Hypothesis 2
Hypotheses
Direct effect
Std. estimate
Result
Pooled Data
H2
NSI  BAA
-0.092*
Not Supported
Pakistan Sample
H2
NSI  BAA
0.082
Not Supported
New Zealand Sample
H2
NSI  BAA
-0.192
*
Not Supported
*p< 0.10
To further explore this relationship, we did multi-group analysis, to know the effect of
demographic variables (gender, age, education and monthly family income) on this
relationship. We found no significant effect of gender and education on this relationship. But
age and monthly family income have significant effect on the relationship in Pakistan sample.
We have found in multi-group analysis that respondents less than age 25 years and belonging
to lower income families i.e. monthly family income less than Rs.80,000, they have
significant positive relationship between negative social influence and brand avoidance
attitude at p<0.10 and p<0.05 respectively (Table 6.12).
Table 6.12: Socio-Demographic effects
Demographic Variables
Direct Effect
n
Std.
Estimate
P-value
Age < 25 years
NSI  BAA
74
0.303
0.082
Age > 25 years
NSI  BAA
204
0.079
0.315
Income < Rs. 80,000
NSI  BAA
122
0.238
0.043
Income > Rs. 80,000
NSI  BAA
156
0.003
0.971
The same behavior is also supported by one interviewee’s statement such that:
182
Data Analysis
Chapter 6
“In my opinion, it was a temporarily created incident and similar event does occur
but they are for specific time period. It was propaganda against them, and there is no
truth behind this. At that time I was immature and I don’t know much about that and I
avoided Nike but now I purchase it”. GA (Male, 27)
Statement clearly tells that before he was under the influence of others but as he grows up, he
learnt and started using the brand which he was avoiding before.
Kulviwat et al. (2009) found that people are not independent, they choose to perform a
behavior which they themselves do not want to perform in their free opinion but they do it
just to comply their behavior with the reference groups’ desirability. Our study supported this
phenomenon in a way that in avoidance of certain brands the negative social influence of
others plays an important significant role when the consumer is not mature.
It is found in earlier studies that consumers who are dissatisfied with a purchase or rejected or
discontinued using a product are telling friends about the experience and influence them to
avoid it (Bodur & Day, 1978; Lau & Ng, 2009; Leonard-Barton, 1985). We have found this
phenomenon empirically true that negative influence of others affect the decision making of
choosing or avoiding brand selection especially in the early age. Our findings are also in line
with Richins’s (1983, 1984) work that negative word of mouth spread by those who are not
satisfied with product/brand or services will have a substantial impact on consumers choice.
We also observed in multi-group analysis that the relationship is significant for the group of
people with low family income, whereas the relationship is not significant for the people from
higher family income (Table 6.12). It seems quite logical that people from lower income
family are very much afraid of losing money in buying brands, particularly, expensive brands,
because of that reason they consult about their decision with important others in family or
with their friends before taking final decision about buying brands. Whereas, on the other
183
Data Analysis
Chapter 6
hand, people belong to higher family income are not conscious about losing money. That’s
why they consult less about their decision and they perceive less influence of others in making
decision about buying a brand or avoiding a brand.
6.7.3
Perceived animosity and brand avoidance attitude
The third hypothesis stated that consumers’ perceived animosity towards the country-oforigin positively influences consumers’ brand avoidance attitude. In pooled data and Pakistan
sample, regression estimates are positive and significant at p<0.05 (Table 6.13). This shows
that perceived animosity is a positive predictor of brand avoidance attitude. The higher the
perceived animosity with the country-of-origin, the higher is the brand avoidance attitude. It
supports H3. In New Zealand sample, the regression estimate is positive but not significant at
p<0.05.
Hypotheses
Direct effect
Std. estimate
Result
Pooled Data
H3
A  BAA
0.101*
Supported
Pakistan Sample
H3
A  BAA
0.145*
Supported
New Zealand Sample
H3
A  BAA
0.104
Not Supported
*p< 0.05
Our findings supported the previous study conducted by Klein et al. (1998) in China. In their
study of animosity model testing on foreign product purchase, they found negative significant
relationship between animosity and willingness to purchase foreign products. Our study has
strengthened this relationship that people have positive significant relationship between
animosity and brand avoidance attitude. It proved that perceived animosity is a strong
predictor of brand avoidance attitude.
184
Data Analysis
Chapter 6
Past research on animosity is mainly focused on the link between perceived animosity and
consumer behavior, by studying the effect of anger on their purchase decisions and proved
that angry consumers do not want to purchase the products from those countries even if the
product itself is held in high esteem (Russell & Russell, 2010). Our study confirms the
findings in Pakistan sample that consumers do have animosity towards certain country and
avoid their brands.
We found support from Yang and Tso (2007) who studied the animosity relationship between
Chinese and Taiwanese and found significant negative attitude by Taiwanese consumers
towards TV programs imported from China. Our results are similar to Russell & Russell’s
(2010) study of French consumers’ attitudes toward US brands, their finding were that
animosity towards a country is directly and negatively related to attitudes towards brands that
are strongly stereotypic of that country. We found that consumers’ animosity is directly and
positively related to avoidance attitudes towards brands.
In New Zealand this relationship is not significant; the possible reason could be its
geographical location southwestern pacific ocean with only one neighbor Australia. People
from both the countries do not have harsh feelings towards each other. On the other hand,
Pakistan is located in south Asia region among the countries like India, China, Iran and
Afghanistan. Pakistani people have more animosity feeling towards India as they had fought
three wars in last fifty years. Also Pakistan has recent diplomatic disputes with US because of
its military action in Afghanistan.
We also proposed in our hypothesis seven that familiarity with COO moderates the
relationship between perceived animosity and brand avoidance attitude; the higher the
familiarity with COO the stronger the relationship and vice-versa. The unstandardized
regression estimates are significant in all three data sets at p<0.05 (Table 6.14). This shows
185
Data Analysis
Chapter 6
that familiarity with COO moderates the relationship between animosity and brand avoidance
attitude. The results from New Zealand partially supported our hypothesis, as moderation
does exist but contrary to our hypothesized moderation relationship. This might be due to the
fact that the average response for familiarity with COO from New Zealand sample is 4.38
(where 1= not familiar and 7= very familiar). The low average shows that the consumers are
not very familiar with the COO.
Hypotheses
Indirect effect
Un-stand.
estimates
Result
Pooled Data
H7
FCO interaction
-0.039***
Supported
Pakistan Sample
H7
FCO interaction
0.086*
Supported
New Zealand Sample
H7
FCO interaction
0.057*
Supported
*p<0.05, **p<0.01, ***p<0.001
In Pakistan and pooled data, our findings are similar to Jiménez and Martin’s (2010) finding
who found in their study of automobiles in Spain that familiarity with the origin of product
and brands has significant moderation effect on the relationship between animosity and the
trust towards the firms.
Similarly, we proposed in our H8 that consumer ethnocentrism moderates the relationship
between perceived animosity and brand avoidance attitude; the higher the consumer
ethnocentrism the stronger the relationship and vice-versa. The unstandardized regression
estimates are significant in pooled data and New Zealand sample at p<0.001 and p<0.10
respectively (Table 6.15). This shows that consumer ethnocentrism moderates the relationship
between animosity and brand avoidance attitude. In Pakistan the interaction term is not
significant. This might be due to the fact that average score for ethnocentrism is 3.64 (where
1= strongly disagree and 7= strongly agree). The low average shows that consumers are not
high ethnocentric in our sample.
186
Data Analysis
Chapter 6
Hypotheses
Indirect effect
Un-stand.
estimates
Result
Pooled Data
H8
ETH interaction
-0.032***
Supported
Pakistan Sample
H8
ETH interaction
0.143
Not Supported
New Zealand Sample
H8
ETH interaction
0.178+
Supported
+p<0.10, ***p<0.001
In New Zealand and pooled data, our findings are consistent with the literature as Alden et al.
(2006) found that the more ethnocentric consumers are, the more they demonstrate negative
attitude toward global brands. We have found that the more ethnocentric consumers are the
more they have positive brand avoidance attitude.
6.7.4
Perceived unethical corporate citizenship and brand avoidance
attitude
H4 proposed that perceived unethical corporate citizenship (PUCC) positively influences
consumers’ brand avoidance attitude. Regression estimates are not significant in all three data
sets at p< 0.05 (Table 6.16). it does not support H4. Unethical behavior perceived by
consumers does not motivate them to avoid the brands of that company, or people may not be
concerned with the company’s role in society in general.
Although the estimates are not significant but the negative sign in Pakistan shows
contradiction to our proposed relationship.
Hypotheses
Direct effect
Std. estimate
Result
Pooled Data
H4
PUCC  BAA
-0.069
Not Supported
Pakistan Sample
H4
PUCC  BAA
-0.095
Not Supported
New Zealand Sample
H4
PUCC  BAA
0.042
Not Supported
187
Data Analysis
Chapter 6
We can conclude that either people do not consider about the company’s ethical role in
society seriously or if the corporate citizenship is perceived unethical even then it would not
effect on consumers’ purchase decisions.
Fan (2007) argued that it is not guaranteed that ethical branding provides a success in
marketplace; or on the other hand, we can see that a popular or successful brand is not
perceived ethical. Consumers generally have ethical concern with brands or organizations but
these concerns do not necessarily play a role in their actual purchasing behavior. Carrigan and
Attalla (2001) concluded in their study on UK respondents, that although consumers are more
sophisticated today, but this is not essentially true that this will translate into a behavior that
favors ethical companies over unethical ones.
Titus and Bradford (1996), in their study on US respondents, found that in recent times, with
more choices available in the market place, consumers’ sophistication is declining rather than
increasing and as a result the unsophisticated consumers are more likely to reward unethical
business practices and punish ethical business behavior.
According to some marketing scholars, in recent times consumers are better informed, more
educated and well aware about consumer rights and product requirements at least in western
society (Barnes & McTavish, 1983; Hirschman, 1980) but it is not necessary that possessing
‘consumer sophistication’ enforce them to behave wise or ethical buying practices (Titus &
Bradford, 1996). Further, Titus and Bradford (1996) emphasized that there is a clear
distinction between sophisticated consumer characteristics and sophisticated consumer
behavior which has to be recognized in marketing ethics literature.
In ethical purchasing behavior, an attitude-behavior gap has been identified by Roberts
(1996). He found that generally consumers have socially responsible attitudes but only 20
percent of the consumers have actually purchased something in the last year because the
188
Data Analysis
Chapter 6
product was associated with a good cause. People express their willingness to make ethical
purchases to make their decision linked to good reputation but in reality social responsibility
is not their most dominant criteria in making purchase decision (Boulstridge & Carrigan,
2000).
Carrigan and Attalla (2001) referred brand marketer’s survey results that they found 57
percent of their sample with the opinion that they would stop buying that brand if they knew
that child labor had been employed but only 21 percent supported the actions against those
companies which they perceived as unethical. There is a huge difference between supporting
an action and actually carrying it out oneself. Hence, ethical phenomenon is yet to be explored
in true sense.
6.7.5
Perceived risk and brand avoidance attitude
We proposed in the fifth hypothesis that consumers’ perceived risk positively influences
consumers’ brand avoidance attitude. Results are positive and significant at p< 0.001 for
pooled data and in New Zealand (Table 6.17). It confirms that perceived risk motivates
consumers to avoid risky brands. It supports H5. In Pakistan sample, the regression estimate
is positive but not significant at p<0.05.
Hypotheses
Direct effect
Std. estimate
Result
Pooled Data
H5
PR  BAA
0.275***
Supported
Pakistan Sample
H5
PR  BAA
0.041
Not Supported
New Zealand Sample
H5
PR  BAA
0.610***
Supported
***p< 0.001
Our findings are similar to the results of Lee (2009), he found the effect of perceived risk on
attitude and intention towards adoption of internet banking in Taiwanese environment. The
189
Data Analysis
Chapter 6
results showed that perceived risk in different dimensions (security, financial, social and
performance) has a negative effect on attitude and intention in adoption of online banking.
Our results, in another way prove the same thing that perceived risk has positive effect on
brand avoidance attitude and is a strong predictor of brand avoidance attitude.
While looking on the brands list (cf. Appendix XI) of the respondent in Pakistan sample,
majority of the brand are of mobile phones and laptops. Since all the brands are of foreign
made and literature tells that consumers from developing countries perceive foreign brands as
being of higher quality than domestic brands (Raju, 1995). This phenomenon also depicted in
one of our interviewee’s verbatim:
“…it included many things for example design, color, tastes and many more. If all
these things are good in our national brand then I will go for our national brand. I
would purchase our national product without compromising on these quality related
things.”AJ (Male, 38)
It is clear in his statement that he is buying foreign brands and considers it as high quality.
This might be the reason that why people from Pakistan do not consider perceived risk as a
predictor of brand avoidance for global brands.
6.7.6
Brand avoidance attitude and brand avoidance intention
We proposed in hypothesis six that consumers’ brand avoidance attitude is positively linked
to their brand avoidance intention. The positive and significant regression estimates at
p<0.001, in all three data sets show that brand avoidance attitude leads to brand avoidance
intention (Table 6.18). It supports H6.
190
Data Analysis
Chapter 6
Hypotheses
Direct effect
Std. estimate
Result
Pooled Data
H6
BAA  BAI
0.858***
Supported
Pakistan Sample
H6
BAA  BAI
0.832***
Supported
New Zealand Sample
H6
BAA  BAI
0.806***
Supported
***p< 0.001
The significance of this relationship is consistent with previous studies which show that
attitude towards buying affects consumers’ purchase decisions. Our study shows that attitude
towards avoidance also affects consumers’ purchase decisions. Further, it shows that this
relationship exists on negative side as well. i.e. the avoidance attitude towards brands also
affects avoidance intentions. Ang et al. (2001) found that consumers’ attitude toward piracy is
a significant predictor of their purchase intention and consumers who have a positive
disposition toward certain brands are more likely to purchase those brands. Contrary to them,
on the similar lines, we propose that negative attitude towards certain brands also predicts the
negative intentions towards those brands which restrict them to buy those brands.
Our findings are similar to the result of Wee et al. (1995) that more unfavorable a consumers’
attitude toward a brand the lower would be their intention to purchase such brands. Vis-à-vis,
if consumers have more avoidance attitude for certain brands the more will be their avoidance
intention for those brands.
6.8
Conclusion
The summary of results of all hypotheses in all three data sets (Pakistan, New Zealand and
Pooled samples) is provided in Table 6.19.
191
Data Analysis
Chapter 6
Table 6.19: Summary of Results
Hypotheses
Direct effect
Std. estimate
p-value
Result
Pooled Data
H1
US  BAA
0.684
***
Supported
(n = 476)
H2
NSI  BAA
-0.092
0.104
Not Supported
H3
A  BAA
0.101
0.033
Supported
H4
PUCC  BAA
-0.069
0.165
Not Supported
H5
PR  BAA
0.275
***
Supported
H6
BAA  BAI
0.858
***
Supported
H7
FCO Interaction
-0.039
a
***
Supported
H8
ETH Interaction
-0.032
a
***
Supported
Pakistan Sample
H1
US  BAA
0.581
***
Supported
(n = 278)
H2
NSI  BAA
0.082
0.247
H3
A  BAA
0.145
0.025
Supported
H4
PUCC  BAA
-0.095
0.216
Not Supported
H5
PR  BAA
0.041
0.616
Not Supported
H6
BAA  BAI
0.832
H7
FCO Interaction
0.086
Partially Supported
***
Supported
a
0.045
Supported
a
0.448
Not Supported
H8
ETH Interaction
0.143
New Zealand Sample
H1
US  BAA
0.548
***
Supported
(n = 198)
H2
NSI  BAA
-0.192
0.070
Not Supported
H3
A  BAA
0.104
0.144
Not Supported
H4
PUCC  BAA
0.042
0.626
Not Supported
H5
PR  BAA
0.610
***
Supported
H6
BAA  BAI
0.806
H7
FCO Interaction
0.057
H8
ETH Interaction
0.178
b
***
Supported
a
0.046
Partially Supported
a
0.100
Supported
US= Undesired Self, NSI= Negative Social Influence, A= Animosity, PUCC= Perceived Unethical Corporate
Citizenship, PR= Perceived Risk, FCO= Familiarity with Country-of-origin, ETH= Ethnocentrism, BAA=
Brand Avoidance Attitude BAI= Brand Avoidance Intention.
a
b
=Unstandardized regression estimate
=Multi-group analysis with demographic variables
192
General Discussions
Chapter 7
Chapter 7:
7
General Discussions & Conclusion
“A conclusion is the place where you got tired of thinking”
---Martin H. Fischer
7.1
This chapter concludes the whole discussion. It discusses the principle findings of the study,
then it highlights the possible theoretically contributions and managerial implications of this
study. Moreover, this chapter also identifies some limitations and future research avenues
opened by this study have been recommended.
7.2
Summary of the research findings
We discuss the principle findings of the study based on all three data sets (Pooled data,
Pakistan sample and New Zealand sample).
The purpose of our study was to find out the influential determinants (factors) of brand
avoidance, and further among those factors, the relative importance to understand that which
factor is more important for this avoidance behavior. We have chosen two countries Pakistan
(a developing country) and New Zealand (a developed country) to have diversity in the
consumer markets. Based on qualitative interviews conducted in Pakistan, we found some
factors (Undesired self, negative social influence, animosity, perceived unethical corporate
citizenship, perceived risk, familiarity with country-of-origin, and ethnocentrism) that have
potential to influence consumers to avoid certain brands. We proposed a hypothetical model
based on these factors, with supports from literature. The empirical studies conducted in
193
General Discussions
Chapter 7
Pakistan and New Zealand show that among those factors only undesired self, animosity and
perceived risk are the significant direct influential factors, and among them undesired self is
relatively the most important pre-purchase determinant of brand avoidance.
We find that undesired self is a strong predictor of brand avoidance even if you have no direct
experience with the brands; it is an extension of Ogilvie’s (1987) findings that undesired self
is a concrete self image based on past experience of failure. Our finding proved that as self
congruence (ideal or actual) is important for the brand selection (Sirgy, 1982), similarly
undesired self is also important for brand avoidance.
We also find the animosity as it perceived and studied in literature, it has a negative
relationship towards willingness to buy products or brands from the certain countries with
which consumers have anger feelings (Russell & Russell, 2010). The moderation effects
proved that the animosity relationship with brand avoidance attitude is stronger when the
familiarity with the country-of-origin is high (Jiménez & Martín, 2010) and also when the
consumer is high ethnocentric.
Negative social influence as it is expected to influence in collectivistic culture (Pakistan),
where people are likely to do shopping together, and seek approval from others before making
the final purchase. We have found no significant effect on brand avoidance attitude. Yet, this
relationship is proved significant for consumers less than 25 years and belonging to the family
of income less than Rs. 80,000 in subgroup analysis.
Although companies are more concerned about their corporate social responsibilities and their
role in society, but the effect of ethical and unethical practices perceived by consumers is still
a question. We have found no significant effect of perceived unethical corporate citizenship
on brand avoidance attitude in both countries. The respondents from both the countries
(developing and developed) exhibit same behavior.
194
General Discussions
Chapter 7
Perceived risk has been studied frequently in technology acceptance model and found
significant relationship with technology adoption attitude (Lee, 2009). We have found similar
results in New Zealand that consumers perceive high risk in certain brands and avoid those
brands. Whereas in Pakistan our findings also supported by Raju (1995) who found that
people from developing countries have the feeling of foreign brands as high quality and
perceive less risks against those brands.
7.3
Contributions
In the introduction chapter, we have identified the practical problems faced by brand
managers and the academic gaps in existing brand avoidance research, particularly, prepurchase brand avoidance.
A brand can be avoided for many different reasons which
consumer perceive unmatched with him, and awareness to these reasons may be helpful in
maintaining or improving in brand related issues. The knowledge about anti-consumption
attitude and behaviors has become an important part to marketing scholars and practitioners.
Our purpose of this study was to enhance the understanding of anti-consumption domain and
to fill the literary gap exists in the area of pre-purchase brand avoidance. The primary
objective was to provide a more integrated and comprehensive understanding of the subject
area by offering a conceptual modal, empirically tested in two countries (Pakistan and New
Zealand).
Since, no prior study specifically has been done to address a similar set of objectives. So, we
tried to find out that what are the specific reasons behind this avoidance of specific brands?
What brands consumers avoid? The list of brands avoided in both the countries has been
provided in Appendix XI. In Pakistan, majority of the brands are global and related to
electronics brands such as mobile phones and laptops, against which consumers have shown
195
General Discussions
Chapter 7
avoidance behavior, whereas, in New Zealand, people showed avoidance behavior against
brands from multiple categories of products.
It is argues that brand avoidance behavior is a phenomena exercised in developed countries
where consumers have full choice of selection and avoidance against brands. Our findings
from Pakistan showed that the people from developing countries also exhibit brand avoidance
behavior. We have found significant effect of family income on the brand avoidance attitude
in Pakistan that shows that as the income increases, the people exercise more avoidance
behavior.
The brand avoidance phenomenon has been identified qualitatively through in-depth
interviews with the consumers and empirically tested on a variety of consumers from both the
countries. We have samples, a mixture of students and professionals from Pakistan whereas,
our sample from New Zealand consist of university student only.
Our study contributed to the area of brand avoidance in many ways:

It provided the evidence, qualitatively and empirically, of pre-purchase brand

We developed and verified a conceptual model based on the factors (undesired self
congruence, negative social influence, perceived animosity, perceived unethical
corporate citizenship, and perceived risk) affecting pre-purchase brand avoidance,
emerged through in-depth interviews and supported by literature.

We provided a generalized perception of pre-purchase brand avoidance by combining
both the samples (Pakistan sample and New Zealand sample) by using sample weights
to represent the samples matched with socio-demographic characteristics of target
populations.
196
General Discussions

Chapter 7
To further understand the difference between countries, a separate country-wise
perception was provided and difference between the perceptions was addressed.

We provided the evidence of moderating role of familiarity with country-of-origin on
brand avoidance attitude and empirically proved it in all three sets of data.

Our study discussed the role of ethnocentrism as a moderator on brand avoidance
attitude, empirically proved in New Zealand sample.
By explicitly exploring the reasons why consumers avoid brands before consumption, this
thesis will contribute to brand avoidance area in different ways. First, from a managerial
point of view, it is very important for the managers to know why consumers are not willing to
buy their product for the first time; although they have fulfilled the pre-requisites of the
purchase regarding the price, quality level, after sales services, etc. Second, from an academic
point of view, a significant amount of literature is available on positive attitude towards brand
but, on the negative side, literature is still not enough to answer managers’ question. So in the
form of this thesis, a conceptual framework is developed by joining together the existing
literature available in the anti-consumption area. Our study offers, for the first time, a
comprehensive framework based on conceptual and empirical findings of pre-purchase brand
7.4
Implications
Our study has academic and practical implications, from the academic perspective, to have
knowledge that why people avoid certain brands is interesting; the major reason of avoidance
is the unmet expectation or dissatisfaction. While our study provides insight of the brand
avoidance phenomenon even people do not have first-hand experience with brands. From
practical perspective, to be aware of brand avoidance, the practitioners should accept the state
197
General Discussions
Chapter 7
of flux between positive and negative brand equity (Lee, 2007) which may create a problem
for company in longer time period.
Companies may take advantage over its competitors by having this knowledge and translating
it into appropriate strategies to cope with the consumers who have pre-purchase negative
attitudes and behaviors towards their brand. The focus of relational marketing is to understand
consumer and build customer relationship, so companies in general and marketing managers
should take it as a long term objective to work on those consumers and to mitigate prevailing
negative feelings about their brand or company.
From a practical point of view, it is not always easy to know about the consumers, who are
avoiding your brands. Even if there are many consumers who exhibit brand avoidance
attitudes, the number of such consumers is often fewer than the consumers who hold positive
attitudes towards the brand. When the companies look on the sales figures they do not take
into account the consumers holding negative attitude towards their brands, as this number is
not enough to affect on the immediate sales as compared to the satisfied consumers with
brands (Reich, 1998). It is also worth noticing that in financial terms the effect may be
negligible if the brand is selected by majority of the consumers in target markets, so not all
the brand managers should be concerned about a small chunk of consumers avoiding their
brands.
Ironically, this should not be surprising that for one segment of the population the brand may
generate negative results and simultaneously in another segment it has positive results, as the
brands and the associated promises are not created to satisfy entire populations (Ward, Larry,
& Jonathan, 1999). Hooley et al. (2005) argued that an established brand having good repute
in the market can predicts superior market performance irrespective of consumer’s
satisfaction or dissatisfaction. It’s also true that despite of brand avoidance attitudes and
198
General Discussions
Chapter 7
behaviors of some consumers, there are many brands that continue to create value for their
shareholders.
In small to medium enterprises (SMEs), it is relatively easy to identify the affect of
consumer’s avoidance as compared to large organizations, where due to existing large market
share with strong business relationships and established successful distribution channels, the
immediate effects of avoidance behaviors of end-users are not as noticeable. Probably, for
large organizations, these initial effects of brand avoidance may take years to translate into a
noticeable financial loss.
Despite of the above mentioned advantages for large organizations, the researcher believes
that if sufficient consumers continue to avoid a brand and the firm is not responding, then the
brand could be in trouble and develop negative brand equity for the firm. If this situation
persists over a longer period of time, the brand usually considered as market-based asset,
could become the market-based liability. But the questions remain there for brand managers
that whether it is possible and necessary to change certain features of their brand in order to
reduce this negative brand equity. Hence, brand managers could take advantage of having
knowledge and understanding of why their brand is faltering.
In terms of marketing implications, the knowledge about the factors that prevent brand
avoidance such as lack of alternatives, product involvement, and social influence can also be
useful information to attract competitor’s clientele in target markets. For instance, if the lack
of alternatives is the main restriction in switching of brand, in that situation competitor could
advertise itself as a suitable alternative. Similarly, this could be competitive advantage for the
companies that by anticipating the factors that prevent brand avoidance, they can implement
appropriate strategies to make their customers avoid competition.
199
General Discussions
Chapter 7
The researcher in convinced that despite the managerial implications of brand avoidance, it
must be discussed in academic literature by the marketing researchers, in order to provide
comprehensive perspectives of brand consumption.
7.5
Limitations
No research is without limitation, so this subsection will address some of the limitations of
this thesis. The first and obvious limitation is small sample sizes in both the samples,
particularly; New Zealand sample size was relatively small.
Although, from Pakistan, the researcher succeeded in collecting a more heterogeneous sample
from different fields of life, by taking almost reasonable respondents form all the age-groups,
professionals and students data. But the New Zealand sample was solely student based sample
of one university. So the results obtained from New Zealand limits the generalizability of the
findings. Although, Douglas and Craig (1997) argued that the youth globally exhibits a high
degree of homogeneity across multiple cultures, because of their high exposure to global
telecommunications and technologies.
Similarly, samples were also skewed, i.e. Pakistan sample is male dominant with male
(73.0%); and New Zealand sample is female dominant with female (61.1%). However, in
pooled sample data analysis, this skewness has been reduced by allocating sampling weights
to make the samples representatives of the target populations. Yet a large sample including
equal number of consumers in genders would improve the generalizability of the findings for
separate country wise analysis.
The research is exploratory in nature and we used a convenience sample in data collection
through online web-survey. This may be a potential objection of true representation of the
population in both the samples and limit generalizability. A wider range of participants by
200
General Discussions
Chapter 7
using other sampling techniques (e.g. random sampling or stratified random sampling) may
increase the generalizability of the findings.
The respondents are contacted through internet in Pakistan, by using social media networks.
That is the reason that respondents were highly educated such as education level ‘master or
above’ (72.8%). The findings should only be interpreted as being representative of the
participants’ attitudes and intentions. A heterogeneous sample representing true population
would increase the generalizability.
We have found that in Pakistan sample major avoidance is against cell phones and laptops
(83.4%). Pakistan is a developing country and it has shown tremendous progress in last
decade, particularly telecommunication sector has shown significant growth. According to
Pakistan Telecommunication Authority (PTA), Pakistan has more than 100 million cellular
users (cf. Table 4.1). Similarly, with the increase of internet users (cf. Table 4.1) the need for
laptop has also increased. That is why people have more avoidance in that category. The
results of Pakistan sample should be interpreted according to the product category avoided by
the respondents.
Past research has indicated that particular product categories may generate specific
quality/value judgments or product evaluations can be attribute-specific (Charles, 2012).
Product specific research was not the objective of the study; however, it is possible that
specific product categories may yield further insight into variations in consumption attitudes
within cross-cultural populations.
Although, before data collection from Pakistan, the researcher confirmed that English
language is better understood and recommended to be used in questionnaire, yet the results
would lack generalizability as our majority of the sample consist of highly educated
consumers of the country. For less educated consumers, this may create some comprehension
201
General Discussions
Chapter 7
problems. Future study can address this issue by collecting data in native language (Urdu)
which may improve the results.
Country of origin have many connotation such as country of origin, country of design,
country of assembly, country of manufacture, or country of ownership provide a variety of
perspectives on product-country relationships. Our study did not attempt to differentiate
among these various points of view of brand/product origin. Although, Charles (2012) argued
that various dimensions only make the decision making process more complicated, but It is
possible that consumers may have varying attitudes and interpretations about their
avoidance/support of foreign product purchase.
7.6
Future research
Anti-consumption is a growing area of research and there are much opportunities for other
researchers to contribute in this domain. Limitations of current study can be better addressed
in future studies.
Findings of the current study show that undesired self congruence, negative social influence,
animosity, and perceived risk are the major determinants of the pre-purchase brand avoidance
attitudes leading to intention; whereas familiarity with country-of-origin and ethnocentrism
are the moderators affecting on the relationship between animosity and brand avoidance
attitude. There may be several other factors influencing consumers’ brand avoidance attitude,
such as lack of after sales services and unavailability of spare parts, word-of-mouth, religious
and politically motivated factors which can be investigated in future studies. Moreover, the
cultural factors like social norms and being collectivistic/individualistic may also exert
influence on consumers brand avoidance attitude.
202
General Discussions
Chapter 7
In developing countries, availability of counterfeit brands has been increased. Attitude
towards counterfeit brands, the opinion of peer group or friends who already had bought and
consumed a counterfeit brand or the risk of purchasing a counterfeit brand on internet would
be an interesting dimension to explore the factors affecting pre-purchase brand avoidance.
The current study explored brand avoidance for general brands. The future study can test this
model or explore new dimensions (factors) for specific product categories/sectors such as
automobile, electronics products (mobile, laptop, TVs, cameras), food and beverages,
clothing, perfumes, shoes, and sports products. Pre-purchase avoidance factors may be
different for conspicuous and inconspicuous products. Service sector can also be explored in
future studies for pre-purchase avoidance.
Methodologically, in-depth interviews and survey questionnaire were the most widely used
approaches. However, there is always a gap between what people say and what they do or gap
between attitudes and behaviors (Ajzen & Fishbein, 1969; Fishbein & Ajzen, 1972). One
possible way to know actual behavior is observational study as Taylor and Bogdan (1998 p.
90) states that “no other method can provide more depth of understanding than directly
observing people … at the scene”. Future study can explore the actual avoidance behavior at
the time consumption in marketplace.
As the study of brand avoidance is interested in the deliberate rejection of brands when
affordability is not the issue, therefore, the data based on the people who had the financial
power to make choices, particularly Pakistan sample shows consumers from middle to upper
class. However, low income consumers are able to make choices among brands in most of
developed countries, so it is observed that they have the capacity to exercise brand avoidance.
In future research, it would be interesting to compare the findings of consumers from lower
socio-economic class of developed nations with the findings of current study.
203
General Discussions
Chapter 7
The focus of the current study is to explore the reasons affecting brand avoidance. Future
studies could examine which avoidance remedies are easy to establish and maintain.
Past studies on negative and positive WOM have confirmed that negative WOM may have
stronger effect than positive WOM. Hence, future study could investigate that which of the
two is more influential in consumer decision making.
Finally, in our study we interviewed the end-user consumers, so the consumer perspective
about brands is the main focus of this study. Thus, future study may investigate orientations
from other than the end-user consumers, such as managers in business-to-business context,
may also provide considerable managerial implications.
7.7
Conclusion
It is evident from literature that conventional brand management inclined towards positive
side and focuses on brand loyalty, consumer preferences and brand selection. Whereas, the
other source, unhappy and dissatisfied consumers, has not been focused that can provide a
rich source of learning for both academics and practitioners. In this thesis, we tried to address
the gap in research by identifying the possible reasons that why consumers avoid certain
brands, particularly, even if they do not have direct experience with them.
Exploring themes through 13 in-depth interviews and empirically testing the proposed
research model in two different countries (Pakistan= 278 and New Zealand= 198) with latest
statistical techniques shows the motivation of the researcher to understand the phenomenon of
brand avoidance.
Literature provides evidence that self congruence (ideal or actual) is a strong predictor of
brand selection; we found that consumer’s undesired self congruence is the key determinant
of pre-purchase brand avoidance attitudes and behaviors. Past studies show that country of
204
General Discussions
Chapter 7
origin is a good signal to use in promotion of the brands if the country image is perceived in
terms of high quality/values. Our study proposes that for global brands, it is essential to
understand the relationship between the home country of the brand and the country of target
consumers. The results may not be favorable if the country is perceived in terms of low
quality/values or the consumer perceives animosity towards the country of origin.
We found that although the consumer is more sophisticated today, well informed about
consumer rights and product requirements, have more information about companies and
brands, and have more choices available in market place, even then it is not sure that
consumer will prefer ethical company over unethical ones.
In terms of marketing implications, theoretical knowledge about the factors that motivate
consumers for brand avoidance cautions marketing managers to think for possible strategies
to ensure that their brands remain healthy.
In the end, by increasing knowledge in the domain of anti-consumption, this study offers an
overall contribution to consumer behaviors research. In conclusion, this thesis provides a
valuable answer to the vital question: why do people avoid brands even they have no direct
experience with them?
“Life is not always what one wants it to be, but to make the best
of it as it is, is the only way of being happy.”
--- Jennie Jerome Churchill
205
References
8 References
Aaker, D. (1996). Building Strong Brands. New York: Free Press.
Aaker, J. (1999). The Malleable Self: The Role of Self-Expression in Persuasion. Journal of
Marketing Research, 36(1), 45–57.
Abougomaah, N. H., Schlacter, J. L., & Gaidis, W. (1987). Elimination and choice phases in
evoked set formation. Journal of Consumer Marketing, 4(4), 67–72.
Abrams, D., & Hogg, M. A. (1988). Comments on the motivational status of self-esteem in
social identity and intergroup discrimination. European Journal of Social Psychology,
18(4), 317–334.
Ahmed, Z. U., Johnson, J. P., Yang, X., Fatt, C. K., Teng, H. S., & Boon, L. C. (2004). Does
country of origin matter for low-involvement products? International Marketing
Review, 21(1), 102–120.
Aikin, L. S., & West, S. G. (1991). Multiple Regression: Testing and Interpreting
interactions. Thousand Oaks, CA: Sage.
Ajzen, I. (1991). The theory of planned behaviour. Organizational behavior and Human
Decision Process, 50, 179–211.
Ajzen, I., & Fishbein, M. (1969). The prediction of behavioral intentions in a choice situation.
Journal of Experimental Social Psychology, 5(4), 400–416.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior.
Prentice-Hall.
Alba, J. W., & Hutchinson, J. W. (1987). Dimensions of Consumer Expertise. The Journal of
Consumer Research, 13(4), 411–454.
References
Albert, N., Merunka, D., & Valette-Florence, P. (2008). When consumers love their brands:
Exploring the concept and its dimensions. Journal of Business Research, 61(10),
1062–1075.
Alden, Steenkamp, & Batra. (2006). Consumer attitudes toward marketplace globalization:
Structure, antecedents and consequences. International Journal of Research in
Marketing, 23(3), 227–239.
AMA, American Marketing Association. (2011). Dictionary of Marketing Terms, vol. 2011.
Ambler, T., & Styles, C. (1996). Brand development versus new product development:
towards a process model of extension decisions. Marketing Intelligence & Planning,
14(7), 10–19.
Anderson, J., & Gerbing, D. (1988). Structural Equation Modeling in Practice: A Review and
Recommended Two-Step Approach. Psychological Bulletin, 103(3), 411–423.
Ang, S. H., Cheng, P. S., Lim, E. A. C., & Tambyah, S. K. (2001). Spot the difference:
consumer responses towards counterfeits. Journal of Consumer Marketing, 18(3),
219–235.
Bagozzi, R. P. (1992). The Self-Regulation of Attitudes, Intentions, and Behavior. Social
Psychology Quarterly, 55(2), 178–204.
Bagozzi, R. P., & Baumgartner, H. (1994). The Evaluation of Structural Equation Models and
Hypothesis Testing (Vol. 1, pp. 386–422).
Bagozzi, R. P., Gopinath, M., & Nyer, P. U. (1999). The Role of Emotions in Marketing.
Journal of the Academy of Marketing Science, 27(2), 184–206.
Bagozzi, R. P., & Yi, Y. (1989). The Degree of Intention Formation as a Moderator of the
Attitude-Behavior Relationship. Social Psychology Quarterly, 52(4), 266.
Bagozzi, R., Yi, Y., & Phillips, L. (1991). Assessing Construct Validity in Organizational
Research. Administrative Science Quarterly, 36(3).
207
References
Bahaee, M., & Pisani, M. J. (2009). The use of the consumer ethnocentrism and consumer
animosity scales in Iran: A research note. Thunderbird International Business Review,
51(2), 143–150.
Baker, M. J., & Churchill, G. A. (1977). The Impact of Physically Attractive Models on
Advertising Evaluations. Journal of Marketing Research, 14(4), 538–555.
Balabanis, G., Mueller, R., & Melewar, T. C. (2002). The Relationship Between Consumer
Ethnocentrism and Human Values. Journal of Global Marketing, 15(3), 7.
Banister, E. N., & Hogg, M. K. (2001). Dislikes, distastes and the undesired self:
conceptualising and exploring the role of the undesired end state in consumer
experience.
Banister, E. N., & Hogg, M. K. (2004). Negative symbolic consumption and consumers’ drive
for self-esteem: The case of the fashion industry. European Journal of Marketing,
38(7), 850–868.
Barnes, J. G., & McTavish, R. (1983). Segmenting Industrial Markets by Buyer
Sophistication. European Journal of Marketing, 17(6), 16–33.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51(6), 1173–1182.
Batra, R., Boush, D. M., Chung-Hyun, K., & Kahle, L. R. (2000). Cynicism and Conformity
as Correlates of Trust in Product Information Sources. Journal of Current Issues &
Research in Advertising, 15(2), 71–79.
Bauer, R. A. (1960). “Consumer Behavior as Risk Taking.” Dynamic Marketing for a
Changing World, Chicago. American Marketing Association, 398–98.
Bearden, W. O., & Teel, J. E. (1983). Selected Determinants of Consumer Satisfaction and
Complaint Reports. Journal of Marketing Research, 20(1), 21–28.
208
References
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin,
107, 238–246.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis
of covariance structures. Psychological Bulletin, 88(3), 588–606.
Berkowitz, L., & Lutterman, K. G. (1968). The traditional socially responsible personality.
Public Opinion Quarterly, 32(2), 169 –185.
Bettman, J. R. (1973). Perceived Risk and Its Components: A Model and Empirical Test.
Journal of Marketing Research, 10(2), 184–190.
Bhattacharya, C. B., & Elsbach, K. D. (2002). Us versus Them: The Roles of Organizational
Identification and Disidentification in Social Marketing Initiatives. Journal of Public
Policy & Marketing, 21(1), 26–36.
Bloch, P. H. (1995). Seeking the Ideal Form: Product Design and Consumer Response. The
Journal of Marketing, 59(3), 16–29.
Blodgett, J. G., Granbois, D. H., & Walters, R. G. (1993). The effects of perceived justice on
complainants’ negative word-of-mouth behavior and repatronage intentions. Journal
of Retailing, 69(4), 399–428.
Bodur, M., & Day, R. L. (1978). consumer response to dissatisfaction with services and
intangibles. Advances in Consumer Research, 5(1), 263–272.
Borgmann, A. (2000). Reflections and reviews: the moral complexion of consumption.
Journal of Consumer Research, 26, 418–422.
Bosnjak, M., & Rudolph, N. (2008). Undesired self-image congruence in a low-involvement
product context. European Journal of Marketing, 42(5/6), 702–712.
Bougie, R., Pieters, R., & Zeelenberg, M. (2003). Angry Customers don’t Come Back, They
Get Back: The Experience and Behavioral Implications of Anger and Dissatisfaction
in Services. Journal of the Academy of Marketing Science, 31(4), 377 –393.
209
References
Boulstridge, E., & Carrigan, M. (2000). Do consumers really care about corporate
responsibility? Highlighting the attitude—behaviour gap. Journal of Communication
Management, 4(4), 355–368.
Bourdieu, P. (1984). Distinction. Routledge, (London), Kegan and Paul.
Bowen, H. R. (1953). Social Responsibility of the Businessman. Harper & Row, New York.
Brown, T. J., Barry, T. E., Dacin, P. A., & Gunst, R. F. (2005). Spreading the Word:
Investigating Antecedents of Consumers’ Positive Word-of-Mouth Intentions and
Behaviors in a Retailing Context. Journal of the Academy of Marketing Science,
33(2), 123 –138.
Brown, T. J., & Dacin, P. A. (1997). The Company and the Product: Corporate Associations
and Consumer Product Responses. The Journal of Marketing, 61(1), 68–84.
Campbell, B. M. (1969). The Existence and Determinants of Evoked Set in Brand Choice
Behavior. Ph.D. Diss., Columbia University.
Carrigan, M., & Attalla, A. (2001). The myth of the ethical consumer – do ethics matter in
purchase behaviour? Journal of Consumer Marketing, 18(7), 560–578.
Carroll, A. B. (1998). The Four Faces of Corporate Citizenship. Business and Society Review,
100(1), 1–7.
Carroll, A. B. (1999). Corporate Social Responsibility. Business & Society, 38(3), 268 –295.
Carroll, B. A., & Ahuvia, A. C. (2006). Some antecedents and outcomes of brand love.
Marketing Letters, 17(2), 79–89.
Carty, V. (2002). Technology and Counter-hegemonic Movements: The case of Nike
Corporation. Social Movement Studies: Journal of Social, Cultural and Political
Protest, 1(2), 129.
210
References
Carver, C. S., Lawrence, J. W., & Scheier, M. F. (1999). Self-Discrepancies and Affect:
Incorporating the Role of Feared Selves. Personality and Social Psychology Bulletin,
25(7), 783–792.
Charles, W. R., Jr. (2012). Consumer Demographics as Antecedents in the Animosity Model
of Foreign Product Purchase. International Journal of Business and Social science,
3(4), 13–21.
Chaudhuri, A. (1997). Consumption Emotion and Perceived Risk: A Macro-Analytic
Approach. Journal of Business Research, 39(2), 81–92.
Cheng, T. C. E., Lam, D. Y. C., & Yeung, A. C. L. (2006). Adoption of internet banking: an
empirical study in Hong Kong. Decision Support Systems, 42(3), 1558–1572.
de Chernatony, L., & Dall’Olmo Riley, F. (1998). Defining A “Brand”: Beyond The
Literature With Experts’ Interpretations. Journal of Marketing Management, 14(5),
417–443.
Cherrier, H. (2009a). Anti-consumption discourses and consumer-resistant identities. Journal
of Business Research, 62(2), 181–190.
Cherrier, H. (2009b). Anti-consumption discourses and consumer-resistant identities. Journal
of Business Research, 62(2), 181–190.
Churchill, G. (1979). A paradigm for developing better measures of marketing constructs.
Churchill, G. A. (1999). Marketing Research: Methodological Foundation (7th Ed.). London:
The Dryden Press.
Clarke, I., Shankarmahesh, M. N., & Ford, J. B. (2000). Consumer ethnocentrism,
materialism and values: a four country study. AMA Winter Educators’ Conference
Proceedings, San Antonio, TX.
211
References
Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco: W.H. Freeman &
Company.
Cox, D. F., & Bauer, R. A. (1964). Self-Confidence and Persuasibility in Women. The Public
Opinion Quarterly, 28(3), 453–466.
Craig-Lees, M., & Hill, C. (2002). Understanding voluntary simplifiers. Psychology and
Marketing, 19(2), 187–210.
Cunningham, S. M. (1967). “The major dimensions of perceived risk”, in Cox, D.F. (Ed.),
Risk Taking and Information Handling in Consumer Behavior. Boston University
Press, Boston, MA.
Dalli, D., Romani, S., & Gistri, G. (2006). Brand dislike: representing the negative side of
consumer preferences. Article, .
Davis, D. L., Guiltinan, J. P., & Wesley, H. J. (1979). Service Characteristics, Consumer
Search, and the Classification of Consumer Services. Journal of Retailing, 15(3).
Davis, K. (1973). The Case for and against Business Assumption of Social Responsibilities.
The Academy of Management Journal, 16(2), 312–322.
Day, D. S. (1978). Are consumer Satisfied? In D. A. Aaker & G.S. Day (Eds.), Consumerism:
Search for the consumer interest, 3rd Edition.
Debbabi, S., Daassi, M., & Baile, S. (2010). Effect of online 3D advertising on consumer
responses: the mediating role of telepresence. Journal of Marketing Management,
26(9-10), 967–992.
Dill, W. R. (1958). Environment as an Influence on Managerial Autonomy. Administrative
Science Quarterly, 2(4), 409–443.
Dobscha, S., & College, B. (1998). The lived experience of consumer rebellion against
marketing. Advances in Consumer Research, 25, 91–97.
212
References
Dodds, W. B., Monroe, K. B., & Grewal, D. (1991). Effects of Price, Brand, and Store
Information on Buyers’ Product Evaluations. Journal of Marketing Research, 28(3),
307–319.
Douglas, S. P., & Craig, C. S. (1997). The changing dynamic of consumer behavior:
implications for cross-cultural research. International Journal of Research in
Marketing, 14(4), 379–395.
Dowling, G. R., & Staelin, R. (1994). A Model of Perceived Risk and Intended RiskHandling Activity. The Journal of Consumer Research, 21(1), 119–134.
Duhacheck, A., Zhang, S., & Krishnan, S. (2007). Anticipated group interaction : Coping with
valence asymmetries in attitude shift. Journal of consumer research, 34(3), 395–405.
Durvasula, S., Andrews, J. C., & Netemeyer, R. G. (1997). A Cross-Cultural Comparison of
Consumer Ethnocentrism in the United States and Russia. Journal of International
Consumer Marketing, 9(4), 73–93.
Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational Images and Member
Identification. Administrative Science Quarterly, 39(2), 239–263.
East, R., Hammond, K., & Lomax, W. (2008). Measuring the impact of positive and negative
word of mouth on brand purchase probability. International Journal of Research in
Marketing, 25(3), 215–224.
El-Adly, M. I. (2010). The Impact of Advertising Attitudes on the Intensity of TV Ads
Avoiding Behavior. International Journal of Business and Social science, 1(1), 1–14.
Elliot, A. J. (1999). Approach and avoidance motivation and achievement goals. Educational
Psychologist, 34(3), 169.
Elliott, G. R., & Cameron, R. C. (1994). Consumer Perception of Product Quality and the
Country-of-Origin Effect. Journal of International Marketing, 2(2), 49–62.
213
References
Elsbach, K. D., & Bhattacharya, C. B. (2001). Defining Who You Are by What You’re Not:
Organizational Disidentification and the National Rifle Association. Organization
Science, 12(4), 393–413.
Engel, J. F., Kegerreis, R. J., & Blackwell, R. D. (1969). Word-of-Mouth Communication by
the Innovator. The Journal of Marketing, 33(3), 15–19.
Englis, B. G., & Solomon, M. R. (1995). To Be and Not to Be: Lifestyle Imagery, Reference
Groups, and “The Clustering of America.” Journal of Advertising, 24(1), 13–28.
Englis, B. G., & Solomon, M. R. (1997). I am not therefore, i am: the role of avoidance
products in shaping consumer behavior. Advances in Consumer Research, 24, 61–63.
Escalas, J. E., & Bettman, J. R. (2005). Self-Construal, Reference Groups, and Brand
Meaning. Journal of Consumer Research: An Interdisciplinary Quarterly, Journal of
Consumer Research: An Interdisciplinary Quarterly, 32(3), 378–389.
Esrock, S., & Leichty, G. (1998). Social Responsibility and Corporate Web Pages: SelfPresentation or Agenda-Setting? Public Relations Review, 24(3), 305–319.
Evans, R. H. (1982). Measuring Perceived Risk: A Replication and an Application of Equity
Theory. Advances in Consumer Research, 9, 550–555.
Evrard, Y., Roux, E., & Bernard, P. (2003). Market: étude et recherches en marketing (3rd
ed.). Paris: Dunod.
Fan, Y. (2007). Ethical branding and corporate reputation. Corporate Communications: An
International Journal, 10(4), 341–350.
Featherman, M. S., & Pavlou, P. A. (2003). Predicting e-services adoption: a perceived risk
facets perspective. International Journal of Human-Computer Studies, 59(4), 451–
474.
Firat, A. F., & Venkatesh, A. (1995). Liberatory Postmodernism and the Reenchantment of
Consumption. The Journal of Consumer Research, 22(3), 239–267.
214
References
Fishbein, M., & Ajzen, I. (1972). Attitudes and Opinions. Annual Review of Psychology,
23(1), 487–544.
Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An Introduction to
Theory and Research. Addison-Wesley Pub (Sd).
Fiske, S. T. (1980). Attention and weight in person perception: The impact of negative and
extreme behavior. Journal of Personality and Social Psychology, 38(6), 889–906.
Folkes, V. S. (1988). The Availability Heuristic and Perceived Risk. The Journal of
Fornell, C., & Larcker, D. F. (1981). Evaluating Structural Equation Models with
Unobservable Variables and Measurement Error. Journal of Marketing Research,
18(1), 39–50.
Fournier, S. (1998). Consumers and Their Brands: Developing Relationship Theory in
Consumer Research. Journal of Consumer Research, 24(4), 343–353.
Fournier, S., Dobscha, S., & Mick, D. G. (1998). Preventing the premature death of
relationship marketing. Harvard Business Review, 76(1), 42–51.
Fournier, S., & Mick, D. G. (1999). Rediscovering Satisfaction. The Journal of Marketing,
63(4), 5–23.
Fraenkel, J. R., & Wallen, N. E. (2006). How to Design and Evaluate Research in Education
(6th ed.). McGraw-Hill, New York.
Freeman, R. E. (1994). The Politics of Stakeholder Theory: Some Future Directions. Business
Ethics Quarterly, 4(4), 409–421.
Friedman, M. (1985). Consumer Boycotts in the United States, 1970-1980: Contemporary
Events in Historical Perspective. Journal of Consumer Affairs, 19(1), 96–117.
Frooman, J. (1997). Socially Irresponsible and Illegal Behavior and Shareholder Wealth.
Business & Society, 36(3), 221 –249.
215
References
Garretson, J. A., & Clow, K. E. (1999). The influence of coupon face value on service quality
expectations, risk perceptions and purchase intentions in the dental industry. Journal
of Services Marketing, 13(1), 59–72.
Gentry, J. W., Putrevu, S., & Shultz II, C. J. (2006). The effects of counterfeiting on
consumer search. Journal of Consumer Behaviour, 5(3), 245–256.
Granbois, D. H., Walters, R. G., & Blodgett, J. G. (1993). The effects of perceived justice on
complainants’ negative word-of-mouth behavior and repatronage intentions. Journal
of Retailing, 69(4), 399–428.
Grewal, R., Cline, T. W., & Davies, A. (2003). Early-Entrant Advantage, Word-of-Mouth
Communication, Brand Similarity, and the Consumer Decision-Making Process.
Journal of Consumer Psychology, 13(3), 187–197.
Griffith, D. A., & Chen, Q. (2004). The Influence of Virtual Direct Experience (VDE) on OnLine Ad Message Effectiveness. Journal of Advertising, 33(1), 55–68.
Grover, R., & Vriens, M. (2006). The Handbook of Marketing Research. SAGE Publications,
Inc.
Grubb, E. L., & Harrison L. Grathwohl. (1967). Consumer Self-Concept, Symbolism and
Market Behavior: A Theoretical Approach. The Journal of Marketing, 31(4), 22–27.
Gupta, B., Dasgupta, S., & Gupta, A. (2008). Adoption of ICT in a government organization
in a developing country: An empirical study. The Journal of Strategic Information
Systems, 17(2), 140–154.
Hair, J. F. J., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate Data
Analysis (5th Ed.). New Jersey: Prentice-Hall.
Hair, J. F. J., Anderson, R. E., Tatham, R. L., & Black, W. C. (2009). Multivariate Data
Analysis (7th Ed.). New Jersey: Prentice-Hall.
216
References
Halstead, D. (1989). Expectations and disconfirmation beliefs as predictors of consumer
satisfaction, repurchase intention, and complaining behavior: An empirical study.
Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 2, 17–
21.
Hamilton, D. L., & Huffman, L. J. (1971). Generality of impression-formation processes for
evaluative and nonevaluative judgments. Journal of Personality and Social
Psychology, 20(2), 200–207.
Hamzaoui, L., & Merunka, D. (2006). The impact of country of design and country of
manufacture on consumer perceptions of bi-national products’ quality: an empirical
model based on the concept of fit. Journal of Consumer Marketing, 23(3), 145–155.
Han, C. M. (1988). The Role of Consumer Patriotism in the Choice of Domestic versus
Foreign Products. Journal of Advertising Research, 25–32.
Han, C. M. (1990). Testing the Role of Country Image in Consumer Choice Behaviour.
European Journal of Marketing, 24(6), 24–40.
Harrison-Walker, L. J. (2001). The Measurement of Word-of-Mouth Communication and an
Investigation of Service Quality and Customer Commitment As Potential Antecedents.
Journal of Service Research, 4(1), 60 –75.
Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions in OLS
and logistic regression: SPSS and SAS implementations. Behavior Research Methods,
41(3), 924–936.
He, Y., Merz, M. A., & Alden, D. L. (2008). Diffusion of Measurement Invariance
Assessment in Cross-National Empirical Marketing Research: Perspectives from the
Literature and a Survey of Researchers. Journal of International Marketing, 16(2),
64–83.
217
References
Heeter, C., & Greenberg, B. S. (1985). Profiling the zappers. Journal of Advertising Research,
25(2), 15–19.
Herr, P. M., Kardes, F. R., & Kim, J. (1991). Effects of Word-of-Mouth and ProductAttribute Information on Persuasion: An Accessibility-Diagnosticity Perspective. The
Journal of Consumer Research, 17(4), 454–462.
Hirschman, A. (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms,
Organizations, and States. Harvard University Press.
Hirschman, E. C. (1980). Innovativeness, Novelty Seeking, and Consumer Creativity. Journal
of Consumer Research, 7(3), 283–295.
Hirschman, E. C. (1992). The Consciousness of Addiction: Toward a General Theory of
Compulsive Consumption. Journal of Consumer Research, Journal of Consumer
Research, 19(2), 155–79.
Hofstede, G. (1980). Cultural’s Consequences: International Differences in work-related
Issues. Sage, Beverly Hills, CA.
Hofstede, G. (2011). Geert Hofstede Cultural Dimensions. Retrieved July 19, 2011, from:
http://www.geerthofstede.com/hofstede_dimensions.php?culture1=66&culture2=63#compare.
Hogg, M. (1998). Anti-Constellations: Exploring the Impact of Negation on Consumption.
Journal of Marketing Management, 14(1), 133.
Hogg, M. K., Banister, E. N., & Stephenson, C. A. (2009). Mapping symbolic (anti-)
consumption. Journal of Business Research, 62(2), 148–159.
Hogg, M. K., Cox, A. J., & Keeling, K. (2000). The impact of self-monitoring on image
congruence and product/brand evaluation. European Journal of Marketing, 34(5/6),
641–667.
218
References
Hogg, M., & Michell, P. (1996). Identity, self and consumption: a conceptual framework.
Journal of Marketing Management, 12, 629–644.
Holt, D. B., Quelch, J. A., & Taylor, E. L. (2004). How global brands compete. Harvard
Business Review, 82(9), 68–75, 136.
Honkanen, Verplanken, B., & Olsen, S. O. (2010). Ethical values and motives driving organic
food choice. Journal of Consumer Behaviour, 430(5), 420–430.
Hooley, G. J., Greenley, G. E., Cadogan, J. W., & Fahy, J. (2005). The performance impact of
marketing resources. Journal of Business Research, 58(1), 18–27.
Horn, J. L. (1991). Comments on “Issues in Factorial Invariance” in Best methods for the
Analysis of Change,. Ed. Linda M. Collins and John L. Horn, Washington, DC:
American Psychological Association, 114–125.
Howard, J. A., & Sheth, J. N. (1969). The theory of buyer behavior. Wiley.
Hui, C. H., & Triandis, H. C. (1985). Measurement in Cross-Cultural Psychology A Review
and Comparison of Strategies. Journal of Cross-Cultural Psychology, 16(2), 131–152.
Iyer, R., & Muncy, J. A. (2009). Purpose and object of anti-consumption. Journal of Business
Research, 62(2), 160–168.
Jacoby, J., Chestnut, R. W., & Fisher, W. A. (1978). A Behavioral Process Approach to
Information Acquisition in Nondurable Purchasing. Journal of Marketing Research,
15(4), 532–544.
Jacoby, J., & Kaplan, L. B. (1972). “The Components of Perceived Risk”, in Proceedings of
the Third Annual Conference of the Association for Consumer Research. Association
for Consumer Research, 382–393.
Jacoby, J., & Kyner, D. B. (1973). Brand Loyalty vs. Repeat Purchasing Behavior. Journal of
Marketing Research, 10(1), 1–9.
219
References
James, L. R., & Brett, J. M. (1984). Mediators, Moderators, and tests for Mediation. Journal
of Applied Psychology, 69(2), 307–321.
Janda, S., & Rao, C. P. (1997). The effect of country-of-origin related stereotypes and
personal beliefs on product evaluation. Psychology and Marketing, 14(7), 689–702.
Jiménez, N. H., & Martín, S. san. (2010). The role of country-of-origin, ethnocentrism and
animosity in promoting consumer trust. The moderating role of familiarity.
International Business Review, 19(1), 34–45.
Johnson, H. H. (2003). Does it pay to be good? Social responsibility and financial
performance. Business Horizons, 46(6), 34–40.
Jones, D. (1997). Good works, good business. USA Today, (April 25), 1B.
Jöreskog, K. (1971). Simultaneous factor analysis in several populations. Psychometrika,
36(4), 409–426.
Kanar, A., Collins, C., & Bell, B. (2010). A Comparison of the Effects of Positive and
Negative Information on Job Seekers’ Organizational Attraction and Attribute Recall.
Articles and Chapters.
Kaplan, L. B., Szybillo, G. J., & Jacoby, J. (1974). Components of perceived risk in product
purchase: A cross-validation. Journal of Applied Psychology, 59(3), 287–291.
Keller, K. L. (1993). Conceptualizing, Measuring, and Managing Customer-Based Brand
Equity. The Journal of Marketing, 57(1), 1–22.
Keller, K. L., & Lehmann, D. R. (2006). Brands and Branding: Research Findings and Future
Priorities. MARKETING SCIENCE, 25(6), 740–759.
Kelman, H. C. (1958). Compliance, Identification, and Internalization: Three Processes of
Attitude Change. The Journal of Conflict Resolution, 2(1), 51–60.
220
References
Kent, R. J., & Allen, C. T. (1993). Does Competitive Clutter in Television Advertising
“Interfere” with the Recall and Recognition of Brand Names and Ad Claims?
Marketing Letters, 4(2), 175–184.
Kinnear, T. C., & Taylor, J. R. (1996). Marketing Research An Applied Approach (5th Ed.).
New York: McGraw Hill.
Klein, J., & Dawar, N. (2004). Corporate social responsibility and consumers’ attributions and
brand evaluations in a product-harm crisis. International Journal of Research in
Marketing, 21(3), 203–217.
Klein, J., & Ettenson, R. (1999). Consumer animosity and consumer ethnocentrism: An
analysis of unique antecedents. Journal of International Consumer Marketing, 11, 5–
24.
Klein, J. G. (1996). Negativity in Impressions of Presidential Candidates Revisited: The 1992
Election. Personality and Social Psychology Bulletin, 22(3), 288 –295.
Klein, J. G. (2002). Us Versus Them, or Us Versus Everyone? Delineating Consumer
Aversion to Foreign Goods. Journal of International Business Studies, 33(2), 345–
363.
Klein, J. G., Ettenson, R., & Morris, M. D. (1998). The Animosity Model of Foreign Product
Purchase: An Empirical Test in the People’s Republic of China. The Journal of
Marketing, 62(1), 89–100.
Klein, J. G., Smith, N. C., & John, A. (2004). Why We Boycott: Consumer Motivations for
Boycott Participation. Journal of Marketing, 68(3), 92–109.
Klein, N. (2000). No Logo: Taking Aim at Brand Bullies. London: Flamingo.
Kline, R. B. (2010). Principles and Practice of Structural Equation Modeling, Third Edition
(Third ed.). The Guilford Press.
221
References
Korn, E. L., & Graubard, B. I. (1995). Examples of Differing Weighted and Unweighted
Estimates from a Sample Survey. American Statistician, 49(3), 291.
Kotler, P. (1991). Marketing Management: Analysis, Planning, Implementation, and Control
(8th ed.). Prentice Hall.
Kotler, P., & Armstrong, G. (2007). Principles of Marketing, 12th Edition (12th ed.). Prentice
Hall.
Kotler, P., Keller, K., Brady, M., Goodman, M., & Hansen, T. (2009). Marketing
management. Pearson Education.
Kozinets, R. V. (2002). Can Consumers Escape the Market? Emancipatory Illuminations from
Burning Man. The Journal of Consumer Research, 29(1), 20–38.
Kozinets, R. V., & Handelman, J. (1998). Ensouling consumption: A netnographic
exploration of the meaning of boycotting behavior. Advances in Consumer Research
Volume, 25, 475–480.
Kozinets, R. V., & Handelman, J. M. (2004). Adversaries Of Consumption: Consumer
Movements, Activism, And Ideology. Journal of Consumer Research, 31(3), 691–
704.
Kulviwat, S., Bruner II, G. C., & Al-Shuridah, O. (2009). The role of social influence on
adoption of high tech innovations: The moderating effect of public/private
consumption. Journal of Business Research, 62(7), 706–712.
Kwong, K. ., Yau, O. H. ., Lee, J. S. ., Sin, L. Y. ., & Tse, A. C. . (2003). The Effects of
Attitudinal and Demographic Factors on Intention to Buy Pirated CDs: The Case of
Chinese Consumers. Journal of Business Ethics, 47(3), 223–235.
Laczniak, R. N., DeCarlo, T. E., & Ramaswami, S. N. (2001). Consumers’ Responses to
Negative Word-of-Mouth Communication: An Attribution Theory Perspective.
Journal of Consumer Psychology, 11(1), 57–73.
222
References
Lai, V. S., Chau, P. Y. K., & Cui, X. (2010). Examining internet banking acceptance: a
comparison of alternative technology adoption models. International Journal of
Electronic Business, 8(1), 51 – 79.
Langtry, B. (1994). Stakeholders and the Moral Responsibilities of Business. Business Ethics
Quarterly, 4(4), 431–443.
Laroche, M., Papadopoulos, N., Heslop, L. A., & Mourali, M. (2005). The influence of
country image structure on consumer evaluations of foreign products. International
Marketing Review, 22(1), 96–115.
Lastovica, J. (2006). Frugal Materialists. Paper presented at the anti-consumption seminar,
International Centre for Anti-Consumption Research, 20–21.
Lau, G. T., & Ng, S. (2009). Individual and Situational Factors Influencing Negative Wordof-Mouth Behaviour. Canadian Journal of Administrative Sciences / Revue
Canadienne des Sciences de l’Administration, 18(3), 163–178.
Lee, M. S. W. (2007). Brands we love to hate: An exploration of brand avoidance. Thesis, .
Lee, M. S. W., Fernandez, K. V., & Hyman, M. R. (2009). Anti-consumption: An overview
and research agenda. Journal of Business Research, 62(2), 145–147.
Lee, M. S. W., Motion, J., & Conroy, D. (2009). Anti-consumption and brand avoidance.
Journal of Business Research, Journal of Business Research, 62(2), 169–180.
Lee, M.-C. (2009). Factors influencing the adoption of internet banking: An integration of
TAM and TPB with perceived risk and perceived benefit. Electronic Commerce
Research and Applications, 8(3), 130–141.
Leonard-Barton, D. (1985). Experts as Negative Opinion Leaders in the Diffusion of a
Technological Innovation. The Journal of Consumer Research, 11(4), 914–926.
Levy, S. (1959). Symbols for sales. Harvard Business Review, 37(4), 117–24.
223
References
Levy, S. J. (1999). Brands, Consumers, Symbols, and Research: Sydney J. Levy on Marketing.
Sage Publications, Thousand Oaks, CA.
Lewis, I. M. (1985). Social Anthropology in Perspective: The Relevance of Social
Anthropology (2nd Revised ed.). Cambridge University Press.
Lin, W.-B. (2008). Investigation on the model of consumers’ perceived risk--integrated
viewpoint. Expert Systems with Applications, 34(2), 977–988.
Little, T. D., Bovaird, J. A., & Widaman, K. F. (2006). On the Merits of Orthogonalizing
Powered and Product Terms: Implications for Modeling Interactions among Latent
Variables. Structural Equation Modeling: A Multidisciplinary Journal, 13(4), 497–
519.
Luce, R. A., Barber, A. E., & Hillman, A. J. (2001). Good Deeds and Misdeeds: A Mediated
Model of the Effect of Corporate Social Performance on Organizational
Attractiveness. Business & Society, 40(4), 397 –415.
Lusch, R. F., & Vargo, S. L. (2004). Evolving to a New Dominant Logic for Marketing.
Journal of Marketing, 68(1), 1–17.
Lutz, R. J. (1975). Changing Brand Attitudes Through Modification of Cognitive Structure.
The Journal of Consumer Research, 1(4), 49–59.
Maignan, I., & Ferrell, O. C. (2001). Antecedents and benefits of corporate citizenship: an
investigation of French businesses. Journal of Business Research, 51(1), 37–51.
Maignan, I., Ferrell, O. C., & Hult, G. T. M. (1999). Corporate Citizenship: Cultural
Antecedents and Business Benefits. Journal of the Academy of Marketing Science,
27(4), 455–469.
Malhotra, N. K. (1999). Marketing Research: An Applied Orientation (3rd Ed.). London:
Prentice-Hall International, Inc.
224
References
Matthiesen, I., & Phau, I. (2005). The “HUGO BOSS” connection: Achieving global brand
consistency across countries. Journal of Brand Management, 12(5), 325–338.
McWilliams, A., & Siegel, D. (2001). Corporate Social Responsibility: A Theory of the Firm
Perspective. The Academy of Management Review, 26(1), 117–127.
Micheletti, M. (2003). Political Virtue and Shoping. New York: Palgrave Macmillan.
Micheletti, M., Follesdal, A., & Stolle, D. (2003). Politics, Products, and Markets: Exploring
Political Consumerism Past and Present. New Brunwinsk, NJ: Transaction Press.
Mitra, K., Reiss, M. C., & Capella, L. M. (1999). An examination of perceived risk,
information search and behavioral intentions in search, experience and credence
services. Journal of Services Marketing, 13(3), 208–228.
Mizerski, R. W. (1982). An Attribution Explanation of the Disproportionate Influence of
Unfavorable Information. The Journal of Consumer Research, 9(3), 301–310.
Moorman, C., Diehl, K., Brinberg, D., & Kidwell, B. (2004). Subjective Knowledge, Search
Locations, and Consumer Choice. The Journal of Consumer Research, 31(3), 673–
680.
Morgan, R. M., & Shelby D. Hunt. (1994). The Commitment-Trust Theory of Relationship
Marketing. The Journal of Marketing, 58(3), 20–38.
Mullen, M. R. (1995). Diagnosing Measurement Equivalence in Cross-National Research.
Journal of International Business Studies, 26(3), 573–596.
Muniz, A. M., & Hamer, L. O. (2001). Us versus Them: Oppositional Brand Loyalty and the
Cola Wars. Advances in Consumer Research, 28, 355–361.
Murray, K. B., & Vogel, C. M. (1997). Using a hierarchy-of-effects approach to gauge the
effectiveness of corporate social responsibility to generate goodwill toward the firm:
Financial versus nonfinancial impacts. Journal of Business Research, 38(2), 141–159.
225
References
Myers, M. B., Roger, J. C., Thomas, J. P. J., & Charles, R. T. (2000). An Application of
Multi-Group Causal Models in Assessing Cross-Cultural Measurement Equivalence.
Journal of International Marketing, 8(4), 108–121.
Nagashima, A. (1977). A Comparative “Made in” Product Image Survey among Japanese
Businessmen. Journal of Marketing, 41, 95.
Narayana, C. L., & Markin, R. J. (1975). Consumer Behavior and Product Performance: An
Alternative Conceptualization. The Journal of Marketing, 39(4), 1–6.
Narver, J. C., & Slater, S. (1990). The Effect of a Market Orientation on Business
Profitability. Journal of Marketing, october(1), 20–35.
Netemeyer, R., Durvasula, S., & Lichtenstein, L. R. (1991). A cross-national assessment of
the reliability and validity of the CETSCALE. Journal of Marketing Research, 28,
320–7.
Nijssen, E. J., & Douglas, S. P. (2004). Examining the animosity model in a country with a
high level of foreign trade. International Journal of Research in Marketing, 21(1), 23–
38.
Nunnally, J. C. (1978). Psychometric Theory (2nd ed.). McGraw-Hill, New York.
Nyer, P. U., & Gopinath, M. (2005). Effects of complaining versus negative word of mouth
on subsequent changes in satisfaction: The role of public commitment. Psychology
and Marketing, 22(12), 937–953.
Ogilvie, D. M. (1987). The Undesired Self: A Neglected Variable in Personality Research.
Journal of Personality and Social Psychology, 52(2), 379–385.
Oliva, T. A., Oliver, R. L., & MacMillan, I. C. (1992). A Catastrophe Model for Developing
Service Satisfaction Strategies. The Journal of Marketing, 56(3), 83–95.
Oliver, R. L. (1980). A Cognitive Model of the Antecedents and Consequences of Satisfaction
Decisions. Journal of Marketing Research, 17(4), 460–469.
226
References
Oliver, R. L. (1999). Whence Consumer Loyalty? The Journal of Marketing, 63, 33–44.
Olsen, S. O. (2004). Antecedents of Seafood Consumption Behavior. Journal of Aquatic Food
Product Technology, 13(3), 79–91.
Papadopoulos, N., Marshall, J. J., & Heslop, L. A. (1988). Strategic implications of product
nd country images: a modeling approach. Marketing Productivity, European Society
for Opinion and Marketing Research, 69–90.
Park, C. W., Jaworski, B. J., & Maclnnis, D. J. (1986). Strategic Brand Concept-Image
Management. The Journal of Marketing, 50(4), 135–145.
Parkes, L. P., Schneider, S. K., & Bochner, S. (1999). Individualism-collectivism and selfconcept: Social or contextual? Asian Journal of Social Psychology, 2(3), 367–383.
Patterson, M. J., Hill, R. P., & Maloy, K. (1995). Abortion in America: A Consumer-Behavior
Perspective. Journal of Consumer Research, 21(4), 677–694.
Patton, M. Q. (1990). Qualitative research and evaluation methods (2nd ed.). Sage, Newbury
Park, California.
Penaloza, L., & Price, L. L. (1993). Consumer Resistance: A Conceptual overview. Advances
in Consumer Research, 20, 123–128.
Peter, J. P., & Ryan, M. J. (1976). An Investigation of Perceived Risk at the Brand Level.
Peter, J. P., & Tarpey, L. X. (1975). A Comparative Analysis of Three Consumer Decision
Strategies. The Journal of Consumer Research, 2(1), 29–37.
Pfeffermann, D. (1993). The Role of Sampling Weights when Modeling Survey Data.
International Statistical Review, 61, 317–337.
Pfeffermann, D. (1996). The Use of Sampling Weights for Survey Data Analysis. Statistical
Methods in Medical Research, 5(3), 239–261.
227
References
Phau, I., Teah, M., & Lee, A. (2009). Targeting buyers of counterfeits of luxury brands: A
study on attitudes of Singaporean consumers. Journal of Targeting, Measurement and
Analysis for Marketing, 17(1), 3–15.
Phillips, A. G., Silvia, P., & Paradise, M. J. (2007). The undesired self and emotional
experience: a latent variable analysis. Journal of Social and Clinical Psychology,
26(9), 1035–1047.
Piacentini, M. G., & Banister, E. N. (2009). Managing anti-consumption in an excessive
drinking culture. Journal of Business Research, 62(2), 279–288.
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method
biases in behavioral research: a critical review of the literature and recommended
remedies. The Journal of Applied Psychology, 88(5), 879–903.
Povey, R., Wellens, B., & Conner, M. (2001). Attitudes towards following meat, vegetarian
and vegan diets: an examination of the role of ambivalence. Appetite, 37(1), 15–26.
Raju, P. S. (1984). Exploratory brand switching: An empirical examination of its
determinants. Journal of Economic Psychology, 5(3), 201–221.
Raju, P. S. (1995). Consumer behavior in global markets: the A-B-C-D paradigm and its
application to eastern Europe and the Third World. Journal of Consumer Marketing,
12(5), 37–56.
Reich, R., B. (1998). The new meaning of corporate social responsibility. California
Management Review, 40(2), 8–17.
Richins, M. L. (1983). Negative Word-of-Mouth by Dissatisfied Consumers: A Pilot Study.
The Journal of Marketing, 47(1), 68–78.
Richins, M. L. (1984). Word of Mouth Communication as Negative Information. Advances in
228
References
Riefler, P., & Diamantopoulos, A. (2007). Consumer animosity: a literature review and a
reconsideration of its measurement. International Marketing Review, 24(1), 87–119.
Roberts, J. A. (1996). Will the real socially responsible consumer please step forward?
Business Horizons, 39(1), 79–83.
Roselius, T. (1971). Consumer Rankings of Risk Reduction Methods. The Journal of
Marketing, 35(1), 56–61.
Rossiter, J. R., & Percy, L. (1987). Advertising and Promotion Management. New Yok:
McGraw-Hill Book Company.
Roth, M. S., & Romeo, J. B. (1992). Matching Product Catgeory and Country Image
Perceptions: A Framework for Managing Country-Of-Origin Effects. Journal of
International Business Studies, 23(3), 477–497.
Rozin, P., & Fallon, A. E. (1987). A perspective on disgust. Psychological Review, 94(1), 23–
41.
Rumbo, J. D. (2002). Consumer resistance in a world of advertising clutter: The case
ofAdbusters. Psychology and Marketing, 19(2), 127–148.
Russell, C. A., & Russell, D. W. (2010). Guilty by stereotypic association: Country animosity
and brand prejudice and discrimination. Marketing Letters, 21(4), 413–425.
Russell, D. W., & Russell, C. A. (2006). Explicit and implicit catalysts of consumer
resistance: The effects of animosity, cultural salience and country-of-origin on
subsequent choice. International Journal of Research in Marketing, 23(3), 321–331.
Rust, R. T., Varki, S., & Oliver, R. L. (1997). Customer delight: Foundations, findings, and
managerial insight. Journal of Retailing, 73(3), 311–336.
Sandelowski, M. (1995). Focus on qualitative methods:sample size in qualitative research.
Research in Nursing and Health, 18, 179–183.
229
References
SandIkci, Ö., & Ekici, A. (2009). Politically motivated brand rejection. Journal of Business
Research, 62(2), 208–217.
Schiffman, L. G. (1972). Perceived Risk in New Product Trial by Elderly Consumers. Journal
of Marketing Research, 9(1), 106–108.
Schlossberger, E. (1994). A New Model of Business: Dual-Investor Theory. Business Ethics
Quarterly, 4(4), 459–474.
Sekaran, U. (2002). Research Methods for Business - A Skill Building Approach (4th Ed.).
Wiley.
Seva, R. R., & Helander, M. G. (2009). The influence of cellular phone attributes on users’
affective experiences: A cultural comparison. International Journal of Industrial
Ergonomics, 39(2), 341–346.
Shankarmahesh, M. N. (2006). Consumer ethnocentrism: an integrative review of its
antecedents and consequences. International Marketing Review, 23(2), 146–172.
Sharma, S., Shimp, T. A., & Shin, J. (1995). Consumer ethnocentrism: A test of antecedents
and moderators. Journal of the Academy of Marketing Science, 23(1), 26–37.
Shaw, D., & Clarke, I. (1999). Belief formation in ethical consumer groups: an exploratory
study. Marketing Intelligence & Planning, 17(2), 109–120.
Shaw, D., & Newholm, T. (2002). Voluntary simplicity and the ethics of consumption.
Psychology and Marketing, 19(2), 167–185.
Shepherd, R., & Raats, M. M. (1996). Attitude and beliefs in food habits. In H. L. Meiselman,
and H. J. H. MacFie (Eds.), Food choice, acceptance and consuption, 346–364.
Shimp, T. A., Dunn, T. H., & Klein, J. G. (2004). Remnants of the U.S. Civil War and
modern consumer behavior. Psychology and Marketing, 21(2), 75–91.
Shimp, T. A., & Sharma, S. (1987). Consumer Ethnocentrism: Construction and Validation of
the CETSCALE. Journal of Marketing Research, 24(3), 280–289.
230
References
Shin, M. (2001). The Animosity Model of Foreign Product Purchase - Does It Work In
Korea? Journal of Empirical Generalizations in Markating Science, 6(1), 1–14.
Shoham, A., Davidow, M., Klein, J. G., & Ruvio, A. (2006). Animosity on the Home Front:
The Intifada in Israel and Its Impact on Consumer Behavior. Journal of International
Marketing, 14(3), 92–114.
Sirgy, J. (1982). Self-Concept in Consumer Behavior: A Critical Review. The Journal of
Sirgy, M. J. (1985). Using self-congruity and ideal congruity to predict purchase motivation.
Journal of Business Research, 13(3), 195–206.
Sirgy, M. J., Grewal, D., Mangleburg, T. F., Park, J.-O., Chon, K.-S., Claiborne, C. B., et al.
(1997). Assessing the Predictive Validity of Two Methods of Measuring Self-Image
Congruence. Journal of the Academy of Marketing Science, 25(3), 229 –241.
Skowronski, J., & Carlston, D. (1989). Negativity and extremity biases in impression
formation: A review of explanations. Psychological Bulletin, 105(1), 131–142.
Solomon, M. R. (1983). The Role of Products as Social Stimuli: A Symbolic Interactionism
Perspective. Journal of Consumer Research: An Interdisciplinary Quarterly, Journal
of Consumer Research: An Interdisciplinary Quarterly, 10(3), 319–29.
Solomon, M. R. (2002). Consumer Behavior: Buying, Having, and Being (5th ed.). Prentice
Hall.
Speck, P. S., & Elliott, M. T. (1997). Predictors of Advertising Avoidance in Print and
Broadcast Media. Journal of Advertising, XXVI(3), 61–76.
Spector, P. E. (2006). Method Variance in Organizational Research Truth or Urban Legend?
Organizational Research Methods, 9(2), 221–232.
231
References
Srivastava, R. K., Fahey, L., & Christensen, H. K. (2001). The resource-based view and
marketing: The role of market-based assets in gaining competitive advantage. Journal
of Management, 27(6), 777 –802.
Steenkamp, J.-B. E. M., & Baumgartner, H. (1998). Assessing Measurement Invariance in
Cross-National Consumer Research. Journal of Consumer Research, 25(1), 78–90.
Steiger, J. (1990). Structural Model Evaluation and Modification: An Interval Estimation
Approach. Multivariate Behavioral Research, 25(2), 173–180.
Stern, B. B. (2006). What Does Brand Mean? Historical-Analysis Method and Construct
Definition. Journal of the Academy of Marketing Science, 34(2), 216 –223.
Stone, R. N., & Grønhaug, K. (1993). Perceived Risk: Further Considerations for the
Marketing Discipline. European Journal of Marketing, 27(3), 39–50.
Sumner, G. A. (1906). Folkways. New York: Ginn Custom Publishing.
Tabachnick, B. G., & Fidell, L. S. (2007). Using Multivariate Statistics. Boston: Pearson
International Edition.
Tajfel, H. (1978). Intergroup behaviour: Individualistic perspectives. In H. Tajfel & C.
Fraser, Introducing social psychology. London: Penguin.
Tam, J. L. M. (2008). Brand familiarity: its effects on satisfaction evaluations. Journal of
Services Marketing, 22(1), 3–12.
Taylor, J. W. (1974). The Role of Risk in Consumer Behavior. The Journal of Marketing,
38(2), 54–60.
Taylor, S., & Bogdan, R. (1998). Introduction to qualitative research methods : a guidebook
and resource (3rd ed.). New York: Wiley.
Thakor, M. V. (1996). Brand origin: conceptualization and review. Journal of Consumer
Marketing, 13(3), 27–42.
232
References
Thompson, C. J., & Arsel, Z. (2004). The Starbucks Brandscape and Consumers’
(Anticorporate) Experiences of Glocalization. The Journal of Consumer Research,
31(3), 631–642.
Thompson, C. J., Locander, W. B., & Pollio, H. R. (1989). Putting Consumer Experience
Back into Consumer Research: The Philosophy and Method of ExistentialPhenomenology. Journal of Consumer Research, 16(2), 133–146.
Thompson, C. J., Rindfleisch, A., & Arsel, Z. (2006). Emotional Branding and the Strategic
Value of the Doppelgänger Brand Image. Journal of Marketing, 70(1), 50–64.
Thomson, M., MacInnis, D. J., & Whan Park, C. (2005). The Ties That Bind: Measuring the
Strength of Consumers’ Emotional Attachments to Brands. Journal of Consumer
Psychology, 15(1), 77–91.
Titus, P. A., & Bradford, J. L. (1996). Reflections on Consumer Sophistication and Its Impact
on Ethical Business Practice. Journal of Consumer Affairs, 30(1), 170–194.
Triandis, H. C., & Albert, R. D. (1987). Cross-Culutral perspectives. In Jablin, F.M.
Putman,L.L. Roberts, K.H. Porter, L.W. (Eds) Handbook of Organizational
Communication. Beverly Hills, CA: SAGE.
Turker, D. (2009). Measuring Corporate Social Responsibility: A Scale Development Study.
Journal of Business Ethics, Journal of Business Ethics, 85(4), 411–427.
Verbeke, W., & Vackier, I. (2005). Individual determinants of fish consumption: application
of the theory of planned behaviour. Appetite, 44(1), 67–82.
Ward, S., Larry, L., & Jonathan, G. (1999). What High-Tech managers need to know about
brands. Harvard Business Review, 77(4), 85–95.
Watson, J. J., & Wright, K. (2000). Consumer ethnocentrism and attitudes toward domestic
and foreign products. European Journal of Marketing, 34(9/10), 1149–1166.
233
References
Wee, C.-H., Ta, S.-J., & Cheok, K.-H. (1995). Non-price determinants of intention to
purchase counterfeit goods: an exploratory study. International Marketing Review,
12(6), 19–46.
Whang, Y. O., Allen, J., Sahoury, N., & Zhang, H. (2004). Falling in Love with a Product:
The Structure of a Romantic Consumer-Product Relationship. Advances in Consumer
Research, 31, 320–327.
Wilk, R. (1994). I Hate Pizza, Distaste and Dislike in the consuming lives of Belizeans,
Presented at the Annual Meeting of the American Anthropology Association, Atlanta.
Wilk, R. (1995). Learning Distate The Social Importance of Not wanting, prepared for the
conference Learning to consume, Lund University,, 18–20.
Wilk, R. (1996). A Critique of Desire: Distaste and Dislike in Consumer Behavior. Special
Session Paper. Association for Consumer Research.
Wilk, R. (1997). Special session summary: Learning not to want things. Advances in
Consumer Research, 24, 61–63.
Wright, K. B. (2005). Researching Internet-Based Populations: Advantages and
Distadvantages of OnineSurvey Research, Online Questionaire Authoring Software
Packages, and Web Survey Services. Journal of Coputer-Mediated Communication,
10(3).
Yang, K. C. C., & Tso, T. K. (2007). An Exploratory Study of Factors Influencing Audience’s
Attitudes Toward Imported Television Programs in Taiwan. International Journal on
Media Management, 9(1), 19–27.
Yansaneh, I. S. (2005). “Construction and Use of Sample Weights”, in United nation
Statistics Division (ed.), Designing Household Surveys Samples: Practical Guidelines.
New York: United Nations.
234
References
Zahorik, A. J. (1994). A nonhierarchical brand switching model for inferring market structure.
European Journal of Operational Research, 76(2), 344–358.
Zavestoski, S. (2002a). The social-psychological bases of anti-consumption attitudes.
Psychology and Marketing, 19(2), 149–165.
Zavestoski, S. (2002b). Guest editorial: Anti-consumption attitudes. Psychology and
Marketing, 19(2), 121–126.
Zeithaml, V. (1988). Consumer perceptions of price, quality, and value: A means-end model
and synthesis of evidence. Journal of Marketing, 52(3), 2–22.
235
Appendices
9 Appendices
Appendix I: Participants’ detail-Pre-testing in Qualitative study
Master
Professional
Income
Status
Academic
Rs. 25,001-40,000
Single
31
Master
Academic
Rs. 25,001-40,000
Married
M
32
Master
Academic
Rs. 25,001-40,000
Married
AN
M
36
Master
Banker
Rs. 40,001-60,000
Married
5
SG
M
30
Master
Business
Rs. 80,001-100,000
Married
6
RA
M
32
Master
Banker
Rs. 40,001-60,000
Married
7
HH
M
32
Master
Job
Rs. 60,001-80,000
Married
8
IA
F
30
Master
Job
Rs. 25,001-40,000
Married
9
FG
M
28
Master
Job
Rs. 25,001-40,000
Married
10
MH
M
30
Master
Job
Rs. 25,001-40,000
Married
11
SQ
F
32
PhD
Academic
Rs. 80,001-100,000
Single
12
KH
M
48
Master
Job
Rs. 80,001-100,000
Married
13
AG
M
50
Engineer
Job
Rs. 80,001-100,000
Married
14
JZ
M
51
Engineer
Job
Rs. 80,001-100,000
Married
15
TA
M
44
Master
Job
Rs. 61,001-80,000
Married
16
NT
F
25
Bachelor
Job
Rs. 25,001-40,000
Single
17
AR
M
45
Master
Job
Rs. 41,001-60,000
Married
18
SS
F
28
Bachelor
Job
Rs. 25,001-40,000
Married
19
MA
M
47
Master
Job
Rs. 41,001-60,000
Married
20
ZA
M
31
Bachelor
Job
Rs. 40,001-60,000
Married
21
KA
M
27
Bachelor
Student
Rs. 1-25,000
Single
22
BN
M
25
Intermediate Student
Rs. 1-25,000
Single
23
IH
M
29
PhD
Student
Rs. 80,001-100,000
Married
24
GA
M
27
PhD
Student
Rs. 80,001-100,000
Married
25
AA
F
27
PhD
Student
Rs. 80,001-100,000
Single
26
HY
F
27
Master
Student
Rs. 1-25,000
Married
27
AJ
M
38
PhD
Student
Rs. 80,001-100,000
Married
No.
Initial
Gender
Age
1
AR
M
32
2
RR
M
3
YH
4
Education
Status
Appendices
Appendix II: Preliminary Analysis of Sample 1
a) Univariate Detection of Outliers
Descriptive Statistics
N
Minimum
Maximum
N
Minimum
Maximum
ZBA1
286
-1.95331
1.34903
ZFCO
286
-2.44392
1.69462
ZBA2
286
-1.88618
1.30897
ZA1
286
-2.97302
1.03460
ZBA3
286
-2.19155
1.30678
ZA2
286
-3.15509
1.09129
ZBA4
286
-2.08101
1.21328
ZA3
286
-2.47739
1.20969
ZBA5
286
-2.54147
.95305
ZA4
286
-2.18803
1.21910
ZBA6
286
-2.58484
.97502
ZA5
286
-1.94580
1.25247
ZBA7
286
-2.43607
.97920
ZUC1
286
-1.83138
2.38129
ZUS1
286
-2.12467
1.35760
ZUC2
286
-1.95383
2.31179
ZUS2
286
-1.99920
1.56324
ZUC3
286
-1.92096
2.32146
ZUS3
286
-1.36374
1.82018
ZUC4
286
-1.68355
2.24703
ZUS4
286
-1.79822
1.72432
ZUC5
286
-1.57853
1.86773
ZUS5
286
-1.98036
1.45573
ZPR1
286
-2.01694
1.47556
ZNSI1
286
-1.62897
1.83057
ZPR2
286
-1.75274
1.50588
ZNSI2
286
-1.82043
1.82043
ZPR3
286
-1.30006
2.22986
ZNSI3
286
-1.60274
1.73096
ZPR4
286
-1.62915
1.78848
ZET1
286
-1.55319
1.86383
ZET2
286
-1.90717
1.32354
ZET3
286
-1.69893
1.51182
ZET4
286
-2.21501
1.08730
Threshold value of Z-Score > + 3.29 (Tabachnick and Fidell, 2007)
237
Appendices
b) Detection of Univariate Outliers through Box Plots
No significant univariate outlier present (no value marked by star *)
238
Appendices
c) Detecting Multivariate Outliers
Multivariate Outliers
PD
Value
163.6903 0.0000
Leverage
Hat
0.5744
31
Cook’s
Case
Distance
No.
0.078
183
0.0382
65
183
117.0377 0.0000
3
65
205
100.3428 0.0000
265
0.0334
0.3429
4
233
99.7951
0.0000
0.0309
97.9742
0.0000
6
245
186
65
271
5
96.9815
0.0001
27
0.0263
7
123
96.699
0.0001
1
0.0263
8
219
94.4733
0.0001
261
0.0241
9
231
94.3319
0.0001
231
0.0210
10
258
93.6440
0.0001
279
0.0210
11
94
93.2945
0.0001
72
0.0200
12
164
91.8923
0.0002
14
0.0192
13
241
91.1327
0.0002
0.0187
14
244
90.2736
0.0003
245
15
15
198
89.1966
0.0004
255
0.0163
16
261
89.1587
0.0004
2
0.0143
17
72
88.8301
0.0004
18
97
88.3605
0.0005
19
284
86.6438
0.0007
20
48
82.1046
0.0021
Sr. No.
1
2
Case
No.
183
Cut-Points
Case
No.
2
Prob. of D2 (p) < 0.001
11
0.4107
0.0271
0.0166
Cook’s D > 0.0159
Leverage Hat > 0.3671
Cd > 4/n-k-1 & Lh > 3(k+1)/n; Here n = 286, k = 34; (Hair et al. 2009)
239
Appendices
Appendix III: Common Method Bias for Sample 1
Total Variance Explained
Component
Initial Eigenvalues
Total
% of Variance
Extraction Sums of Squared Loadings
Cumulative %
Total
% of Variance
Cumulative %
1
8.659
25.469
25.469
8.659
25.469
25.469
2
5.891
17.326
42.794
5.891
17.326
42.794
3
2.310
6.794
49.588
2.310
6.794
49.588
4
2.172
6.387
55.975
2.172
6.387
55.975
5
1.983
5.833
61.808
1.983
5.833
61.808
6
1.625
4.780
66.588
1.625
4.780
66.588
7
1.210
3.559
70.147
1.210
3.559
70.147
8
.971
2.855
73.002
9
.705
2.072
75.075
10
.669
1.967
77.042
Extraction Method: Principal Component Analysis.
240
Appendices
Appendix IV: Preliminary Analysis of Sample 2
a) Univariate Detection of Outliers
N
Minimum
Maximum
N
Minimum
Maximum
ZBA1
199
-2.62525
1.30276
ZFCO
199
-2.08956
1.52479
ZBA2
199
-2.63456
1.06796
ZA1
199
-2.67357
.99990
ZBA3
199
-2.44613
1.10701
ZA2
199
-2.84225
1.02736
ZBA4
199
-1.88765
1.31203
ZA3
199
-2.14699
1.06945
ZBA5
199
-2.99104
.80015
ZA4
199
-2.47009
1.06199
ZBA6
199
-2.75004
.85034
ZA5
199
-2.11054
1.08336
ZBA7
199
-3.09141
.82700
ZUC1
199
-2.19565
1.81906
ZUS1
199
-2.97000
.93893
ZUC2
199
-2.16651
1.74697
ZUS2
199
-2.92597
.98624
ZUC3
199
-2.26943
1.60157
ZUS3
199
-1.78886
1.16945
ZUC4
199
-2.01457
1.82791
ZUS4
199
-1.96205
1.35151
ZUC5
199
-1.93658
1.48902
ZUS5
199
-2.65486
1.23459
ZPR1
199
-1.38256
1.70300
ZNSI1
199
-1.56045
1.98846
ZPR2
199
-1.65892
1.42636
ZNSI2
199
-1.79205
1.96842
ZPR3
199
-1.12349
2.22177
ZNSI3
199
-1.75970
2.13841
ZPR4
199
-1.67316
1.70713
ZET1
199
-1.26624
2.29955
ZET2
199
-1.49490
1.89203
ZET3
199
-1.40143
2.23619
ZET4
199
-1.58297
1.75048
Threshold value of Z-Score > + 3.29 (Tabachnick and Fidell, 2007)
241
Appendices
b) Detection of Univariate Outliers through Box Plots
One significant univariate outlier present (marked by star *)
c) Detecting Multivariate Outliers
Multivariate Outliers
Sr. No.
1
2
Case
No.
85
182
Cut-Points
98.6020
PValue
0.0005
95.9612
0.001
D2
Case
No.
36
Prob. of D2 (p) < 0.001
Cook’s
Case
Distance
No.
0.0511
85
Cook’s D > 0.0244
Leverage
Hat
0.498
Leverage Hat > 0.5276
Cd > 4/n-k-1 & Lh > 3(k+1)/n; Here n = 199, k = 34; (Hair et al. 2009)
242
Appendices
Appendix V: Common Method Bias for Sample 2
Total Variance Explained
Component
Initial Eigenvalues
Total
% of Variance
Extraction Sums of Squared Loadings
Cumulative %
Total
% of Variance
Cumulative %
1
9.095
26.749
26.749
9.095
26.749
26.749
2
3.515
10.338
37.087
3.515
10.338
37.087
3
2.963
8.713
45.801
2.963
8.713
45.801
4
2.664
7.834
53.635
2.664
7.834
53.635
5
2.185
6.428
60.062
2.185
6.428
60.062
6
2.023
5.950
66.012
2.023
5.950
66.012
7
1.015
2.986
68.998
1.015
2.986
68.998
8
.947
2.786
71.784
9
.888
2.610
74.394
10
.827
2.431
76.825
Extraction Method: Principal Component Analysis.
243
Appendices
Appendix VI: Questionnaire for Pakistan
Dear Respondent,
I am a PhD candidate in the Marketing department at IAE-Graduate School of Management,
Aix en Provence, France. As a part of my PhD. study, I am conducting a survey on brand
avoidance. The survey measures the perceptions of consumers towards the brands that they
avoid. Through your participation, I eventually hope to better understand the reasons behind
brand avoidance.
I am interested in your perception of ‘affordable brands’ which you would NEVER
consider using/buying, and do not have the intention to use/buy in future.
There is no right or wrong answer in this survey. The survey will take about 05 – 10 minutes
to complete. All your answers to the survey questions will be kept confidential.
If you have any questions or concerns about questionnaire, or if you are interested in the
research findings, please contact me at [email protected]
Thanks for your cooperation!
Sincerely,
Muhammad Asif KHAN
PhD Candidate
Marketing Department
IAE-Graduate School of Management
University Paul Cézanne, Aix-Marseille III, France
244
Appendices
Consider in your mind the 'Brand' you avoid and answer the following questions.
[Remember: You can afford that brand but you never considered it to use/buy before]
Strongly
Agree
Agree
Slightly
Agree
Neutral
Slightly
Disagree
Disagree
Following questions are concerning your purchasing of that
brand
Strongly
Disagree
Brand Name: _________________________________________
6
7
2
1
2
3
4
5
6
7
3
1
2
3
4
5
6
7
4
Using that brand is not pleasant
1
2
3
4
5
6
7
5
1
2
3
4
5
6
7
6
1
2
3
4
5
6
7
7
I would rather buy any other brand than this brand
1
2
3
4
5
6
7
Following questions are concerning to your personal feeling
about that brand
Strongly
Agree
5
Agree
4
Slightly
Agree
3
Neutral
2
Slightly
Disagree
1
Disagree
Using that brand is not a good idea
Strongly
Disagree
1
6
7
2
I associate that brand with something unappealing, which I do
not want to become
1
2
3
4
5
6
7
3
I do not want to be associated with the type of people using
that brand
1
2
3
4
5
6
7
4
1
2
3
4
5
6
7
5
1
2
3
4
5
6
7
Following questions are concerning the role of others in not
purchasing that brand
Strongly
Agree
5
Agree
4
Slightly
Agree
3
Neutral
2
Slightly
Disagree
1
Disagree
Strongly
Disagree
1
1
1
2
3
4
5
6
7
2
1
2
3
4
5
6
7
3
People who influence my behavior think that I should not use
1
2
3
4
5
6
7
245
Appendices
Strongly
Agree
Agree
Slightly
Agree
Neutral
Slightly
Disagree
Disagree
Following questions are concerning to your personal feeling
about foreign brands
Strongly
Disagree
that brand
1
2
3
4
5
6
7
2
We should purchase products manufactured in our country
instead of letting other countries get rich off us
1
2
3
4
5
6
7
3
We should not buy foreign products/brands because this hurts
our business and cause unemployment
1
2
3
4
5
6
7
We should buy from foreign countries only those
4 products/brands that we cannot obtain within our own country
1
2
3
4
5
6
7
2
3
4
5
6
7
Strongly
Agree
Agree
Slightly
Agree
Following questions are concerning to your perception about the
country from which that brand originates.
Neutral
1
Slightly
Disagree
Strongly
Disagree
1
Disagree
Please indicate your level of familiarity with the country of that
brand
Very
Familiar
It is not right to purchase foreign products/brands because it
puts us out of job
Not
Familiar
1
6
7
2
1
2
3
4
5
6
7
3
1
2
3
4
5
6
7
4
1
2
3
4
5
6
7
5
1
2
3
4
5
6
7
Following questions are concerning the role in society of that
company whose brand you avoid
Strongly
Agree
5
Agree
4
Slightly
Agree
3
Neutral
2
Slightly
Disagree
1
Disagree
Strongly
Disagree
1
1
The company does not participate in activities which aim to
protect and improve the quality of the natural environment
1
2
3
4
5
6
7
2
The company does not contribute to campaigns and projects
that promote the well-being of the society
1
2
3
4
5
6
7
3
The company does not emphasize the importance of its social
responsibilities to the society
1
2
3
4
5
6
7
246
Appendices
6
7
5
The company does not provide full and accurate information
about its products to its customers
1
2
3
4
5
6
7
Following questions are concerning to your risks attached in
buying that brand …
Buying that brand is a financial risk for me because of such
1 things as its poor warranty, high maintenance costs, or highly
monthly payments
Strongly
Agree
5
Agree
4
Slightly
Agree
3
Neutral
2
Slightly
Disagree
1
Disagree
The company does not respect consumer right
Strongly
Disagree
4
1
2
3
4
5
6
7
2
Buying that brand is a performance risk for me because it
would run extremely poorly
1
2
3
4
5
6
7
3
Buying that brand is a physical risk for me because it would
hurt me
1
2
3
4
5
6
7
4
Buying that brand is too risky in general for my liking
1
2
3
4
5
6
7
Demographic characteristics:
Gender:
Male
Female
Age:
Less than 20 years
20-25 years
26-30 years
31-35 years
36-40 years
More than 40 years
Education level and Area (highest level of education completed):
Matric
Intermediate
Bachelor
Master or Above
Monthly Income of your family is:
Below or Rs. 60,000
Between Rs. 61,000 to Rs. 80,000
Between Rs. 81,000 to Rs. 100,000
Above Rs. 100,000
247
Appendices
Appendix VII: Questionnaire for New Zealand
The same questionnaire (cf. Appendix VI) has been used for New Zealand sample except the
following changes in demographic information:
Education level and Area (highest level of education completed):
Secondary School Certificate
Under Graduate Tertiary Degree/Diploma
Post Graduate Tertiary Degree/Diploma
Master Degree or Above
Monthly Income of your family is:
Less or $ 30,000
$31,000 to $50,000
$51,000 to $70,000
$71,000 to $90,000
$91,000 and Above
248
Appendices
Appendix VIII: Moderation effect
a) For Pakistan Sample
249
Appendices
b) For New Zealand Sample
250
Appendices
Appendix IX: Sample Weights
Sample weights calculations for Pakistan
Gender
Male
Female
Education
Less than Master
Master or Above
Sub Total
Less than Master
Master or Above
Sub Total
Sample= 278
No
%
42
0,207
161
0,793
203
1,000
34
0,453
41
0,547
75
1,000
Population= 190 million
millions
%
Gender %
0,657
0,740
0,512
0,231
0,260
0,888
1,000
0,585
0,791
0,488
0,155
0,209
0,74
1,000
The weights are calculated by dividing population percentage with respective sample percentage:
Gender
Male
Female
Education
Less than Master
Master or Above
Less than Master
Master or Above
Weights
3.576
0.328
1.744
0.383
Note that our target population is people living in urban areas in Pakistan because of the data collection method of online web-based survey.
Appendices
252
Appendices
253
Appendices
254
Appendices
Source : http://www.pbs.gov.pk/content/labour-force-survey-2010-11
255
Appendices
Sample weights calculations for New Zealand
Gender
Male
Female
Sample= 198
No
%
66
0,857
11
0,143
77
1,000
94
0,777
27
0,223
121
1,000
Age
Less than 25 years
More than 25 years
Sub Total
Less than 25 years
More than 25 years
Sub Total
Population= 4.3 million
millions
%
Gender %
8,156
0,490
0,49
8,488
0,510
16,644
1,000
8,488
0,490
0,51
8,834
0,510
17,322
1,000
The weights are calculated by dividing population percentage with respective sample percentage:
Gender
Male
Female
Age
Less than 25 years
More than 25 years
Less than 25 years
More than 25 years
Weights
0.572
3.570
0.631
2.285
People employed, unemployed, and not in labour force
By sex
Seasonally adjusted series
(1)
Labour force
Employed
Not in labour force
Unemployed
Total
(000)
256
Labour force
participation rate
Employment
rate
(%)
Unemployment
rate
Appendices
Males
Series ref: HLFQ
Quarter
2009
2010
2011
Dec
Mar
Jun
Sep
Dec
Mar
Jun
Sep
Dec
S1A1S
1 144
1 156
1 154
1 173
1 164
1 175
1 175
1 183
1 179
S1B1S
R
R
R
R
R
R
R
R
85
72
86
72
79
78
80
79
81
S1Z1S
R
R
R
R
R
R
R
R
S1C1S
1 229
1 228
1 240
1 245
1 243
1 253
1 255
1 262
1 259
R
R
R
R
R
R
R
R
422
425
425
422
431
423
431
425
433
S1E1S
R
R
R
R
R
R
R
R
74,4
74,3
74,5
74,7
74,2
74,7
74,4
74,8
74,4
S1H1S
R
R
R
R
R
R
R
R
69,3
69,9
69,3
70,3
69,5
70,1
69,7
70,1
69,7
S1F1S
R
R
R
R
R
R
R
R
6,9
5,8
6,9
5,8
6,4
6,2
6,4
6,3
6,4
R
R
R
R
R
R
R
R
Females
Series ref: HLFQ
Quarter
2009
2010
2011
Dec
Mar
Jun
Sep
Dec
Mar
Jun
Sep
Dec
S1A2S
1 014
1 016
1 016
1 020
1 023
1 036
1 038
1 034
1 042
S1B2S
R
R
R
R
R
R
R
R
76
70
75
78
77
77
75
78
70
S1Z2S
R
R
R
R
R
R
R
R
1 090
1 087
1 091
1 098
1 100
1 113
1 113
1 112
1 112
S1C2S
R
R
R
R
R
R
R
R
660
665
669
668
671
659
665
670
674
S1E2S
R
R
R
R
R
R
R
R
62,3
62,0
62,0
62,2
62,1
62,8
62,6
62,4
62,3
S1H2S
R
R
R
R
R
R
R
R
57,9
58,0
57,8
57,8
57,8
58,4
58,4
58,0
58,4
S1F2S
R
R
R
R
R
R
R
R
7,0
6,5
6,9
7,1
7,0
6,9
6,7
7,0
6,3
R
R
R
R
R
R
R
R
Total
Series ref: HLFQ
Quarter
2009
2010
Dec
Mar
Jun
S1A3S
2 158
2 173
2 170
S1B3S
R
R
R
161
142
161
S1Z3S
R
R
R
2 319
2 314
2 331
S1C3S
R
R
R
257
1 082
1 090
1 094
S1E3S
R
R
R
68,2
68,0
68,1
S1H3S
R
R
R
63,5
63,8
63,4
S1F3S
R
R
R
6,9
6,1
6,9
R
R
R
Appendices
2011
Sep
Dec
Mar
Jun
Sep
Dec
2 193
2 186
2 211
2 213
2 217
2 221
R
R
R
R
R
150
156
155
155
157
150
R
R
R
R
R
2 343
2 343
2 367
2 368
2 374
2 371
R
R
R
R
R
1 090
1 102
1 083
1 097
1 095
1 107
R
R
R
R
R
1. All previously published figures are subject to revision when the seasonal adjustment program is run each quarter.
Symbol:
R revised
Source: Statistics New Zealand
258
68,2
68,0
68,6
68,3
68,4
68,2
R
R
R
R
R
63,9
63,5
64,1
63,9
63,9
63,9
R
R
R
R
R
6,4
6,7
6,6
6,6
6,6
6,3
R
R
R
R
R
Appendices
People employed. unemployed. and not in labour force
By age group(1)
Labour force
Employed
Unemployed
Not in labour
force
Total
Labour force
participation
rate
Working-age
population(2)(3)
(000)
SAA3AA
Series ref: HLFQ
Quarter
2010
2011
Dec
Mar
Jun
Sep
Dec
Absolute sampling error
Quarter
Dec
Mar
Jun
Sep
Dec
Series ref: HLFQ
R
7.6
213.3
214.1
209.8
209.6
211.7
9.7
SAA3AC
39.5
41.4
39.3
31.3
36.2
R
R
7.1
SAB3AC
15–19 years
SAC3AA
R
164.3
167.2
174.1
181.2
164.4
7.3
SAB3AB
27.0
33.5
26.4
29.1
28.6
154.7
150.9
142.2
133.8
149.3
R
R
76.5
72.4
85.7
84.6
85.1
R
8.2
25–29 years
SAC3AC
259
319.0
318.2
316.3
315.1
313.7
SAE3AA
R
...
20–24 years
SAC3AB
8.2
SAZ3AC
SAD3AA
7.3
SAZ3AB
240.3
247.7
236.2
238.7
240.3
Unemployment
rate
(%)
SAZ3AA
4.7
SAA3AB
Series ref: HLFQ
2010
2011
115.3
109.5
102.9
102.5
113.1
SAB3AA
Employment
rate
R
...
SAD3AC
R
2.3
SAD3AB
316.8
320.1
321.9
323.4
325.4
48.5
47.4
45.0
42.5
47.6
SAH3AA
R
2.5
SAE3AC
R
2.4
SAE3AB
75.9
77.4
73.4
73.8
73.9
36.1
34.4
32.5
32.5
36.1
SAF3AA
R
3.0
SAH3AC
R
3.1
SAH3AB
67.3
66.9
65.2
64.8
65.1
25.5
27.5
27.6
23.4
24.2
SAF3AB
R
11.2
13.5
11.2
12.2
11.9
2.9
SAF3AC
R
Appendices
Quarter
2010
2011
Dec
Mar
Jun
Sep
Dec
Quarter
2010
2011
204.1
210.6
Jun
213.7
Sep
Dec
211.1
Series ref: HLFQ
Dec
Mar
Jun
Sep
Dec
sampling error
Series ref: HLFQ
7.8
Dec
Mar
Absolute
R
17.0
18.6
13.7
15.0
16.7
R
3.2
SAA3AD
Series ref: HLFQ
Quarter
2010
2011
217.9
221.1
220.2
222.0
224.2
R
224.7
R
R
R
SAF3AD
75.9
77.9
6.4
5.3
78.9
82.8
R
77.8
4.8
R
6.1
215.2
10.4
225.7
47.0
272.7
82.8
78.9
4.6
5.2
2.6
5.0
5.0
...
1.8
1.9
1.1
35–39 years
SAC3AE
SAA3AE
SAB3AE
SAZ3AE
235.9
229.7
229.5
223.2
224.3
10.6
12.0
10.7
11.3
11.1
246.5
241.7
240.3
234.5
235.4
5.7
SAA3AF
R
2.5
SAB3AF
R
4.7
SAZ3AF
R
49.2
51.4
49.9
52.8
49.3
4.7
40–44 years
SAC3AF
260
R
SAD3AE
SAE3AE
SAH3AE
SAF3AE
295.7
293.1
290.2
287.3
284.6
83.4
82.5
82.8
81.6
82.7
79.8
78.4
79.1
77.7
78.8
4.3
5.0
4.5
4.8
4.7
...
SAD3AF
R
R
R
R
1.4
SAH3AD
82.8
271.4
7.2
7.8
5.9
6.3
6.9
2.7
81.0
82.2
270.9
46.7
R
SAE3AD
269.0
270.4
46.5
R
75.3
75.9
75.5
76.0
76.3
2.2
SAD3AD
51.0
48.0
224.4
R
R
81.2
82.2
80.2
81.2
81.9
…
30–34 years
SAC3AD
218.0
222.4
10.7
R
289.2
291.4
291.5
292.0
294.0
6.3
SAZ3AD
13.9
11.8
13.6
54.3
51.7
57.6
55.0
53.1
6.3
SAB3AD
R
234.9
239.6
233.9
237.0
240.9
1.7
2.0
1.1
SAE3AF
SAH3AF
SAF3AF
R
R
Appendices
Quarter
2010
2011
Absolute
Dec
Mar
Jun
Sep
Dec
sampling error
Series ref: HLFQ
Quarter
2010
2011
Absolute
Dec
Mar
Jun
Sep
Dec
sampling error
Series ref: HLFQ
Quarter
2010
2011
Absolute
Dec
Mar
Jun
Sep
Dec
sampling error
Series ref: HLFQ
255.1
258.0
250.9
252.0
256.0
R
5.0
11.7
12.6
14.0
14.6
11.3
R
2.4
266.7
270.6
264.8
266.6
267.3
R
4.7
SAB3AG
SAZ3AG
267.6
265.9
267.5
267.8
265.6
11.2
11.9
13.0
12.5
11.6
278.8
277.8
280.5
280.3
277.2
4.9
R
2.4
4.7
SAZ3AH
243.8
243.7
251.7
249.5
250.2
9.8
10.4
7.6
10.2
10.2
253.5
254.1
259.3
259.8
260.4
4.9
SAA3AI
2.1
SAB3AI
4.6
SAZ3AI
R
R
40.2
41.7
37.7
38.7
39.5
R
261
4.4
4.6
5.3
5.5
4.2
R
0.9
SAD3AG
SAE3AG
SAH3AG
SAF3AG
320.6
319.9
318.6
317.1
316.0
87.0
86.8
88.1
88.4
87.7
83.5
83.1
84.0
84.5
84.1
4.0
4.3
4.6
4.5
4.2
R
R
R
1.5
1.5
0.8
SAD3AH
SAE3AH
SAH3AH
SAF3AH
293.7
295.8
297.0
298.4
299.9
86.3
85.9
87.3
87.0
86.8
83.0
82.4
84.7
83.6
83.4
3.8
4.1
2.9
3.9
3.9
4.6
55–59 years
SAC3AI
R
82.1
82.9
80.6
80.9
82.2
1.6
...
50–54 years
SAC3AH
85.9
87.0
85.1
85.6
85.8
1.5
4.7
SAB3AH
R
41.8
42.1
38.0
36.7
38.8
R
...
45–49 years
SAC3AG
R
SAA3AH
R
R
310.5
311.2
311.4
311.6
311.6
4.7
SAA3AG
R
43.8
40.6
46.5
45.0
44.3
...
SAD3AI
R
1.6
SAE3AI
R
R
1.6
SAH3AI
0.8
SAF3AI
R
R
R
Appendices
Quarter
2010
Dec
198.0
7.7
205.6
46.9
252.5
81.4
78.4
3.7
2011
Mar
198.9
8.2
207.1
47.1
254.2
81.5
78.3
3.9
Jun
202.8
7.6
210.4
45.3
255.7
82.3
79.3
Sep
206.6
Dec
209.0
7.7
216.6
42.6
259.3
83.6
80.6
3.5
4.5
1.9
4.2
4.2
...
1.6
1.7
0.9
SAD3AJ
SAE3AJ
SAH3AJ
SAF3AJ
Absolute
error
sampling
Series ref: HLFQ
R
SAA3AJ
6.0
R
SAB3AJ
212.6
SAZ3AJ
R
44.8
R
60–64 years
SAC3AJ
257.4
R
82.6
R
80.3
3.6
R
2.8
R
Quarter
2010
2011
Absolute
error
Dec
157.1
5.3
162.4
69.5
232.0
70.0
67.7
3.3
Mar
159.9
4.1
164.0
70.2
234.3
70.0
68.3
2.5
Jun
156.6
Sep
157.7
Dec
159.2
5.9
165.2
71.1
236.2
69.9
67.4
3.6
5.0
1.5
5.1
5.1
...
2.2
2.1
0.9
SAD3AK
SAE3AK
SAH3AK
SAF3AK
sampling
Series ref: HLFQ
4.9
R
SAA3AK
5.2
161.5
R
162.9
73.9
R
73.2
235.4
R
65 years and over
SAZ3AK
SAC3AK
SAB3AK
236.1
68.6
R
69.0
66.5
R
66.8
3.0
R
3.2
R
Quarter
94.9
2.1
97.1
446.7
543.8
17.8
17.5
2.2
Mar
98.4
2.3
100.7
448.4
549.1
18.3
17.9
2.3
Jun
102.8
1.9
104.7
447.5
552.2
19.0
18.6
Sep
104.4
Dec
108.5
S
109.5
453.2
562.7
19.5
19.3
S
7.2
0.7
7.2
7.2
...
1.3
1.3
0.7
SAE3AZ
SAH3AZ
SAF3AZ
2010
Dec
2011
Absolute
error
sampling
Series ref: HLFQ
SAA3AZ
R
2.2
SAB3AZ
R
106.6
SAZ3AZ
R
450.3
Total all ages
SAC3AZ
262
R
557.0
SAD3AZ
R
19.1
R
18.8
1.8
R
2.1
R
Appendices
Quarter
2010
Dec
2 203.0
155.6
2 358.6
1 084.2
3 442.8
68.5
64.0
6.6
2011
Mar
2 209.9
166.7
2 376.7
1 080.9
3 457.6
68.7
63.9
7.0
Jun
2 208.3
149.9
2 358.2
1 102.9
3 461.1
68.1
63.8
Sep
2 206.5
Dec
2 237.0
150.7
2 387.7
1 088.3
3 476.0
68.7
64.4
6.3
24.1
11.6
22.8
22.8
...
0.7
0.7
0.5
Absolute
error
sampling
R
151.0
R
2 357.5
R
1 109.2
R
3 466.6
R
1. As these figures are not seasonally adjusted. Statistics New Zealand recommends using annual comparisons.
2. Civilian. non-institutionalized, usually resident New Zealand population aged 15 years and over.
3. There is no sampling error in these cells. because of the sample design and the estimation methods used. There will be a small
error arising from the population estimates used in post-stratification.
Symbols:
R revised (Please refer to the 'Data quality' section of this release for more information.)
S suppressed (Please refer to the 'Data quality' section of this release for more information.)
… not applicable
Source: Statistics New Zealand
http://www.stats.govt.nz/
263
68.0
R
63.6
6.4
R
6.4
R
Appendices
Appendix X: Discriminant Validity with Weighted Sample
a) Discriminant validity for Pakistan sample
BAA
US
NSI
A
PUCC
PR
BAI
BAA
0.621
0.457
0.147
0.033
0.001
0.101
0.711
US
NSI
A
PUCC
PR
BAI
0.532
0.278
0.020
0.052
0.260
0.310
0.737
0.031
0.147
0.270
0.185
0.718
0.227
0.061
0.025
0.676
0.208
0.001
0.627
0.081
0.774
b) Discriminant validity for New Zealand sample
BAA
US
NSI
A
PUCC
PR
BAI
BAA
0.660
0.373
0.205
0.032
0.156
0.304
0.745
US
NSI
A
PUCC
PR
BAI
0.592
0.210
0.111
0.319
0.092
0.381
0.685
0.178
0.069
0.504
0.139
0.863
0.036
0.080
0.047
0.839
0.071
0.220
0.661
0.174
0.653
264
Appendices
Appendix XI: List of Brands
a) List of Brand for Sample 1 (Pakistan)
No. Brand Name
No. Brand Name
No. Brand Name
1
Acer
26
IPhone
51
Sony Ericsson
2
Airtel
27
Kodak
52
Sufi
3
Aiwa
28
Lenovo
53
Super Asia
4
Alcatel
29
LG
54
Suzuki
5
Amazon Kindle
30
Marlboro
55
Toshiba
6
Apple
31
Microsoft Windows
56
UPS
7
Aptech
32
Mitsubishi
57
VCR
8
Archos
33
Motorola
58
Vodafone
9
Asus
34
Nike
59
Waves
10
Bata
35
Nikon
60
Westwood
11
Belkin
36
Nobel
61
Wireless Connection
12
BenQ
37
Nokia
13
Black Berry
38
Orange
14
Bosh
39
Orient
15
Caltus
40
Packardbell
16
Casio
41
Panasonic
17
Chine Mobile
42
Parada
18
Citizen
43
PEL
19
Club Mobil
44
Philips
20
Compaq
45
Q Mobil
21
Daewoo Electronics
46
Remington
22
Dell
47
Rolex
23
HP
48
Samsung
24
HTC
49
Samsung Laptop
25
IPad
50
Sony
265
Appendices
b) List of Brand for Sample 2 (New Zealand)
No. Brand Name
1 Acer
No. Brand Name
26 GUESS
No. Brand Name
51 Quicksilver
2
Adidas
3
Apple
27 Home Brand
28 Honda
52 Radox shower product
53 Ralph Lauren
4
Avon
5
BIC
29 HTC
30 IBM
54 Rip Curl
55 Rugby World Cup
6
Blackberry
7
Budget Brand
31 KFC
32 Kia
56 Samsung
57 Sanyo
8
Centaur Saddles
9
Citroen
33 LG
34 Lion red beer
58 Signature Range
59 Sony Vaio
10 Coach
11 Coca Cola
35 Louis Vuitton
36 LV, Dior, Chanel, et Burberry
60 Subaru
61 Supré Clothing
12 Converse
13 Crest Toothpaste
37 MAC cosmetic
38 Marlboro Cigarettes
14 Daewoo
15 DoDo Communications
39 Maybelline (Make-up)
40 McDonalds fast food
62 Telecom
63 The Natural
The Warehouse
64 Confectionery
16 Dotti
17 Dove
41 Monster Energy Drink
42 Motorolla
18 Ed hardy
19 Esquires
43 Nestle
44 Neutrogena Skincare
20 Fila
21 Ford
45 Nike
46 Nivea
22 G-Unit
23 Gap
47 NUTELLA
48 Oak
24 Garage
25 Gold Knight
49 Pantene shampoo
50 Playboy
266
65 Vegemite
66 Vero insurance
67 Vevo (plasma tv brand)
68 Woolworths
69 Xbox
Appendices
Appendix XII: Discriminant Validity-SEM based approach
Pakistan Sample: Discriminant Validity between US & NSI
Pakistan Sample: Discriminant Validity between US & A
267
Appendices
Pakistan Sample: Discriminant Validity between US & PUCC
Pakistan Sample: Discriminant Validity between US & PR
268
Appendices
Pakistan Sample: Discriminant Validity between NSI & A
Pakistan Sample: Discriminant Validity between NSI & PUCC
269
Appendices
Pakistan Sample: Discriminant Validity between NSI & PR
Pakistan Sample: Discriminant Validity between A & PUCC
270
Appendices
Pakistan Sample: Discriminant Validity between A & PR
Pakistan Sample: Discriminant Validity between PUCC & PR
271
Appendices
Pakistan Sample: Discriminant Validity between BAA & BAI
New Zealand Sample: Discriminant Validity between US & NSI
272
Appendices
New Zealand Sample: Discriminant Validity between US & A
New Zealand Sample: Discriminant Validity between US & PUCC
273
Appendices
New Zealand Sample: Discriminant Validity between US & PR
New Zealand Sample: Discriminant Validity between NSI & A
274
Appendices
New Zealand Sample: Discriminant Validity between NSI & PUCC
New Zealand Sample: Discriminant Validity between NSI & PR
275
Appendices
New Zealand Sample: Discriminant Validity between A & PUCC
New Zealand Sample: Discriminant Validity between A & PR
276
Appendices
New Zealand Sample: Discriminant Validity between PUCC & PR
New Zealand Sample: Discriminant Validity between BAA & BAI
277
Appendices
Appendix XIII: Résumé de thèse en français
9.1
Introduction
9.1.1
Contexte de la recherche
Les consommateurs sont indépendants dans leurs choix de produits ou de marques. Ils
peuvent aimer certaines marques, être indifférents ou exprimer des opinions négatives à
l'égard d'autres. Un travail considérable a été fait par les chercheurs concernant les réponses
positives liées à la marque. Par exemple, l’amour pour la marque (Albert, Merunka, &
Valette-Florence, 2008; Carroll & Ahuvia, 2006; Whang, Allen, Sahoury, et Zhang, 2004),
l’attachement à la marque (Thompson, Rindfleisch, & Arsel, 2006; Thomson, MacInnis , &
Whan Park, 2005), la passion pour la marque (Fournier, 1998), la satisfaction pour la marque
(Fournier et Mick, 1999), et la joie de la marque (Rust, Varki, & Oliver, 1997). Les
consommateurs ont tendance à acheter des marques en congruence avec leur image de soi, ou
celles qui donnera un sens désiré à leur vie (Aaker, 1999; Grubb & Harrison L. Grathwohl,
1967; Levy, 1959; Sirgy, 1982; Salomon , 1983).
Alors que la littérature sur le côté négatif qui étudient les raisons pour les lesquels le
consommateur évite certaines marques est limitée ou étroitement définie. (Dalli, le romani, et
Gistri, 2006). Cette inégalité est compréhensible parce que les entreprises et les institutions
sont très intéressés par les résultats pratiques du coté positif du savoir. Elles sont intéressées
par ce que veulent les consommateurs et ce qui sont prêts à acheter (Dalli et al., 2006). En
même temps, il est également difficile de justifier cette inégalité sur le plan théorique, dans le
but de mieux comprendre les comportements d'achat et de consommation, il est essentiel de
prendre les deux aspects positifs et négatifs dans le même cadre (Dalli et al, 2006.). Lorsque
nous portons une attention particulière à la littérature existante, nous pouvons voir que les
278
Appendices
études majeures sur l'Anti-consommation se concentrent sur l'insatisfaction liées aux produits
ou services (Banister & Hogg, 2004; Oliver, 1980), ou sur le phénomène de la contre-culture
comme la simplification volontaire (Craig-Lees & Hill, 2002; Piacentini & Banister, 2009),
les boycotts (Klein, Smith, & John, 2004; Kozinets & Handelman, 1998) et la résistance des
consommateurs (Cherrier, 2009a; Hogg, 1998; Kozinets, 2002; Zavestoski, 2002a).
9.1.2
Identification des problèmes
Le consommateur est indépendant; il / elle peut ou ne peut pas choisir une marque. La
question fondamentale est que veut le consommateur du produit ou de la marque? Bien sûr, il
veut répondre à son besoin en consommant le produit ou la marque. Si le produit est capable
de satisfaire son besoin de base avec un niveau de qualité acceptable alors le consommateur
est satisfait, il va continuer à consommer ce produit ou cette marque (Jacoby & Kyner, 1973).
Mais maintenant le monde est devenu un village global et les consommateurs peuvent trouver
une gamme de marques différentes dans presque toutes les catégories pour remplir le même
désir avec le niveau de qualité requis. La recherche souligne qu'en dehors de la satisfaction
des besoins de base, les consommateurs veulent aussi se représenter lui / elle-même en
consommant un certain produit ou marque (Aaker, 1999; Grubb & Harrison L. Grathwohl,
1967; Levy, 1959; Sirgy, 1982; Solomon, 1983). Pour cette raison les responsables marketing
créent une image acceptable de leur produit / marque avec laquelle les consommateurs
peuvent sentir une affiliation et peuvent tirer une signification symbolique.
Les responsables marketing sont toujours intéressés par l'acceptation de leur marque sur le
marché, de l’augmentation de leur part de marché et de la maximisation du profit. Pour que
leurs produits / marques soient acceptés par le consommateur et pour augmenter leur part de
marché ils doivent offrir des produits / marques avec un niveau de qualité acceptable pour
répondre aux besoins du consommateur. Depuis que le monde est devenu une communauté
279
Appendices
globale la technologie est à son apogée, et dans cette ère technologique, presque toutes les
entreprises/organisations ont la capacité de satisfaire le besoin fondamental du consommateur
avec un niveau de qualité acceptable.
Alors la question qui se pose que doit faire une entreprise pour l'acceptation du
produit/marque? Ils doivent différencier leur produit/marque des autres disponibles. Pour cela,
ils doivent connaître le soi du consommateur et de créer une image acceptable. Un grand
nombre de recherches ont été effectué sur les facteurs positifs qui aident les responsables
marketing à atteindre leurs objectifs. Parallèlement, il est tout aussi important de connaître les
facteurs négatifs qui créent problème dans l'acceptation de leur produit. Dans notre étude,
l’objectif est d'explorer ces raisons et facteurs qui restreignent le consommateur d'utiliser le
produit / marque au-delà des attributs qui lui sont liés. Plus précisément, quels sont les
déterminants pré-achats de l'évitement de marque?
9.1.3
Importance de l'étude
En explorant explicitement les raisons pour lesquelles les consommateurs tendent à éviter les
marques avant de les consommer, en effet cette thèse contribuera à l'évitement de marque de
différentes manières. Tout d'abord, d'un point de vue managérial, il est très important pour les
gestionnaires de savoir pourquoi les consommateurs ne sont pas prêts à acheter leur produit
pour la première fois; bien qu'ils aient rempli les pré-requis de l'achat sur le prix, niveau de
qualité services après-vente, etc . Deuxièmement, d'un point de vue académique, une quantité
importante de la littérature est disponible sur l'attitude positive à l'égard de la marque, mais,
sur le côté négatif, la littérature n'est pas encore suffisante pour répondre à la question des
gestionnaires. Donc, dans cette présente thèse, un cadre conceptuel est élaboré en réunissant
la documentation existante disponible dans le domaine de l’anti-consommation. Notre étude
propose, pour la première fois, un cadre global fondé sur les conclusions théoriques et
280
Appendices
empiriques du phénomène de l’évitement de marque au pré-achat. La connaissance des
attitudes et des comportements de l’évitement de la marque au pré-achat contribue
théoriquement ainsi que pratiquement à un domaine intéressant et sous-étudié. Le point de
vue traditionnel des responsables du marketing se concentre sur la persuasion du
consommateur pour choisir leurs marques. Alors que cette recherche démontre que le
consommateur tend à éviter certaines marques avant de les consommer et confirme qu’il est
essentiel de prendre connaissance du côté négatif de la marque. Les efforts limités ont été
déployés sur ce côté .Thompson et al. (2006) ont souligné l'importance de l'évitement de la
marque et ont expliqué que le résultat d'ignorer ou de ne pas faire attention aux attitudes et
comportements de l’évitement de la marque peut conduire à une «crise de marque». Cela
peut aussi conduire à une perte non identifiée dans les bénéfices, une réduction de la société
ou des parts de marché de la marque, ainsi qu’une dégradation de la réputation de l'entreprise
(Lee, 2007).
Les entreprises conscientes de cet effet négatif de la marque peuvent prendre avantage sur
leurs concurrents en adoptant des stratégies appropriées pour faire face à des consommateurs
qui ont des attitudes et des comportements négatifs au pré-achat envers leur marque.
L'objectif du marketing relationnel est de comprendre le consommateur et de construire une
relation avec le client, donc les entreprises et les responsables marketing doivent le prendre
comme un objectif à long terme et travailler sur ces consommateurs et atténuer ces sentiments
négatifs qui éprouvent au sujet de leur marque /société.
9.1.4
Objectifs académiques et pratiques
La recherche dans l’anti-consommation est en croissance et se concentre principalement sur le
côté de «contre» la consommation. Notre étude fournira une approche pertinente dans la
compréhension du comportement du consommateur et en général elle aidera dans le
281
Appendices
discernement des habitudes de consommation. L'évitement de la marque est le phénomène qui
étudie le comportement d'évitement des consommateurs envers certaines marques, même s’ils
ont la capacité financière de les consommer (cf. définition de l’évitement de la marque). Par
conséquent, l'évitement de la marque pourrait être un lien entre la recherche dans la marque
et l'anti-consommation. Dans la littérature de la consommation symbolique, de nombreuses
études ont conclu que cognitivement le consommateur a une compréhension plus claire de ce
qu'il ne veut pas, contrairement à ce qu'il veut ou désire (Banister & Hogg, 2001, 2004; Hogg,
1998; Hogg, Cox, & Keeling, 2000; Hogg & Michell, 1996; Ogilvie, 1987). De la même
manière, certains chercheurs ont fait valoir que ce qu'un consommateur n'aime pas est tout
aussi important pour sa prise de décision que ce qu’il aime, et que les informations négatives
sur les marques a le même effet que les informations positives (Aaker, 1996; Bourdieu, 1984
; Herr, Kardes, & Kim, 1991; Lutz, 1975; Wilk, 1994, 1997).
La littérature de la formation d’impression indique que les informations négatives ont un
impact plus fort sur les impressions que les informations positives (Kanar, Collins, et Bell,
2010). Les gens comptent plus sur les informations négatives, car elles sont plus de nature
diagnostique et ont plus de valeur dans la discrimination entre les jugements que les
informations positives (Skowronski et Carlston, 1989). Ces perspectives soulignent
l'importance de l'évitement de la marque, mais la motivation derrière l'étude de l'évitement de
la marque est de démontrer que les gens peuvent se définir avec leurs goûts de consommation,
mais ils peuvent aussi obtenir l'image souhaitée en restant loin des choses qu'ils ne veulent
pas consommer ou évitent de consommer (Wilk, 1996). Aujourd’hui dans l’environnement du
marketing, la préoccupation du consommateur dans la sélection des marques a augmenté.
Comme le nombre de marques augmentent sur le marché le consommateur est de plus en plus
sceptique et difficile à satisfaire (Firat et Venkatesh, 1995; Fournier, Dobscha, & Mick, 1998;
282
Appendices
Zavestoski, 2002a). La publicité massive a contesté la fidélité à la marque et a engendré un
changement de marque fréquent (Aaker, 1996; Banister & Hogg, 2001; Rumbo, 2002).
Les attitudes et les comportements du consommateur sont contre les désirs des marketers.
Nous avons vu ces contres comportements dans le passé, sous la forme de protestations
(Kozinets & Handelman, 2004), les boycotts (Klein et al, 2004;. Kozinets & Handelman,
1998), «culture jamming» (Rumbo, 2002; Thompson &Arsel, 2004), et «BurningMan»
(Kozinets, 2002). La plupart du temps, les consommateurs réguliers rejettent ces
produits /marques qui ne présentent pas une congruence avec leur propre soi (Fournier et al,
1998; Hogg, 1998) ou les marques /produits ou qui ne parviennent pas à répondre à leurs
attentes (Halstead, 1989; Oliver, 1980).
9.2
Cadre théorique et hypothèses
9.2.1
Définition de l’Evitement de la marque
Le terme « évitement de la marque » sera discuté ici de manière à spécifier les paramètres de
l’étude. La littérature en psychologie a étudié en détail les motivations à l’approche et à
l’évitement. Elliot (1999) nous apprend qu’en matière de motivation à l’approche, le
comportement
est
« initié
ou
dirigé
par
un
évènement
ou
une
possibilité
positive/désirable » et qu’en ce qui concerne la motivation à l’évitement, le comportement est
« initié ou dirigé un évènement ou une possibilité négatif/indésirable ». Le concept
d’évitement de la marque n’a fait l’objet d’une définition qu’à partir des travaux de Lee
(2007) et se définit de la façon suivante : « Le rejet conscient, délibéré et actif d’une marque
que le consommateur peut s’offrir, cela dû à la signification négative associée à la marque »
Aussi, en tenant compte de la définition psychologique de l’évitement, les caractéristiques de
l’évitement de la marque peuvent-ils être définies comme « les attitudes et les comportements
du consommateur envers une marque, motivé par la signification et les conséquences
283
Appendices
négatives associées à la marque » Une étude approfondie de la littérature met en avant
l’existence d’études en nombre limité telles que celles de Thompson et al. (2006) et d’Oliva et
al. (1992) qui utilisent explicitement le terme d’ « évitement de la marque ». L’évitement de
la marque a été décrit comme l’antithèse de la fidélité à la marque (Oliva et al. 1992). Il
propose d’ailleurs que la satisfaction
conduit à la fidélité à la marque alors que
l’insatisfaction mène à l’évitement de la marque ou au changement de marque. Les deux
concepts étant des phénomènes différents. Le terme d’évitement à la marque a également été
discuté par Thompson et al. (2006). Ils utilisent le terme ‘Doppelgänger brand image’ qu’ils
définissent comme famille d’images et d’histoires désobligeantes sur une marque qui s’ancre
dans la culture populaire par un réseau librement organisé de consommateurs activistes antimarque, de bloggers, et de leaders d’opinion et des médias d’information et de
divertissement ». Leur étude suggère que les promesses faites aux consommateurs sur les
marques et qui reposent sur des fondements émotionnels sont vues comme non-authentiques.
En conséquence de quoi, les consommateurs commencent à éviter la marque, ceci conduisant
les gestionnaires à en étudier les raisons et trouver des solutions visant à « guérir » leur
marque. Les concepts proposés par Narayana and Markin (1975) sont pertinents pour notre
étude. Ils proposent trois ensembles sous la conceptualisation du comportement du
consommateur. Ils divisent en deux la décision faite par les consommateurs concernant
l’ensemble des marques disponibles sur le marché : l’ensemble conscient et l’ensemble
inconscient. Si la marque appartient à l’ensemble inconscient, les chances d’être vu comme
telle marque sont nulles à ce moment précis. Mais si la marque appartient à l’ensemble
conscient, elle a des chances d’être considérée comme telle. Narayana and Markin (1975)
identifient également trois ensembles : l’ensemble évoqué, l’ensemble inerte et l’ensemble
inepte.
284
Appendices
9.2.2
Approche qualitative
Notre objectif de l'étude est d'explorer les raisons du comportement des consommateurs pour
éviter les marques. Donc compatible avec la recherche précédente en utilisant un
phénoménologique (Patterson, Hill & Maloy, 1995) approche, nous avons utilisé des
entrevues personnelles comme la technique de collecte de données primaires (Hirschman,
1992; Thompson, Locander, & Pollio, 1989) pour comprendre du consommateur processus de
pensée et d'explorer les raisons de sa décision prise de ne pas acheter certaines marques.
9.2.3
Méthode d'analyse
En raison de la nature exploratoire de la recherche des entrevues en profondeur sont utilisés
pour explorer les raisons sous-jacentes des comportements des consommateurs d'évitement de
la marque. Un groupe d'auto-sélectionné de treize (8 hommes, 5 femmes), les participants sont
interviewés jusqu'au point de saturation où le thème de départ à répétition (Taylor & Bogdan,
1998).
Les données sont analysées en utilisant QSE NVivo 8. C'est un outil d'analyse puissant qui
aide les chercheurs dans la gestion et l'analyse des données qualitatives telles que les
transcriptions des entrevues. Conformément à la conception de l'entrevue classique en
profondeur, un guide est mis au point pour obtenir la réponse sur les questions de recherche
principale.
9.2.4
Thèmes émergents
Voici les thèmes ressortent de l'analyse des entrevues :

La Congruence du soi non désirée

L’influence sociale négative

Animosité
285
Appendices

Familiarité avec le Pays d’origine

Ethnocentrisme du consommateur

Citoyenneté non éthique perçue de l’entreprise

Risque perçu
9.2.5
Les déterminants pré-achats de l’évitement de la marque
A partir de notre étude qualitative, fondée sur des entretiens en profondeur (h=8 et f=5) ; nous
avons trouvé plusieurs facteurs prédisant l’attitude d’évitement de la marque. Dans notre
étude nous identifions parmi les déterminants, deux catégories : les déterminants pré-achats et
post-achats. Les premiers reposent sur des besoins personnels, sociaux et sociétaux ; alors que
les seconds sont liés à des expériences personnelles que le consommateur a avec le produit ou
la marque (Lee, Motion, et al., 2009).
Notre étude est centrée autour des raisons qui motivent les consommateurs à ne pas
consommer une marque particulière au-delà des attributs du produit. La littérature identifie
une pléthore de raisons qui contribue à la formation d’attitudes négatives constituant un frein
à l’usage de la marque.
9.2.6
La congruence du soi non désirée
Les consommateurs forment leur concept de soi et identifient leurs groupes sociaux de
référence à travers ce qu’ils choisissent de ne pas consommer comme ce qu’ils choisissent de
consommer (Banister & Hogg, 2001; Englis & Solomon, 1997; Hogg, 1998; Hogg et al.,
2000). Englis and Solomon (1995) postulent que « les consommateurs peuvent éviter l’achat,
la possession et l’usage de tels produits ou activités, cela dû à la réticence d’être associé à un
groupe objet qui veulent éviter ». En ce sens, les consommateurs définissent le « non moi » en
refusant l’ « anti-constellation » afin de rester à l’ écart d’une association avec le dit
286
Appendices
stéréotype (Hogg, 1998; Hogg & Michell, 1996). L’anti-constellation « représente la
complémentarité des choix négatifs à travers une catégorie de produits multiples (Hogg,
1998).
Le concept du soi non désiré a été introduit par Ogilvie (1987), le décrivant comme l’identité
la moins désirée. Il consiste en la somme des traits dont la valence est négative, des souvenirs
d’expériences déplaisantes, des situations embarrassantes, des évènements effroyables et des
émotions non voulues, l’individu est constamment motivé par l’évitement (Bosnjak &
Rudolph, 2008). Le soi non désiré est un construit psychologique et pertinent dans le cadre de
l’évitement de la marque. Ogilvie (1987) suggère que le soi non désiré est un ensemble
d’associations et de valeurs avec lequel les personnes ne souhaitent pas être confondues en
raison du soi négatif que cela engendrerait. D’autres chercheurs soutiennent qu’en
comparaison à ce que les personnes veulent être, ils ont une idée assez claire de ce qu’ils ne
veulent pas être, et s’écarter du soi non désiré de cette façon peut être plus efficace que de
tenter de s’approcher du soi idéal (Lee, 2007; Ogilvie, 1987). Une étude précédente
mentionne la réponse d’un participant : « tu peux t’en sortir sans avoir une parfaite opinion
positive simplement en évitant d’avoir une image négative » (Banister & Hogg, 2001).
En réalité, nous ne pouvons différencier la consommation de la non-consommation. Il y a
deux types de choix : les choix fondés sur le souhait et le besoin et les choix fondés sur
l’aversion et le dégoût. Néanmoins, les deux ne sont pas des alternatives ou des opposés. Les
deux ensembles peuvent être vus comme étant alternativement dépendant de valences
socialement ou individuellement positives ou négatives, comme étant matérialistes ou
spirituels, actifs ou passifs (Wilk, 1996). De la même façon, ne pas avoir, ne pas porter, ne
pas manger peuvent être une forme de conformité ou de distinction sociale comme le sont ;
287
Appendices
le fait d’avoir, de porter et de manger. Dans cette étude, nous posons les hypothèses suivantes
pour s’intéresser aux relations entre le soi non désiré et l’évitement de la marque.
H1 : La congruence du soi non désiré influence positivement l’attitude d’évitement de la
marque du consommateur.
9.2.7
L’influence sociale négative
Les consommateurs croient qu’ils sont capables de construire leurs propres opinions mais ce
que les autres pensent peut aussi les influencer, les changements sont plus particulièrement
susceptibles de se produire sous l’effet des personnes avec des attitudes négatives (Duhacheck
et al., 2007). Les attitudes des consommateurs envers les produits/marques sont influencées
par les autres autour d’eux. Dans l’ère actuelle des réseaux sociaux, la société devient plus
interconnectée via les réseaux en ligne tels que Facebook, Twitter et Myspace. Cette influence
devient même un conducteur significatif
des attitudes des consommateurs individuels
(Duhacheck et al., 2007).
L’influence sociale est définie comme le degré auquel les membres d’un groupe social de
référence influencent les comportements et les expériences des uns et des autres par la
pression sociale pour réaliser un comportement spécifique (Kelman, 1958; Kulviwat et al.,
2009). De même, dans cette étude le terme influence sociale négative peut être considéré
comme
le degré auquel les membres
d’un groupe de référence influencent les
comportements et les expériences des uns et des autres par la pression sociale pour ne pas
réaliser un comportement spécifique. Dans leur étude, Duhacheck et al. (2007) soutiennent
que les opinions négatives ont une influence particulièrement forte sur les attitudes des
autres. Ils soutiennent également que les gens avec des opinions sur des produits ont tendance
à devenir plus négatifs dans des discussions de groupe.
288
Appendices
Richins (1983, 1984) discutent l’importance et l’impact du bouche à oreille négatif (negative
word of mouth NWOM) qui le définit comme une communication interpersonnelle entre des
consommateurs concernant une organisation marketing ou un produit, qui dénigrent l’objet de
cette communication (1984, p.697). L’auteur souligne que le NWOM continue et organisé par
un large groupe de consommateur qui n’est pas satisfait d’un produit, d’une marque, d’un
service a un impact important sur le choix du consommateur. Dans notre étude, nous
cherchons à savoir si les effets de l’influence sociale négative envers l’évitement de certaines
marques. Donc nous posons l’hypothèse suivante :
H2 : L’influence sociale négative influence positivement l’attitude d’évitement de la marque
du consommateur.
9.2.8
Animosité
La littérature concernant l’animosité a établi que les consommateurs évitent les produits et les
marques de certains pays, non pas par rapport a leur qualité, mais par rapport à l’animosité
qu’ils ressentent envers leur pays d’origine (Charles, 2012; Klein et al., 1998). A l’origine de
cette attitude nous pouvons avoir plusieurs raisons comme l’histoire d’un conflit, les positions
politiques, religieuses, ethniques ou des différences culturelles, ainsi que des actes
de
violence ou terrorisme assimilés à un pays. Ces raisons peuvent engendrer des sentiments
d’animosité des consommateurs d’un pays envers un autre pays spécifique et entrainer alors
une restriction des la consommation des produits ou services originaires de ces derniers.
Klein et al. (1998), introduisent le concept de « Animosité » à la littérature en développant et
en présentant le modèle d’animosité pour l’achat des produits étrangers, qu’ils ont testé et
validé avec l’attitude des consommateurs chinois envers les produits fabriqués au Japon. Ils
définissent le construit d’animosité comme « les résidus d’antipathie liée à des évènements
présents ou passés dans les contextes militaires, politiques ou économiques », qui affectent le
289
Appendices
comportement des consommateurs dans les marchés internationaux. Dans le contexte actuel
de globalisation économique, les compagnies ont plus d’opportunités pour livrer leurs
produits aux autre pays, et au même temps, les consommateurs de ces pays ont la liberté de
choisir les produits et les marques qu’ils consomment.
Dans le modèle original (Klein et al. 1998) les effets des avis et des attitudes des
consommateurs n’ont pas été pris en compte, même si des autres recherches prouvaient que
l’animosité joue un rôle dans la réponse évaluative des consommateurs (Russell & Russell,
2010). Shoham et al. (2006) étudient l’opinion des consommateurs israéliens juifs envers les
produits israéliens arabes, leur résultats indique que l’animosité impacte également les
jugements de la qualité des produits et la propension à l’achat. Similairement, Yang and TSO
(2007) montrent que l’animosité envers la Chine est significativement liée à l’attitude des
consommateurs taiwanais envers les programmes de télévision originaires de Chine. Par
rapport à la discussion précédente, qui concerne l’animosité envers le pays d’origine, nous
proposons l’hypothèse suivante :
H3 :
L’animosité perçue des consommateurs envers le pays d’origine influence
positivement l’attitude d’évitement de marque du consommateur.
9.2.9
Citoyenneté non éthique perçue de l’entreprise
Le rôle des compagnies dans la société a été le sujet de plusieurs discussions depuis le milieu
du dernier siècle (Turker, 2009). Les préoccupations par rapport au bien être de l’humanité et
de l’environnement ont augmenté considérablement parmi les consommateurs de tout le
globe. Aujourd’hui, les parties prenantes des communautés nationales et internationales
plaident pour une utilisation plus responsable et plus équilibrée des entreprises. Dans la
littérature ce phénomène est appelé par certains chercheurs Responsabilité Sociale de
l’Entreprise (RSE), autres auteurs l’appel étique de l’entreprise, et dans ces derniers temps la
290
Appendices
performance sociale des entreprises a été appelé « Citoyenneté de l’Entreprise » (Carroll,
1998).
Carroll (1998) divise la Citoyenneté de l‘Entreprise en quatre parties : légale, économique,
éthique et philanthropique. De la même façon, Esrock et Leichty (1998) soulignent que la
RSE a quatre éléments centrales qui sont, « des pratiques morales, honnêtes et justes ; des
produits sûres et fiables,
le bon traitement des employés ; et l’amélioration de
l’environnement ».
Comment l’entreprise est vue et évaluée par les parties prenantes constituent la base de cette
interaction. Même si le concept de la Citoyenneté de l’Entreprise est habituellement employé
par les hommes d’affaires il n’est pas encore formalisé par les académiciens. La recherche
dans le domaine de la RSE suggère que la citoyenneté de l’entreprise doit être défini comme
« l’étendu dans lequel les entreprises doivent assumer les responsabilités dans les domaines
économiques, légaux, éthiques et discrétionnaires qui s’impose à leur égard par leurs parties
prenantes. (Maignan & Ferrel, 2001). Encore plus importante, est l’idée de savoir si
l’entreprise ou organisation est perçue comme étant éthique et socialement responsable, et de
savoir si les gestionnaires de l’entreprise s’occupent de ce sujet. (Freeman, 1994; Langtry,
1994; Schlossberger, 1994). Nous supposons que quand un consommateur perçoit la
citoyenneté de l‘entreprise comme non éthique, il serait motivé a éviter cette compagnie ou
leurs produits et marques. Alors, notre hypothèse est :
H4 :
Une mauvaise perception de la citoyenneté de l’entreprise de la part du
consommateur (PUCC) influence positivement l’attitude envers l’évitement de marque.
291
Appendices
9.2.10 Risque perçu
Etant donné que les résultats des échanges commerciaux sont incertains, les consommateurs
visent à réduire les risques dans leurs achats. Depuis les années 60 (Bauer, 1960), la théorie
du risque perçu a été mobilisée pour expliquer le comportement du consommateur. La
littérature présente plusieurs définitions du concept de risque perçu. Peter et Ryan (1976) le
définit comme une perte subjective attendue et Featherman and Plavlou le définit comme une
perte possible quand l’individu cherche à atteindre un résultat désiré (Lee, 2009). Plusieurs
chercheurs ont étudie l’impact de ce risque dans le processus traditionnel de prise de décision
des consommateurs (Lin, 2000 ; Mitra, 2008 ; Reis & Capella, 1999). Certains chercheurs
soutiennent que le risque perçu est un concept bidimensionnel qui est lié à la fois à
l’incertitude de la décision d’achat et aux conséquences qui émanent du fait de prendre une
action adverse (Bettman, 1973; Cunningham, 1967; Mitra et al., 1999; Schiffman, 1972).
Notre recherche est alignée avec celle de Stone et Grønhaug (1993), nous divisons le concept
de risque perçu en plusieurs domaines qui sont : le financier, de performance, physique, social
et psychologique, qui se définit comme :

Risque financier: la probabilité que le résultat d’un achat soit une perte d’argent ou
un coût soutenu de manutention du produit.

Risque de performance: la possibilité qu’un produit ne fonctionne pas comme prévu
ou au niveau attendu et qu’il n’arrive pas à livrer les résultats attendus.

Risque physique: la probabilité qu’un produit acheté puisse être la source d’une
menace à l’intégrité physique ou à la vie d’une personne.

Risque général: une mesure générale du risque perçu quand toutes les mesures sont
prises en compte au même temps.
292
Appendices
9.2.11 Attitude d’évitement de marque et Intention d’évitement de marque
La littérature concernant le comportement du consommateur (Fishbein & Ajzen, 1975; Lutz,
1975) soutient que l’attitude du consommateur influence leurs choix. Selon Fishbein & Ajzen
(1975) l’attitude d’un individu est le réflexe de ses sentiments positifs ou négatifs à l’égard de
la réalisation d’un comportement particulier. L’intention est définie à leur tour comme
l’indication du niveau d ‘effort requis pour qu’un individu soit prêt à réaliser
un
comportement (Ajzen, 1991). Nombreuses études passées ont fournies l’évidence d’une
relation positive entre l’attitude et l’intention (Povey et al., 2001; Verbeke & Vackier, 2005).
L’attitude des consommateurs envers les marques joue un rôle significatif dans l’influence
de leur intention d’achat (Phau, Teah,&Lee, 2009). Une prise de décision non éthique, et
même l’achat d’une marque qui fait des publicités considérées comme peu éthiques, est
majoritairement expliqué par l’attitude, sans tenir en compte du type du produit (Phau et al.,
2009 ; Wee, Ta &Cheok, 1995).
Si les consommateurs ont une attitude favorable ou positive à l’égard de certaines marques
ils préféreront acheter ces marques, de l’autre côté, si les consommateurs ont une attitude
négative ou défavorable envers les marques, ils auront moins tendances à les acheter (Wee et
al., 1995). Suite à cette discussion nous présentons l’hypothèse suivante:
H6 :
L’attitude d’évitement de marque des consommateurs est positivement
liée
à leur
intention d’évitement de marque.
9.2.12 Variables modératrices
Nous discutons l’effet modérateur de deux variables (Familiarité avec le pays d’origine et
l’ethnocentrisme du consommateur) dans la relation entre l’animosité et l’attitude de
l’évitement de marque.
293
Appendices
9.2.13 Familiarité avec le pays d’origine
Dans cette période de globalisation économique, les compagnies ont l’opportunité de
distribuer leurs biens aux consommateurs à travers le monde en entier. Du point de vue du
consommateur, cela leur donne plus d’options en ce qui concerne leur décision parce qu’ils
ont une gamme plus grande de produits et services dans presque toutes les catégories. Alors,
le pays d’origine (COO) est devenu un facteur clé à tenir en compte si nous voulons étudier
l’évaluation que les consommateurs font des produits étrangers (Jiménez & Martín 2010). Les
chercheurs qui s’intéressent à étudier les effets des pays d’origine ont plusieurs résultats, cela
est peut-être du à la diversité des domaines, conceptions et contextes dans lesquelles ces
recherches ont été réalisées. Dans un sens plus large, le Pays d’origine est un élément de
l’image qui est construite par les caractéristiques du produit (comme l’innovation, la
technologie, la fiabilité, le prix, la qualité générale, les produits typiques) et les associations
aux caractéristiques du pays (Jiménez & Martin, 2010).
Quant à d’ autres études, le Pays d’origine est décrit comme un aspect de la marque qui donne
une idée aux consommateurs pour l’identification du siège de l’entreprise non pas le pays ou
le produit a été fabriqué (Ahmed et al., 2004; Shankarmahesh, 2006; Thakor, 1996).
Des recherches précédentes ont montré que la familiarité influence le processus de prise de
décisions des consommateurs, surtout les évaluations des produits du pays d’origine
(Moorman et al., 2004). Dans leur étude des automobiles en Espagne, Jiménez et Martin
(2010), trouve un effet modérateur significatif de la familiarité avec le pays d’origine des
produits et de la marque dans la relation entre l’animosité et la confiance envers l’entreprise.
Les consommateurs puissent ne pas prendre en compte une marque étrangère non connu parce
qu’ils font des inférences négatives concernant la qualité du produit (Han, 1990). Nous
n’avons trouvé aucune évidence dans la littérature
294
concernant l’effet modérateur de la
Appendices
familiarité et du pays d’origine dans la relation entre l’animosité et l’évitement de la marque.
Par rapport à cette discussion nous postulons l’hypothèse suivante :
H7 :
La familiarité avec le Pays d’origine modère la relation entre l’animosité perçue et
l’attitude envers l’évitement de la marque ; plus la familiarité avec le pays d’origine est forte
plus la relation est forte, et vice versa.
9.2.14 Ethnocentrisme du consommateur
L’ethnocentrisme est un concept très ancien (Sunner, 1906), les consommateurs sont
considérés comme « ethnocentriques » quand ils ont une tendance à acheter les produits
fabriqués localement (Sharma, Shimp & Shin 1995). Shimp &Sharma (1987) ont introduit
l’échelle de mesure de la tendance ethnocentrique du consommateur (CETSCALE). Ils
appliquent l’ethnocentrisme à l’étude du marketing et au comportement des consommateurs,
ils ont mis au jour le concept de « tendances ethnocentriques du consommateur » (CET) pour
représenter les croyances que les consommateurs ont par rapport à la pertinence et la moralité
de l’achat des produits fabriqués à l’étranger.
L’ethnocentrisme des consommateurs a été étudié vastement comme modérateur dans
plusieurs contextes comme les Etats-Unis (Han, 1988; Klein & Ettenson, 1999; Shimp &
Sharma, 1987), la Chine (Klein et al., 1998), la Corée (Sharma et al., 1995), la France,
l’Australie et le Mexique (Clarke et al., 2000) et la Turquie (Balabanis et al., 2002).
Les consommateurs qui ont des convictions ethnocentriques fortes ont une tendance plus
prononcée à évaluer négativement les produits étrangers si nous le comparons à de
consommateurs qui n’ont pas ces convictions. Ceux qui considèrent qu’il est mauvais
d’acheter des produits étrangers, perçoivent ses produits comme étant d’une qualité inférieure
que les produits fabriqué localement ; les consommateurs ethnocentriques préfèrent les
295
Appendices
produits locaux non pas seulement par convictions économiques ou morales, mais aussi parce
qu’ils sont persuadés que les produits locaux sont les meilleurs produits disponibles. Il est
possible aussi qu’un consommateur ressente de l’animosité envers un pays sans forcement
considérer que les produits de ce pays sont de mauvaise qualité. Ce consommateur peut ne pas
vouloir acheter des produits venant du pays pour lequel il ressent de l’animosité, mais croit
toujours en sa bonne qualité. Par rapport à cette discussion, nous présentons l’hypothèse
suivante :
H8 :
L’ethnocentrisme du consommateur modère la relation entre l’animosité perçue et
l’attitude d’évitement de marque ; plus l’ethnocentrisme est élevé plus la relation est forte et
vice versa.
9.2.15 Caractéristiques sociodémographiques
Notre étude tient en compte quatre caractéristiques sociodémographiques du consommateur
(sexe, âge, éducation et ressources mensuelles de la famille) utilisées en tant que prédicateurs
de l’attitude
et du comportement à l’égard de l’évitement de marque.
296
Appendices
9.2.16 Modèle de recherche
En tenant comme base les discussions précédentes nous proposons le modèle de recherche
suivant pour être testé empiriquement à travers une enquête en ligne:
Pre-Purchase Factors
Familiarity with
COO
Undesired Self
Congruence
H1
Negative Social
Influence
H2
Perceived
Animosity
Perceived Unethical
Corporate
Citizenship
H7
H3
Brand Avoidance
Attitude
H4
H5
H8
Perceived
Risk
Ethnocentrism
Figure a : Modèle de recherche
297
H6
Brand Avoidance
Intention
Appendices
9.3
Méthodologie de recherche
9.3.1
Contexte de recherche
Le contexte principal de notre recherche est le Pakistan, basé sur la facilité des chercheurs
dans la collection des données. Pour tester le modèle proposé dans plusieurs marchés et
valider des résultats, la recherche est réalisée en Nouvelle Zélande. Pourquoi avons-nous
choisi la Nouvelle Zélande ? Le premier objectif est de collecter les données du pays où
l’anglais est la langue maternelle parce que la traduction des questionnaires en d’autre langue
peut causer des problèmes. La deuxième raison est l’obtention de la diversité dans les résultats
en considérant un pays émergent (Pakistan) et un pays développé (Nouvelle Zélande).
9.3.2
Plan de recherche
L’étude présente est réalisée en deux phases :

La première phase: l’étude qualitative (interviews approfondies) a pour l’objectif de
comprendre le concept dans le contexte Pakistanais.

La deuxième phase: l’étude quantitative a été réalisée en Pakistan et en Nouvelle
Zélande en utilisant le design transversal avec l’enquête en ligne.
9.3.3
Collection des données
Nous collectons l’échantillon de 286 consommateurs de Pakistan et de 199 consommateurs de
Nouvelle Zélande. Le questionnaire final inclut également des variables sociodémographiques : genre, âge, niveau d’étude (le plus haut niveau atteint), et le revenu
mensuel familial. Le questionnaire complet est mis en ligne avec une lettre de motivation qui
explique l’objectif de cette recherche et assure l’anonymat des personnes sondées, a été créé
en utilisant le service gratuit, fourni par Kwiksurveys (http://www.kwiksurveys.com/). Le lien
de l’enquête a été envoyé aux sondés potentiels par deux façons:
298
Appendices

Par mail

Par le système social comme des sites sociaux de média (Facebook and Twitter)
La technique de boule de neige est utilisée dans la collecte des données, c’est une méthode
simple et efficace mais la représentativité de l’échantillon est un problème que le chercheur
doit compromettre.
9.3.4
Dimension de construction
Dans cette partie, nous avons donné des détails des instruments de recherche utilisée pour
mesurer des variables impliqués dans cette étude:
299
Table a: Questionnaire Items with References
S.No.
Reference
S.No.
Cheng et al. 2006
1
Familiarity with Country-of-Origin (COO)
Laroche et al. 2005
1
I think that using that brand is not a good idea
2
3
Cheng et al. 2006
1
Klien et al. 1998
4
I think that using that brand is not pleasant
Cheng et al. 2006
2
Klien et al. 1998
3
Klien et al. 1998
Self made
Reference
Animosity
1
Baker and
Churchill 1977
4
2
Baker and
Churchill 1977
5
3
I would not prefer to buy that brand over any other brand
Baker and
Churchill 1977
Self made
1
2
I associate that brand with something unappealing, which I do not
want to become
Self made
3
I do not want to be associated with the type of people using that
brand
Self made
4
5
Self made
Klien et al. 1998
Klien et al. 1998
1
The company does not participate in activities which aim to protect and improve
the quality of the natural environment
2
The company does not contribute to campaigns and projects that promote the
well-being of the society
Turker 2009
3
The company does not emphasize the importance of its social responsibilities to
the society
Turker 2009
4
The company does not respects consumer right
5
The company does not provide full and accurate information about its products
and its customers
Self made
300
Turker 2009
Turker 2009
Turker 2009
Perceived Risk
1
Kulviwat et al 2009
1
Buying of that brand is a financial risk for me because of such things as its poor
warranty, high maintenance costs, or highly monthly payments.
2
Kulviwat et al 2009
2
Buying of that brand is a performance risk for me because it would run
extremely poorly.
3
People who influence my behavior think that I should not use that
brand
Gupta et al. 2008
3
Buying of that brand is a physical risk for me because it would hurt me
4
Buying of that brand is too risky in general for my liking
1
It is not right to purchase foreign products because it puts our
people out of job
Shimp and Sharma
1987
2
We should purchase products manufactured in our country instead
of letting other countries get rich off us
Shimp and Sharma
1987
3
We should not buy foreign products because this hurts our
business and cause unemployment
Shimp and Sharma
1987
4
We should buy from foreign countries only those products that we
cannot obtain within our own country
Shimp and Sharma
1987
301
Peter & Tarpey
1975
Peter & Tarpey
1975
Peter & Tarpey
1975
Self made
Appendices
9.4
Analyse des données et les résultats
L’analyse des données de deux échantillons différents (échantillon 1: Pakistan & échantillon
2: Nouvelle Zélande) est représentée séparément pour la précision. Elle est divisée en 3
parties: la première partie aborde le nettoyage préliminaire des données. La deuxième partie
approche l’analyse factorielle confirmatoire (AFC), en utilisant la modélisation d’équations
structurelles (MES) pour analyser la dimension des propriétés et la validité convergente et
discriminante des échelles. Par la suite, le modèle structurel de la régression est analysé pour
tester les hypothèses proposées. La deuxième partie du chapitre explique les effets modérés de
la familiarité des consommateurs avec le pays d’origine et l’ethnocentrisme des clients. Après
le nettoyage préliminaire des données, nous avons des échantillons finaux (l’échantillon de
Pakistan = 278 et de Nouvelle-Zélande = 198) pour plus de l’analyse.
Nous avons aussi rassemblé les données de deux pays (476=278+198). Le poids de
l’échantillonnage a pour l’objectif de réduire l’effet de dissymétrie démographique des
échantillons et des matrices de covariances sont utilisées comme des inputs dans l’AMOS
après application le poids dans SPSS. Nous évaluons l’invariance des dimensions entre les
deux échantillons avant de les rassembler.
9.4.1
Procédure et méthodes analytiques
La modélisation d’équation structurelle (SEM) est utilisée principalement pour tester la
relation hypothétique de ce modèle de recherche proposé. Ensuite, ce sont des index
convenables utilisés pour l’évaluation de l’adéquation de ce modèle :

Le Chi-deux (χ2/df) est un test statistique avec le degré liberté correspondant et un
niveau de signification (Kline, 2010)

Le Chi-deux normalisé (χ2/df)

L’indice d’ajustement comparative [CFI;(Bentler, 1990)]
Appendices

Tucker-Lowis Index [TLI; (Bentler & Bonett, 1980)]

La moyenne quadratique résiduelle standardisée [SRMR; (Kline, 2010)]

L’erreur de la moyenne quadratique de l'approximation [RMSEA;(Steiger, 1990)]
9.4.2
L’analyse factorielle confirmatoire
Pour la simplicité, une AFC séparée pour les variables dépendantes et indépendantes est
pratiquée pour tester les propriétés des mesures des structures latentes sous-jacentes. Nous
analysons également et établissons la validité convergente et discriminante de toutes les
mesures.
9.4.3
Hypothèses testées et résultats de l’étude
Après avoir développé le modèle des mesures acceptables et établi la validité convergente et
discriminante pour ces deux échantillons, nous confirmons le modèle structurel selon nos
hypothèses proposées, et dans la partie prochaine nous discutons les résultats des hypothèses
testées.
9.4.4
Effets sociodémographiques
Nous vérifions les effets des variables sociodémographiques (genre, âge, éducation et revenue
mensuel familial) sur la variable dépendante. Nous vérifions séparément l’effet dans les deux
échantillons. Dans l’échantillon de Pakistan, nous trouvons que l’âge, l’éducation et le revenu
mensuel familial ont des effets significatifs sur le comportement d’évitement de la marque
tandis que le genre n’en a pas. Dans l’échantillon de la Nouvelle-Zélande, nous ne trouvons
pas un effet significatif des variables démographiques sur le comportement d’évitement de la
marque. Nous discutons et interprétons des résultats de notre étude pour tous les trois groupes
de données, c’est-à-dire Pakistan, Nouvelle-Zélande et l’échantillon rassemblé.
303
Appendices
9.4.5
Congruence du soi indésirable et comportement d’évitement de
marque
La signification des estimations standardisées à p<0.001 dans les trois groupes de données
montre que le soi indésirable est l’indice positive de l’évitement de la marque. Il supporte H1.
Les résultats confirment le phénomène de rester loin de l'image de soi indésirable ou d'un
groupe de personnes.
Les résultats obtenus prouvent fortement nos hypothèses, la congruence du soi indésirable
influence positivement le comportement d’évitement de marque. Les découvertes confirment
les recherches antérieures (Englis & Solomon, 1995; Ogilvie, 1987; Phillips et al., 2007) qui
indiquent que les consommateurs peuvent s’abstenir d’acheter ou d’utiliser
des
produits/marques pour ne pas s’identifier à l’image ou groupe qui tendent à éviter . Le support
clair est fourni aux conclusions de Banister & Hogg (2004) où les personnes sondées ont des
sensations négatives envers des images du soi indésirable.
9.4.6
Influence social négative et le comportement d’évitement de la
marque
Nous avons obtenu des résultats mixtes dans les trois groupes de données. Dans l’échantillon
de Pakistan, l’estimation standardisée est positive mais pas significative p<0.05, ce qui ne
supporte pas H2. Dans les données rassemblées et les données de Nouvelle-Zélande, nous
avons trouvé les estimations de la régression significatives p<0.10 mais dans la direction
opposée de la relation proposée dans notre modèle. Ici, H2 n’est pas également supportée.
Pour explorer plus ce rapport, nous avons fait l’analyse multi-groupe pour connaitre l’effet
des variables démographiques (genre, âge, éducation et la revenue mensuelle familiale) dans
cette relation. Nous n’avons pas trouvé l’effet significatif du genre et de l’éducation dans cette
304
Appendices
relation. Mais l’âge et le revenu mensuel familial ont l’effet significatif sur cette relation sur
l’échantillon de Pakistan. Dans l’analyse multi-groupe, nous avons constaté que les personnes
sondées moins de 25 ans qui ont le revenu familial plus bas, c’est-à-dire celui inférieur à Rs
80,000, ont la relation positive significative entre l’influence sociale négative et le
comportement d’évitement de marque respectivement à p<0.10 et p<0.05.
Kulviwat et al. (2009) ont trouvé que les gens ne sont pas indépendants, ils choisissent
exprimer des comportements qui normalement ne les expriment pas dans l’état libre mais le
font juste pour correspondre leur comportement avec la désirabilité du groupe de référence.
Notre étude a supporté ce phénomène dans la façon où éviter certaines marques, l’influence
sociale négative des autres jouent un rôle important significatif quand le consommateur n’est
pas mûr.
9.4.7
Animosité perçue et le comportement d’évitement de marque
Dans l’échantillon des données rassemblées et de Pakistan, les estimations de régression sont
positives et significatives à p<0.05. Cela montre que l’animosité perçue est l’indice positif du
comportement d’évitement de la marque. Plus l’animosité perçue avec le pays d’origine est
élevée, plus le comportement d’évitement de marque est élevé. H 3 est supporté. Dans
l’échantillon de Nouvelle-Zélande, l’estimation de la régression est positive mais pas
significative à p<0.05. Nos découvertes ont soutenu l’étude antérieure réalisée par Klein et
al. (1998) en Chine. Dans ses études du modèle de l’animosité réalisée sur l’achat des produits
étrangers, ils ont trouvé la relation négative significative entre l’animosité et la volonté
d’acheter des produits étrangers. Notre étude a renforcé la relation entre l’animosité et le
comportement d’évitement de la marque. Nous avons trouvé l’appui de Yang et TSO (2007)
qui ont étudié la relation de l’animosité entre les Chinois et les Taïwanais et ont découvert le
comportement négatif significatif des consommateurs taïwanais contre des programmes de
305
Appendices
télévision importés par la Chine. Nous avons également proposé l’effet modérateur de la
familiarité avec l’origine du pays, hypothèse sept. Les estimations unstandardisées de
régression sont significatives dans les trois groupes de données à p<0.05. Cela montre que la
familiarité avec le pays d’origine modère la relation entre l'animosité et l'attitude d'évitement
de la marque.
Les résultats de Nouvelle-Zélande supportent partiellement notre hypothèse, la modération
existe, mais inversement à notre relation de modération proposée. Cela peut être dû au fait que
la réponse moyenne pour la familiarité avec le pays d’origine de l’échantillon de NouvelleZélande est de 4,38 (où 1 = pas familier et 7 = très familier). La moyenne faible montre que
les consommateurs ne sont pas très familiers avec le Pays d’origine. De même, nous avons
proposé dans l’hypothèse H8 que l’ethnocentrisme du consommateur modère ces relations.
Les coefficients unstandardisé de régression sont significatifs sur les données rassemblées et
sur l’échantillon de Nouvelle-Zélande (p < .0001, et p < .10).
Ceci montre que
l’ethnocentrisme du consommateur modère la relation entre l’animosité et l’attitude
d’évitement envers la marque. Sur l’échantillon de Pakistan, l’interaction n’est pas
significative. Cela pourrait être dû la moyenne du score pour l’ethnocentrisme qui est de 3.64
(sur l’échelle de mesure allant de 1 = très désaccord à 7 – très d’accord). La faible moyenne
montre que les consommateurs ne sont pas très « ethnocentrique » dans notre échantillon.
306
Appendices
9.4.8
Citoyenneté non éthique perçue de l’entreprise et
l'attitude
d'évitement de marque
Pour cette hypothèse (H4), les indices estimés de régression ne sont pas significatifs sur les
trois échantillons (p
< 0.05). H4 n’est donc pas validée. Le comportement contraire à
l'éthique perçue par les consommateurs ne les incite pas à éviter les marques de l’entreprise,
ou les gens peuvent ne pas être préoccupés par le rôle de l'entreprise dans la société en
général.
Fan (2007) a fait valoir qu'il n'est pas garanti que la marque éthique fournit une réussite dans
le marché, ou d'autre part, nous pouvons voir qu'une marque populaire ou ayant du succès
n'est pas perçue éthique. Les consommateurs ont généralement le souci sur l’éthique avec des
marques ou des organisations, mais ces préoccupations ne jouent pas nécessairement un rôle
dans leur comportement d'achat réel. Carrigan et Attalla (2001) ont conclu dans leur étude sur
les répondants du Royaume-Uni, que même si les consommateurs sont aujourd’hui plus
sophistiqués, cela ne se traduira pas forcément par un comportement qui favorise les
entreprises éthiques sur celles contraires à l'éthique.
Titus et Bradford (1996), dans leur étude portant sur les répondants des États-Unis, ont
constaté que dans ces derniers temps, avec plus de choix disponibles sur le marché, la
sophistication des consommateurs baisse plutôt qu’augmente et, par conséquent les
consommateurs non avertis sont plus susceptibles de récompenser les pratiques commerciales
inéthiques et de punir le comportement éthique.
9.4.9
Le risque perçu et de l'attitude d'évitement de marque
Les résultats de l'hypothèse (H5) sont positifs et significatifs (p < 0,001) sur les données
rassemblées et sur l’échantillon de la Nouvelle-Zélande. Il confirme que le risque perçu
307
Appendices
motive les consommateurs à éviter les marques risquées. H5 est donc valide. Dans
l'échantillon du Pakistan, l'estimation de régression est positive mais non significative à p <
0,05.
Nos résultats sont similaires aux résultats de Lee (2009) qui a trouvé l'effet du risque perçu
sur l'attitude et l'intention de l'adoption de services bancaires par Internet dans un
environnement taïwanais. Les résultats ont montré que le risque perçu dans différentes
dimensions (sécurité, financier, social et performance) a un effet négatif sur l'attitude et
l'intention de l'adoption de services bancaires en ligne. Nos résultats, d'une autre manière
prouvent la même chose, que le risque perçu a un effet positif sur l'attitude d'évitement de la
marque et est un puissant prédicateur de l'attitude d'évitement de la marque.
En regardant sur la liste des marques citées par des répondants dans l'échantillon du Pakistan,
la majorité des marques sont des téléphones mobiles et d’ordinateurs portables. Comme toutes
les marques sont d'origine étrangère et la littérature dit que les consommateurs des pays en
développement perçoivent les marques étrangères comme étant de meilleure qualité que les
marques nationales (Raju, 1995).
9.4.10 L’attitude d’évitement de la marque et l’intention d’évitement de la
marque
Les estimations de régression positives et significatives à p <0,001, dans les trois ensembles
de données montrent que l'attitude d'évitement de marque conduit à l'intention d'évitement de
la marque. H6 est valide.
L'importance de cette relation est cohérente avec les études antérieures qui montrent que
l'attitude envers l'achat influence les décisions d'achat des consommateurs. Notre étude
montre que l'attitude envers l'évitement affecte aussi les décisions d'achat des consommateurs.
308
Appendices
En outre, il montre que cette relation existe sur le côté négatif aussi. C’est-à-dire que l'attitude
d'évitement des marques affecte aussi les intentions d'évitement des marques. Ang et al.
(2001) a constaté que l'attitude des consommateurs envers le piratage est un facteur prédictif
significatif de leur intention d'achat et les consommateurs qui ont une attitude positive envers
certaines marques sont plus susceptibles d'acheter ces marques. Contrairement à eux, sur les
lignes similaires, nous proposons que l'attitude négative à l'égard de certaines marques
prévoie également des intentions négatives à l'égard de ces marques qui les limitent à acheter
ces marques.
Nos résultats sont similaires aux résultats de Wee et al. (1995) qui démontrent que plus
l'attitude des consommateurs est défavorable envers des marques, moins serait leur intention
d'acheter ces marques. Vis-à-vis, plus les consommateurs ont une attitude pour éviter
certaines marques plus serait leur intention d'évitement ces marques.
9.5
Discussions générales et Conclusion
Ce chapitre conclut la discussion. Il met en lumière les contributions possibles théoriques et
les implications managériales de cette étude. En outre, ce chapitre identifie également
certaines limites et ouvre des pistes de recherches futures.
9.5.1
Contributions
Le phénomène d'évitement de la marque a été identifié qualitativement grâce à des entretiens
en profondeur avec les consommateurs et testé empiriquement sur des consommateurs
originaires de deux pays. Nous avons dans l’échantillon en provenance du Pakistan, un
mélange d'étudiants et de professionnels alors que l’échantillon de Nouvelle-Zélande se
composent d’étudiants universitaires.
309
Appendices
Notre étude a contribué à appréhender en profondeur l’évitement de la marque avec plusieurs
manières :

Elle a fourni la preuve, qualitativement et empiriquement du phénomène d'évitement
de la marque au pré-achat.

Nous avons élaboré et vérifié un modèle conceptuel basé sur les facteurs (la
congruence de soi indésirable, l'influence sociale négative, l'animosité perçue, la
citoyenneté inéthique perçu de l'entreprise et du risque perçu) qui affectent l'évitement
de la marque au pré-achat. Ces facteurs ont émergé grâce à des entretiens en
profondeur et une étude de la littérature.

Nous avons fourni une perception généralisée d'évitement de la marque au pré-achat
en combinant les deux échantillons (Pakistan et Nouvelle-Zélande) à l'aide de poids de
l'échantillon pour représenter les échantillons appariés avec les caractéristiques sociodémographiques des populations cibles.

Pour mieux comprendre la différence entre les pays, une perception séparée de chaque
pays a été fournie, et la différence entre les pays a été établie.

Nous avons fourni la preuve du rôle modérateur de la familiarité avec le pays d'origine
sur l'attitude d'évitement de la marque ? Ce résultat est empiriquement prouvé dans
les trois ensembles de données.

Notre étude a examiné le rôle de l'ethnocentrisme comme un modérateur sur l'attitude
d'évitement de la marque, empiriquement prouvé pour l’échantillon de la NouvelleZélande.
9.5.2
Implications
Notre étude a des implications théoriques et pratiques, du point de vue académique et apporte
des connaissances expliquant pourquoi les gens évitent certaines marques; la principale
310
Appendices
raison de l’évitement est l'attente non réalisée ou l’insatisfaction. Bien que notre étude fournit
un aperçu du phénomène d'évitement de la marque les gens n’ont pas une première expérience
avec les marques.
Du point de vue pratique, être conscient de l'évitement de la marque, les praticiens devraient
accepter le flux entre l'image de marque positive et négative (Lee, 2007), qui peut créer un
problème pour la société à long terme.
Les entreprises peuvent prendre l’avantage sur leurs concurrents en ayant cette connaissance
et ainsi mettre en place une stratégie appropriée pour faire face à des consommateurs qui ont
des attitudes et comportements négatifs au pré-achat envers leur marque. L'objectif du
marketing relationnel est de comprendre les consommateurs et de construire une relation
client. Les entreprises en général, et les responsables marketing doivent en tenir compte sur le
long terme afin de travailler sur ces consommateurs et atténuer les sentiments négatifs.
Le chercheur est convaincu que, malgré les implications managériales de l'évitement de la
marque, elle doit être discutée dans la littérature académique par les chercheurs en marketing,
afin de fournir des perspectives globales de consommation de marque.
9.5.3
Limites
Aucune recherche n'est sans limites. Ce paragraphe présente certaines limites de cette thèse.
La première limite est évidente, c’est la faible taille des échantillons dans les deux
échantillons, en particulier, celui de la Nouvelle-Zélande.
Des échantillons ont également été biaisés, l’échantillon pakistanais est composé de 73%
d’homme et au contraire, celui de la Nouvelle-Zélande de 61,1% de femmes. Toutefois, dans
l'échantillon de données lors de l’analyse groupée, cette asymétrie a été réduite en attribuant
des poids d'échantillonnage afin de représenter au mieux les échantillons des populations
311
Appendices
cibles. Pourtant, un grand échantillon, représentatif en genres permettrait d'améliorer la
généralisation des résultats pour l’analyse distincte de chaque pays.
La recherche est de nature exploratoire. Nous avons utilisé un échantillon de commodité dans
la collecte de données par le Web-sondage en ligne. Cela peut être une objection potentielle
de la représentation réelle de la population présente dans les échantillons et de la
généralisation. Un large éventail de participants en utilisant d’autres techniques
d'échantillonnage (par exemple, un échantillonnage aléatoire ou aléatoire stratifié) peut
augmenter la possibilité de généraliser les résultats.
Pays d'origine a beaucoup de connotations comme le pays d'origine, pays de conception, pays
d'assemblage, ou le pays de fabrication. Notre étude n'a pas tenté de faire la différence entre
ces différents points de vue de la marque / produit d'origine. Bien que, Charles (2012) a fait
valoir que quelques dimensions seulement rendent le processus décisionnel plus compliqué,
mais il est possible que les consommateurs peuvent avoir différentes attitudes et des
interprétations au sujet de leur évitement / soutien de l'achat des produits étrangers.
9.5.4
Les recherches futures
L’anti-consommation est un domaine qui fait l’objet de plus en plus de la recherche. De plus,
il existe beaucoup d’autres possibilités pour les chercheurs de contribuer à son
développement. Les limites de l'étude actuelle peuvent être mieux prises en compte dans les
études futures.
Dans les pays en développement, la disponibilité des marques contrefaites a augmentée.
L’attitude à l'égard des marques contrefaites, l'avis du groupe des pairs ou des amis qui
avaient déjà acheté et consommé une marque contrefaite ou le risque d'acheter une marque
contrefaite sur internet pourraient être des pistes intéressantes à explorer pour comprendre les
312
Appendices
facteurs qui influent sur l'évitement de marque au pré-achat. L'objectif de la présente étude
est d'explorer les raisons qui touchent à l'évitement de la marque. Des études futures
pourraient examiner quels remèdes d'évitement sont faciles à établir et à entretenir.
Des études antérieures sur le bouche à oreille négatif et positif ont confirmé que le bouche à
oreille négatif peut avoir un effet plus fort que le bouche à oreille positif. Ainsi, une étude
future pourrait enquêter sur ce qui est plus influent dans la prise de décision des
consommateurs.
Enfin, dans notre étude, nous avons interrogé les consommateurs finaux, le point de vue des
consommateurs sur les marques est ainsi l'objectif principal de cette étude. Une étude future
pourra enquêter sur les orientations des consommateurs autres que finaux, tels que les
gestionnaires dans un contexte business-to-business, ce qui pourrait également fournir des
implications managériales considérables.
9.5.5
Conclusion
Il est évident que la gestion de la littérature classique de marque est orientée vers le côté
positif, et met l'accent sur la fidélité aux marques, les préférences des consommateurs et la
sélection de la marque.
Alors que les consommateurs mécontents et insatisfaits n’ont pas été ciblés pour fournir une
source riche d'apprentissage pour les universitaires et les praticiens. Dans cette thèse, nous
avons essayé de combler l'écart en matière de recherche en identifiant les raisons possibles
pour lesquelles les consommateurs évitent certaines marques, en particulier, s’ils n'ont pas eu
d'expérience directe avec elles. En termes d'implications marketing, les connaissances
théoriques sur les facteurs qui motivent les consommateurs à éviter certaines marques pour
313
Appendices
permettre aux responsables des marques de penser aux stratégies possibles pour s'assurer que
leurs marques restent en bonne forme.
En fin de compte, en augmentant les connaissances dans le domaine de l'anti-consommation,
cette étude offre une contribution globale à la consommation des comportements des
consommateurs. En conclusion, cette thèse apporte une réponse valable à la question vitale:
pourquoi les gens évitent les marques, même ils n'ont eu aucune expérience directe avec
elles?
314

Thesis_Muhammad Asif KHAN

Transcription

Documents pareils

Client Brief Launch Date: Marketing Objective

Master of Science HES-SO en Business Administration – diplômé-es

gerard darel

Playtime, French exhibitors in 2014

Offre VIE TRADE MARKETING ASSISTANT

Abortion overview - Classes

1 BTS session 2010 Anglais Corrigé indicatif I. COMPRÉHENSION

The FCDE has invested €11m in the De Fursac group.

Year 9 French Planning grid for use of activities on linguascope.com

surfboard - Villard de Lans