|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001966417
007 cr mnu|||uuuuu
008 081021s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002228
Harmon, Tracy R.
Essays on multiple identities and motivated consumption :
b exploring the role of identity centrality on self-brand connections
h [electronic resource] /
by Tracy R. Harmon.
[Tampa, Fla.] :
University of South Florida,
ABSTRACT: This dissertation consists of three essays on the role of identity centrality in the formation of consumer self-brand connections. It contributes to a better understanding of how consumers negotiate multiple identities in the marketplace when making brand choices. This is significant as much of the research on the self-concept and consumer behavior has focused on isolated self-dimensions or have examined single consumer identities in isolation.^ Theoretically grounded in identity process theory (Breakwell 1986), which suggests individuals construct their identity through multiple identity motives influencing identity centrality, enactment, and affect; this dissertation addresses these gaps by answering two specific questions: 1) What are the various identity motives that influence a consumer's individual and group identity centrality leading to enhanced self-brand connections? 2) How does identity centrality influence reference group brand associations in the formation of self-brand connections? In Essay 1, a framework for conceptualizing the influence of multiple identity motives on self-brand connections is proposed driven by findings from consumer in-depth interviews. The framework suggests identity centrality mediates the relationship between the satisfaction of multiple identity motives on self-brand connections, and moderates self-brand connections when reference group brand associations are considered.^ Fourteen propositions are presented, and are empirically tested in Essays 2 and 3. In Essay 2, identity motives from identity process theory along with others identified in Essay 1 are empirically validated, using both hierarchical linear modeling and hierarchical multiple regression. The findings support the influence of two identity motives informing identity centrality, namely: recognition and continuity. This is significant, as prior research in consumer behavior has largely focused on the self-esteem and self-consistency motives (Grub and Grathwohl 1967; Sirgy 1982). Essay 3 investigates the moderating effect of identity centrality on the formation of self-brand connections as reference group brand associations are considered. It is found that the when the ingroup identity is highly central, stronger self-brand connections result. On the contrary, when the ingroup identity is low in centrality self-brand connections are mitigated.^ The differential effects of self-brand connections due to identity centrality provide insight into intra-group differences when the brand is consistent with the ingroup image. The results support a general importance of the role of identity centrality at both the individual and group levels, providing a catalyst for future studies examining the role of the self-concept in consumer behavior.
Dissertation (Ph.D.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 184 pages.
Co-adviser: Anand Kumar, Ph.D.
Co-adviser: David J. Ortinau, Ph.D.
t USF Electronic Theses and Dissertations.
Essays on Multiple Identities and Motivated Consumption: Exploring the Role of Identity Centrality on Self-Brand Connections by Tracy R. Harmon A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Marketing College of Business Administration University of South Florida Co-Major Professor: Anand Kumar, Ph.D. Co-Major Professor: David J. Ortinau, Ph.D. James R. Stock, Ph.D. Americus L. Reed, III, Ph.D. Date of Approval: February 15, 2007 Keywords: different selves, intrinsic motives, automobiles, qualitative study, reference groups Copyright 2007, Tracy R. Harmon
Dedication This dissertation is dedicated to the memory of Joyce A. Harmon. She was my greatest teacher, and ul timately the love of my life. I carry her memory in my spirit and her strength in my heart. Without her I would have never known how to complete this monumental task. Without her, I would have allowed my own understanding to lead me astray. Without her, I would not have known how to fight for what is right within my soul. Without her, I would have never learne d how to balance emotion and ambition. Without her, I would have taken this crown and let it characterize me, but her teachings have kept me grounded and true. Mommy, this is for you with love.
Acknowledgements This dissertation, unlike many others ha s been one of the greatest learning experiences of my life. Over the course of four years, I have come to face to face with my inner self, acknowledging every fear, weakness, strength, and desire that I have. For the weak-hearted this may led to a series of se ssions on a therapists couch, but for me, it has led to one defining moment after the next. There have been many persons who I would like to acknowledge for those many defining moments. I will star t with the order of meeting them. Dr. James Stock was the professor of my second marketing seminar of Marketing Theory and Thought. From the beginning I instantly respected him because he was a fair, honest man full of integrity; and from all accounts, very genuine and a will ing to mentor Ph.D. students. For a firstyear Ph.D. student that made all of the diffe rence in the world. His class was tough but he taught me the importance of rigor, and made me appreciate non-conformi ty in intellectual pursuit. He is a profound scholar who dese rves a place among the Whos Who among marketing faculty. He is the epitome of a good colleague, faculty member, and Christian. Dr. Robert Nixon, a management professo r, who I met in the hallways. He is another inspiration. He always had a sm ile on his face and something nice to say, whenever I saw him. He always checked in on me to see how I was doing and never hesitated to offer words of encouragemen t. He became my personal coach during my time at USF, and I needed him more than he probably ever knew. He learned me quickly and taught me how to be me and manage me simultaneously. He became my source of
advice, encouragement, and friendship. My e xperience at USF would have not been the same if he werent there. He is a true Godsend, and I will forever be indebted to him for helping me navigate this process. Dr. David Ortinau, my third marketing seminar professor. He is the champion professor of students, one of the few faculty th at I have come across, that will put student success above his own. He sacrificed his time over and over again for not just me, but any Ph.D. student who needed assistance. He was always willing to listen, teach, and on occasion sit down for a drink. He was a stickl er for editing, and I along with the other Ph.D. students, knew that whenever we recei ved a paper back from him it would usually be full of red ink. I would usually go to the pages where there was no red ink first, then read his comments last. He taught me to be thorough and consistent in my writing. For that I am appreciative. Dr. Anand Kumar was a newcomer to the department. He was friendly and unassuming, which is why I think he agreed to originally chair my dissertation. Perhaps had he been there in 2003, when I started he would have thought twice about working with me. He carries a wealth of knowledge, a nd encouraged me to think about my ideas a little more in-depth than perhaps I would have without him. He was a great listener for conceptual ideas and patiently allowed me to tr ip over my ideas until I was finally able to stand on a concrete idea. Without him, I woul d have been spinning ideas forever without any conclusive ending. Dr. Americus Reed III for inspiring my inte rest in identity research. I ran across his paper (Psychology and Marketing 1998) in the summer of 2004. I had never heard the term social identity, but after reading his article, the ideas began to develop and I knew
that I had found a research area. Americus ha s an excellent vision for research, and has continuously provided me with much needed clarity on my dissertation efforts. I look forward to learning more from him and hope to achieve his level of scholarly accomplishments within the future. I can I could say that want to be an Americus Reed, but instead Ill just say that I want to be Tracy Harmon with an Americus Reed publication record a nd scholarly wisdom. I would like to thank the ot her faculty within the marke ting department especially Dr. Stamps for encouraging me to apply to the program at the Ph.D. Project conference. Had I not met her, I would have never applie d to the USF program. She was a continuous source of advice, serving as an advisor to my matriculation at USF, and for that I am grateful. Yancy Edwards, what a great guy; always so warm and friendly, and a great lunch partner. I will miss you dearly; I can only hope to find colleagues like you as I move forward. Ivan Lapuka, my true cohort buddy and I guess an ex-fianc of sorts (I am laughing as I type this). I will miss you d early. You are an amazingly good person, and I can only hope that I will always have you as a friend. Remember to pray always, and keep smiling. Thank you for the wonderful laug hs over the years (e.g. Strategy Seminar), and perhaps one day, I will eventually add the hedonic motive to my model. Merlyn Griffiths and the PhD Project fam ily, they were so very encouraging and supportive. They made this journey much more bearable. The womens bible study group, I always needed a refreshing, uplif ting break from the department. Carol Anderson, my marketing professor during busin ess school, you were such an inspiration, I love your spirit and your generosity, I can only hope to be like you in the profession.
Thank you for your generosity and teachings Jessica Vazquez, for your continued assistance across the years, you were so very helpful and I dont know what I would have done without you. All of the wonderful docto ral students in the various departments across USF who made this journey memorable. Florida A&M University for teaching me about my own identity and what it means to be African American at a Historical Black University. You are the sole inspirat ion for this study. Lets Go Rattlers! And last but not least my family, my sister and brother, my nephews, my girlfriends, their parents, church members, and Jeffrey David. Thank you all for listening to me vent about my research and for ta king my mind off of my studies. But most importantly, thank you for pushing me pass the limits I set for myself. I love all of you.
i Table of Contents List of Tables................................................................................................................. ....iv List of Figures................................................................................................................ ....vi Abstract....................................................................................................................... ......vii Chapter 1: Introduction........................................................................................................ 1 1.1. Introduction.....................................................................................................1 1.2. Problem Statement..........................................................................................3 1.3. Research Purpose and Importance..................................................................5 1.4. Theoretical Perspective...................................................................................6 1.5. Concepts and Definitions................................................................................6 1.6. Research Questions.......................................................................................10 1.7. Dissertation Research Agenda......................................................................11 Chapter 2: Literature Review................................................................................................. 2.1. Introduction...................................................................................................14 2.2. The Self-Concept..........................................................................................14 2.2.1. The Perceived Self.........................................................................14 220.127.116.11. Development of the Perceived Self.................................15 2.2.2. The Ideal Self.................................................................................15 18.104.22.168. Development of the Ideal Self........................................16 2.2.3. The Social Self...............................................................................16 22.214.171.124 Development of Social Identities....................................17 2.3. Multiple Selves.............................................................................................17 2.4. Self-Concept in Consumer Behavior............................................................18 2.4.1. One-Dimensional Self-Concept Research.....................................19 2.4.2. Two-Dimensional Self-Concept Research.....................................22 2.4.3. Multidimensional Self-Concept Research.....................................24 2.5. Research Gaps...............................................................................................27 2.6. Chapter Summary.........................................................................................28 Chapter 3: Essay 1 Isolating Multiple Selves: Exploring the Role of Identity Motives on Identity Centrality in th e Formation of Self-Brand Connections...............29 3.1. Introduction...................................................................................................29 3.2. Theoretical Framework.................................................................................31 3.3. Research Method and Study Design.............................................................33 3.4. Data Analysis................................................................................................36 3.5. How Identity Motives Lead to Identity Centrality........................................41
ii 3.5.1. Self-Esteem Motive.......................................................................41 3.5.2. Distinctiveness Motive...................................................................43 3.5.3. Continuity Motive..........................................................................44 3.5.4. Self-Efficacy Motive......................................................................45 3.5.5. Belonging Motive..........................................................................47 3.5.6. Meaning Motive.............................................................................49 3.5.7. Recognition Motive.......................................................................50 3.5.8. Consistency Motive.......................................................................52 3.5.9. Security Motive..............................................................................54 3.5.10. Identity Centrality..........................................................................56 3.5.11. Reference Group Brand Associations............................................58 3.5.12. Brand Symbolism..........................................................................60 3.5.13 Self-Brand Connections.................................................................60 3.6. Conceptual Framework.................................................................................61 3.7. Discussion and Implications.........................................................................65 3.8. Research Limitations and Future Research...................................................66 3.9. Chapter Summary.........................................................................................67 Chapter 4: Essay 2 Multiple Motives, Multiple Selves: Why Self-Esteem and Self-Consistency Doesnt Fully Expl ain Self-Referent Consumption..........................68 4.1. Introduction...................................................................................................68 4.2. Theoretical Framework.................................................................................70 4.2.1. Identity Centrality..........................................................................70 4.2.2. Identity Motives.............................................................................71 4.2.3. Self-Esteem Motive.......................................................................73 4.2.4. Distinctiveness Motive...................................................................73 4.2.5. Continuity Motive..........................................................................74 4.2.6. Self-Efficacy Motive......................................................................75 4.2.7. Belonging Motive..........................................................................75 4.2.8. Meaning Motive.............................................................................76 4.2.9. Recognition Motive.......................................................................77 4.2.10. Consistency Motive.......................................................................78 4.2.11. Security Motive..............................................................................79 4.2.12. Self-Brand Connections.................................................................80 4.2.13. Brand Symbolism..........................................................................82 4.3. Methodology.................................................................................................84 4.3.1. Determining the Study Context.....................................................85 4.3.2. Participants.....................................................................................86 4.3.3. Non-Response Bias........................................................................86 4.3.4. Procedure.......................................................................................89 4.4. Data Analysis and Results............................................................................92 4.4.1. Dependant Variables......................................................................92 4.4.2. Independent Variables...................................................................93 4.4.3. Satisfying the Assumptions of Multiple Regression......................95 4.4.4. Hypothesis 1...................................................................................98 4.4.5. Hypothesis 2...................................................................................99
iii 4.4.6. Hypothesis 3.................................................................................101 4.4.7. Hypothesis 4.................................................................................103 4.4.8. Hypothesis 5.................................................................................104 4.5. Discussion...................................................................................................105 4.6. Managerial Implications.............................................................................111 4.7. Research Limitations and Future Research.................................................112 4.8. Chapter Summary.......................................................................................113 Chapter 5: Essay 3 When the Ingr oup Fails to Indicate Brand Meaning: Exploring the Role of Identity Ce ntrality in Self-Brand Connections........................114 5.1. Introduction.................................................................................................114 5.2. Theoretical Framework...............................................................................116 5.2.1. Social Identity Theory vs. Identity Theory..................................116 5.2.2. Reference Groups and Their Influence on Brand Meaning.........118 5.2.3. Identity Centrality and Reference Groups...................................119 5.2.4. The Role of Self-Construal in Brand Associations......................121 5.2.5. Brand Symbolism........................................................................123 5.3. Methodology...............................................................................................125 5.3.1. Participants...................................................................................125 5.3.2. Procedure.....................................................................................127 5.4. Data Analysis and Results..........................................................................134 5.4.1. Hypotheses 1................................................................................134 5.4.2. Hypotheses 2................................................................................138 5.4.3. Hypotheses 3................................................................................140 5.5. Discussion...................................................................................................143 5.6. Managerial Implications.............................................................................145 5.7. Research Limitations and Future Research.................................................146 5.8. Chapter Summary.......................................................................................147 Chapter 6: Conclusion and Implications..........................................................................148 6.1. Introduction.................................................................................................148 6.2. Theoretical and Conceptual Contributions.................................................148 6.3. Managerial Contributions...........................................................................151 6.3.1. Which Motives to Pursue?...........................................................151 6.3.2. How to Proceed Strate gically and Tactically?.............................153 6.3.3. How Can the Framework Help Firms Attract More Customers?...................................................................................154 6.3.4. How Can Identity Motives/Id entity Centrality Efforts be Evaluated and Assessed?.............................................................156 6.4. Limitations of the Research........................................................................157 6.5. Future Research..........................................................................................158 6.6. A Closing Note...........................................................................................160 References..................................................................................................................... ...161 Appendices..................................................................................................................... ..178
iv Appendix A: Interview Questions.......................................................................179 Appendix B: Coded Transcriptions.....................................................................180 About the Author...................................................................................................End Page
v List of Tables Table 2.1 Review of One-Dimensional Se lf-Concept Literature in Marketing...............21 Table 2.2 Review of Two-Dimensional Se lf-Concept Literature in Marketing..............23 Table 2.3 Review of Multidimensional Se lf-Concept Literature in Marketing...............26 Table 3.1 Informant Demographics.................................................................................34 Table 3.2 Sample Interview Questions............................................................................35 Table 3.3 Steps in Thematic Analysis..............................................................................36 Table 3.4 Sample Representati on of Coded Data Extracts..............................................38 Table 3.5 Research Propositions for Figure 3.3...............................................................64 Table 4.1 List of Identity Motives...................................................................................72 Table 4.2 Product Categories Fr equency Count Distribution..........................................85 Table 4.3 Summary and Comparison of Characteristics of First Wave Respondents to Second Wave Respondents....................................................88 Table 4.4 Summary and Comparison of Key Study Measures of First Wave Respondents to Second Wave Respondents....................................................89 Table 4.5 Identity Motives Descriptive St atistics and Coefficients of Reliability...........93 Table 4.6 Correlations Between Ratings of Consumer Identities for Identity Centrality and Each Hypothesized Identity Motive.........................................95 Table 4.7 Examples of Consumer Iden tities Influencing Automobile Brand Choice ...........................................................................................................100 Table 4.8 Regression Results for Identity Motives Predicting Identity Centrality........101 Table 4.9 Regression Results for Iden tity Motives Predicting Self-Brand Connections....................................................................................................102
vi Table 4.10 Regression Results for Id entity Centrality Mediation...................................103 Table 4.11 Moderated Regression Re sults for Self-Brand Connections.........................104 Table 5.1 Hypothesis Testing for St udent vs. Non-Student Population........................127 Table 5.2 Example of Reference Groups and Brands Listed by Participant..................131 Table 5.3 Self-Construal Measures................................................................................133
vii List of Figures Figure 1.1 Identity Motivated Se lf-Brand Connection Framework..................................12 Figure 3.1 Initial Thematic Map........................................................................................39 Figure 3.2 Final Thematic Map.........................................................................................40 Figure 3.3 Conceptual Model of the Role of Identity Centrality in Self-Brand Connections......................................................................................................63 Figure 4.1 Conceptual Model of Identity Motives Influe ncing Identity Centrality and Self-Brand Connections............................................................................84 Figure 4.2 Normal P-P Plot of Regressi on Standardized Residual for Self-Brand Connections......................................................................................................97 Figure 4.3 Plot of Residuals Showing Homoscedasticity.................................................98 Figure 5.1 Self-Brand Connections by Group Type by Brand Image Match..................135 Figure 5.2 Ingroup Self-Brand Connections by Brand Image Match by Identity Centrality........................................................................................................137 Figure 5.3 Outgroup Self-Brand Connections by Brand Image Match by Identity Centrality........................................................................................................138 Figure 5.4 Outgroup Self-Brand Connectio ns by Brand Image Match by SelfConstrual Type...............................................................................................139 Figure 5.5 Ingroup Self-Brand Connectio ns by Brand Image Match by SelfConstrual Type...............................................................................................140 Figure 5.6 Ingroup Self-Brand Connectio ns by Brand Image Match by Brand Symbolism.....................................................................................................141 Figure 5.7 Outgroup Self-Brand Connectio ns by Brand Image Match by Brand Symbolism.....................................................................................................142
viii Essays on Multiple Identities and Motivated Consumption: Exploring the Role of Identity Centrality on Self-Brand Connections Tracy R. Harmon ABSTRACT This dissertation consists of three essays on the role of identity centrality in the formation of consumer self-brand connections. It contributes to a bett er understanding of how consumers negotiate multip le identities in the marketplace when making brand choices. This is significant as much of th e research on the self-concept and consumer behavior has focused on isolated self-dimen sions or have examined single consumer identities in isolation. Theo retically grounded in identity process theory (Breakwell 1986), which suggests individua ls construct their identity through multiple identity motives influencing identity centrality, enactment, and affect; this dissertation addresses these gaps by answering two specific questions : 1) What are the various identity motives that influence a consumers individual and group identity centrality leading to enhanced self-brand connections? 2) How does identity centrality influen ce reference group brand associations in the formati on of self-brand connections? In Essay 1, a framework for conceptualiz ing the influence of multiple identity motives on self-brand connections is proposed driven by findings from consumer indepth interviews. The framework suggests iden tity centrality mediat es the relationship between the satisfaction of multiple identi ty motives on self-brand connections, and moderates self-brand connections when reference group brand associations are
ix considered. Fourteen propositions are presented, and are empirically tested in Essays 2 and 3. In Essay 2, identity motives from identity process theory along with others identified in Essay 1 are empirically valid ated, using both hierarchical linear modeling and hierarchical multiple regression. The findi ngs support the influence of two identity motives informing identity centrality, namely: recognition and continuity. This is significant, as prior research in consumer behavior has larg ely focused on the self-esteem and self-consistency motives (Gr ub and Grathwohl 1967; Sirgy 1982). Essay 3 investigates the moderating effect of identity centrality on the formation of self-brand connections as reference group brand associations ar e considered. It is found that the when the ingroup identity is highly central, st ronger self-brand connections result. On the contrary, when the ingroup identity is low in centrality self-brand connections are mitigated. Th e differential effects of se lf-brand connections due to identity centrality provide insight into in tra-group differences when the brand is consistent with the ingroup image. The results support a general importance of the role of identity centrality at both the individual and group levels, providing a ca talyst for future studies examining the role of the self-concept in consumer behavior.
1 Chapter 1 Introduction 1.1. Introduction Marketers have long recognized that meanings are asso ciated with brands, and individuals tend to purchase those products th at support and develop their self-image to express who they are (Aaker 1999; Belk 1988; Fournier 1998; Gr ub and Grathwohl 1967; Kleine, Kleine, and Kernan 1993; Laverie, Kl eine, and Kleine 2002). Much of the current work on self-expression has examined self-congr uity, the extended self, identity creation, consumers relationships with brands, consum er brand meaning, and most recently selfbrand connections. A limitation of many of thes e studies pertaining to self-expression is that they have sought to understand the c onsumers self-definition by focusing on one identity component or one sp ecific identity. In doing so, researchers have failed to consider a consumers multiple identities a nd their impact on brand-related outcomes, such as self-brand connections. This is relevant, as a central theme in contemporary social psychology is the multiplicity of identity, whereby individuals have multiple roles and group memberships with which they identify and derive meaning (Settles 2004). This means that at any given time an individual can hold multiple identities each with a unique image and level of importance. A recent article in the New York Times mentioned that when Time Warner Cable sells high-priced bundles of televi sion, telephone, and Internet services, 60 percent of the time mothers decided whic h package to buy. So we are showing the
2 packages through a mothers lens and on moth ers day (Deutsch 2006, p.5). They do this by crafting advertisements that features a mother in different activities throughout the house (e.g. reading to her children, getting dinne r ready for her family). By targeting her mother identity instead of her career woman, ho memaker, or wife identities, marketers at AOL are relying on the centrality of her moth er identity influencing purchase of their bundled services. The consumer behavior literature is beginning to recognize the multidimensional nature of the self-concept, moving beyond the established Cartesian view that assumes the thinking mind as indivisible and uni fied (Aaker, Firat, and Schultz 1997; 2001; Huffman, Ratneshwar, and Mick 2000; Klei ne, Kleine, and Kernan 1993; Laverie, Kleine, and Kleine 2002; Mandel 2003). Within the multiple selves tradition, consumer selves are studied as multiple self-concepts actual, ideal, and social (Belch and Landon 1977; Sirgy 1983), situational self (Hogg a nd Savolainen 1998; Schenk and Holman 1979); malleable self (Aaker 1999), fragmented self (Firat and S hultz 1997, 2001; Firat and Venkatesh 1995; Gould 1991); and role iden tities (Arnett, Ge rman, and Hunt 2003; Kleine et al. 2003; Laverie et al. 2002). These studies, along with others, draw from the dynamic self-concept paradigm in order to understand the relationshi ps between salient identities and consumption congruent with those identities (Aaker 1999; Forehand, Deshpande and Reed 2002; Kleine, Kleine, a nd Kernan 1993; Laveri e, Kleine, Kleine (2002). Although insightful, re search on the consumption behaviors of individuals through their multiple selves, has failed to address many pertinent questions, such as: How does identity centrality vary across a consumers multiple identities in a consumption setting? How do multiple selves interact with each other and give meaning
3 to brand consumption experiences? Under what conditions will the reference group identity become central to behavioral outcomes when ot her identities are present? Such is the motivation behind the current research. Identity centrality, the importance or psychological attachment i ndividuals place on thei r identities (Settles 2004), can aid in explaining how individuals negotiate multiple identities exacerbating one and buffering others. This is of particul ar importance to consum er behavior as the presence of multiple identities suggests indi viduals readily associate with a number of social groups, roles, and cat egories depending on the needs of the individual or the contextual environment (Crawford 2004). Theref ore, various motivations driving identity centrality and symbolic brand choice may coexist. Understanding this link between motivation and behavior is cr itical to gaining insight in to the relationship between identity centrality and self-brand connections. 1.2. Problem Statement Despite the pervasiveness of the research on self-concept and consumer behavior, the consideration of multiple identities in consumption experiences has been vastly under-realized in the marketing li terature. Three short-comings in the disciplines use of the self-concept in marketing can be identifie d. First, the application of multiple selves (working self theory) has been limited both conc eptually and substantively. As mentioned above, most consumer research on self-concep t and consumption cons iders identities in isolation, or limits the domain of identity consideration. The cons ideration of multiple identities in the formation of self-brand connections has been overlooked; creating a gap in marketers understanding of the motivati ons and consequences competing consumer
4 identities have on consumption related out comes (i.e. self-brand connections, purchase intentions, attitudes). A second short-coming of current research of the self-concept and consumption concerns the influence of identity motives While other researchers have addressed individual motives in consumer behavior (e.g., Escalas and Bettman 2005; Erdem and Swait 2004; Argo, Dahl, Manchanda 2005) their relevance to iden tity centrality has either been omitted or weakly linked. In order to address these shortcomings, alternative identity motives beyond what has been proposed in prior models of how the self-concept influences consumer behavior (Sirgy 1982; Grub and Grathwohl 1967; Levy 1959; Belk 1988; Kleine et al. 1993) should be considered. This is rele vant as there is neither a conceptualization nor empiri cal findings that address th e simultaneous influence of multiple identity motives on brand choice and related behavioral outcomes. Meanwhile, research dedicated to brand choice has inde pendently examined reference group influence (Escalas and Bettman 2005), self efficacy (Erdem and Swait 2004); self-meaning (Fournier 1998); and self-pre sentation (Argo, Dahl, Manc handa 2005). Taken together these studies suggest multiple motives influence the same outcome (i.e. brand choice). While Vignoles et al. (2002; 2006) has examined the influence of mu ltiple motives on the identity construction of Anglican priests, the study of multiple motives on brand directed behavior has yet to be investigated. The third and final shortcoming concer ns the centrality (e.g. psychological importance) of consumer identities. Although previous research has examined identity salience and associated cognitions and behaviors, identity centrality has been virtually ignored. Identity centrality may be an impor tant consideration in the formation of
5 consumer attitudes behavior. For instance prior research in consumer behavior has shown that reference groups influence self-bra nd connections (Escalas and Bettman 2003; 2005); serve as extensions of th e self (Belk 1988); and activate self goals when others are perceived (Kenrick, Maner, and Butler 2002) However, these studies have failed to consider the centrality of those reference group identitie s; leading to questions surrounding the differential attitudes and cogni tions within the group. This is supported by the social psychology literature which suggests individuals assign a level of significance to their identities, such that th is level of significance directly impacts group driven attitudes, behaviors, and cognitions. 1.3. Research Purpose and Importance The need to expand our knowledge of th e self-brand phenomena beyond current findings is the motivation for this research, o ffering identity centrality as a meaningful construct for analysis. While it appears that the idea of identit y centrality is both applicable and acceptable there has been vi rtually no study of the relationship per se. Also, there exists no integrative theoretical account of brand-identity centrality phenomenon. Thus, great untapped potential lies in applying identity centrality to the study of self-brand connections. Based on this, th e purpose of this research is to examine the influence of multiple motives on identity centrality shaping the formation of selfbrand connections. This i nvolves considering the range of motives that drive an individuals construction and maintenance of a positive and consistent self-image. The present research has implications for brand managers with regard to positioning strategies. They can improve the e ffectiveness of their brand by positioning it on the multiple motives that engender a central identity for the consumer. This is
6 supported by Graeff (1996) who argues that br and managers can manage the effects of image congruence such that consumers shoul d have a favorable attitude and purchase intention towards brands that are perceived to be similar to their desired self-image. This research will also provide insight as to how consumers manage multiple identities in the marketplace and how the centrality of a particular identity in fluences a consumers selfbrand connection. This is relevant to market ers as effective marketing strategies that address the influence of multiple identity motives will likely lead to stronger self-brand connections, when a central identity is evoked. 1.4. Theoretical Perspective This dissertation employs Identity Pr ocess Theory (Breakwell 1983; 1986) to understand how multiple identity motives lead to a consumers central identity affecting brand choice. Identity Process Theory asserts there is an interaction of multiple identity motives relevant to both individual and group identity processes. These processes lead to identity construction through cognitive, beha vioral and affective processes. This theoretical perspective is relied upon to con ceptualize the relations hip between identity motives, identity centrality and self-brand conne ctions. By grounding this research in this theoretical perspective, the re searcher is able to gain a deeper understanding of motives driving multiple consumer selves, and related centrality. Other theoretical perspectives such as role theory and social identity theo ry are used in tandem with Identity Process theory to support the formulati on of the research hypotheses. 1.5. Concepts and Definitions Before proposing a framework to explai n the relationship between identity motives and self-brand connections, it is necessary to define the key concepts that will be
7 presented in the current research. This is particularly important given the multifaceted and inconsistent meanings given to terms such as self-concept a nd identity in previous studies. According to seminal work by James (1890), the self-concept is multifaceted and is comprised of the spiritual self which he defines as the en tire stream of our personal consciousness, (p. 296); the material self all those aspects of material existence in which we feel a strong sense of ownership, our bodies, our families, and our possessions; the social self, defined as the recognition which he (a man) gets from his mates (p.294); and the bodily self which is defined as the attributes of physical body. This dissertation will focus on the social self because it houses the many identities consumers possess and it is appropriate for examining reference group influences in symbolic consumption. This focus is supported by James (1890) who considers the multiplicity of social selves; and states a man has as many social selves as there are individuals who recognize him and carry an image of him in their mind (p.294). There exists many possible selves that co mprise the overall self concept (Markus and Kunda 1986), and those possible selves lead to an individuals identity. Identity theory suggests that the core of an identity is the categorization of the self as an occupant of a role, and the incorp oration of the meanings and expectat ions associated with that role and its performance (Thoits 1986). Accord ing to Stryker (1980), identity, is an internalized positional designation for each position or role relationship they have in society. This means that the overa ll self is organized into multiple parts, each of which is tied to various aspects of society. This vi ewpoint is further su pported by Markus and Kunda (1986) who suggest that multiple selves are the source of an individuals identity.
8 In a similar vein, Stets and Burke (2000) suggest that when individuals self-categorize or identify with a particular role, an identity is formed. In this sense, the relationship between identity and self-concept become s coupled as the self-concept becomes classified in order to relate to other soci al categories. For instance, having a social identity means a person has affiliated themse lves with a socially categorized group who is similar to their self-concept through a process of social co mparison (e.g. AfricanAmerican, woman, Jewish, Muslim). The previous definitions of the self-concept and identity lead to the concept that represents the thrust of this dissertation, identity motives which are defined as pressures toward certain identity states and away from others which guide the processes of identity construction and inherently is a function of an individuals overall self-concept (Vignoles, Regalia, Manzi, Golledge, and Scabini 2006). It might appear that Maslows (1954) hier archy of needs is conceptually the same as identity motives. However, according to Jame s (1890), individuals have three levels of needs: 1) material (physiological, safety), 2) social (belongingness, esteem), and 3) spiritual; which serves as an impetus fo r Maslows (1954) hierarchy (Daniels 2001). Maslows (1954) model demonstrates how th e same product can satisfy different needs implying that motivation energizes and directs goal-oriented behavior In turn, different levels of motives specify be nefits marketers should emphasize. However, this study examines social needs due to the social natu re of identity constr uction. Other needs such as spiritual, material and f unctional needs are all beyond the scope of this dissertation. Given the social demands of self-image congruence, defined as how an individual perceives themselves in relation to others, soci al needs are fitting fo r the current research.
9 Maslows hierarchy does not capture the full ra nge of identity motives considered in the current study because it only conceptually addresses two proposed identity motives, specifically belongingness and esteem. In additi on within Maslows (1954) hierarchy, the various needs are like empty tanks to be filled sequentially; only when a more basic category of need is fulfilled do individuals proceed to the next higher-order need category (Belk, Ger, and Askegaard 2003, p. 100). Maslows (1954) hierarchy undervalues the true typology of identity motives that will be discussed in this present research. Belk, Ger, and Aske gaard (2003) further suggest th at needs or mere wants oftentimes mask the passion that consumer s experience in connection with certain consumption activities. Alternative motivational conceptualizations challenging Maslows hierarchical model include consumers relationships with pr oducts and services as extensions of the self (Belk 1988); symbolic consumption (Hirschman and Holbrook 1982); and among others hedonic consumption (Holbrook and Hirschman 1981). These conceptualizations, like identity motives, connect an individual s most intimate feeling and goals with consumption. The conceptualization of iden tity motives, can be thought as a type of desire, which addresses a more passionate mo tivation than Maslows (1954) needs. As such, features of current identities that sa tisfy a consumers identity motives may be associated with positive affect and can be accentuated in self-presentation. Lastly, identity centrality is defined as the importance an individual attaches to a given identity (Settles 2004). In the present st udy, identity centrality is associated with a single identity. That is not to say that iden tity centrality cannot be relevant for multiple identities forming a single identity germ ane to a specific context (e.g. auto-buying
10 identity, fashion-buying identity). Howeve r, in the present study, the centrality of collective multiple identities is not explored. Nonetheless, identity centrality requires conscious awareness and is usually measur ed by asking individuals to rank different identities according to thei r importance (Rane and McBride 2000). Conceptually, identity centrality is different from identity salience. Salience is the likelihood that a particular identity will be invoked in any given situa tion in comparison to th e likelihood that other identities might be invoked (Rane and McBr ide 2000; Stryker and Serpe 1994). Salience is not a part of an individu als consciousness but simply re flects the probability that an identity will be enacted (Rane and McBr ide 2000) and usually is measured by asking individuals to name the first thing they would tell someone about themselves (e.g., Minton and Pasley 1996; St ryker and Serpe 1994). 1.6. Research Questions The current research will address two fundame ntal research questions. First, what are the various identity motives that influen ce a consumers individual and group identity centrality? Self-brand connections? Sec ond, how does identity centrality influence reference group brand associa tions in the formation of self-brand connections? These research questions will be addressed across three essays. In Essay 1, identity motives driving identity centrality is qua litatively explored to answer the research question, what are the various identity motives that influe nce a consumers individual and group identity centrality (i.e. importance of the ingroup iden tity)? Essay 2 will empirically test the relationship between the identified identity motives and identity centrality in an automobile context. Lastly, Essay 3 will exam ine the influence of identity centrality on the relationship between reference group br and associations a nd related self-brand
11 connections, and seeks to answer the ques tion, how does identity centrality influence reference group brand associa tions in the formation of self-brand connections? The format of each essay will offer an intr oduction, followed by a theoretical framework supporting the research propositions (Essay 1) and hypotheses (Essay 2 and 3). This will be followed by research methodology and a brief discussion of the findings. 1.7. Dissertation Research Agenda The overall goal of the dissertation is to achieve a better understanding of what it means for a consumer to establish centrality among their various identities. A focus on the managerial utility of us ing identity centrality to understand consumer self-brand connections is maintained throughout the disse rtation. Specifically, the researcher seeks to empirically demonstrate that establishing cent rality of identities with a consumer pays off, and to provide insight into how th e marketing manager can affect self-brand connections. To achieve these objectives, a multi-method research program has been designed. The intent of the dissertation is to develop a solid conceptual foundation from which identity centrality theory can be cultivated and to test portions of this theory as a way of demonstrating managerial utility of the construct of identity centrality as a whole. Before the research agenda can be begin, the legitimacy of considering identity centrality is established. In the next chapter, evidence in support of the identity centrality construct in consumer behavior is presented. The goal of the discussion is to make salient the many motives that are germane to ident ity centrality so that that reader will understand centrality as something more endur ing than identity salience in self-brand connections. In order to generate primary data for these motives, Chapter 3 reveals insights from a qualitative exploration of iden tity motives and identity centrality in the
12 domain of self-brand connections. The data support the basic contention that consumers establish centrality of identity with regard to their brands, and yield an understanding of the various motives driving this phenome non. Identity centrality emerges from the analysis as a key self-b rand connection mechanism. Figure 1.1 presents the framework organizi ng this dissertation. Id entity centrality comprises the core of the model and its role as both a dependent and independent variable is explored. The outcome variable, se lf-brand connection re flects the primary psychological and behavioral benefits that accrue from the brands ability to aid in identity construction. In Chapter 4, identity centrality is formally investigated on the individual level. As a final step, group level factors affecting identity centrality levels are considered in Chapter 5. Finally, Chapter 6 concludes the dissertation by outlining the contributions, limitations, and future research. Figure 1.1 Identity-Motivated Self-B rand Connection Framework Identity Motives Identity Centrality Self-brand connection Reference Group Brand Associations Brand Symbolism
13 Collectively, the conceptual and empiri cal components of the dissertation provide a sound test of the contributory value of identity centrality in the fo rmation of self-brand connections. It is the hope of th e author that the in sights herein are capa ble of stimulating future research in the area, and guiding cu ltivation of a theory of Brand Identity Centrality.
14 Chapter 2 Literature Review 2.1. Introduction This chapter reviews the background literat ure on the self-concep t. The chapter is in four sections. First, a brief introduc tion to multiple selves is presented and its characterization in the domain of consumer behavior is offered. Second, a theoretical overview of the overall self-concept in soci al psychology is presented; followed by a review of self-concept studies in marketing. Finally gaps in the marketing self-concept literature are identified, and later explored in this dissertation. 2.2. The Self-Concept The self-concept is comprised of three interrelated self-dom ains (Higgins 1987): the perceived self the ideal self and the actual self. Each one of these elements plays a central role in how the self concept relates to construc ting, organizing, and influencing consumer behavior. 2.2.1. The Perceived Self William James (1890) saw the self as cons isting of whatever the individual views as belonging to himself or herself, which incl udes a material, a social, and a spiritual self. Perceptions of the material self includ e those views of ones own body, family, and possessions. The social self includes observation others have of the individual, and the spiritual self includes perceptions of ones emotions and desires. Gecas (1982) asserts that the content of the self-concept is deri ved from perceptions of social and personal
15 identities, traits, attributes, and possessi ons. Traits are considered broad reaction tendencies and express relatively perman ent patterns of beha vior (Cattell 1965). Individuals demonstrate their values primar ily through their behavi or and through their speech. This particular element of the perceive d self is concerned with the set of values that the individual considers guides his or her decisions and actions. 126.96.36.199. Development of the Perceived Self Self perceptions are esta blished through interactions with ones environment and feedback; whereby processes of attitude form ation, attitude change (Ajzen and Fishbein 1980) and self attribution (Jones 1990) contribut e to the development of a set of self perceptions. Feedback concerns the response an individual de rives from the behavior and communication, be it verbal or non-verbal of othe rs. When feedback to the target is clear, plentiful and consistent, a set of strongly held self perceptions is formed; whereas ambiguous, lacking, or inconsistent feedback results in weakly held self perceptions (Leonard, Beauvais, and Scholl 1995). 2.2.2. The Ideal Self The ideal self represents the set of tra its, competencies and values an individual desires to possess (Rogers 1959). In this defini tion, possess refers to the aspirations of the individual to believe that he or she act ually has a particular trait, competency, or value. The conception of the ideal self is similar to Schlenkers (1985) idealized image (i.e. the ultimate person one would like to be ). This element of the self-concept is the higher-order goal that most individuals st rive for when consuming symbolically.
16 188.8.131.52. Development of the Ideal Self As an individual continues to interact with a reference group he or she has received feedback from; the individual will in turn internalize the feedback based on the relevant traits, competencies, and values that are important to the reference group (Leonard, Beauvais, and Scholl 1995). In this regard the individual becomes innerdirected using the internalized elements of the self as measures of their successes or failures. Internalized competencies and valu es have been suggested as the basis of the ideal self (Higgins, Bond, Kl ein and Strauman 1987) and as an internal standard for behavior (Bandura 1977). In c ontrast if the individual recei ves negative feedback or positive, but conditional feedback the individual may not internalize the feedback or may partially internalize th e traits, competencies, and values of the reference group. In other words, this type of individual becomes other-directed and will either retreat from the group or require continuous feedback fr om group members (Leonard et al. 1995). 2.2.3. The Social Self The social self translates into an indivi duals social identity. Tajfel and Turner (1985) defined social identification as a proces s by which individuals classify themselves and others into different soci al categories such as woman, Baptist, and student. The process of classification serves to orde r and segment the social environment enabling the individual to locate or de fine himor herself within a given social context. Thus social identities are those aspects of an indi viduals self concept that derive from the social categories to which he or she percei ves as belonging to (T ajfel and Turner 1985).
17 184.108.40.206. Development of Social Identities Individuals can establish their social identities through involvement with reference groups in social situations. As indivi duals continue to engage in their particular reference groups, the group becomes the basis for identification. Naturally the success or failure of the reference group as a whole beco mes a source of feedback for the individual. When an individual identifies with a socially referenced group, he/she perceives the fate of the group as his or her own (Foote 1951; Tolman 1943). In summary the self is comprised of several domains which are interrelated and are dynamically involved, this means each dom ain motivates the others development. Therefore it is not th e researchers position that these self-concept domains (e.g. ideal, perceived, and social selves) operate in oppos ition to one another. Rather, the various domains operate in tandem, each contributing to the creation of an ove rall self-concept. 2.3. Multiple Selves Markus and Kunda (1986) put forth the term malleable (or working) selfconcept, which refers to a host of self-concep tions (e.g. ideal self, perceived self, social self) that can be made accessible at a given moment. They suggest that an individuals set of self-conceptions are possible selves, the selves one would li ke to be or is afraid of becoming. These selves functi on as incentives for behavior providing images of the behavior for the future. They also function to provide an interpretative and evaluative context for the current view of th e self (Markus and Wurf 1987). The conceptualization of the malleable self has two important implications for the present study. First, the self is multifaceted and dynamic. Any particular self conception can be activated at any given time particularly due to social cues a nd situations, such as
18 ones hopes, fears, goals and identities (Aak er 1999). Second, conflicting traits may exist in an individuals self-concept. For example, a consumer might think of himself as both highly intelligent and unaccomplished. The relati ve accessibility of the two traits in a given situation determines which trait w ill be expressed (Linville and Carlston 1994). Markus and Kunda (1986) argue that a trait becomes accessible if it was just activated before an event, if it was evoked by an experien ce or a memory, or if it has been elicited by the social situations at a part icular point in time. To this point, the self is regarded as stable, while also being mall eable (Markus and Kunda 1986). The concept of multiple selves is not ne w in the marketing discipline, but the degree of interest in multiple selves is new (Rowan and Cooper 1999). Although William James (1890) was one of the earliest scholars to discuss the divided nature of the selfconcept, only recently has there been an incr ease in academic work that portrays the self as fragmented and fluid. In the consumer behavior literature, multiple selves have been characterized as the fragmented self (Emmons 1992), multiphrenic self (Firat and Shultz 1997, 2001), malleable self (Aaker 1999), and multiple identities (Kleine et al. 1993). While social scientists, as well as marketer s do not have a clear consensus regarding the nature of the self, more scholars now agree that the self is plural and dynamic. Outside of the marketing discipline, this idea has been ex amined by a number of social psychologists (Gergen 1991; Baumgartner 2002; Rosenberg 1979; Brewer and Ga rdner 1996; Markus and Kunda 1986), among others. 2.4. Self-Concept in Consumer Behavior In attempting to organize this vast and di vergent literature on the self-concept, it is perhaps more fruitful to discuss the fi ndings based on their level of dimensionality.
19 Self-concept researchers have viewed the self as one-dimensional, two-dimensional, and multidimensional. One-dimensional refers to an individuals overall-self concept (or the actual self), as ones perception and eval uation about oneself (e.g. Grub and Grathwohl 1967; Birdwell 1968). Two-dimensional refers to dual self-concepts of actual self and the ideal self, as ones perception and evaluation of wh at one would like to be (e.g. Landon 1974; Belch and Landon 1977; Zinkhan and Hong 1991). Lastly, multidimensional refers to three or more self-concepts. Individuals belong to a host of social groups, each with its own distinct id entity. As noted earlier, the multidimensional nature of the self has been regarded as the situational self, malleable self, possible selves, social identities, role identities, wo rking self, and the fragmented self. Using this organizing framework, a review of the self-concept studies in marketing is offered in the next section. 2.4.1. One-Dimensional Self-Concept Research Congruency between self-concept and c onsumption behaviors were initially examined by consumer behavior researcher s such as Grub and Grathwohl (1967) and Birdwell (1968). Not only were their studies among the first to introduce the self-concept into the marketing discipline, they establishe d the value of the self -concept in consumer behavior. Grub and Grathwohl (1967) conceptu alized the model of consuming behavior, which asserts that because the self concept is of value to the individual, their behavior will be directed toward the protection and enhancement of their self-concept. They suggest a more specific approach in examin ing the selfproduct image relationship by introducing self-theory, stating, the concept of the self is more restricted than personality, which facilitates measurement and centers on the critical element of how the
20 individual perceives himsel f (Grub and Grathwhol 1967, p. 23). Birdwells (1968) study found self-image to be more congruent with the owners brand of au tomobile than with seven alternative brands. Differences were found among ownership groups perception of all of the automobile brands studied. Am ong the eight brands tested, substantial differences in the images of each car wa s found. Specifically, true differences exist between the images held of low-priced car s by owners of medium-priced and prestige cars. Lastly, Grub and Hupp (1968) developed a methodology to test the relationship between a consumers self-c oncept, automobile brand and brand strategies. Their methodology along with their proposed theory was supported, showing consumers of different brands in a product class perceive th emselves to have signi ficant differences in self-concepts. A strong criticism of these in itial studies is th at they failed to measure causality, because self-product image congruence was measured after purchase (Evans 1968; Landon 1974). Therefore, the participants answ ers may have been fueled by their need to reduce dissonance. Another major criticism of this work was not all consumers were interested in reveali ng their actual self with their pur chases (Landon 1974). This suggests that consumers have differing motivations for purchases beyond their one-dimensional actual self. Belk (1988) extended the understanding of the self-concept with his notion of the extended self, which includes body, internal pr ocesses, ideas and experiences, and those persons, places, and things to which one feels attached ( p. 140). According to Belk, possessions vary in their importance to th e individual and can be seen as forming multiple layers around, what he refers to as th e core self. His theory promotes a fluid
21 explanation of the self -concept, inclusive of culture, pe ople and time, and has become the foundation for much of the work in consum er behavior on the role of consumption experiences in identity crea tion. More recently Gould ( 1991) conducted an open-ended survey of consumers and their self-concepts. His results suggest th e presence of a single self-concept, which was inconsistent with other studies promoting the multidimensionality of the self-concept. In a similar study, Firat and Schultz (2001) provide contrasting results, revealing the fr agmented nature of the self-concept. These inconsistencies and ambiguities led to th e consideration of multiple identities as witnessed in two-dimensiona l and multi-dimensional resear ch. Table 2.1 summarizes the one-dimensional self-concept studies f ound within the marketing literature. Table 2.1 Review of One-Dimensional Self -Concept Literature in Marketing One-dimensional Self-Concept Studies Author Objective Theory/Paradigm Grub and Grathwohl 1967 Examined relationship between consumer selfconcept and general consumption behavior Self-theory and symbolic interactionism Birdwell 1968 Examined relationship between self image and product image and auto perceptions across car owners. Personality Theory Belk 1988 Examined consumer possessions as extensions of the self-concept. Multiple Literatures Grub and Hupp 1968 Examined the relationship between consumer selfconcept, automobile brands and brand stereotypes. Self-theory Gould 2001 Examined consumers definition of their selfconcept, dynamic selfconcept. Self-theory
22 2.4.2. Two-Dimensional Self-Concept Research While research on the one-dimensional self provided great insi ght into the selfconcept and product-image congruity, there wa s still a need to re solve a few of the ambiguities found in earlier studies. In res ponse, researchers introduced the dual selfconcept, comprised of an actual self and ideal self (e.g. Landon 1974; Belch and Landon 1977). Understanding which selfactual or id eal influenced the purchase decision became the focal point of these studies. Landon (1974) concluded that across all products, some participants were actualizers, while others were perfectionists. Ac tualizers were highly correlated with their actual self than with their ideal selves. Perfectionists were highly correlated with ideal selves, than with their actual selves. A lacks of consistent results, highlighted the need to understand the condi tions under which self (actual vs. ideal) would operate (Landon 1974). Belch and Landon (1977) found ownership to affect product ratings based on the two-dimensional self-concept. Specificall y, product owners had higher correlations between purchase intentions and their ideal self-image, compared to correlations between purchase intention and their actual self-ima ge. However, non-owners failed to show a high correlation between purchase intentions a nd both their actual a nd ideal self image. Other researchers attempted to prove the infl uence of the dual self-concept, but were yet unsuccessful (e.g. Ross 1971). This is not to s uggest that two-dimens ional research ended in the 1970s. In 1995, Hong and Zinkhans study results indicate that brand memory is not mediated by the extent to which advertis ing expressions are cong ruent with viewers self-concept. However, brand preference a nd purchase intention were shown to be influenced by the self-congruency of an ad. Aaker and Lee (2001) conducted four
23 experiments showing goals associated with approach and avoidance needs influence persuasion and that the accessibility of distinct self-views (independent vs. interdependent) moderates these effects. Th ey find individuals with an accessible independent self-view are more persuaded by promotion-focused information that is consistent with an approach goal. Interd ependent individuals are more persuaded by prevention focused information that is cons istent with an avoidance goal. Table 2.2 summarizes the two-dimensional self-con cept studies found within the marketing literature. Table 2.2 Review of Two-Dimensional Self-Concept Literature in Marketing Two-dimensional Self-Concept Studies Author Objective Theory/Paradigm Ross 1971 Examined relationships between actual and ideal selfconcepts, along with consumer brand preference. Interpersonal self concept literature in social psychology Landon 1974 Examined relationships between actual and ideal selfconcepts, along with purchase intentions. Self-concept and Personality Theory Belch and Landon 1977 Examined the effects of social desirability and product ownership on actual self and ideal self, along with purchase decisions. Self-concept Hong and Zinkhan 1995 Examined the effects of the actual self and ideal self on advertising effectiveness as a function of ad recall and attitude towards the ad. Self-schema Theory Aaker and Lee 2001 Examined the effects of selfconstrual on approach and avoidance needs. Self-construal Theory and Cultural Psychology
24 2.4.3. Multidimensional Self-Concept Research Moving away from the two-dimensional stud ies and the associated difficulties in generating the prominence of the actual self ove r the ideal self, researchers began looking at the multidimensional self. Schenk and Holm an (1979) conceptualized the situational self, as attitudes, pe rceptions, and feeling the individual wishes other individuals in the situation to formulate about his/her character. They introduced this term to include the influence that various situations may have on the enactment of specific selves. By allowing consumer preferences to vary with the self-image they want to express in a specific situation, the situationa l self has become widely used in scenario based consumer research. The malleable self concept, or the wo rking self-concept (Markus and Kunda 1986) has been widely used and accepted in consumption of aesthetic cosmetic surgery (Schouten 1991), general consumption beha viors (Morgan 1993), brand choice (Aaker 1999), and risk taking be haviors (Mandel 2003). Social identity theory and identity th eory have both aided researchers in conceptualizing multiple identiti es in consumer behavior. Social Identity Theory suggests an individual identity emerges from a reflexiv e activity of self-categorization in social groups to which they belong (Tajfel and Tu rner 1985); whereas Identity Theory is concerned with the associated meanings, res ources, and expectation with an individuals roles (Stets and Burke 2000). Consumer behavi or research has used these theories to reveal that consumption experien ces are associated more strong ly with specific identities and roles than with the global self (Kleine et al. 1993; Lave rie et al. 2002; Arnett et al. 2003). Along these lines, identity salience has also been s hown to influence attitudes
25 toward congruous brands (Forehand, Deshpa nde, and Reed 2002; Dimofte, Forehand, and Deshpande 2003) and the frequency in which identity-related consumption behaviors are enacted (Kleine et al. 1993; Laveri e et al. 2002; Arnett et al. 2003). More recently, contemporary researchers ar e using consumer narratives to assess the multidimensionality of their self-concep ts (Fournier 1998; Escalas 2003; Ahuvia 2005). Thompson (1997) uses hermeneutics to interpret consumer stories detailing their experiences with brands, services, and general shopping. The stories reveal how consumers perceive their identity and how their perceptions are made manifest in everyday consumption activities. Fourniers study (1998) on c onsumer relationships with their brands finds that individuals buy multiple brands in support of multiple dimensions of their self-concept. Ahuvia (2005) examines the role of loved possessions and activities in the construction of a coherent identity narrative. He found individuals use three strategies, namely labeled demarcating, co mpromising, and synthesizing for resolving identity conflicts. However, he never reported the multiple identities among the respondents, leaving a need for the consid eration of multiple id entities, and their associated meanings. Table 2.3 summarizes the multidimensional self-concept studies found within the marketing literature.
26 Table 2.3 Review of Multidimensional Self -Concept Literature in Marketing Multidimensional Self-Concept Studies Author Objective Theory/Paradigm Schenk and Holman 1979 Examined the relationship between a consumers situati onal self and brands Symbolic Interactionism Sirgy 1982 Examined self-concept motives leading to purchase intention. Product-Image Congruity Theory Solomon 1983 Examined the relevance of product symbolism in self-definition and role performance. Symbolic Interactionism Shouten 1991 Examined the harmonious self-concept through the consumption of aesthetic plastic surgery. Working Selfconcept Theory, Social roles Morgan 1993 Examined brand choice relevant to possible selves as elements of the selfschema. Working Selfconcept Kleine, Kleine, and Kernan 1993 Examined salient role identities their related behaviors. Social Identity Theory and Role Theory Firat and Venkatesh 1995 Examined the postmodern fragmentation of the self. Postmodern Discourse Thompson 1997 Examined multip le identity construction and maintenance through consumption. Hermeneutics Hogg and Savolainen 1998 Examined the relationship between the situational self and brand image, along with the influence of public and private situations on brand choice. Symbolic Interactionism Aaker 1999 Examined the relationship between malleable self and various consumption scenarios. Working-self Theory Laverie, Kleine, Kleine 2002 Examined the role of appraisals on discourse and identity importance. Identity Theory, Appraisal Theory, and Symbolic Interactionism Arnett, German, and Hunt 2002 Examined charitable giving relative to identity and related behaviors. Identity Theory Mandel 2003 Examined self-cons trual on financial and social risk taking. Working Selfconcept Reed II 2004 Examined relationship between salient social identities, self-importance and purchase intentions. Social Identity Theory Ahuvia 2005 Examined the relationship between loved objects and possessions to construct a coherent narrative identity. Narrative Processing
27 2.5. Research Gaps The self-concept studies in consumer behavior have undoubtedly expanded our knowledge of multiple selves. However the li terature is fragmented and dominated by studies investigating relationships between a specific self-concept dictated by the researchers and related behavior s (Baumgartner 2002) leading to two distinct gaps in the literature. Gap1: The current body of research on se lf-concept based consumer behavior examines multiple selves that are either measured or made salient in an experimental setting, prohibiting participants to expre ss other selves important to them beyond the research environment. By testing the relationship between identity motives and identity centrality, a process for identity importance emerges in the analysis (Breakwell 1993). This is contrary to specifying a particular identity for the participant, which can sometimes mask the conditional effects of rese arch stimuli. In broadening the study of multiple identities through ident ity centrality, researchers wi ll be able to understand how consumers negotiate multiple identities that are important to them in a consumption setting. Gap2: While identity centrality may provide insights into how multiple identities are managed in the marketplace, the sec ond gap is concerned with the behavioral consequences of identity centrality in cons umer attitudes and behavior. Current studies have failed to examine the importance a particular identity has for the participant, creating a gap between identity importance and behavioral out come. By investigating the level of identity centrality in behavioral outcomes, researchers can establish causality between the identity and behavi or. This is based on the premised that individuals who
28 assign a high level of importance to the identity are more likely to be committed to behaviors supporting the identity, and are likely to have enhanced behavior, attitudes, and cognitions in support of that identity. 2.6. Chapter Summary As shown in this chapter, research on the self-concept in marketing has been examined from the one-dimensional, two-di mensional, and multi-dimensional views of the self. While this literature has expanded our knowledge in this field, the consideration of identity centrality has been virtually ignored. This omission provides the rationale behind the research gaps addressed in this dissertation. By addressing these gaps, it is hoped that a broadened view of the multiple self will emerge. To address the first research gap rega rding the link between various identity motives and identity centrality, a qualita tive study focused on identity-motivated consumption was conducted. The next chapter re ports the findings of this study offering evidence for nine identity motives which ar e presented along with research propositions, tested in Chapters 4 and 5 of this dissertation.
29 Chapter 3 Essay 1-Isolating Multiple Selves: Exploring the Role of Identity Centrality in the Formation of Self-Brand Connections In this chapter, the re sults of a qualitative st udy among 13 informants are reported. Informants were interviewed to e xplore how the self-concept is implicated in their consumer behavior and to what extent identity motives play a role in self-brand interactions. Using thematic analysis, evidence for nine id entity motives is found and incorporated into a framework of iden tity-motivated consumption. The framework consists of two main components: (1) identity motives leading to a central identity, and (2) the moderating influences of refere nce group brand associations and brand symbolism. The framework contends that thes e two components influence the degree to which a brand is incorporated into an individuals self-con cept (e.g. self-brand connection). Testable research propositions are put forth and implications for marketing mangers are discussed. 3.1. Introduction Self-brand connections has been examin ed through an individuals narrative processing (Escalas 2004), reference groups (Escalas and Bettman 2003; 2005), and from the perspective of children and adolescen ts (Chaplin and Roedder-John 2005). These studies along with prior rese arch suggest self-brand connections are formed through a matching process whereby a consumer identifie s a product or a brand that is congruent with their overall self-image be it desired or actual (Grub and Grathwohl 1967; Levy 1959; McClelland 1951; Sirgy 1981; 1982). As product images are brought to mind
30 similar images about the self are triggere d, and a comparison between the brand and the individuals self-concept is made. A positive comparison between the individual and the brand results in the individual perceiving the brand congruent with his or her overall selfconcept. However it is unclear as to how self-brand connections are formed when multiple identities are releva nt to a particular brand. Self-brand connections need only occur be tween the brand and one aspect of the self, with more schematic aspects of the self resulting in stronger connections (Escalas 2004, p. 170). This point is illustrated with an example of a consumer who has multiple self aspects, relating her professional asp ect to the Burberry brand (luxury line of handbags, accessories, shoes and clothing) a nd her mother aspect to the Gymboree brand (clothing line for child ren), with both aspects bei ng connected to the self independently. While this illustration offers clarity to the multiplicity of the self and selfbrand connections, how an individual es tablishes importance among the different aspects of their selves has yet to re ceive full consideration. To address this shortcoming, identity centrality defined as the psychological importance assigned to an identity (Stryker and Serpe 1994; Settles 2004 ) is used to explain how individuals manage multiple identities, and how different identities become associated with particular brands. Identity centrality is hypot hesized to be associated with motivation (Bagozzi, Bergami and Leone 2003). Relatedly identity motives have been found to guide identity centrality (Vignoles, Regalia, Manzi, Golle dge and Scabini 2006, Breakwell 1986, 1993) This is significant because if an individual is motivated to purchase a brand to support a particular aspect of their self-concept, then similar motivations should guide not only
31 whether or not a brand is incorporated into an individuals self-concept, but also the degree to which the integrat ion occurs. To date, the re lationship between identity importance and identity motivations has ye t to be examined, leaving a gap in the literature on self-brand interactions. The research reported in this study was unde rtaken with the objective to address the research gaps identified above. Specifically, the primary objective of this research was to substantiate the existence of multiple identity motives posited to influence selfbrand interactions and identity centralit y. A second but related objective was to demonstrate the external validity of the id entity centrality proposition grounding this dissertation. Both objectives we re accomplished by (1) integr ating theory and research on the self-concept and consumer behavior to id entify specific identity motives, (2) reporting the qualitative findings insights obtained in an exploratory investigation of identitymotivated consumption, (3) advancing a fram ework showing the relationship between identity motives, identity centrality, a nd self-brand connections, and (4) offering propositions to stimulate future research re lated to identity ce ntrality and self-brand connections The discussion to follow culminates in an articulation of a research agenda capable of investigating these potential insights further. 3.2. Theoretical Framework Identity process theory pr ovides an integrated model of social psychological processes and motivational principles, referre d to as identity motives (Breakwell 1986). These motives guide the identity construction processes that dictate what endstates are deemed desirable for the structure of ident ity. Identity motives also and determine what changes will be made within identity and ar e associated with social influences and
32 interactions (Breakwell 1993). The Identity pr ocess theory was developed in response to the boundaries of the social identity theory proposed by Tajfel (1974) It is argued that the predictive reliability of the social identity theory is low due to its limited focus of one identity process, namely se lf-esteem (Breakwell 1986). Mi ndfully, self-esteem is not considered an identity process in this study. Rather, self-esteem is conceptualized as an identity motive. According to the identity process theory, there are two processes involved in the building of an identity. The first is the assimilation-accommodation process, and the second is the evaluation process. Assim ilation-accommodation is a memory system1; it absorbs new elements of identity (e.g. values, attitudes, style, or interpersonal networks) and adjusts the existing identity to place th em; whereas evaluation involves the allocation of value to identity elements. Both proce sses are deemed information processing systems biased towards self-interest rather than accu racy in constructing identity structures (Breakwell 1986, 1993). The overall relevance of individual id entity motives is evidenced in their outcomes, or structures. Identity structures are manifestations of identity motives and are influenced by the degree of motive satisfac tion. It has been shown that different self evaluation motives are relevant to predictions of cognitive, affective, and behavioral outcomes (Dauenbeimer, Stablberg, Spr eemann, Sedikides 2002; Swann, Griffin, Predmore, and Gaines 1987). As it relates to identity, these are distinguished as perceived centrality, affect, and enactment. Perceived centrality is the cognitive dimension of identity. It addresses the chief importan ce each identity motive has on a consumers overall identity. Identity aff ect is how happy or unhappy an identity makes the individual.
33 Lastly, identity enactment is the extent to wh ich an individual will show who he or she is to others. Because individuals are likely to engage in behaviors and be more committed to identities that carry high levels of impor tance, the study focused on identity centrality in self-brand interactions. Using identity process theory as a guide, the current study propos es that identity motives guide identity centrality in self-brand interactions. 3.3. Research Method and Study Design The objectives of this resear ch guided the selection of method and articulation of the study design. First, the exploratory goals of the study dictated the use of semistructured depth interviews. This method is better suited to the goa l of understanding the relationship between identity centrality and self-brand connections than other methodological approaches (e.g. unstructured in terviews, survey inte rviews, participant observation). This is because semi-structured depth interviews allow subjective meanings to be gathered or articulate d. Survey interviews typically have a fixed set of questions and the researcher generally keeps the orde r of questions the same. Given this, the researcher is limited in his or her ability to probe the participants on their responses. Participant observation allows rich meaning to emerge from the data. In some cases, the researcher is becomes a full participant and in some cases a spectator of sorts. This type of data collection was not feasible given th e exploratory objectives of the research. Thirteen interviews were conducted. The sampling included both students and non-students; gender and ethnic affiliations were also va ried. The interviews were conducted with 7 students, 6 non-students across three ethnic groups, namely African American, Caucasian, and Hispanics. Informants were chosen randomly to participate in 1 As originally presen ted by the author.
34 the study. Table 3.1 provides an overview of the demographic variables collected for each participant. Interviews lasted on aver age about 30 to 50 minutes. They were audio recorded and in some cases videotaped when an audio recorder was not available. Table 3.1 Informant Demographics Participant Age Gender Race Occupation Marital Employment Nikki 19 Female White Student Single Part-time Tonya 20 Female Black Student Single Part-time Erica 21 Female Black Student Single Unemployed Erroll 33 Male Black Financial Analyst Single Full-time David 37 Male Black Firefighter Divorced Full-time Laramie 25 Male Black Civil Engineer Single Full-time Danielle 21 Female White Student Single Unemployed Rebekkah 23 Female Hispanic Student Single Part-time Stephanie 24 Female White Student Single Unemployed Stedman 18 Male Bi-racial Student Single Unemployed Candace 33 Female White Health Information Manager Married Full-time Norman 46 Male White Marketing Director at an Engineering Firm Married Full-time Marcela 37 Female Bi-racial Graduate Admissions Officer Single Full-time All interviews began with a brief in troduction of the re searcher and the respondent. Respondents were th en provided with a brief de scription of the research project and were told that the interview dealt with how they shop and choose brands within the marketplace. Next, the terms a nd constructs used in the interview were explained (i.e. brand, identity). The research er then led a warm-up discussion of the informants general interest in brands by allo wing participants to talk freely about their favorite brands (unspecific to any product ca tegory). This was followed by a series of
35 questions addressing their (a) beliefs con cerning the influence of the self-concept on brand consumption (b) perceptions of how they view themselves (c) perceptions of how others view them in the marketplace (based on the brands they use), (d) motivations influencing brand consumption and (e) fact ors beyond the self-concep t that guide brand self-brand interaction. A list of sample questions posed in the interviews is presented below in Table 3.2. A full listing of que stions is available in the Appendix. Table 3.2 Sample Interview Questions Domain: Brand Preference Can you name your favorite brands ? (in any particular category) What are your thoughts on others who purchase things that are really expensive? Name a brand that you would absolutely not wear. What does a particular brand say about you? When you think of Brand X, what comes to mind? What does Brand X mean to you? What does the (specific feature) of Brand X mean to you? Doe Brand X help communicate your identity to others (whether you know them or not)? Does Brand X help you become the person you want to be? Interview protocols were modified during data collection to take advantage of emerging themes (Spradley 1979). This was facil itated by a series of probing questions to further investigate these emerging themes. Fo r example, if a participant responded to one of the questions and introduced a theme th at the researcher was unfamiliar with (e.g. identity threat, need for authentication), probing questions were asked to gain
36 clarification. Once the resear cher obtained enough informati on from the participant to understand the theme, the interview gui deline was once was again assumed. 3.4. Data Analysis Verbatim transcripts of the interviews we re prepared by the author and served as the data set for this study. Over 60 pages of single-spaced text were generated by the thirteen informants. Thematic analysis was used to evaluate the data. This technique is known to provide a flexible (in terms of theory independence) approach in analyzing rich, detailed, and complex data. It is also a method that is typically inde pendent of theory and epistemology, and can be applied across a range of theoretical and epistemological approaches (Braun and Clarke 2006). Analys is followed the general procedures of thematic analysis as presented in Table 3.3. Table 3.3 Steps in Thematic Analysis Order of Thematic Process Explanation 1. Familiarizing yourself with your data Tran scribing the data, reading the data, noting down your initial ideas. 2. Generating initial codes Coding interesting features. 3. Search for themes Collating codes into the data relevant to each potential theme 4. Reviewing themes Check if the themes work in relation to the coded extracts (level 1) and the entire data set (level 2), generating a thematic map of the analysis. 5. Defining/naming themes Ongoing analysis to refine the specifics of each theme, and the overall story the analysis tells, generating clear definitions and names for each of them, and conducting inter-rater reliability. 6. Producing the report The final opportunity for analysis. Selection of vivid, compelling extract examples, final analysis of selected extracts, relating back to the analysis to the research question and literature, producing scholarly report of the analysis. Source: Braun and Clarke (2006), Adapted from Labov (1972, p.363)
37 The first round of analysis involved a w ithin-person analysis The goal of this analysis was a holistic interpre tation of motivated consumpti on, as it is manifest within the individual consumer. The analysis asks whether there exists a theme or themes that capture similarities in descriptions of the informants motivations for brand selection and lend coherence to obtained res ponses. The within person analys is began with a reading of the transcripts in which recurrent, behavioral tendencies were identified. Specific brand related narratives were then considered for their manifestations of these themes as evidence of identity motivat ed consumption emerged. Broad higher-order themes (e.g. motivated consumption) helped to provide a general overview of the direc tion of the intervie w, while detailed lower order codes (e.g. specific identity motives) enabled fine distinct ions to be made, both within and between cases (King 2004). This hierarchi cal coding allowed the analysis of text at different levels of specificity. As Dey (1993) explains, codes must be m eaningful with regards to the data but also meaningful in relation to other categor ies. Codes identify a feature of the data (semantic content or latent) that appears intere sting to the researcher and refers to the most basic segment, or element of the raw da ta or information that can be assessed in a meaningful way regarding the phenomenon (Boyatzis 1998, p.63). The coded data differs from the units of analysis which are considered themes (Braun and Clarke 2006). Themes are broader than coding and to some extent depend on whether ideas are more data-driven or the ory-driven. The current research was coded based on the goal of the researcher to identi fy particular features of the data set, addressing motivations of self-brand interactions If portions of the transcription failed to
38 contain information germane to this topic, the data was not extracted. Therefore, the researcher systematically worked through the en tire data set giving attention to each data item. Throughout this process, interesting as pects within the date were identified and formed the basis of repeated patterns (theme s) across the entire data set. Table 3.4 shows sample data extracted and coded from the fu ll transcripts are coded. Fully coded extracted data are available in the Appendix. Table 3.4 Sample Representation of Coded Data Extracts Data Extracted Coded for: But if they were really cute, I would never tell any of my friends, and I would never let them see the inside label cause it will say Payless. And if we had to go somewhere to take off our shoes, I wouldnt take them off, Id be too embarrassed. I cant have my friends thinking I shop at Payless. 5 Brand prestige 6 Conspicuous brand use 7 Reference groups 8 Belonging motive 9 Security motive For where I am in my life right now, those cars are more reflective of the fact that I have reached a certain level. So I wouldnt buy an Acura because its looks just like a Toyota and everybody has a Toyota. I wouldnt buy a Cadillac, cause that is my dads car. I wouldnt by an Infiniti or Jaguar, because Jaguars are an old retired mans car. Its not a girl car; its a guy car, an old man car. And I am not the SUV type, even though some of them look nice, I am not the truck type. 1. Life cycle stage, brand congruency 2. Distinctiveness motive 3. Brand user imagery 4. Family brand associations 5. Self-consistency motive 6. Identity threat 7. Brand-identity congruency The second level of interpretation involv ed an across-person analysis. The goal here was to discover convergent themes cap turing commonalities and patterns within the data, across individuals. In a ddition to the motives, key cons tructs were extracted from the data, coded, and organized for subsequent analysis. An initial thematic map resulted
39 in two main themes, purchase motivations a nd brand influences (see Figure 3.1). Central ideas that were coded at this stage include d basic purchase such as special occasions, identified needs, other consumers, and th e availability of di sposal income. Brand influences dealt with internal or external factors impacting an indi viduals brand choice. These influences included brand apathy, brand parity, brand symbolism, self-concept, life cycle stage, aesthetic appeal, affo rdability, and other consumers. Figure 3.1 Initial Thematic Map After refinement, the thematic map was reduced to one primary theme, brand influences. This theme was captured by eight sub-themes: brand apathy, aesthetic appeal, life cycle stage, brand parity, brand symbolism, affordability, other consumers, and self-
40 concept based influences (see Figure 3.2). Th e reduction of the thematic map was based primarily on the research objective, which was to identify motivations of self-brand interactions. Thus the broad order theme of brand influen ces was isolated, and the subthemes of other consumers and self-con cept based influences were further broken down into individual factors. Based on the refined thematic map, the current research focuses on the motives identified in the anal ysis along with the re lationship between the self-concept and motivated brand consumption. Each motive serves as a lower order code to the sub-theme self-concept based brand in fluence, while brand in fluences served as the higher-order theme. Figure 3.2 Final Thematic Map
41 3.5. How Identity Motives Lead to Identity Centrality Based on the final thematic map in Figure 3.2, the focus of data analysis surrounded self-concept based brand influences. In particular, the transcriptions were coded for identity motives and the presence of multiple identities. The results reveal identity motives were influential in self-b rand interactions and oftentimes played a significant role in a specific identity becomi ng central. This is because, identity motives beyond self-esteem and self-consistency aided them in narrowing their evoked set of brands. The qualitative findings show individuals used brand a ssociations to satisfy their motivations of: self-esteem distinctiveness continuity belonging self-efficacy belonging meaning recognition consistency and security These identity motives were based on a review of identity influences in both the marketing and social psychology literatures, and further supported by the res pondents interviewed. Next, each motive is identified, defined, and its relevance to self-b rand connections is pr ovided in the form of testable research propositions. 3.5.1. Self-Esteem Motive The self-esteem motive is defined as the motivation to maintain and enhance a positive conception of oneself (Gecas 1982, p. 20) This definition is similar to that of Sirgy (1982) who defined self-esteem as the te ndency to seek experiences that enhances an individuals self-concept. Self-esteem may be increa sed directly through selfenhancement or indirectly through self-improve ment. This motive is typically maintained through self-verification (Sed ikides and Strube 1997).
42 Sirgy (1982) suggested the e xperience of building self-esteem allows individuals to enhance their overall self-concept. This is evidenced by informant Tonya2, who talks about her fashion brands: Baby Phat outfits get me compliments; they make me feel good about myself. They say stuff like... oh she got some money, cause she got a Baby Phat outfit on. Since wearing this particular brand of clothing generated a positive level of self-worth through apprecia tion and compliments, creatin g self-esteem through fashion was important as others perceived her as weal thy. Self-esteem helped foster a coherent view of herself that was beyond fashion; such that others who perc eived her would draw other implications about her self-concept (e.g. wealth) beyond the image of the Baby Phat brand. Beyond receiving compliments and feeli ng appreciated by other individuals, another informant talked about what it means to have low self-esteem in fashion. For her self-esteem was a matter of conspicuous cons umption, and internally driven. Her, selfworth was not based on how others viewed her, but rather, her internal sense of confidence. When questioned about ideal wa y to exhibit self-esteem through fashion, Nikki states: Wearing what you like, not thinking about what other people are going to judge what you are wearing, and I think someone who has self-esteem doesnt have to wear something really revealing, promote a bad message. Something that maybe it doe snt look good to you, but if they like it (meaning the person who is wearing it) then they are happy wearing it. Like a shirt from Target. Tonyas and Nikkis responses represent tw o distinct sources of self-esteem. One in which an individuals wort h is assessed based on external assessment from others. The other in which self-esteem is intrinsically assessed independent of others perceptions. 2 The names of all informants have been changed to conceal their true identity.
43 The external view (e.g. Tonya) represents a form of self-verification, which allows her to maintain her self-esteem (Sed ikides and Strube 1997) thr ough fashion. The internal view (e.g. Nikki) is tangential to se lf-affirmation theory (Steele 1988), whereby an individual is motivated to maintain a level of integr ity within their self-concept. Despite their differences in how they view self-esteem, it is recognized that this motive can drive selfbrand interactions that enhan ces ones self esteem; or in Nikkis case serve as a motive for the omission of brands that give others the opportunity to a ssess her self-esteem. 3.5.2. Distinctiveness Motive The distinctiveness motive refers to the establishment and maintenance of a sense of differentiation from others. (Brewe r 1991; Vignoles, Chryssochoou, and Breakwell 2000). When distinctiveness needs are threatened or frustrated, individu als will engage in cognitive or behavioral c oping strategies to restore a sense of di stinctiveness (Branscombe, Ellemers, Spears, and Doos je 1999; Breakwell, 1988; Brewer 1991). The distinctiveness motive was used more frequen tly by respondents to reveal motivations for consuming in conjunction with a given social identity compared to a person identity. With regard to her social identit y, Rebeccas satisfies motivations of distinctiveness within her re ference group. When asked if sh e desires to be distinctive among her close friends, or with strangers, sh e suggests that group di stinctiveness doesnt really happen, and continues to talk about her social group of friends: So you have people that dress very punk rockish and they are trying to be different (the y wear black th ings and wear chains) . gothic. But then ther e is a whole group of them, so they are not different. For me it is more of the friends that I come into contact with on a daily basis, Id rather be distinctive among my friends versus people I didnt know. .Some people have to buy a br and and it has to have the name all over it, like BEBE in big bold letters, I cant stand
44 that. Like the Louis Vuitton bags, the same thing, I want high fashion, but not like what everyone else has. I dont want to go around looking like them. Rebeccas thoughts about the Gothic indi viduals, highlights the distinctiveness of an overall group. In this scenario, member s of the group strive to be collectively distinctive from the larger population. This is considered a form of intergroup distinctiveness (Tajfel and Turner 1979). On an individua l level, Rebecca talks about her interest in wearing high fash ion clothing, however she doesnt want to be branded like everyone else in her group. She prefers to wear clothing that ha s covert labeling. Her behavior is supported by the Optimal Distinc tiveness Theory (Brewer 1991), which states individuals strive for both assimilation a nd distinctiveness within groups. Rebeccas responses suggest she strives for within group distinctiveness because it is more relevant to her identity goals. Conclusively, distinctiveness is an important motive of identity presentation because it serves as a means of self-evaluation through social comparison. 3.5.3. Continuity Motive The continuity motive is defined as an individuals motivation to maintain a sense of connection across time and situati on within identity (Breakwell 1988, p. 24). This motive represents an identifiable concep tual thread uniting th e past, present, and future within a persons iden tity across time (Bre akwell 1988). James (1892) was one of the first to promote continuity as a chief feature of the identity. This motive focuses on a particular memo ry, or an identity that unites their current self-concept with an experience from th eir past, or perhaps, a desired image in the future. This behavior can be thought of as a t ype of identity nostalgia. Take for example, Candace, she is a Health Information Systems Manager, and is expecting her first child soon. Its a girl, so here she talks about the things she would like to share with daughter
45 from her childhood, and even the products she used as child that she will also purchase for her daughter. She explains why she still uses the same brand of moisturizer from a child, and how she will use the same brand on her daughter. Palmers Cocoa Butter, I used it as a moisturizer, and have used it ever since. With the impending stretch marks, I have been putting it on my belly tw ice a day. And when she is old enough, Ill use it on her ( unborn daughter). Lets see what else. Oh yes, barrette s and beads in my hair, my mother would braid our hair and make it pretty with the beads and barrettes. I cant recall any brands of them, because I was so little, bu t I want to do her (unborn daughter) hair the same way. I dont really know how to braid but I will learn. I want he r to experience a lot of the same things I did when I was a little girl (smiling). Candaces current use of the Palmers Cocoa Butter stems from her use of the product as a child. Even as she goes through her pregnanc y, she continues to use the brand. As an adult she has access to a host of different bra nds of moisturizer, but she has opted to continue to the Palmers brand. Her comm itment to use a product from her childhood unites her childhood self to her adult self through continued consumption of the brand. Her plans to introduce her daughter to the bra nd and braid her hair in the same manner as her mom did to her, shows she is establis hing continuity between her childhood identity and her soon-to-be mother identity. 3.5.4. Self-Efficacy Motive The self-efficacy motive is defined as an indivi duals desire to maintain and enhance feelings of competence and cont rol (Breakwell 1993, p. 205). Bandura (1977) argues the role of self-efficacy beliefs in hum an functioning is that "people's level of motivation, affective states, and actions are based more on what they believe than on what is objectively true" (p. 2). For this re ason, how individuals behave can often be
46 better predicted by the beliefs they hold about their capabilities th an by what they are actually capable of accomplishing. With the increase use of self-service t echnologies and increased choices in the marketplace, consumers will increasingly become motivated to show feelings of competence and control. Depending on the identity in question, the need for competence and control can manifest itself in a host of different ways. For example, older consumers like Norman, exhibit competen ce by relying not only on thei r experience with the brand, but also their knowledge of th e product category. When asked why he continues to be a loyal Volvo customer, he talks about hi s knowledge and experience with the Volvo brand. Well like I said I like to shop based on my smarts, and if I am going to invest 30 grand into a car, I want it to last at least until my retirement ki cks in (laughing). Volvos are good cars. They last. My stati on wagon, I have had that car for almost 12 years now. And it runs pretty well. With the exception of a few minor repair s, I havent really had any major problems. His mentioning of his smarts, length of ow nership and the small number of repairs are indicators of his self-efficacy. In the next excerpt, he explains his role as a father and husband in his decision to buy Volvos: Being the responsible father, I wanted Sascha (his daughter) to have a car that was going to last her a while, at least until she was able to put a dent into her student loans. She would have a reliable car, and wouldnt have to worry about buying a new one. Well not at least until she was in a position to afford a new one. If she is a smart girl like her father, shed buy a Volvo, when that time came around (smiling). My wife, well she r uns a Jewelry store, and she is price conscious like me. But she wanted a nice car without us both having to spend a lot of money. So the solution to that is . I kept the old station wagon, and she
47 drives the newer sedan. We ar e both sold on the quality of Volvos. I dont need to convince her much. Normans commitment to providing reli able transportation for his family is exemplified through his choice of the Volvo brand for both his wife and daughter. As a husband and father his desire to exhibit comp etence in finding automobiles for the two of them is manifest in his second quote, where he talks about giving his daughter a Volvo for her graduation gift, and giving his wife th e new Sedan. His experience with the Volvo brand for 12 years reveals his knowledge of the brand. Thus he is not only motivated to satisfy feelings of self-efficacy of autos and/or Volvos for his daughter and wife, but he is also motivated to demonstrate his comp etency and knowledge of the Volvo brand. Self-efficacy perceptions help determine what individuals do with the knowledge and skills they have in selecting a specific brand. More importantly, high levels of selfefficacy allow individuals to feel skilled in the marketplace. The self-efficacy motive appears to be important for identity presenta tion due to the enduri ng beliefs that motivate behavioral capabilities of competence and mastery. 3.5.5. Belonging Motive The belonging motive is defined as the need to maintain or enhance feelings of closeness to, or acceptance by other people, wh ether in dyadic relationships or within a group Baumeister and Leary (1995). Thes e authors identified this motive as a fundamental human motivation with two main features (p.47). Firs t, individuals need frequent personal contact or interaction with other individuals. These interactions are affectively positive or pleasant, but more im portantly they should be free from conflict and negative affect. Second, indi viduals need to perceive that there is an interpersonal
48 bond or relationship marked by stability, aff ective concern, and continuation into the foreseeable future. The belonging motive captures two important aspects of group influences, social embeddedness and intimacy. Individuals are so cially embedded in their groups and use their group memberships as a source of reference during consumption experiences (Bearden and Etzel 1982; Escalas and Bettman 2005). A variety of affiliation-based scenarios emerge based on this intention, lead ing to a particular identity to becoming central. Erroll, a financial advi sor, explains his need for in timacy with family and friends when considering what type of automobile he would buy: I associate the F150 with friends, because with friends we think of intimacy. With the regular cab, not the extended cab, but the regular cab, you can pretty much fit one person, maybe two. They have close seats which provides for a lot of intimacy between people. And whats one thing that friends ask people to do, and that is to go out. You know when you got a truck, people ask can you haul this and haul that. Also, the practicality of the truck. His mentioning of the intimacy, as well as the notion of being needed by his friends to haul items satisfies his need to belong. The Ford F-150 will enable him to be needed by his friends, as well as provide intim acy for him and others when riding in the cab. The need to belong with so cially identified others, in cluding family and friends evolves around some level of reciprocal action, associative n eeds, and repeated interactions. Brand associations that satis fy an individuals belonging motive aids in identity centrality, as witnessed in the case of Erroll. This motive is important for its affective value, and as a means of s ubjective well-being through group affiliation.
49 3.5.6. Meaning Motive The meaning motive refers to the need to find significance or purpose in ones own existence (Baumeister 1991). Marketers have long considered an individuals search for meaning a critical feature of consumer behavior. The depiction of ones existence as meaningful is a principal attribute of ps ychological well-being (McCracken 1986; 1988), self presentation (Schau and Gilly 2003), and narrative processing (Escalas 2004). A consumers search for meaning plays an esse ntial role in brand choice (Ng and Houston 2006), and brand relationships (Fournier 1998) Stephanie, a recent college graduate, explains why she wears the Ab ercrombie and Fitch brand ( the researcher begins the conversation and the dialogue continues thereafter ): When you see others wearing A and F and you dont have it on, what do you think? What do I think? Well most pe ople I see are younger girls, and I think what a brat, their mom buys them all of their clothes (laughing). But the r eason that I like A and F is really stupid, but I like the moose on their shirts. Because I like moose. Why? Because its cute, and Ill buy anything with birds and moose on it. Also my mom co llects Christmas moose and things like that. What if Louis Vuitton made a handbag that has its LVs all over it, and then they had moose on it, would you buy it? (The informant mentioned earlie r that she refused to carry Louis Vuitton handbags becaus e they are carried by everyone, and they were typically fake handbags.) If it was cute then yeah, because I like moose. I always say like for Christmas people either have a snowman or Santa Clause, my mom used to be all snowman and Santas, but I am going to be reindeer and moose. We usually have one (moose) that we decorate every year.
50 Any other reasons you like Abercrombie and Fitch, besides the moose? Probably because it is the highest ( most expensive ) out of American Eagle and all of them, and Hollister. The significance of brands can take on an assortment of meanings. As witnessed with the case of Stephanie, meaning can originate from various experiences and can permeate a host of consumption experience s (i.e. Abercrombie and Fitch, Christmas moose). What is particularly interesting about her significance for moose is that it is not only based on a family tradition surrounding Ch ristmas, but has also extended into her choice of fashion brands. The level of significance a consumer a ssigns to a brand bui lds a historical narrative of identity across time. In this way, the importation of meaning aids in identity maintenance and creation. Thus a brand that evokes meaning for the consumer becomes imbedded in their identity (LaTour, LaTour, and Zinkhan 2007; Escalas and Bettman 2005), and is likely to have a high level of importance during self -brand interactions. 3.5.7. Recognition Motive The recognition motive is defined as the need to be acknowledged or rewarded. Acknowledgement can come in the form of bene fits directly relate d to the individuals actual or desired self concept and can increase the individuals attr activeness, power, and commitment to the brand (Elliot and Wattana suwan 1998). Alternativel y, as in the case of Marcela, an older woman who is a Graduate Admissions Director at a small liberal arts college, acknowledgement can be negatively rela ted to the individuals actual or desired self-concept. She was asked if others assess her based on the type of car she drives. She replies yes. I ask her to share an experience that has led to her c onclusion. She explains:
51 Right now I have a twelve year old Mitsubishi Gallant, Its a black car, but the paint on the car has peeled off on the back of the trunk and part of the roof and it looks speckled like a black and white car. It makes the car look beat up. But its a good car. People see me in that car and make judgments that I am poor, and in an economically low social class, otherwise I woul dnt be driving in such a low end car. If you pull up at a stop light, and someone is next to you in a Mercedes, they look at you in your car and you can tell by their facial e xpression that they are so not digging you. I thought for a long time it was because of my skin color, but people of my own skin color do the same thing when they drive fancy cars. But the funny thing is people who are in cars more beat up than mine, they say hi to me.(Laughing) In an earlier part of he r interview she stated that the Lexus IS350 are driven by mothers and IS250s are driven by college st udents. Marcellas experiences of being acknowledged by consumers whose cars are in wors e shape than her, is contrasted by her desire to not be associated with people w ho drive the Lexus IS350 and IS250. I ask her to explain why she doesnt like to be acknowle dged by mothers who drive the Lexus IS350, but doesnt mind being acknowledged by a Mexi can driver, whose car is in worse shape than hers. In order to get at this feeling, the respondent was asked to place the Mexican driver in a Lexus IS250 or IS350 and talk about her re sponse if she were to see them at a stop light or at a st op sign. She explains: If I pull up at a stop sign a nd they look Mexican I dont care what they think because their car is more beat up than mine. But if a Mexican pulls up in a Mercedes I think they will judge me based on my car. These thoughts are real! I am not making this up. The sad part is it never really bothered me until I got into a diffe rent salary level. Up until I hit the six figure salary le vel, I didnt mind. I dont know. I just dont want people t o .. No! I dont want people to think of me as just anothe r statistic as a poor African, blacklooking statistic. I want them to see me an upwardly mobile successful woman.
52 What is interesting about her comparis on between a Mexican i ndividual driving a car in worse shape than her own versus a Le xus, is that her need to be recognized is motivated by her salary level. She wants to be recognized as a six-figure professional woman who is upwardly mobile. Since she is in the market for a new car, her goal is to buy a car that will encourage others to recogn ize her as the upwardl y mobile successful woman. This can be thought of as her cen tral identity. The r ecognition motive occurs through social comparisons both with similar and different others, and is important to identity centrality due to its self-evaluative properties. 3.5.8. Consistency Motive The consistency motive is defined as an i ndividuals motivation to maintain uniformity of identity across situation and time. Consumers are faced with many choices of brands in the marketplace and each brand represents something unique in the minds of the c onsumer. From brand positioning to brand attributes, these images conjure up corre sponding images for the consumer. If a consumer perceives a direct parallel betw een the brand and thei r self-image, they are more likely to purchase it (Dolic h 1969; Sirgy 1981; 1982). However, if the brand in some way conflicts with an i ndividuals self-image, the individual is more likely to bypass that brand for one that is more in line with their self-image. Marcella speaks in very specific terms of automobiles brands being representative of certain stereotypes. Below, she explai ns why none of the brands of cars are consistent with her self-image.
53 They (Lexus IS350) are not s porty. They look like someone who has a family of four. I wa nt something that is sporty, that makes me look upwardly mobile and professional. Something that says Ive ma de it. Im single and I dont have a family. I dont want peopl e to think that I am older person. You know, so what I do is . while I am driving, I look at cars and I look at who is in them, and how the cars look, and pretty much everything except for the IS350 I see mature looking people in them, looking like they have kids. Lexus also has an IS250 but the people that I see in the 250 are like young collegey kids. Its fast and sporty, but the young college kids are in them. But I dont want anyone to think Let me correct myself, I dont want people to assume that I have kids. Th ats why I dont only want to buy that model. I also will not buy the IS250 because I dont want anyone to think that I am one of those just trying to get my hands around lif e college kids. Does that sound kind of vain? It is evident that Marcellas views are based strongly on her perceived user imagery of the two types of Lexus bra nds (IS350 and IS250). In one regard she doesnt want to be associated with older consumers who have families, but she wants to avoid association with young colle giate students. This is theorized and supported by other researchers who suggest user stereotypes do in fact shape nonuser attitudes (Grub and Hupp 1968; Gr ub and Stern 1971; Chaplin and RoedderJohn 2005). This comparison between the consumer and the brand image is conceptualized in Sirgys (1982) Produc t-Image Congruity Theory, which states consumers are likely to form purchase in tentions based on their need for selfesteem and self-consistency. Consistency is an important motive with regard to identity presentation. In any instance it informs an indivi duals behavior, and in a consumption context, if informs brand choice.
54 3.5.9. Security Motive The security motive is defined as the motivation of an individual to protect their overall self-concep t from internal and external threats (Ashmore, Deaux, and McLaughlin-Volpe 2004). In particular, th is motive addresses an individuals need for understanding experiences a nd outcomes that construct identity, including direct responses to identity threat. Branscombe, Ellemers, Spears, and Doosje (1999) proposed four types of so cial identity threat within groups. The first threat deals with categorization wh ereby individuals are categorized against their will. The second type of threat deals with socially categorized groups. Here a threat exists when the value of the in -group is undermined. The third type of threat is that which contests an indi viduals status as a good (or prototypical) member of the group. Lastly, the four th type of threat addresses the distinctiveness of the in-group, a nd when information challenges the distinctiveness of the in-group compared with the out-group, high identifiers will feel threatened by any perceived sim ilarity with the relevant out-group. When a consumer is exposed to external threats (e.g. derogation, alienation), or internal threats (e.g. low self -esteem, lack of confidence) he/she will be motivated to protect their self-concept. This is achieved through certain be haviors that ensure the selfconcept is expressed accurately and is shield ed from external threats. In the case of Stefanie, when asked about brands comparab le to A and F that she would consider wearing those brands, she replied:
55 Hollister is like more the middle-school aged children, and A and F is for high school ki ds and older and Rule, well no one knows about Rule, and th ey only have a few stores, but no one knows about it. So you would well assuming that you can fit the clothes in Hollister. Would you wear them? Yeah, I can fit them, but I rea lly dont go in there because its for high schoolers, but every now and then, Ill buy a tshirt out of there. Stefanies desire not to go into the Hollis ter store because she views it as mostly for high school aged consumers is indicativ e of her attempting to protect her recentcollege graduate identity. One could argue that her desire no t to be associated with highschoolers could be motivated by her need for distinctiveness versus security. However, the distinctiveness motive operates primarily within an individual s referenced ingroup (Brewer 1991). In Stefanies case, high-sc hoolers would be an outgroup for her for two reasons. First, she is a recen t college graduate and is looking forward to her first professional job. Second, the Hollister brand is positioned for high-schoolers aged 14-18; and the Abercrombie and Fitch brand is targeted at young people aged 18-22 (www.abercrombie.com\ourbrands.html). The security motive deals with an indivi duals need to protect their overall selfconcept including direct responses to identity threat (Branscombe et al 1999). Given this, the security motive appears to be equally important for both identity preservation and presentation. This motive can in centrality as identities increase in importance to mitigate or ward off identity threat. The nine identity motives presented above illustrate the various motivations that influence self-brand interactions and lead to identity centrality. The depth interviews
56 provide rich evidence of the highly contextual nature of identity motives. As these motives are satisfied, brand in teractions are facilitated th rough the self-concept. The motives presented here are not meant to be an exhaustive list, but rather a formidable start of multiple identity motive research. 3.5.10. Identity Centrality Individuals are motivated by multiple iden tities which each have their own level of significance in a consumption environm ent (Kleine, Kleine, and Kernan 1993). Significance of identity is conceptualized as identity centrality, defined as the level of importance an individual assigns to a partic ular identity (Thoits 1986, Stryker and Serpe 1994). This construct was initia lly proposed in the social psychology literature (Strkyer and Serpe 1994). However its relevance to the marketing domain has yet to be explored. In this section, a case is made for the iden tity centrality in self-brand interactions. A consumers identity is consider ed central when that specific identity is important to not only to ones global definition of the self but al so dominates other iden tities in ones view of the brand. It is put forth that identity cen trality will play a critical role in the brand considerations and the formation of associated attitudes and behaviors. Identity centrality is posited to be driv en by identity motives which are shaped by a consumers ideal self in accordance with their higher-order ident ity goals (Breakwell 1983). As a brand primes a higher-order identity goal, corresponding identity motives are satisfied by brand images and/or associations which in turns lead to identity centrality. One illustration of how identity centrality influences the choice of brands is shown through David, the contract firefighter in Ir aq. When asked what will be the primary reason for the type of automobile he c hooses, he explains hi s answers below:
57 ( Researcher ) So when you purchase your car, what is going to be the driving factor on w hy you purchase that car? Is it going to be the fact that you ar e a father, or the fact that you are a fireman? (David) On whether I buy it, or ( Researcher ) The brand that you choose? (David) It would be for quality, it will be umm probably for luxury, status. ( Researcher ) Is that related to your black male identity? (David) What the status symbol? ( Researcher ) Yes. Or is it more related to your father identity? (David) Well thats more of my black male identity not really my father, if thats th e case then I would get a Volvo Station wagon. (laughing). It s a little bit of both all wrapped up. You know I cant say which one it would be . Well the car will be more for status. So if I really wanted just straight up stat us, I would choose the Lexus, and if I was going for the responsible, competent father, then Id choose the Volvo or the Honda. Davids various identities and roles are al igned with certain automobile brands. In one sense, he is focusing on the status of the car which corresponds to his overall self definition. On the other hand, he is talking a bout a more specific level of brand-identity congruence. His multiple ident ities are implicated in his explanation of why he would consider the Lexus (e.g. status), Volvo a nd Honda (e.g. responsible competent father) brands. This is supported by Settles (2004) who concludes central identities provide social validation and offer a framework fo r interpreting the world. Similarly, Thoits (1986) suggests that a centra l identity may also provide i ndividuals with scripts (or
58 guides) for how to behave. In this case identity centrality guides the decision-making process. 3.5.11. Reference Group Brand Associations Reference groups can serve to socially va lidate an individu als self-concept through social comparison (Folkes and Kies ler 1991), reference group members are a source of information for arriving at and eval uating an indviduals be liefs about the world (Escalas and Bettman 2003). The congruency between group membership and brand usage (Bearden, Netemeyer, and Teel 1989, Moschis 1985) has been found to influence brand choice and self-brand connections (Escalas and Bettman 2003; 2005). However because reference groups can serve to soci ally validate an individuals self-concept through social comparison (Folkes and Kies ler 1991), reference group members are a source of information for arriving at and eval uating their beliefs about the world (Escalas and Bettman 2003). Informant, David talks about why he has a penchant for Escalades: Just straight up me! Me being a black male, umm just, just strictly frontin. But the fact that it will be a big enough vehicle for my family would be a plus, but thats not really what I am buying that for. Im buying that for status. For a big, pretty vehicle thats it. By placing his desire for an Escalade with in the context of other black men, he is using reference group associa tions to associate himself with the Escalade brand. The brand association of black men is what comes to mind when he thinks of Escalades, and since he is also a black male, he feels as if the vehicle is congruen t with his identity. Another informant Marcella uses her gi rlfriends as her reference group. They not only influence her purchase of high end brands but they also influence her concealment of low-end brands. I asked about another product category beyond au tomobiles that is
59 symbolic of who she is, and she replies that shoes are just as symbolic as automobiles. She also goes on to express how she intentionally tells the name brand of her shoes to her girlfriends: So I like designer shoes, but they dont all have to be designer, they just have to look cute. Sometimes I drop the name of the brand in front of my girlfriends like Minolo Blanick or Via Spiaga. Some of my friends think theyre all that and they know fashion, so because they act that way, we always talk about what we have on. They say oh this suit is Chinese Laundry, this suit is Gucci, or I just picked up this shirt from the Armani Exchange. When I get around them, I will let them know, that it is something expensive. So my friends who arent that snooty, I will say I just got this at a bargain pr ice, and I will tell them how much I paid for it. But with my snooty friends, I will always lie about the price and ne ver tell the true price of it. And say that I bought it at another store, and never an outlet. When asked if she would ever wear a pair of Payless shoes, she replies: Yes. But only if they were re ally cute, I w ould never tell any of my friends, and I would never let them see the inside label cause it will say Payl ess. And if we had to go somewhere to take off our shoe s, I wouldnt take them off, Id be too embarrassed. I cant have my friends thinking I shop at Payless. Marcellas use of reference groups to both influence her choice of luxury brands and to conceal her use of bargain brands shows the magnitude of influence closely affiliated reference groups have on brand consumption. This is different than Davids reference group, which was mu ch broader, and the group was not as closely affiliated. In this way, choosing a known brand help s individuals explain their actions to themselves and others. If the persons refere nce group is familiar with a brand and its associations, it may also indicate to others in the group that the individual has made a
60 good decision based on the brands consistent image with both the individual and the group. Or as in Marcellas case, brand associa tions may also indicate to others in the group that the individual has made a poor decisi on, by selecting a brand inconsistent with the group. 3.5.12. Brand Symbolism The symbolic associations of brands play an important role in shaping the selfconcept. This is largely due to an indivi duals social identity and the brand image. According to Levy (1986) indivi duals engage in symbolic be havior to boast consistency in their self-perception. This means that indi viduals will behave and consume in ways that suit their identity and enhance their self-e steem. Specifically, it means that symbolic consumption is the outcome of how a consumer interprets what is needed to support an identity (Levy 1986; Rook 2001). This is ak in to identity commitment (Foote 1951). When individuals encounter a brand, their re action to it depends on its meaning to them; and its meaning depends on the brand associ ations (McCracken 1986), user imagery or prototypically (Keller 1993); and psychologi cal benefits the brand may offer (Aaker 1991; Fournier 1998). From this perspective it can be implied that when individuals engage in symbolic consumption they are in fact establishing self-image congruency, which is defined as a matching between how an individual sees himself in relation to others and the brand image (Levy 1986). This congruency is posited to be based on an assortment of identity motives and meanings ascribed to the brand. 3.5.13. Self-Brand Connections Brand associations are presumably more m eaningful if they are linked to the selfconcept (Escalas and Bettman 2003). This linkage is manifest through a self-brand
61 connection, defined as the extent to which individuals have in corporated brands into their self-concept (Escalas 2004; Es calas and Bettman 2000). Indivi duals use brands to create and communicate their self-concepts, an in the process create self-bra nd connections. It is posited that self-brand connections reveal a prominent part of the individuals selfconcept and are based on an individuals cent ral identity relevant to the brand. For David the firefighter, he stated that he would purch ase the Escalade because that is the car that most black men drive, but he also goes on to say that that the Escalade is strictly me. This statement exemplifies what self-brand c onnections are all about; the integration of a brand into ones self-concept. In the case of strong self-brand connections, the individuals begin to define who they are in terms of the brand, I am a Coach woman. 3.6. Conceptual Framework The proposed model was derived from the interview findings and consideration of the Identity Process Theory (Breakwell 1986; 1993). It is pos ited that brand associations aid in identity presentation and are more meani ngful if they are linked to the self (Escalas and Bettman 2003). This linkage is manifest through self-bra nd connections, defined as the extent to which individuals have incorpor ated brands into their self-concept (Escalas and Bettman 2003; 2005). Thus it is posited that self-brand connections reveal a prominent part of the indivi duals self-concept and are base d on an individuals central identity relevant to a brand. As the model in Figure 3.3 shows, identity centrality mediates the formation of self-brand connections. The ability of a sp ecific identity to become central by the satisfaction of multiple identi ty motives is posited to infl uence self-brand connections. Given the dynamic interaction of an individual s multiple identities, one might intuitively
62 expect that self-brand connections will be enhanc ed if a given identity is central to brand choice. This is due to the fact that indi viduals have multiple identities based on their traits, roles, and membersh ips in socially categorized groups (Ashmore, Deaux, and McLaughlin-Volpe 2004) with va rying levels of importance. If an identity is not considered central to an indivi duals overall self-concept, then it is assumed that identity ranks low on the identity hierarchy. Thus it will not influence self -brand connections as much. Form a process standpoint, not only is identity centrality posited to be driven by individual identity motives (P1 P9), but it is proposed that identity centrality can further be driven by the simultaneous influe nce of multiple identity motives (P10). Therefore it is predicted that consideration of the brands image results in the activation of multiple identity motives allowing a central identity to dominate. It is also pr edicted that as these motives are satisfied, they will influence self-brand brand connections. Hence, identity centrality may serve as a mediator between the satisfaction of identity motives, brand choice, and related self-brand connections. Th e relationships between the constructs are further supported by the aforementioned inform ant responses and are integrated into a propositional inventory for futu re research (see Table 3.2.).
63 Figure 3.3 Conceptual Model of the Role of Identity Centrality in Self-Brand Connections Sel f esteem (P1 ) Distinctiveness (P2 ) Continuit y (P3 ) Efficacy (P4 ) Belonging (P5 ) Meaning (P6 ) Recognition (P7 ) Consistenc y (P8 ) Security (P9 ) Identity Centralit y (P1 1) Brand Choice Sel f -brand Connection (P1 4) Reference Group Brand Associations (P 12 ) Brand Symbolism (P1 3 ) Identity Motives (P10 )
64 Table 3.5 Research Propositions for Figure 3.3 Number Identity Motive/ Construct Proposition 1 Self-esteem The extent, to which a consumer is motivated to enhance their self-esteem, will influence identity centrality. 2 Distinctiveness The extent, to which a consumer is motivated to maintain distinctiveness from others, will influence identity centrality. 3 Continuity The extent, to which a consumer is motivated to establish continuity in hisor her self-concept, will influence identity centrality. 4 Self-Efficacy The extent, to which a consumer is motivated to demonstrate efficacy in hisor her self-concept, will influence identity centrality. 5 Belonging The extent, to which a consumer is motivated to meet their desire to belong to a reference group, will influence identity centrality. 6 Meaning The extent, to which a consumer is motivated to establish meaning in their self-concept, will influence identity centrality. 7 Recognition The extent to which a consumer is motivated to receive recognition from others, will influence identity centrality. 8 Consistency The extent, to which a consumer is motivated to gain consistency in his or her self-concept, will influence identity centrality. 9 Security The extent, to which a consumer is motivated to protect his or her self-concept, will influence identity centrality. 10 Multiple Motives The extent, to which consumers are driven by multiple identity motives, will influence identity centrality. 11 Reference Group Associations The extent to which a reference group identity is central will influence self-brand connections when brand associations are consistent with a reference group. 12 Brand Symbolism The extent to which a brand has symbolic properties will moderate the relationship between identity centrality and self-brand connections, when brand associations are consistent with the reference group.
65 3.7. Discussion and Implications The findings provide a starting poi nt for understanding how self-brand interactions develop when multiple identitie s driven by the simultaneous influence of multiple motives influence brand consumption. The analysis of the informant responses suggests that consumers associate a future or current identity with a particular brand, because that brand not only satisfies multip le identity motives, but also allows the achievement and presentation of a desired self-image. By considering multiple motives and the isolation of consumer identities, mark eters may be better informed to speak to the varied nature of todays consumers. This is important for marketing communications because oftentimes marketer s target a specific market segment while overlooking the heterogeneity within the segment. If the segmented identity is not central for the consumer and a competing identity become s central, then the communication becomes ineffective. By isolating the appropriate identity through cen trality measures ineffective communications can be avoided. From a theoretical perspective, the process of self-brand connections conforms to the Identity Process th eory (Breakwell 1993) such that identity is constructed based on motivational principles (e.g. motives). As id entity motives are satisfied through brand associations, the consumer is likely to inco rporate the brand furt her into their selfconcept. When the informants considered their self-concept in their consumption experiences, their brand choices were driven by the identity motives identified in this present study Therefore, one important theoretical c ontribution of this research is the development of a conceptual framework to accommodate identity centrality in the formation of self-brand connections that bui lds from Breakwells Identity Process theory,
66 and is supported by informant responses. By in troducing Identity Pr ocess Theory in the consumer behavior research, research can be broadened to incorporate the management of multiple identities (through centrality) influencing cons umption. Thus by combining the study of identity motives with centrality, the role of the self-concept in consumer behavior can be extended to accommodate the in fluence of multiple identities. This view of multiple identities is supported by the changing focus of self-concept literature in both consumer behavior and social psychology (s ee Chapter 2 for a review) from a unitary view of the self to a more dynamic and situational self. 3.8. Research Limitations and Future Research The conceptual framework and findings of this research provide important managerial and theoretical insights into the ce ntrality of identity in forming self-brand connections. Notwithstanding these insights, several limitations should be addressed. First, this study was limited to the participants featured in this study; therefore applying the framework to individuals from other cultur es would prevent the generalization of this conceptual model. Certain motives may not be relevant in more collectivist cultures, where group behavior is empha sized over individual behavior. Second, caution must be exercised in drawing conclusions outside the scope of this research. These results are from consumers whose knowledge of brands a nd motivations for specific brands were assessed in a brief interview. Ther efore generalizablity may be limited. Future research is warranted to ad dress these limitations and expand the theoretical validity of the framework. One opportunity for future investigation that deserves further attention is the potential for conducting comparative multiple-case analyses. This could be a series of case studies exploring a va riety of consumption
67 contexts assessing identity centrality within the same consumer over different time periods. These studies could provide insight into how self-brand connections actually form in cultures where brand exposure may be limited. 3.9. Chapter Summary Though obviously exploratory in nature, th ese findings suggest the prominence of identity motives informing identity centralit y. The goals of the rese arch were two-fold: (1) to probe the existence of multiple identity motives in brand choice/consumption and (2) to propose a framework of identity-m otivated consumption. Great potential in applying identity centrality to consumer behavi or research was demonstrated and its use as a mediator in self-brand connections identi fied. A formal test of the conceptual model developed in this chapter forms the remain ing two essays. In the next chapter, the simultaneous influence of multiple identity mo tives is tested using automobiles as the contextual setting.
68 Chapter 4 Essay 2-Multiple Motives, Multiple Selves: Why Self-Esteem and SelfConsistency Doesnt Fully Expl ain Self-Referent Consumption In this chapter, the influence of multipl e identity motives on identity centrality and self-brand connections was empirically te sted. Using multiple regression, the results indicate motives of recogniti on and continuity were positiv e predictors of identity centrality; and the belonging and security mo tives were positive predictors of self-brand connections. Additionally, identity centrali ty was found to mediate the relationship between identity motives and self-brand connections. 4.1. Introduction One of the least contested claims in marke ting is that individuals are motivated to enhance their self-esteem and establish cons istency in their self-concept when engaging in symbolic consumption (Levy 1959; Gr ub and Grathwohl 1967; Sirgy 1982; Levy 1986). However, less attention has been directed towards additional identity motives guiding self-brand interactions. Recently, in the social psychology literature, motives beyond self-esteem and self-consistency were found to aid in identity construction namely: efficacy, meaning, distinctiveness, and belonging (Vignoles, Regalia, Manzi, Golledge, and Scabini 2006). These alternative motives can provide a richer understanding of how and why individuals are motivated to engage in self-brand interactions. Prior research has implicitly viewed the self as a stable construct (Sirgy 1982, 1985) focusing on a pre-established identity (Escalas and Bettman 2005) or multiple
69 identity elements within a single doma in (Sirgy 1982; Sirgy 1985; Sirgy, Grewal, Mangleburg, Park, Chon, Claiborne, Johar, a nd Berkman 1997) weakening the external validity of their research findings. This is also true of the product image congruency theory (Grub and Grathwohl, 1967; Ericksen an d Sirgy 1992; Kleine et al. 1993). Markus and Kunda (1986) put forth the term malleabl e (or working) self-c oncept, which refers to a host of self-conceptions (e .g. ideal self, perceived self, so cial self) that can be made accessible at a given moment. Each self leads to the presentation of different identities, providing social validation and a framework for interpreting the world through different scripts and serves as a guide for behavior (T hoits 1986). It is argued that multiple identity motives are associated with the importance of each identity. Identity centrality, defined as the psyc hological importance an individual places on a given identity (Stryker and Serpe 1994; Settles 2004). Speci fically, identity centrality reveals how a single identity is exacerbated while remaining identities are buffered. This process is akin to identity management, whereby each identity varies in importance to the overall self-definition infl uencing affective, beha vioral and cognitive outcomes (Stets and Burke 2000). Because mu ltiple identities operate through a process of identity centrality, and importance is almo st always associated with some degree of motivation (Bagozzi, Bergami, Leone 2003), it is argued that as multiple identity motives are satisfied a given identity increases in centrality. Taken together, the consideration of a dditional identity motives and identity centrality addresses the shortcom ings of research on self-brand interactions. Based on the arguments presented above, the current study so ught out to answer th e following research
70 question: Which identity motiv es beyond self-esteem and self -consistency guide identity centrality and self-b rand connections? 4.2. Theoretical Framework Self definition has been viewed as the driver of consumption across decades of consumer behavior and psychology resear ch (Levy 1959; Grub and Grathwohl 1967; McClelland 1951; Solomon 1983). Kleine, Kleine, and Kernan (1993) later confirmed this relationship by showing an individual s consumer behavior is stronger among specific roles than the global self. This findi ng is largely based on the symbolic meanings of products often linked to the personal imag es of the product-user (Sirgy et. al.1997). In this regard, consumption stimulates self -reflexive evaluation leading to identity maintenance and definition. But which identity is being maintained and defined? Multiple selves incorporate the self view as a dynamic structure a nd provides individuals with motives, self-relevant information, and goals n eeded to guide their behavior (Markus and Nurius 1986). Kleine et al (1993, p. 210) argues for multidim ensional self by stating: The significance of a product to consumers de pends on which of their ideas it enables and the importance of that identity what it contributes to their overall sense of self. This is further supported by Rook (1987), who suggests that individuals consume products that allow them to support a given identity. 4.2.1. Identity Centrality In studying the outcomes of consumers rela tionships with brands, salience of a particular identity has been considered as a driver of consumer brand outcomes. This implies a shifting of self-categorization (T urner, Hogg, Oakes, Reicher and Wetherell 1987) and reflects a momentary change in the self-concept that guide s social perception
71 and behavior. Individuals re peatedly consume products, in part, to enact identities consistent with their ideal se lf (Huffman, Ratehswar, and Mi ck 2000), therefore it reasons that a single identity should become cen tral in consumption activities. Consumers oftentimes encounter crosssituational scenarios that require an enduring cognitive prominence within their self-concept. This mean s individuals are inclin ed to perceive and act in accordance with a partic ular identity compared to others (Oakes 1987), thereby establishing centrality. Identity centrality is positively related to identity commitment, psychological well-being, and identity relate d performance (Settles 2004). This is supported by Martire, Stephens, and Town send (2000) who found identity centrality buffers some identities and exacerbates othe rs, due to the expectations surrounding each identity. Thus, the following hypothesis is offered: H1: Identity centrality will occur among an individuals multiple identities in a gi ven product setting. 4.2.2. Identity Motives Apart from self-esteem and self-consistenc y, little effort has been dedicated to exploring the range of motives influenc ing consumption of products and brands consistent with a consumers image. In revi ewing theories of the individual self-concept, social identity, identity threat, and general consumer behavior, nine key motives were identified and supported by a qualitative study (see Essay 1) sugges ting individuals are motivated by: self-esteem distinctiveness, c ontinuity self-efficacy, belonging meaning, recognition, consistency, and security. They are inherently a func tion of an individuals overall self-concept and serve as drivers of identity construction. Each motive pulls together motivational assumptions and predic tions regarding identity construction and
72 presentation from several th eoretical perspectives. These motives are summarized in Table 4.1. Table 4.1 List of Identity Motives Identity Motive Definition Application of Motive to Consumer Behavior Research Self-Esteem The desire to maintain and enhance a positive conception of oneself (Gecas, 1982, p. 20). Smeesters and Mandel (2006); Bearden, Hardesty, and Rose (2001); Sirgy (1982) Distinctiveness The desire to establish and maintain a sense of differentiation from others.(Vignoles, Chryssochoou, and Breakwell 2000). Forehand, Deshpande, and Reed (2002); Simonson and Nowlis (2000); Deshpande and Donthu (1986); Continuity The desire to maintain a sense of connection across time and situation within identity (Breakwell, 1983, p. 24). Spangenberg and Sprott (2006); Tian and Belk (2005); Hamilton and Biehal (2005); Chaplin and Roedder John (2005); Agrawal and Maheswaran (2005) Efficacy The desire to maintain and enhance feelings of competence and control (Breakwell 1993, p. 205). Erdem and Swait (2004); Keller 2006 Belonging The desire to maintain or enhance feelings of closeness to, or acceptance by, other people, whether in dyadic relationships or within a group (Baumeister and Leary 1995). Escalas and Bettman (2003; 2005) Meaning The need to find significance or purpose in ones own existence (Baumeister 1991). Ahuvia (2005); Baumgartner (2002); Wooten and Reed (2004); Reed (2004); Krishnamurthy and Sujan (1999); McCracken (1986) Recognition The need to be acknowledged or rewarded by others (Breakwell 1988). Argo, Dahl, Manchanda (2005); Muniz and Schau (2004); Schau and Gilly (2003); Consistency The desire to establish congruency with the self-concept. (Sirgy 1982). Sirgy (1982) Security The desire to protect the overall selfconcept from internal and external threats (Branscombe et al. 1999) Ashmore et al. (2004); Marques and Yzerbyt (1998)
73 4.2.3. Self-Esteem Motive The self-esteem motive is defined as the motivation to maintain and enhance a positive conception of oneself (Gecas 1982, p. 20) This definition is similar to Sirgys (1982) definition, he defined self-esteem as the tendency to seek experiences that enhance an individuals self-c oncept. Self-esteem is an element of Maslows (1954) motivational hierarchy and is often used as a superordinate goal in studying consumer motivation. For this reason it is sometimes view ed as a second-order construct, however, it is treated as a first-order c onstruct is this study. The pursuit of self-esteem is recognized by researchers as one of the most important motivational drivers of consumer behavior and decision-making. This is because a consum ers decisions are regularly made within the context of enhancing or protecting th eir self-esteem (Grub and Grathwohl 1967). Brendl, Chattopadhyay, Pelham, and Carvallo (2 005) find self-esteem threat (i.e. negative feedback) increases the liking for ones name letters, compared to the self-affirmation condition. Micks (1996) study on materialism and social desira bility revealed a negative relationship between material values and self-esteem. The self-esteem motive is associated with a large number of identity pr ocess theories (for a review, cf. Hoyle 1999) and is implicated in intergroup relations. 4.2.4. Distinctiveness Motive The distinctiveness motive refers to the desire to esta blish and maintain a sense of differentiation from others (Brewer 1991; Vignoles, Chryssochoou, and Breakwell 2000). Viewed as a core value of Western cultures, di stinctiveness is also accepted as a universal human need necessary for a meaningful se nse of identity (Vignoles, Chryssochoou, and Breakwell 2002). Marketers understand this need and oftentimes target consumers with a
74 host of stimuli activating thei r need for distinction. Resear ch in consumer behavior clearly demonstrates individuals are motivated to achieve identity distinctiveness in a host of consumption settings. For example Deshpande and Donthu (1989) reveal high Hispanic identifiers are more likely to be br and loyal due to the di stinctiveness of their ethnic identity in comparison to Caucasian consumers. Forehand, Deshpande, and Reed (2002) found that distinctiveness influenced participant judgme nts on a series of advertisements among Asian a nd Caucasian respondents. The distinctiveness motive is featured in the Uniqueness theory (Deci and Ryan 2000); Brewers (1991) Optimal Distinctiveness theory; and the Identity Process Theory (Breakwell 1993). 4.2.5. Continuity Motive The continuity motive refers to an individuals motivation to maintain a sense of connection across time and situation within identity (Breakwell 1988, p. 24). Continuity is not necessarily the absence of change as there is an identifiable conceptual thread uniting the past, present, and future within a persons identity (Breakwell 1988). James (1892) was one of the first to pr omote continuity as a chief feat ure of the identity. It is a strong motive that encourages permanence with in an individuals past, present, and future. The continuity motive has implications for marketing across a number of domains, including gift-giving, fashion, and overall self-definition. Lowrey, Otnes, Ruth (2004) found that trad itional gifts can be filled with specific meaning allowing relational connections over ti me between the gifter and the receiver. Individuals seek value expressi ve fashions which provide a se nse of continuity to various memories, activities, and significant relations hips (Murray 2002). Fu rther, individuals seek or create social contexts which provide self-confirming feedback (Swann 1987,
75 Spangeberg and Sprott 2006) establishing s ubjective continuity. In contrast, when subjective continuity is allayed, negative a ffect ensues (Keller, Lipkus, and Rimer 2002) and a threat to identity arises (Wooten and Reed 2004; Dahl, Manchanda, and Argo 2001). The continuity motive is highlighted in the self-ver ification theory (Swann 1983); self-concept enhancing tactic ian theory (Sedikides and Strube 1997), as well as the identity process theory (Breakwell 1993). 4.2.6. Self-Efficacy Motive Self-efficacy is defined as the tendency for an individual to maintain and enhance his/her feelings of compet ence and control (Bandura 1977; Breakwell 1993). As with previous identity motives, self-efficacy is theorized as a defining feature of identity (Codol 1981), and has been advanced as a primary human motivation (Deci and Ryan 2000). In the consumer behavior domain, the self-efficacy motive has been widely studied. Keller (2006) investigated self-effi cacy on an individuals willingness to perform a new health behavior depending on the role of regulatory focus. Whereas, Chandran and Morowitz (2005) examined the role of efficacy on participative pricing (i.e. auctions). Lastly, Duhachek (2005) found that a lack of self-efficacy is associated with consumers inability to evaluate information when de pressed (Duhacheck 2005). This motive is found in theories such as the self-monitoring th eory (Gangestad and Snyder 2000; Snyder 1974) and the Optimal Distinctiveness theory (1991). 4.2.7. Belonging Motive The belonging motive is defined as the desire to ma intain or enhance feelings of closeness to, or acceptance by other people, wh ether in dyadic relationships or within a group and is identified as a fundamental human motivation (Baumeister and Leary
76 (1995, p. 497). The belonging motive is oftent imes employed in studies of social influences during consumption. In particular, this motive explains how reference groups contribute to the formation of consumer va lues, attitudes, and marketplace behavior (Bearden and Etzel 1982). Es calas and Bettman (2003, 2005) examine reference groups as a source of brand associations reveali ng stronger self-b rand connections for brands used by member and aspiration groups. When a consumers need to belong is threatened, it has been shown that th ey respond with various c oping strategies including identification with more in clusive in-groups (Wooten a nd Reed 2004), self-stereotype (Brendl et al. 2005), and overestimating cons ensus for their own beliefs (Rose and Wood 2005). The belonging motive is featured in the Sociometer theory (Leary and Baumeister 2000), optimal distinctiveness theory (Brewe r 1991), and uniqueness theory (Snyder and Fromkin 1980). 4.2.8. Meaning Motive The meaning motive refers to the desire to find significance or purpose in ones own existence (Baumeister 1991) and within their possess ions. Marketers have long considered an individuals search for mean ing as critical in un derstanding consumer behavior. As the consumer becomes an active partner with the marketer in brand-meaning formation, brand meaning evolves over time and is assigned in ways that make the brand more meaningful to the customer. A consum ers search for meaning has been found to play an essential role in brand choice (Ng and Houston 2006), brand relationships (Fournier 1998), and their participation in brand communities (Muniz and OGuinn 2001). Thus its applicability to consumer be havior is well-supported. The depiction of ones existence as meaningful is a prin cipal attribute of ps ychological well-being
77 (McCracken 1986), self presentation (Schau and Gilly 2003), and narrative processing (Escalas 2003; 2004). These st udies suggest individuals i nvest meaning into not only possessions, but also their identities. The desire for a meaningful identity is linked to selfaffirmation theory (Steele 1988), Turners ( 1985) self-categorizati on theory, and Hoggs (2000) uncertainty reduction principle. 4.2.9. Recognition Motive The recognition motive is defined as the desire to be acknowledged or rewarded (Breakwell 1993). This acknowledgement can co me in the form of benefits directly related to the individuals actua l or aspired self concept a nd can increase the individuals attractiveness, power, and commitment to th e brand. Recognition allows individuals to non-verbally communicate in the marketpla ce, while satisfying the need to be acknowledged. Belk, Bahn, and Mayer (1982) argue recognitions (subjective inferences) based on choice of consumption objects, and are one of the s trongest and most culturally universal phenomena inspired by consumer behavior (p.4). Another way to view recognition is to consider it as a type of esteem that is deri ved from three activities: the mastery of ones environment, realization of ones abilities, and recognition from others with regard to those achievements. This is often associated with self-confidence, pride, creativity, and a strong sense of identity (Breakwell 1993). On the surface, the recognition motive may appear to be synonymous with the self-esteem motive, but these motives inspire different self-definitional goals. For instance, recognition implies a desire to be acknowledge by others; whereas, self-esteem is driven by an evalua tion of ones self-worth. Consumer research employing this motive has focused on brand legitimacy in gay communities (Kates 2004), uniqueness (Tian,
78 Bearden, and Hunter 2001), childrens sym bolic meaning (Belk, Mayer, and Driscoll 1984) and consumption symbolism (Belk, Bahn, and Mayer 1982). The recognition motive is featured in the optimal distinct iveness theory (Brewer 1991) and the uniqueness theory (Snyder and Fromkin 1980). 4.2.10. Consistency Motive The consistency motive is defined as an individuals motivation to maintain uniformity of identity across situation and time (Sirgy 1982). Grub and Grathwohl (1967) were among the first to offe r the self image/product-image congruity theory as a process explanation of self-referent consumption. Th ese authors provided the foundation for a host of research in consumer behavior on the consistency motive. It has also been shown that consumers have more extreme attitudes toward brands that help to express their identities compared to brands that do not aid in identity presentation (Aaker 1999). Meanwhile, other resear chers have shown that product knowledge consistent with an individual s existing self-conc ept received more consideration, is better recalle d, and is perceived as more reliable compared to when product knowledge is inconsistent with an i ndividuals self-concept (Coulter, Price, and Feick 2003). Individuals also tend to commit to issues that help them express and achieve goals consistent with their se lf-view (Shavitt 1990; Wicklund and Gollwitzer 1982), form stronger self-brand connections (Escalas a nd Bettman 2003; 2005) and results in more favorable product evaluation (Sirgy 1985). Lastly, Moorman, Diehl, Brinberg, and Kidwell (2004) find subjective knowledge (i.e perceived knowledge) affects the quality of consumers' choices by altering consumers search location due to self-consistency.
79 This motive is found in the self-discrepan cy theory (Higgins 1987) and the product image-congruity theory (Grub and Grathwohl 1967). 4.2.11. Security Motive The final motive of consideration is the security motive defined as the motivation of an individual to protect their overall self -concept from internal and external threats (Ashmore, Deaux, and McLaughlin -Volpe 2004). Identity thr eat is relevant to how consumers perceive themselves and others, by eliciting motivational states that lead to both individual and group leve l protection. Identity threat among poor migrant Turkish women has been studied by stner and Holt (2007). They conceptu alized a model of dominated acculturation to explain Turkis h consumer culture. Argo, White, and Dahl (2006) examine factors of se lf-threat to demonstrate so cial comparison motivates individuals to lie. In additi on, Tian and Belk (2005) exam ine workplace possessions and find symbolic possessions of future aspirations mitigate corporate identity threat in the event of a buyout. Lastly, Adkins and Ozanne (2005) investigated the low literate consumers self-threat when interacting within the marketplace and their coping strategies when threatened. This motive is f ound in theories such as the self-completion theory (Wicklund and Gollwitzer 1982), Tesse rs (1988) self-evaluation maintenance model, and Steeles (1988) self-affirmation theory. It is recognized that other motives be yond those presented above can drive selfreferent consumption. However, exploration of additional motives is left for future research. Based on the arguments presented for each motive and support from the literature, these nine identity motives are hypothesized to influence identity centrality. Simply
80 stated, as identity motives are satisfied a gi ven identity will be b ecome more prominent leading to identity centrality. Thus the following hypotheses are offered: H2: Multiple identity motives will positively and significantly influence identity centrality. H2A: The self-esteem motive will positively and significantly influence identity centrality. H2B: The self-consistency motive will positively and significantly influence identity centrality. H2C: The distinctiveness motive will positively and significantly influence identity centrality. H2D: The continuity motive will positively and significantly influence identity centrality. H2E: The efficacy motive will positively and significantly influence identity centrality. H2F: The belonging motive will positively and significantly influence identity centrality. H2G: The meaning motive will positively and significantly influence identity centrality. H2H: The recognition motive will positively and significantly influence identity centrality. H2I: The security motive will positively and significantly influence identity centrality. 4.2.12. Self-Brand Connections When brand associations are used to cons truct ones identity or to communicate ones self to others, a self-brand connection is formed (Escalas and Bettman 2003; 2005). Self brand connections measure the degree to which individuals have incorporated brands into their self-concept (Escalas and Bettman 2003; 2005). Presumably, self-brand connections will be enhanced as brands are c hosen to achieve specific identity goals. This
81 is due to identity commitment, which is associ ated with expectations and motivations of behavior relevant to identity goals (Foot e 1951). Identity commitment serves as a boundary criterion in determining how motivatio ns from social influence will be handled. To illustrate, a person committed to the identity of "soccer mom" will interpret marketing stimuli differently than one committed to a "corporate executive" identity. Similarly to the influence of multiple iden tity motives on identity centrality; the influence of multiple identity motives on self-b rand connections is also considered. It is assumed that as identity motives are satis fied, self-brand connections should emerge stronger. Given the exploratory nature of th is relationship and limited research on selfbrand connections, it is unreasonable to di ctate which motive will influence self-brand connections the greatest. Thus all nine identity motives are hypothesized to influence self-brand connections: H3: Multiple identity motives will positively and significantly influence self-brand connections. H3A: The self-esteem motive will positively and significantly influence self-brand connections. H3B: The self-consistency motive will positively and significantly influence self-brand connections. H3C: The distinctiveness motive will positively and significantly influence self-brand connections. H3D: The continuity motive will positively and significantly influence self-brand connections. H3E: The efficacy motive will positively and significantly influence self-brand connections. H3F: The belonging motive will positively and significantly influence self-brand connections.
82 H3G: The meaning motive will positively and significantly influence self-brand connections. H3H: The recognition motive will positively and significantly influence self-brand connections. H3I: The security motive will positively and significantly influence self-brand connections. An individual is likely to support an d be more committed to preserving a particular identity when it is more central (Settles 2004). As st ated earlier, centrality is influenced by multiple identity motives, which are related to a specifi c identity within the individual. Therefore a highl y central identity should l ead to stronger self-brand connections as various motives are satisfi ed. The key determinant of this proposed relationship is brand associat ions congruent with the self -image. Therefore the following hypothesis is offered: H4: Identity centrality will mediate the relationship between identity motives and self-brand connections. 4.2.13. Brand Symbolism Levy (1959) asserts that individuals do not buy products simply for their functional value, but also for their symbo lic meaning. Brands can be symbols whose meaning is used to create and define a consumers self-concept. Brand symbolism enables consumers to form a long-lasting relationship with a particular brand through its emotional and functional utility. The emotional benefits of symbolic brands have been recognized by researchers as a prerequisite for sustaining brand success (Keller 1993). This is because consumers are able to communicate their identities through brand consumption. It follows that some brands ar e better able to communicate an identity better than others. For example, prior consumer research suggests that publicly consumed
83 (vs. privately consumed) and luxury (vs. n ecessity) products are be tter able to convey symbolic meaning about an indivi dual (Bearden and Etzel 1982). A brand that is very popular and used by many different types of people (e.g., a BMW) may have different meanings to consum ers based on the different identities that are presented through brand usage. It is e xpected brand symbolism will moderate the formation of self-brand connections due to the brands ability to communicate something about the individual. It is hypothesized that in cases when cent rality is low, brand symbolism will be the primary source of self -brand connections. Thus it is expected that brand symbolism will moderate the effects of identity centrality on self-brand connections. H5: Brand symbolism will positively moderate the relationship between identity centrality and self-brand connections. The theoretical model guiding this res earch is presented below in Figure 4.1
84 Figure 4.1 Conceptual Model of Identity Motives In fluencing Identity Centrality and SelfBrand Connections 4.3 Methodology The study was designed to test the influe nce of identity motivation, using the methodology developed by Vignoles et al. (2002). Participants freely generated a list of identities then rated each identity for its centrality (dependent variable) and for its association with motivations of self-esteem, distinctiveness, continuity, self-efficacy, belonging, meaning, recognition, consistency, a nd security (independ ent variables). The main analysis was designed to evaluate th e unique contributions of each motive rating to Self-esteem Distinctiveness Continuity Efficacy Belonging Meaning Recognition Consistency Security Identity Centrality Self-brand Connection Brand Symbolism H2 H3 H5 H4
85 predictions to identity cent rality, as shown in Figure 4.1. A secondary goal was to assess the unique contributions to pred ictions of perceived centralit y and self-brand connections. 4.3.1 Determining the Study Context Twenty eight student volunteers participated in a task to identify the appropriate context for this study. Participants were aske d to list four produc t categories consumers use to reveal who they are to others. Each participant prov ided four product categories, for a total of 112 twelve independent re sponses. Among the 112 responses, a total of seven product categories were provided. Each category was rated for its frequency. If every participant (n = 28) indicated automob iles as the product ca tegory individuals use most to show who they are to others, then the automobile category was given a frequency count of 28. The same process was repeated for the other categories. Based on the frequency count of the automobile categor y, this category was chosen for the study context. Examples of other product categorie s include fashion, homes, electronics, jewelry, shoes, and handbags. Table 4.2 shows all the categories and the frequencies. Table 4.2 Product Categories Frequency Count Distribution Product Category Actual Count Automobiles 27 Clothes 20 Homes 18 Electronics 14 Jewelry 12 Shoes 11 Handbags 10
86 4.3.2 Participants Participants were drawn from a c onvenience sample of MBA graduates. Approximately 512 invitations to participate in the online survey were disseminated. Two weeks after the initial inv itation, an email reminder to non-respondents followed. The overall response rate was 17.38% (i.e., 89 ques tionnaires were return ed). Approximately 14 participants attempted the survey but failed to complete it in its entirety; others opted not to participate in the study, leav ing a total of 75 usable surveys. The majority of participants were highl y educated, 74 of them had an advanced degree, while 1 participant had some college, but had yet to receive a degree. Participants were 35% females and 65% males, 19% were single, 65% were married, and 16% were divorced. The sample was comprised of 73% Ca ucasians, 17% Asian or Pacific Islanders, 4% African Americans, 2% Hispanics, 1% Uk rainian, 1% Multiracial, and 2 participants left this question blank. 4.3.3. Non-Response Bias Several methods have been proposed to account for non-response bias data in survey data collection includi ng subjective estimates and extr apolation. The extrapolation method is used in this study to addres s bias between the respondents and nonrespondents. A common method of extrapolati on is the comparison of characteristics for respondents who answer successive waves of a survey (Pace 1939). A wave refers to the response following an invitation to participate in the survey and coul d refer to subsequent invitations to participate in the survey. Ther e were two waves of data collected for this study. Subjects who responded in the later wave were assumed to have responded because of the additional stimulus and were expected to be similar to non-respondents.
87 This method is useful when a survey of non-respondents cannot be conducted, and a test for non-response bias assumes subjects who respond less readily are more like nonrespondents. Less readily is defined as answ ering later or as requiring more prodding to answer (Armstrong and Overton 1977). The extrapolation method tested for significant differences between early respondents and late respondents, with late re spondents being considered a surrogate for non-respondents (Armstrong and Overton 1977). Using this method, responses of the first wave of received surveys were compared to the responses from the second wave of received surveys. All respondent characteristi cs were cross-tabulated between levels of each variable and mailing wave. A Chi-square test of independence was then applied. Results indicated there was not a significant difference between the responses collected during the first mailing and those collected during the second mailing. This result indicated that nonresponse bias is a minor concern. The di stribution of the respondent characteristics for both the first and second wave of mailing is shown below in Table 4.3.
88 Table 4.3 Summary and Comparison of Characteri stics of First Wave Respondents to Second Wave Respondents Variable First Mailing n (%) Second Mailing n (%) Chi-square p-value Employment Status 0.328 Employed full-time 50 (88) 30 (94) Employed part-time 4 (7) 1 (3) Currently Not Employed 3 (5) 1 (3) Gender 0.205 Male 37 (65) 20 (63) Female 20 (35) 12 (37) Age Category 0.254 18 to 25 0 (0) 1 (2) 26 to 35 19 (33) 7 (22) 36 to 45 24 (42) 17 (30) 46 to 55 11 (19) 4 (13) 56 to 69 2 (4) 2 (6) Over 70 1(2) 0 (0) Highest Degree Earned 0.148 Some College 0 (0) 1 (3) Graduate Degree 38 (67) 18 (56) Post-graduate Degree 19 (33) 13 (41) In addition to respondent characteristics, the key measures in the study were also tested for non-response bias. This comp arison was completed by using a one-way ANOVA to compare the means of the key constr uct across the earlier and latter wave of respondents. The key dependent variable fo r this study was self-brand connection, and there was no significant difference between those who responded and earlier and those who were less readily to respond. The same re sults hold true for the identity centrality measure and the brand symbolism measure. Differences were tested on two identity motives (chosen at random), and once again th e results reflect that there is no difference
89 in responses among the sample. The results of the comparisons are shown below in Table 4.4. Table 4.4 Summary and Comparison of Key Study Measures of First Wave Respondents to Second Wave Respondents First Mailing Value (n) Second Mailing Value (n) ANOVA Results Study Measures Self-brand Connections 63.10 (57) 63.25 (32) F (1, 88) = .001, p< 0.974 Identity Centrality 4.94 (57) 4.62 (32) F (1, 88) = 1.132, p< 0.291 Brand Symbolism 4.07 (57) 4.05 (32) F (1, 88) = .835, p< 0.364 Self-esteem Motive 2.99 (57) 2.79 (32) F (1, 88) = 2.412, p< .121 Meaning Motive 2.15 (57) 2.15 (32) F (1, 88) = .001, p< 0.981 4.3.4. Procedure Using a methodology developed by Vignoles et al. (2002) a que stionnaire was design to capture an individuals multiple identities and related motivations. Questionnaires were made available through an online survey website, Vovici, Inc. (http://www.vovici.com). The questionnaire be gan with a brief study introduction and an explanation of what is meant by term identity. Following the study introduction participants were asked to specify freely 6 identities that influenced their consumption decision in the marketplace using a shortened adapted version of the Twenty Statements Test (Kuhn and McPartland 1954). Now thinking about yourself, please list six term descriptors that you feel are relevant and accurately repres ent how you identify yourself when you are out shopping for differ ent types of products and services . The participants were asked to provide only six identities as it was expected that participants would find
90 the subsequent motive ratings too demanding with six identities to rate on each identity motive. Of the 75 participants, 74 pa rticipants provided 6 identities. Next participants explained how each identity influences their purchases, Now for each of your identities that you desc ribed above, please brie fly explain how that specific identity might influence your marketplace shopping behavior. Each of their identities was then rated for its association with motivations of self-esteem distinctiveness continuity efficacy belonging meaning recognition consistency and security. The scale items were generated from the qualitative responses in essay 1, and began with the stem: How much does your identity ? Motives were measured on a scale from 1-6, with 1Not at all and 6Completely as the scale anchors. The scale items were averaged and divided the number of items to yield a standardized mean. All nine motives had acceptable reliabilities a bove 0.70, and showed unidimensionality The scale items for each motive along with coeffici ent of reliability are presented in Table 4.4. Each motive was presented as a question at the top of each page with each identity positioned underneath. Now thinking about your identities you just entered, please respond to each of the following st atements, using the re sponse scale provided. How much does your identity ? (i.e., gives meaning to your life, influences your brand behavior? This was repeated for all six identit ies recorded and each of the nine motives. The scale descriptor s for the nine identity mo tives were anchored by 1Not at all and 7Completely Three questions measured the perceived centrality ( = 0.889) for each identity, anchored by (1)Strongly Disagree and (7)Strongly Agree I often think
91 about being a(n) _______. Being a(n) _______ is an important part of myself image. The fact I am a(n) _______ rarely enters my mind (reversed score) Following the rating of the nine identity mo tives, participants then completed the questions surrounding the auto mobile buying scenario: For the next several questions, we want you to place yourself in the following s ituation. You are given $40,000 to purchase an automobile of your choice, but there is one specific stipulation. You can only use the $40,000 towards the purchase of an automobile. No portion of these monies can be used for any other pur chases. In the space provided, please type in the Make and M odel of the automobile that you would purchase and best repr esents the type of person who you are. Once participants entered the make and mode l of their selected automobiles, they then rated the degree to which they have self-brand connections with the brand by answering seven questions ( = 0.87). The scale was anchored by (1)Strongly Disagree and (7) Strongly Agree : This brand reflects who I am. I can identify with this brand. I feel a personal connection to this brand. I use this brand to communicate who I am to other people. I think this brand help me become the type of person I want to be. I consider this brand to be me (it reflects who I consider myself to be or the way that I want to present myself to others). This brand suits me well. Participants then rated the degree of brand symbo lism for the brand anchored by (1) Strongly Disagree and (7) Strongly Agree This brand communicates something about my personality or identity. This brand symbolizes the ki nd of person, who drives it . This brand communicates who I am to other people. The three items were averaged to form one standardized score ( = 0.71).
92 Following this, participants indicated which of their six identities carried the greatest importance in selecting thei r particular brand of automobile. Now, thinking back to your role-identities (the identities they entered earlier appeared on the screen as the data was piped throughout the survey), please indicate the role-identity that was most influential in your selection of the Make, Mo del. If you feel there was some other roleidentity that influenced your selection of the Make, Model please enter that roleidentity descriptor in the space provided. Following that task, participants res ponded to three questions measuring the perceived centrality of identity3 on brand choice anchored by 1 Not at all and 7Completely: How central is ea ch of your identities to your choice of the Make, Model How important is each of your identities in influencing how you feel about the Make, Model? How much do you see each of your iden tities as being an im portant reflection of who you are when you are driving a (n) Make, Model The three items were averaged to form one standardized score ( = 0.88). This was followed by the collection of demographic information. A debriefing statement was issued to all participants at the end of the survey. The entire procedure took approximately one half-hour. 4.4. Data Analysis and Results 4.4.1. Dependent Variables Self-brand connections was measured us ing seven items (Escalas and Bettman 2003); averaged to form one standardized score ( = 0.87). The mean was 62.79, with a standard deviation of 16.77. 3 In the computer program, part icipants answers were customized, so th at their identity res ponses and brand of utomobile was carried throughout the question stem of the measures.
93 4.4.2. Independent Variables Exploratory factor analysis was (EFA) was used to check the dimensionality of the multi-item identity motive scales. Dimensionality is defined as the number of common factors or latent c onstructs needed to account for the correlation among the variables. The implicit assumption of EFA is that the researcher has a limited idea with respect to the dimensionality of the construct (Netemey er, Bearden, Sharma 2003). The EFA factor loadings along with the descrip tive statistics of the identity motives are presented below in Table 4.5. Table 4.5 Identity Motives Descriptive Statisti cs and Coefficients of Reliability Motive Factor Loadings Standard Mean Standard Deviation Self-Esteem ( = 0.87) 4.44 1.112 Allows you to maintain a positive attitude towards yourself 0.81 Contributes to your satisfaction with yourself 0.84 Makes you feel like a person of worth 0.87 Contributes to your overall self-esteem 0.89 Distinctiveness ( = 0.89) 3.83 1.39 Contributes to your uniqueness 0.87 Allows you to stand out among other people 0.93 Distinguishes you from other people 0.92 Continuity ( = 0.79) 4.10 1.47 Relates to other aspects of your self 0.79 Gives you a sense of continuity in your life between your past, present, and future 0.84 Is associated with your self-concept in the future 0.76 Provides a sense of uniformity in your life 0.75 Self-Efficacy ( = 0.79) 4.02 1.47 Contributes to your confidence in assessing the worth of an outcome or event 0.71 Allows you to exercise control of events that affect your life 0.81 Makes you feel capable of being successful 0.84 Makes you fell effective doing the things you do 0.78
94 Table 4.5 (Continued) Motive Factor Loadings Standard Mean Standard Deviation Belonging ( = 0.79) 3.73 1.46 Gives you a sense of similarity to other people 0.73 Makes you feel as if you belong to a group of people 0.80 Makes you feel close to other people who are similar to you 0.67 Makes you feel accepted by others similar to you 0.72 Determines how much you conform to other people similar to you 0.55 Meaning ( = 0.80) 4.48 1.23 Adds a sense of importance to your life 0.82 Gives meaning to your life 0.87 Give significance to your life 0.75 Gives you a sense of fulfillment/satisfaction with life 0.76 Recognition ( = 0.91) 3.61 1.46 Contributes to your desired level of respect 0.90 Contributes to you need for acknowledgement 0.90 Gives you a sense of appreciation 0.72 Makes you feel admired by other people 0.87 Makes you feel recognized by other people 0.92 Consistency ( = 0.78) 4.61 1.15 Provides stability to who you are 0.76 Gives consistency to who you are 0.82 Brings together who you are as a person 0.70 Aligns with other aspects of overall self-concept 0.82 Security ( = 0.86) 4.31 1.41 Makes you feel secure with your self-concept 0.87 Makes your feel confident with your self-concept 0.91 Makes you feel self-assured 0.88 Identity Centrality was measured using four items; averaged to form one standardized score ( = 0.88). The mean was 3.43, with a standard deviation of 1.07. Brand Symbolism was measured using two items; averaged to form one standardized
95 score ( = 0.71). The mean was 2.40, with a stan dard deviation of 1.03. A correlation of the nine identity motives along with identity centrality is presented below in Table 4.6 Table 4.6 Correlations Between Ratings of Consumer Identities for Identity Centrality and Each Hypothesized Identity Motive Variable 1 2 3 4 5 6 7 8 9 10 11 1. Distinctiveness 1.00 2. Consistency 0.43 1.00 3. Continuity 0.46 0.77 1.00 4. Belonging 0.43 0.57 0.55 1.00 5. Security 0.62 0.81 0.79 0.67 1.00 6. Recognition 0.69 0.63 0.61 0.56 0.84 1.00 7. Meaning 0.52 0.59 0.57 0.30 0.62 0.67 1.00 8. Self-Esteem 0.58 0.73 0.82 0.54 0.88 0.81 0.75 1.00 9. Self-Efficacy 0.49 0.68 0.62 0.38 0. 70 0.62 0.71 0.74 1.00 10. Identity Centrality 0.22 0.33 0.33 0.19 0.30 0.24 0.28 0.31 0.29 1.00 11. Self-brand Connections 0.53 0.47 0.47 0.31 0.55 0.50 0.44 0.52 0.30 0.22 1.00 All correlations are significant at the 0.01 level (2-tailed). 4.4.3. Satisfying the Assumption s of Multiple Regression There are several assumptions the researcher must consider in using multiple regression, namely: linearity, multicollin earity, normality, and homosecdasticity. Linearity concerns the assumed relationship between the X and Y variables are linearly related. A bivariate scatterplot of the variable s of interest (e.g. multiple identity motives and identity centrality) will reveal if the linea rity assumption is violated. If curvature in the relationships is evident, one may cons ider either transforming the variables or allowing for nonlinear components. None of the scatterplots re vealed a non-linear relationship, so this assumption was met. Multicollinearity occurs when two vari ables convey rough the same information. In this case, neither may c ontribute significantly to the model after the other one is
96 included. But together they c ontribute a lot. If both variab les were removed from the model, the fit would be much worse. The correlations between identity motives are presented above in Table 4.6. Values of 0.8 and above indicate that here may an issue of multicollinearity among certain pairs of motives (Cohen and Cohen 1983). As a back-up, the variance inflation factor is calculate d in the regression models to ensure multicollinearity is not an issue in the regr ession models. The Variance Inflation Factor (VIF) is the number of times the variance of the corres ponding parameter estimate is increased due to multicollinearity as compar ed to as it would be in the absence of multicollinearity. Values of VIF exceeding 10 are often regarded as indicating multicollinearity Regression assumes that variables have normal distribution. When variables are skewed, or have substantial outliers, this can distort the relationships among the variables and significance testing. A violation of normality can compromise the estimation of coefficients and the calculati on of confidence intervals. Because parameter estimation is based on the minimization of squared error, a few extreme observations can have a disproportionate influence on parameter es timates. Additionally, calculation of confidence intervals and various significances tests for coefficients are all based on the assumptions of normally distributed errors. If the error distribution is significantly nonnormal, confidence intervals may be too wide or too narrow. There are several ways to test this assu mption such as a visual inspection of the data, histograms, skew, kurtosis, and P-P plot s, while measures such as the KolmogrorovSmirnov tests gives an inferen tial statistic for normality (O sborne and Waters 2002). A way to test for normally distributed errors is a normal probability pl ot of the residuals.
97 This is a plot of the expected normal pr obabilities of occurren ce versus the observed cumulative probabilities of occurrence of the st andardized residuals. If the distribution is normal, the points on this plot should fall cl ose to the diagonal lin e. Shown in Figure 4.2 below is a graph of the normally distribut ed residuals for self-brand connections. Figure 4.2 Normal P-P Plot of Regression St andardized Residual for Self-Brand Connections Observed Cum Prob1.0 0.8 0.6 0.4 0.2 0.0 Expected Cum Prob1.0 0.8 0.6 0.4 0.2 0.0 Lastly, homeoscedasticity is addresses the assumption that the variance of errors is all the same across all levels of the IV. A variance in different leve ls of the independent variable will lead to heterosced asticity, however this condition is said to have litter effect on tests of significance (Berry and Feldman 1985). This assumption can be checked by visual examination of a plot of the standa rdized errors by the regression standardized predicted value. Plot shapes such as a bow tie or a fan is an indication of a violation of
98 this assumption. This assumption is met based on a plot of the errors as below in Figure 4.3. Figure 4.3 Plot of Residuals Showing Homoscedasticity Regression Standardized Residual3 2 1 0 -1 -2 Regression Standardized Predicted Value2 1 0 -1 -2 Based on the above argument, the assumptions for multiple regression have been met. This allows the researcher to help a void an increase in Type I and Type II errors, and increase effect sizes when violations of the assumptions are dealt with prior to analysis (Osborne and Waters (2002). 4.4.4. Hypothesis 1 The primary aim of this study was to establish identity centrality among individuals multiple identities. Among the participants, 67 of the 75 respondents reported a central identity, while the remaini ng eight did not report a central identity. In
99 order to test hypothesis 1, the centrality measure for the identity influencing the automobile brand consumption was compared to the centrality measure for the remaining identities. There were 342 non-central id entities reported and 69 central identities reported (two participants indicated two identitie s as central to their choice of automobile compared to the other participants who i ndicated one identity as being central). Individuals who declared a speci fic identity as central to th e automobile product category rated that identity as more central Mcentral = 4.89 in comparison to the other identities provided Mnot_central= 3.24. The two means were compared using an ANOVA for unequal sample sizes in SPSS. Comparison of the m ean ratings between the identities were significantly different, F (1, 411) = 89.369, p <.000. Thus H1 was supported. An example of the central identities provided by the pa rticipants is shown below in Table 4.7. 4.4.5. Hypothesis 2 This analysis focused on those identities reported as central to the automobile product category. Direct multiple regression was used to test this hypothesis. This method allows the variables to be entered into the model by the discretion of the researcher, providing assessment of the incremental predicti ve ability of any va riable of interest (McQuarrie 1988). Since prior research is limited on the role of identity motives in predicting identity centrality, all nine motives were entered into the model directly. The model was significant F (1, 68) = 3.479, p < .002 with 2 of the nine motives having a significant influence on identity centrality. Th e coefficient of determination, R2, was 0.347 and the adjusted R2 value was 0.247 indicating that 25 per cent of identity centrality is explained by the nine motives of which self-esteem ( = 0.625, t = 1.833, p <0.072) and recognition
100 motives ( = -.517, t = -2.042, p <0.046) were partially signi ficant and significant, respectively. Table 4.8 shows a summary of results for all nine motives. Table 4.7 Examples of Consumer Identities In fluencing Automobile Brand Choice Identities Influencing Consumption Identities Central to Automobile Brands Automobile Brand Participant 1 A 33-Year-Old Man A Cool Guy A Good Looking Guy A Father A Smart Man An American A cool guy Jeep Wrangler Participant 2 Father Practical Scientist Rational Thinker Soccer, Cricket And Tennis Fan Indian Classical Musician Father Mercedes 500SL Participant 3 Father 36 Year Old Automobile Enthusiast Upper Middle Class 35+ Age Group Asian American Automobile enthusiast BMW 335I Participant 4 Father Middle Age Single Business Executive Kids' Taxi Driver Music Listener Traveler Business executive BMW 528I Participant 5 Hr Executive Corporate Coach Weekend Athlete Cool Aunt To My Niece and Nephew Weekend athlete BMW X5 Participant 6 Controller Father Husband Stock Market Technician Financial Analyst Military Supporter Financial Analyst Infiniti M35 Participant 7 Environmentalist Value Seeker Minimalist Down To Earth Negotiator Conservative Environmentalist Lexus GS Hybrid
101 A separate model was run with only self-esteem and recognition motives as predictors of identity centrality. The coefficient of determination, R2, was 0.254, and the adjusted R2 value was 0.231 indicating that 23 pe r cent of identity centrality was explained by the self-esteem ( = 0.805, t = 4.223, p <0.001) and recognition motives ( = -0.440, t = -2.310, p <0.024). Thus H2 was not supported, while H2A was fully supported. Even though the standardized coefficient for the recognition motive was significant, H2H was not was not supported due its negative value. Table 4.8 Regression Results for Identity Moti ves Predicting Identity Centrality Parameter t Sig. Tolerance VIF (Constant) 3.55 Self-Esteem 0.62 1.83 0.07 0.10 10.46 Consistency 0.32 1.53 0.13 0.26 3.83 Meaning -0.10 -0.50 0.62 0.29 3.44 Distinctiveness 0.02 0.15 0.88 0.48 2.10 Continuity -0.16 -0.76 0.45 0.24 4.11 Security -0.03 -0.09 0.93 0.08 12.29 Recognition -0.52 -2.04 0.05 0.17 5.80 Belonging 0.11 0.73 0.47 0.52 1.92 Efficacy 0.25 1. 42 0.16 0.35 2.86 R2 = 0.347 Adjusted R2 = 0.247 Model Fit F(1, 68) = 3.479, p<.002 4.4.6 Hypothesis 3 Hypothesis 3 predicted multiple identity motives would influence self-brand connections. Similar to hypothesis 2, prior re search is limited on the role of identity motives in the prediction of self-brand conn ections. Therefore all nine motives were entered into the model direct ly. The model was significant F (1, 68) = 7.288, p < .000 with 2 of the nine motives having a signifi cant influence on self-brand connections. The
102 coefficient of determination, R2, was 0.526, and the adjusted R2 value was 0.454 indicating that 45 per cent of the variance in self-brand c onnections is explained by the nine motives in which the distinctiveness ( = 0.409, t = 3.147, p <0.003) and efficacy motives ( = -.483, t = -3.193, p <0.02) were both significant. Table 4.9 shows a summary of results for all nine motives. A separate model was run with only the distinctiveness and efficacy motives as predictors of self-brand connections. The coefficient of determination, R2, was 0.337, and the adjusted R2 value was 0.317 indicating that 32 percen t of the variation in self-brand connections was explained by the distinctiveness motive ( = 0.565, t = 4.838, p <0.000). The efficacy motive was insignificant ( = -.028, t = .243, p <0.809). Thus H3 was not supported, while H3C was fully supported. Table 4.9 Regression Results for Identity Moti ves Predicting Self-Brand Connections Parameter t Sig. Tolerance VIF (Constant) 35.47 Self-Esteem 0.38 1.31 0.20 0.10 10.46 Consistency 0.25 1.40 0.17 0.26 3.83 Meaning 0.13 0.78 0.44 0.29 3.44 Distinctiveness 0.41 3.15 0.00 0.48 2.10 Continuity 0.00 -0.02 0.99 0.24 4.11 Security 0.10 0.32 0.75 0.08 12.29 Recognition 0.03 0.14 0.89 0.17 5.80 Belonging -0.11 -0.84 0.40 0.52 1.92 Efficacy -0.48 -3 .19 0.00 0.35 2.86 R2 = 0.526 Adjusted R2 = 0.454 Model Fit F(1, 68) = 7.288, p<.000
103 4.4.7. Hypothesis 4 Hypothesis 4 predicted that identity centrality would mediate the influence of multiple identity motives on self-brand conn ections. This hypothesis was tested by entering all nine motives along with identity centrality into the regression model. The results showed that the effects of identity centrality went away when all nine motives were entered into the model. Similar to hypothesis 3, among the nine motives, the distinctiveness ( = .408, t = 3.116, p<0.003) and efficacy ( = -.499, t = -3.223, p <0.002) motives were the only significant predicto rs of self-brand connections; suppressing influence of identity centrality was not suppor ted. Therefore identity centrality does not mediate multiple identity motive influence on self-brand connections. Thus H4 is not supported. The results of the mediation analysis are shown below in Table 4.10. Table 4.10 Regression Results for Iden tity Centrality Mediation Parameter t Sig. Tolerance VIF (Constant) 32.06 Self-Esteem 0.34 1.14 0.26 0.09 11.06 Consistency 0.23 1.26 0.22 0.25 3.98 Meaning 0.14 0.81 0.42 0.29 3.46 Distinctiveness 0.41 3.12 0.00 0.48 2.11 Continuity 0.01 0.04 0.97 0.24 4.15 Security 0.10 0.33 0.75 0.08 12.29 Recognition 0.06 0.28 0.78 0.16 6.21 Belonging -0.11 -0.89 0.38 0.52 1.94 Efficacy -0.50 -3 .22 0.00 0.34 2.95 Identity Centrality 0.62 0.56 0.58 0.65 1.53 R2 = 0.529 Adjusted R2 = 0.484 Model Fit F(1, 68) = 6.513, p<.000
104 4.4.8. Hypothesis 5 Hypothesis 5 predicted brand symbolism will moderate the effect of identity centrality on self-brand connec tions. In order to analyze th is hypothesis, a moderated multiple regression was run on the data for all participants who specified a central identity. Moderated multiple regression involves hierarchical regression that first tests the relationship of the predictors of interest (e.g., identity central ity, brand symbolism) on the criterion variable (e.g. self-brand connections ), and secondly tests the relationship of a term that carries information about both predictors (the interaction term). The overall model fit was significant F (1, 68) = 33.161, p<.000. However, only the main effect of brand symbolism was significant ( =.0674, t = 2.087, p <0.041). The interaction term was insignificant, ( =.161, t = 0.315, p <0.754). Thus H5 was not supported. The results of the moderated regression mode l are shown below in Table 4.11. Table 4.11 Moderated Regression Results for Self-Brand Connections Parameter t Sig. Tolerance VIF (Constant) 12.31 Identity Centrality -0.08 -0.25 0.61 0.06 17.93 Brand Symbolism 0.67 2.09 0.80 0.06 17.15 Identity Centrality x Brand Symbolism 0.16 0.32 0.04 0.02 42.98 R2 = 0.605 Adjusted R2 = 0.0.587 Model Fit F(1, 68) = 33.161, p<.000
105 4.5. Discussion There are two arguments in this study that have been addresse d: consumers assign importance to their identities and multiple identity motives beyond self-esteem and selfconsistency guide self-brand interactions These arguments were based on the relationship proposed in the Iden tity Process Theory which purports that identity motives lead to identity centrality. The findings reveal full support for H1, and H3, and partial support for H2 and H4. Hypotheses 2 and 4 were partia lly supported due to the limited influence of multiple identity motives. This study failed to provide evidence of the joint influence of self-esteem and self-consistency impacting identity centrali ty. This is inconsistent with Grub and Grathwohls Model of Consuming Behavior (1967) and Sirgys (1982) Product-Image Congruity Theory. However self-esteem by its elf influenced identity centrality in conjunction with recognition. Sinc e self-esteem influenced identi ty centrality, it is safe to say that as brands are chosen to increase an individuals overall self-worth, then that brand is satisfying motives of self-esteem. In other words, the more participants rated an identity as satisfying feelings of self-esteem the more central that identity was to their selection of an automobile. Because self-consistency was not a significant predictor of identity centrality, it begs to question the limits of the product image congruity theory. Specifically one could ask, does the theory have boundary conditions within specific product categories? Or do alternate motives mask the motivation of self-consistency. Overall, the results indicate individual s do in fact assign varying levels of importance to their identities in a consum ption environment. The findings suggest centrality is a viable construct for unders tanding how individuals manage multiple
106 identities in consumption decisions. Particip ants rated as more central, those identities that provided a sense of self-esteem and that gained them recognition with regard to their automobile. With regard to the magnitude of influence towards centrality, the self-esteem ( = 0.805) motive had a greater influence in predicting centralit y than recognition ( = 0.440). This suggests that the identity germane to automobile brand choice satisfied the participants need for self-esteem a great deal more than it does for recognition. For a one unit increase in self-esteem, identity centrality would increase by .805 units, when the recognition motive is held constant. Convers ely for a one unit increase in recognition, identity centrality would d ecrease by .440 units, when the self-esteem motive is held constant. Or stated differently, the self-e steem motive is the key identity driver of identity centrality for the automobile cat egory among this population. Taken together, these motives seem to suggest individuals are motivated more so by internal self aspects (self-esteem) compared to others acknowle dgement of them (recognition). The results also suggest that if an indivi duals need for self-esteem is satisfied, then their motivation for recognition is mitigated in term s of which identity becomes central. The model predicting the influence of multiple motives on self-brand connections showed that the distinctiveness ( = 0.409) and efficacy motives ( = -0.483) were the only significant identity motives. However wh en these two motives were regressed on self-brand connections indepe ndent of the other motives, the efficacy motive became insignificant. Thus it appears that the e fficacy motive was significant due to the associations of the other motives, and this e ffect went away when the other motives were removed from the model. Ultimately, the distinctiveness motive was the only significant predictor of self-brand conn ections. For this population a one unit increase in the
107 satisfaction of the distinc tiveness motives leads to a 0.565 increase in self-brand connections. From an explanation standpoint, mo re information is needed to assess why the distinctiveness motive emerged as the only significant pred ictor of self-brand connections. Perhaps thought protocols and/or open-ended responses will be beneficial to understanding motiv e satisfaction. A glance at the identities in Table 4.4 shows the frequency in which participants listed professional identities. It can be assume d that automobiles are oftentimes a source of differentiation within their social categor ies or otherwise. In essence, automobiles enable them to feel distinctiv e from others. As a result, they in turn integrate the brand into their self-concept. While the interp retation of this finding is limited to this population, it would be advantageous to look at a broader sample and see if this motive is the sole predictor of self-bra nd connections. It may also sugge st that distinctiveness is such a strong motive, that its masks the eff ects of the other motives. More research is needed to make this statement, and analyti cal tools such as stru ctural equation modeling should be employed to assess th e relationship among the motives. The mediation of identity centrality faile d to show significance as hypothesized in H4. If significant this would have indicated that identity centrality reduced the effects of multiple identity motives on self-brand connections. Because mediation is a causal process, it would have been a significant fi nding if identity centrality intervened between motive satisfaction and self-brand connections. This means that identity centrality would have been presumed to cause self-brand conn ections. However, since this did not occur, the significance of the distinctiveness and efficacy motives are of little value within an overall mediation model.
108 Importantly, it would not have been suffici ent to correlate iden tity centrality with self-brand connections, because the two variable s may have been correlated because they are both caused by predictor vari able (s). This hypothesis, if found significant, would have offered a great deal of insight into the formation of self-brand connections. Specifically, because the self-brand connection l iterature is fairly new, it would have been a significant contributi on to understanding the mechanism or process through which identity motives influence self-brand connections. Symbolic benefits of brands operate th rough a signaling process in which what the brand says about the consumer is comm unicated to the consumer and to others (Helgeson and Supphellen 2004). This effect can be based on the image of a typical user of the brand and/or the personality of th e brand itself. When combined with the psychological importance, an id entity that is congruent with brand associations, but is also congruent with the typical user and/or brand personality, stronger connections to the brand are likely. Brand symbolism as a moderator between identity centrality and self-brand connection was not evidenced from the findi ngs. Perhaps this due to the fact that participants indicated high centr ality ratings for more than one identity, which means the effects of identity centrality, may have been masked in the other identities that were not reported as central to the automobile product ca tegory. The solution to th is is to isolate a central identity or have participants establ ish an automobile-buying identity. In this manner, the identities will be appropriately isolated, and the combination of brand symbolism and identity centrality should ultimately lead to stronger self-brand connections. A separate analys is of brand symbolism on self-brand connections revealed
109 that brand symbolism accounts for 60% of th e variance in self-brand connections. Also for a one unit increase in brand symbolis m yields a 0.77 increase in self-brand connections. This finding sugge sts that brand symbolism is a strong predictor of selfbrand connections. Mindfully, it may be appropr iate to consider brand symbolism as a mediator of self-brand connections instead of identity centrality. This potential interactive effect between brand symbolism and identity centrality is worth exploring further, but care must be given to ensure identity centr ality is being accounted for accurately for a specific identity. In the regression models predicting the influence of multiple motives identity centrality (H2) and self-brand connections (H4), not all identity motives were significant. There are several explanations for this outcome, namely the identity motives of the population surveyed and the size of the sample. Beyond the motives that have been widely ac cepted in the literatur e, there is little support for the influence of the typology of mo tives presented in th e automobile category. Therefore it is hard to surmise why most of these motives were insignificant on the dependent variables. One explanation is that the population survey has similar motivations when it comes to purchasing an automobile, so it is likely that all nine motives would not be significant. Or similarly, their motivations are so similar such that the two motives that were found to be signifi cant accurately capture the desires of this population. Even it that were tr ue, other motives should emerge as influencing choice of automobile such as efficacy, because automobiles are high-involvement products. Or perhaps, distinctiveness, mo st automobiles are now customizable, and that feature can speak to the identity needs of individuals who are in the ma rket for an automobile that
110 differentiates them others. This is not to say that all identity motives should have emerged, but the expectation was that ther e would be multiple identity motives, and preferably other than self-esteem and self -consistency that infl uenced both identity centrality and self-brand connections. Alternatively, the small number of re spondents can be the reason why more motives were found to be insignificant. There was simply not enough power in the sample, so the effects did not emerge. While the assumptions of regression were not violated, the small sample size simply tapped into the effect that was hypothesized. More data should be collected so th at there is significant power in the findings in order to draw substantive conclusions a bout the findings here. In addition to the identity motives bei ng insignificant, a few of the identity motives had negative regression coefficients. Th e regression coefficient is the interpreted as the correlation between the independent variable and th e dependent variable (Cohen and Cohen 1983). Therefore, for identity central ity, the meaning, continuity, security all had negative, insignificant regression coefficients. However, recognition was a significant motive which had a negative regression coefficien t. From a quantitative standpoint, this result implies that the rec ognition motive is negativ e correlated with identity centrality. This coul d be due to the recognition be ing a second order motive that is explained by a higher order motive such as self-esteem. Or perhaps identity centrality is motivated by internal se lf-considerations and recogni tion is an external selfconsideration, causing the negative relationship. With regard to self-brand connections, the continuity and belonging motives were insignificant negative motives. However efficacy was the only significant motive that
111 carried a negative coefficient. It is proposed that the efficacy motive is negatively related to self-brand connections to the competency related aspect of the motive. Self-brand connection is the degree to which an individual integrates a brand in to their self-concept, and what this means is that the more an indi vidual satisfies his or her need for efficacy, the weaker the self-brand connection. Perhap s this because efficacy is and extrinsic motive and it doesnt reflect positively on the innate nature of self-brand connections. 4.6. Managerial Implications The identity centrality meas ure can serve several specifi c functions in during selfbrand interactions. Incorporat ed into self-brand studies, id entity centrality allows for consideration of a brands symbolic propertie s consistent with the identity (or related motives) most central to the product categ ory. This relationship is driven by the satisfaction of multiple identity motives as evidenced in this study. Given this, it is necessary for marketers to bette r understand the role of iden tity motivations for brands during the preference-development stage. In doing so, they will improve their understanding of the role of brands in the identity construction process. Particular attention should be paid to th e changing influence of motives across the different phases of an individuals life cycle. It is also useful for marketers and re tailers to consider the situations and circumstances that encourage the prevalence of a particular identity motive. By crafting brand strategies and marketing communications around key motives, a central identity is likely to emerge. Since individuals are committed to identities that have higher levels of importance (Foote 1951), they are likely to enga ge in activities to support that identity.
112 For the marketer this means increased brand involvement, brand now becomes a part of the evoked set, and perhaps brand purchase. 4.7. Research Limitations and Future Research The findings of this research provided impor tant managerial and theoretical insights into the role of identity motives in establ ishing the centrality of identity. Notwithstanding these insights, several limitations should be a ddressed. First, this study was limited to the participants featured in this study. Therefore applying the re sults to a different population of consumers and in a different product ca tegory would prohibit the generalization of these findings. Second, caution must be exercise d for the various identities identified in the automobile context, a measure was not take n to assess whether or not the participants actually owned their vehicle. Th is could have implications for the identity centrality and self-brand connections measures. In additi on, this measure could reveal important insights as to the identi ties guiding their brand c hoice (actual vs. desired). Future research is warranted to ad dress these limitations and expand the theoretical validity of the findings. One oppor tunity for future investigation is the assessment of longitudinal motive ratings. This could be done through a survey taken at two points in time for a given consumption e xperience. By tracking identity centrality over time, the brand manager could consider whether his/her marketing actions are improving or deteriorating consumer self-brand interactions. Thus provide revealing how self-brand connections actually form in cu ltures where brand exposure may be limited. Another consideration is other motives beyond what is presented in the present study. Potential motives include those associ ated with hedonic consumption, as well as motives associated with religious identity, so cial acceptance, and social emulation. Future
113 research is also needed to examine the iden tity motives predicting identity affect and enactment. Marketers are now tr ying to gain insight into how consumers form emotional attachment to brands, as evidenced by the 2007 MSI Conference committed to this topic. Consideration of the motives influencing id entity enactment will allow managers to pinpoint which motives are responsible for ch anging consumption patterns. In this way targeted strategies can be designed to accommod ate enacted identities, as they are guided by specific motives. 4.8. Chapter Summary This chapter reported results of an expl oratory empirical stu dy investigating the ability of various identity mo tives in predicting identity cen trality, enactment, and affect. The goals of the research were two-fold: (1) to empirically test the influence of multiple identity motives on identity centrality a nd (2) to examine the relationship between identity centrality and self-brand connectio ns. Based on our findings, great potential lies in exploring motives beyond self-esteem and self-consistency. Evidence of these alternative motives was demonstrated and their use in self-brand interactions was identified. Generalizing the role of identity centrality beyond individual brand choice, the next chapter examines the role of identity centrality in group br and associations, and related self-brand connections.
114 Chapter 5 Essay 3-When the Ingroup Fails to Indica te Brand Meaning: Exploring the Role of Identity Centrality in Self-Brand Connections In this chapter, the results from a qua si-experimental design study revealed that identity centrality was a positive moderator of the relationship between reference group brand associations and self-brand connectio ns. This moderating influence led to a significant difference in self-brand connec tions between reference group members who rated their identity as low in centrality compared to those who rated their identity as high in centrality. Identity centrality did not significantly impact self-brand connections when self-construal and brand symbo lism were taken into account. 5.1. Introduction Self-brand connections are defined as the degree to which a consumer has incorporated the brand into their self-concept (Escalas and Bettman 2003, 2005). Escalas and Bettman (2005) found that brand associatio ns consistent with an ingroup led to stronger self-brand connections compared to brand associations inconsistent with an ingroup. This is because brands become more meaningful the more closely they are linked to an individuals ident ity. But what happens to this relationship if another identity other than the ingroup identity is associated with th e brand? Or how are self-brand connections influenced if the reference group is associated with multiple brands? This study seeks to address the former question by ex amining the nature of multiple identities in the formation of self-brand connections.
115 Markus and Kunda (1986) put forth the term malleable (or working) selfconcept to refer to an individuals various se lf-conceptions (e.g. ideal self, perceived self, social self) which function to provide an in terpretative and evaluative context for the overall view of the self (Markus and Wurf 1987). To date the consideration of multiple selves has been omitted from research linking reference group influence to self-brand connections (Escalas and Bettman 2005; Ch aplin and Roedder-John 2005). This omission leads to questions surrounding the differen tial attitudes and cognitions within the perceived reference group. This is supported by literature in social psychology which suggests individuals assign a level of signifi cance to their identit ies which directly impacts group driven attitudes, behaviors, and cognitions (Vignoles, Regalia, Manzi, Golledge, and Scabini 2006; Settles 2004; St ryker and Serpe 1994). It is argued that because individuals possess multiple identiti es, an individuals self-brand connection should be reflective of the ident ity that is most central to th e brand. In simpler terms, if the brand is associated with a particular ingroup, then the ingroup identity should carry the most psychological importance in interactions with the brand. Identity centrality, defined as the psyc hological importance one places on a given identity (Settles 2004) can aid in explaining how individuals negotiate multiple identities exacerbating one and buffering others. Identity centrality requires conscious awareness and is usually measured by asking individuals to rank different identities according to their importance (Rane and McBride 2000). Se lf-brand connections are likely to be enhanced when an ingroup identity is central to the brand compared to when it is not. This is due to the functioning of identity commitment, which is hypothesized to be associated with expectations of behavior re levant to identity go als (Foote 1951). It is
116 suggested that depending upon th e centrality of the reference group identity, differential self-brand connections will result thus m oderating the relationship found by Escalas and Bettman (2005). This research shows that th e differential self-bra nd connections are due to the lack of value expressive influe nce by other reference group members when centrality is taken into consideration. 5.2. Theoretical Framework At first glance social identity theory w ould seem to predict that members of a reference group should form strong self-brand co nnections for brands that are consistent with their ingroup as found by Escalas and Bettman (2005). However the case is made that when Social Identity Theory, Iden tity Theory and the personal identity are considered, identity interference may result. 5.2.1. Social Identity Theory vs. Identity Theory Social identity theory (Tajfel and Turn er 1985) suggests that individuals base their identities off of categories or groups; whereas, Identity theory (Stryker 1980) suggests individuals base their id entity off of roles. A social group is defined as a set of individuals who view themselves as member s of the same social category (Hogg and Abrams 1988). To illustrate the difference between the two identity theories, consider the identities of professor a nd student. First, professor and student are roles that are defined with the group (or organization) of a university, such that meanings and expectations are related to each of these ro les. Similarly, professor and student are a social category that constitute ingroups a nd outgroups. This is significant because individuals will always occupy a role and be long to a group simultaneously (Stets and Burke 2000).
117 Other theorists have suggested that the pers on identity also competes with the role and social identity. This is primarily due to the hierarchy of self-categorization. The personal identity is the lowest level of self-categorization (Brewer 1991; Hogg and Abrams 1988) and refers to an individual s categorization of themselves as a unique entity, unique from other indivi duals. This identity is categor ized based on idiosyncratic characteristics and addresses the self meanings that sustain the self as an individual (Stets and Burke 2000). To illustrate the differen ce between a role identity and a person identity: a masculine gender reflects a role identity versus I am a competent person, which reflects a person identity. Within this context, the individual behaves in accordance with his or her personal goals and desires ve rsus as a member of the group or category. In some cases, it may be difficult to sepa rate the role iden tity from the group identity from the person identity. This blurr ing of identities can potentially explain why reference group behavior may be differen t across members w ithin the group. For instance, when the meanings and behaviors asso ciated with a particular role conflict with that of the reference group, an individual may experience psychologi cal tension (Settles 2004). In order to mitigate this tension, an i ndividual may establish a level of importance with the reference group identity over the ro le identity (i.e. identity centrality). In contrast, the opposite can occur, where an i ndividual will establish a higher level of importance for the role identity over the reference group identity. Either way, these identities are always relevant to and influent ial on cognitions, affect and behavior (Stets and Burke 2000).
118 5.2.2. Reference Groups and Their Influence on Brand Meaning Reference groups are defined as a person or group of people that significantly influence the behavior of an individual (Bea rden and Etzel 1982) and can be an important source of brand meaning (Keller 1993; Es calas and Bettman 2005). They have two functions: a normative function that positions and enforces standards for the individual and a comparative function that serves as a point of comparison against which an individual evaluates himself and others (Kelley 1947; Cocanougher and Bruce 1971). Both functions are consistent with processes of self-categor ization and social comparison mentioned in the social identity theory. Building on the work of Kelley (1947) Bearden and Etzel (1982) suggest informational influence from the reference grou p occurs when in the face of uncertainty an individual searches for information and counts on sources with high credibility or experience in order to help make a decisi on. Utilitarian influence occurs when an individual acts according to th e desires of the reference grou p in order to obtain a reward or to avoid punishment. Value-expressive in fluence is characterize d by an individuals acceptance of certain external standpoints give n the psychological need to associate with a person or group. This study limits its fo cus on value-expressi ve reference group influences characterized by the need for psyc hological association w ith a group either to resemble the group or due to a liking for the group. It is has been shown by previous res earchers (Escalas and Bettman 2003; 2005) that if a reference group becomes associated w ith a particular brand, then the associations about the brand may be appropriated by consumer s as they construct th eir self-identities. Similarly, individuals buy brands congruen t with a reference group to enact their
119 reference group identity. Thus, identification with a reference group leads to a number of consequences, namely: 1) the choosing of ac tivities congruent with salient aspects of their reference group identity and 2) positive ev aluations of the group (Ashforth and Mael 1989). Similarly, individuals may avoid associa tions derived from groups to which they do not belong to maintain an accurate por trayal of their self-image. Thus, it is hypothesized: H1: Brand associations consistent with an ingroup (outgroup) will have a favorable (unfavorable) effect on self-brand connections, whereas brand associ ations inconsistent with an ingroup (outgroup) will have an unfavorable (favorable) effect on self-brand connections. H1 is a replication of Escalas and Bettmans (2005) H1A and H1B. 5.2.3. Identity Centrality and Reference Groups According to Hyman (1942) individuals may employ several reference groups in order to evaluate different aspects of their self -image. Internal as well as external conflict may arise when expectations or beliefs conflict between these multiple reference groups. Bearden and Etzel (1982) suggest Hymans (1942) Reference Group Concept provides a way to comprehend why many individuals do not behave like others in their particular social group. Rosenberg (1979) views self components as varying in the degree to which they are central or peripheral part s of the self. Underlying this variation is the importance assigned to the identities from the pers pective of the individual (Rosenberg 1979). Stryker (1980) argues that identities are arranged hierarchically according to their salience and centrality. An individuals ident ity becomes central according to its place in the identity hierarchy, along with the probabi lity of it being invoked in a given situation.
120 Salience is the likelihood that a particular status, role, or identity will be evoked in a given situation in comparison to the likelihood that other statuses, roles, or identities might be evoked (Rane and McBride 2000; St ryker and Serpe 1994). Salience is not a part of one's consciousness but reflects onl y the probability that an identity will be enacted (Rane and McBride 2000). Centrality requires conscious awareness and reflects the importance an individual attaches to a given identity. Prolonged identity salience enhances its centrality and the degree to which it can be li nked with other identities. The stronger the identity's centrality the more committed indivi duals will be to preserving and enhancing that identity (Settl es 2004). Centrality and commit ment influence the strength of an identity, how meaningful it is, and its potential is for shaping attitudes, values, and behaviors (Stryker and Serpe 1994). By introducing identity centrality into the results indica ted by Escalas and Bettman (2005), it is argued that self-brand connections will be moderated by their level of reference group centrality. Thus it would appear that individuals for whom the reference group identity is highly central are likely to have stronger self-brand connections, because they are supporting th e uniform perceptions of the ingroup. In contrast, individuals who are not interested in supporting th e uniform perceptions of the ingroup, or who may have other identity goal s within the group (e.g. personal or role identity goals), are likely to have lower self-brand connections Thus the following hypotheses are offered: H1C: When an ingroup identity is high in centrality, brand associations consistent (inc onsistent) with an ingroup will lead to stronger (weaker) self -brand connections compared to when the ingroup identity is low in centrality.
121 H1D: When an ingroup identity is high in centrality, brand associations consistent (inc onsistent) with an outgroup will lead to weaker (str onger) self-brand connections compared to when the ingroup identity is low in centrality. 5.2.4. The Role of Self-Constru al in Brand Associations How consumers view themselves in relati on to others requires placing self-brand connections within a self-development context. One of the major aims of this study is to show how varying levels of id entity importance (e.g. centralit y) aid in self-definitional goals. However identity centrality is relate d to identity importance within the selfhierarchy and fails to capture individual di fferences in self-definition. Self-construal, which is related to the overall view of the self, captures how an i ndividual views him or herself relative to others. Markus and Kitayama (1991) differe ntiate between independent and interdependent self-construals. Self-construal is conceptualized as the constellation of thoughts, feelings, and actions concerning an indivi duals sense of self in relation to others (Markus and Kitayama 1991; Singelis 19 94). Individuals with independent selfconstrual see the self (i.e., the overall self) as stable and separate from the interpersonal context and value self-promotion, autonom y, assertiveness, and uniqueness. These individuals tend to focus on internal attribut es such as ones own ability, intelligence, unique personality traits, goals, preferences, or attributes that they express in public and verify in private through social comparison. Thei r behavior is consiste nt with and reflects those internal beliefs and valu es (Markus and Kitayama 1991). On the other hand, individuals with an inte rdependent self-const rual perceive the self as more flexible and intertwined with the social context, and value fitting in and maintaining group harmony. Interdependent self-construal refl ects a flexible, variable self
122 whose expression and experience of emotions are significantly shaped by a consideration of the reactions of others in the grou p (Markus and Kitayama 1991). They behave primarily in accordance with the anticipated expectations of others social norms and emphasize the collective welfare and are concer ned with the needs and goals of others. Similar to Escalas and Bettman (2005) it is predicted that individuals who have an interdependent self-construal will form self-brand connections based on some shared aspect of themselves with the ingroup and will be immune to outgroup brand associations (p.380). Individuals who have an independent self -construal will be motivated to establish differe ntiation from the outgroup to c reate a unique self-concept (p.380) which should lead to lower self-b rand connections. Replicating Escalas and Bettman (2005), it is hypothesized: H2A: Brand associations consistent with an outgroup will lead to lower self-brand connections fo r independent self-construals compared to interdependent self-construals. It was hypothesized by Escalas and Bettman (2005) that independent individuals should be immune to outgroup brand associations . However, an increase in centrality should prompt a desire towards establishi ng ingroup differentiation from the outgroup. This is because individuals are likely to display favoritism wh en an ingroup is central to their self-definition and when a given comp arison is meaningful or the outcome is contestable (Tajfel and Turner 1979). Thus it is hypothesized: H2B: When an ingroup identity is high in centrality, brand associations consistent with an outgroup will lead to lower self-brand connections for inte rdependent self-construals compared to independent self-construals.
123 If an individual regards their ingroup identity as high in centrality then comparison to the outgroup will cause them to seek reinforcement from their ingroup. Thus when the ingroup identity is central, it is predicted that the independents will shift their focus of differentiation from the outgr oup to the ingroup. Esse ntially there will be two competing forces: the need for in group affiliation and the need for ingroup differentiation. It is posited that identity cen trality will cause ingroup affiliation needs to be stronger than outgroup differentiation needs, causing independents to form lower selfbrand than interdependents. H2C: When an ingroup identity is high in centrality, brand associations consistent with an ingroup will lead to lower self-brand connections for i ndependent self -construals compared to interdepen dent self-construals. 5.2.5. Brand Symbolism A central thrust of this research is th at consumers are influenced by reference group brand usage, and in part, construct thei r identities based on these associations. This is supported by Levy (1959) who asserts that individuals do not buy products simply for their functional value, but al so for their symbolic mean ing. Brand symbolism enables consumers to form a long-lasting relations hip with a particular brand through the emotional and functional utility of the offe ring. The emotional benefits of symbolic brands have been recognized by marketers as a prerequisite for sustaining brand success (Keller 1993). It follows that some brands are able to communicate an identity better than others. For example, prior consumer research sugge sts that publicly consumed (vs. privately consumed) and luxury (vs. necessity) products are better able to convey symbolic meaning about an individual (Aaker 1991). In addition, Aaker (1991) found that brands
124 that have personality traits similar to that of the consumer yield more favorable attitude towards the brand. It is expected that the degree to which a brand can communicate something about the consumers identity will moderate the relationship between reference group brand associations and self -brand connections. Stronger effects are expected for more symbolic brands (i.e., those brands better able to communicate something about ones self-identity). In line with Escalas and Bettman (2005), the following hypotheses are offered: H3A: More symbolic brands will lead to stronger (weaker) self-brand connections brand associations c onsistent (inconsistent) with an ingroup compared to less symbolic brands. H3B: More symbolic brands will lead to weaker (stronger) self-brand connections brand associations c onsistent (inconsistent) with an outgroup compared to less symbolic brands. If the ingroup identity is central, an individual will behave in a way that establishes congruity with others in th e group. Once this occurs, assuming the more symbolic brands are able to communicate something about the ingroup, then self-brand connections should be higher for individuals who are regard the ingroup identity as central compared to those who percei ve the ingroup as low in centrality. Individuals who view the ingroup as cen tral are committed to establishing congruity between the id entity and behavior, thus outgroup brand associations will have an unfavorable impact on self-brand connections Even if the brand consistent with the outgroup has associations that ar e positively related to the ingroup, the mere fact that it is associated with the outgroup will negativel y impact self-brand connections for individuals who regard their i ngroup as highly central. Assumi ng that brands inconsistent with the outgroup are able to communi cate something about the ingroup, self-brand
125 connections should be higher for individuals who are low in centrality. Thus, the following hypotheses are offered: H3C: When an ingroup identity is hi gh in centrality, more symbolic brands will lead to stronger (weaker) self-brand connections for brand associations consistent wi th an ingroup (outgroup) compared to when the ingroup identity is low in centrality. H3D: When an ingroup identity is hi gh in centrality, more symbolic brands will lead to weaker (s tronger) self-brand connections for brand associations inconsiste nt with an ingroup (outgroup) compared to when the ingroup id entity is low in centrality. 5.3. Methodology The influence of reference group brand asso ciations on self-brand connections due to differences in identity centrality was expl ored. Specifically the goal was to understand the degree to which identity centrality shapes self-brand connections when reference groups brand associations are considered. Br and symbolism was also considered as it may have a stronger effect on self-brand c onnections than identity centrality when reference group images are inconsistent with the ingroup. This is due to a brands ability to help communicate ones identity. A qua si-experimental approach was employed, based on a 2x2x2x2 mixed design with identity centrality (high vs. low) and selfconstrual (independent vs. interdependent) as between-subjects vari ables and group type (ingroup vs. outgroup) and brand image match (c onsistent vs. inconsistent) as withinsubjects variables. 5.3.1. Participants Three hundred and thirteen individuals par ticipated in this st udy. The participant population was selected based on the desire for generalizability across consumer populations. Thus, it was the goal of the re searcher to include both student and non-
126 student populations. As a result, there we re two hundred and fifteen students who participated in this study from a large Sout heastern university in exchange for additional course credit. Student participants were r ecruited through a Basic Marketing course. In addition, ninety-eight non-stude nts participated in the study. They were contacted from an alumni mailing list and participated vol untarily. Data were colle cted using an online data collection website ( http:///www.vovici.com ) as well as in a computer lab on campus. Forty-four participants had to be elimin ated for entering improper or incomplete responses (e.g. listing organizations versus br ands or listing the same brand twice), leaving a total of 269 participants. Sixty-seven percent of part icipants fell into the 1825 age category, and 33% of the participants fell into the 26-35, 26-42, 46-52, and 56-64 age categories. There were 147 females and 122 males. Seventy-seven percen t of participants were single, 14% were married and 9% left this question blank. Si xty-seven percent had some college and had yet to earn a degree, and the remaining 33% ha d obtained a Bachelors degree or higher. All three hypotheses were tested am ong the student group and the non-student group. Support for these hypotheses is shown be low in Table 5.1 Cell sizes are below the self-brand connection values.
127 Table 5.1 Hypothesis Testing for Student vs. Non-Student Population Group Type Image Consistent Image Inconsistent Image Consistent Image Inconsistent H1AIngroup63.3329.83Supported 58.6729.00Supported (n=198)(n=198) (n=71)(n=71) H1BOutgroup31.5046.33Supported 27.5755.00Supported (n=198)(n=198) (n=71)(n=71) Low CentralityHigh Centrality Low CentralityHigh Centrality H1CIngroup55.5066.50Supported 53.1762.83Supported (n= 55)(n=143) (n=31) (n=40) IndependentInterdependent IndependentInterdependent H2A33.1724.50Not Supported 28.8324.83Not Supported (n=23)(n=27) (n=16)(n=7) H2B32.0625.12Not Supported 23.8123.22Supported (n=15)(n=20) (n=12)(n=4) H3ALow Symbolism47.17 33.17Supported 44.6731.83Supported (n=30)(n=69) (n=28)(n=40) High Symoblism72.0326.66 67.6725.20 (n=99)(n=129) (n=43)(n=30) H3BLow CentralityHigh Centrality Low CentralityHigh Centrality Ingroup65.7973.95Supported 59.7973334Supported (n=30)(n=99) (n=18)(n=25) Outgroup27.9429.61Not Supported 31.4829.91Not Supported (n=30)(n=99) (n=18)(n=25) H3COutgroupLow CentralityLow CentralityNot SupportedLow CentralityLow Centrality 42.66675234.000068Not Supported28.5005740.6748Not Supported (n=84)(n=112)(n=40)(n=26)Supported High CentralityHigh CentralityHigh CentralityHigh Centrality 49.1676528.1672326.3338663.0126 (n=111)(n=86)(n=31)(n=45) Note: The differences between means are in he hypothesized direction and significantly different at p<.005 if supported. Student PopulationNon-Student Population 5.3.2. Procedure The procedure was a replication of th at used by Escalas and Bettman (2005). Changes dealt with the order of measures a nd the incorporation of the identity centrality measure.
128 This study used a Visual Basic program that allowed for the customization of the participants' responses. Similar to Es calas and Bettman (2005), the program began with a short study introduction and was follo wed by half of the subjects completing the Singelis (1994) independent and interdependent self-construal scales at the beginning of the study, while the remaining half of the subj ects completed these scales at the end of the study. To facilitate this effort online, tw o versions of the instruments were designed. Participants received a link to complete onl y one version of the instrument. An equal number of invitations were sent out for both versions. Afterwards participants en tered a group to which they belonged (i.e., an ingroup), In the box below, we would like you to type in the name of a group that you belong to and feel a part of. You should feel you are th is type of person and that you fit in with these people. This group shoul d be a tightly knit group, cons isting of individuals who are very similar to one another. Next they entered a group to which they did not belong (i.e., an outgroup) In this box, we would like you to ty pe in the name of a group that you do NOT belong to and do not feel a part of. Y ou should feel you are not this type of person and that you do not fit in with these people. This group shoul d be a tightly knit group, consisting of individuals who ar e very similar to one another. After each group, participants were asked to list one brand that was consistent with the group and one brand that was not. In the box below, we would like you to type in a brand that is consistent with the group that you belong to. This can be a brand that members of the group actually use or it can be a brand that shares the same image as the group. A brand is considered to be a name or sy mbol that distinguis hes one seller's goods from another's. and Now, we would like you to type in a brand that is NOT consistent
129 with the group you belong to. This can be a brand that members of the group would never use or it can be a brand that has the opposite image from the group. Thus, each participant entered four brands, correspondi ng to four group-brand pairs: ingroup-brand matches, ingroup-brand does not match, out group-brand matches, and outgroup-brand does not match. Next, participants completed a series of measures indicating the degree to which they "fit" with each group, anchored by (1)-S trongly disagree (7)-S trongly agree I consider myself to be this type of person. I belong to this group. and I fit in with this group of individuals. These three items were averaged for a standardized score ( = .883). Then they were asked to rate the extent to which an association with each group type would communicate something positive or negative about them. After a short, unrelated filler task de signed to reduce poten tial demand effects, participants rated the degree to which they had self-brand connections (Escalas and Bettman 2003) with these four brands anchored by (0)-Strongly disagree to (100)Strongly agree. This brand reflect s who I am. I can identify with this brand. I feel a personal connection to this brand. I us e this brand to communicate who I am to other people. I think this brand help me b ecome the type of person I want to be. I consider this brand to be me (it reflects who I consider myse lf to be or the way that I want to present myself to others). This brand suits me well. A standardized score was created to form one self-brand connec tion score per participant per brand ( = 0.928). Then they were also asked to rate the brands on a number of dimensions, including the degree to which the brand was able to communicate something symbolic about the brands user using two pr escribed 100-point scale items. To what extent does this brand communicate something sp ecific about the pers on who uses it? anchored by
130 (0)Does not communicate a lot, (100)-Communicates a lot, and How much does this brand symbolize what kind of person uses it? anchored by (0)-Not at all symbolic, (100)Highly symbolic These two items were averaged for a standardized score ( = .890). After which participants answered measures regarding their level of centrality for their ingroup identity (their ingroup identity appear ed on the screen within the question stem based on their earlier re sponses) anchored by (1)Strongly Disagree and (7)Strongly Agree I often think about being a(n) _______ member. Being a(n) _______ member has little to do with how I feel about myse lf in general(reversed score). Being a(n) _______ member is an important part of my self image. The fa ct I am a(n) _______ member rarely enters my mind (reverse scored). The _______ group I belong to is an important reflection of who I am. Overa ll my_______ membership has very little to do with how I feel about myself. (reverse scored) The _______ group I belong to is unimportant to my sense of what kind of person I am(reverse scored). The degree of centrality of the participant s ingroup identity was assessed using the standardized score of the seven items ( = .866). This was followed by the collection of demographic information. A debriefing statement was issued to all participants at the end of the survey. The entire procedure took approximately one half-hour. As noted above, during the study each pa rticipant entered two groups, an ingroup and an outgroup. For each group, participants en tered a brand consistent with the image of the group and a brand not consistent with the image of the group re sulting in a set of four brands. The specific groups and brands are idiosyncrati c to each participant and are not of interest in the analysis. The data are only sorted by group type (ingroup versus
131 outgroup) and brand image match (image ma tches versus image does not match) for analysis. A sample of ingroups and outgroups listed by the participants are shown below in Table 5.2 Table 5.2 Example of Reference Groups and Brands Listed by Participant Participant Number Type of Group Group Listed Brand with Associations Matching Group Brand with Associations Not Matching Group 1 Ingroup Conservatives Jcrew Quiksilver Outgroup Rednecks Wrangler Ralph Lauren 2 Ingroup Dog Lover Land Rover Corvette Outgroup Sports Nut Nike Salvatore Ferragamo 3 Ingroup Soccer Moms Nordstroms Hot Topic Outgroup Young singles Abercrombie and Fitch Ann Taylor 4 Ingroup Busy Moms Target Harley Davidson Outgroup Nascar Fans Armor All Ann Taylor 5 Ingroup Socially Conscious Wild Oats General Motors Outgroup Conservatives Budweiser REI Participants completed the entire self -construal scale (Singelis 1994) for both independent ( = .777) and interdependent ( = .738) self-concepts. The scale is anchored by ( 1)-Strongly Disagree and (7)-Strong Agree Based on a median split of scale scores, participants were divided into high and low groups for each self-construal type. Participants were classified as being interdependent if they scored a 3.5 or higher on this
132 scale and scored less than a 3.5 on the indepe ndent scale. The reverse is true for those participants classified an independent; they scored a 3.5 or higher on the independent scale and below a 3.5 on the interdependent scale. Participants who were high in independent and low in interdependent were considered to be re presentative of the independent self-construal, while participants who were high in interdependence and low in independence were considered to be representative of the interdependent self-construal (Escalas and Bettman 2005; Mandel 2003). Par ticipants who were high on both or low on both scales (e.g. scored a 3.5 or above) were eliminated from the dataset, leaving a total of 77 participants4. By construction, the interdependent participants scored significantly higher on the mean score of the in terdependence items (4.98 vs. 2.97), F (1,75) = 2.259, p < .027 and significantly lower on the mean sc ore of the independence items (5.147 vs. 3.28), F (1,75) = -3.139, p < .003 compared to the independe nt participants. Scale items are presented below in Table 5.3. 4 A similar result was found in Escalas and Bettman (2005). When the authors assigned individuals into self-construal groups, they had to eliminate 168 participants leaving 75 part icipants for the analysis, because they scored high on both self-construal types. In order to retain all individuals, they calculated a self-construal index, and ran the analysis both with and without the index scores. They found the re sults to be virtually id entical in both analyses.
133Table 5.3 Self-Construal Measures Interdependent Self-Construal Items 1. I have respect for the authority figures with whom I interact. 2. It is important for me to maintain harmony within my group. 3. My happiness depends on the happiness of those around me. 4. I would offer my seat in a bus to my professor. 5. I respect people who are modest about themselves. 6. I will sacrifice my self-interest for the benefit of the group I am in. 7. I often have the feeling that my relations hips with others are more important than my own accomplishments. 8. I should take into consideration my parent s advice when making education/career plans. 9. It is important to me to respect decisions made by the group. 10. I will stay in a group if they need me, even when Im not happy with the group. 11. If my brother or sister fails, I feel responsible. 12. Even when I strongly disagree with group members, I avoid an argument. Independent Self-Construal Items 1. Id rather say No directly, than risk being misunderstood. 2. Speaking up during class is not a problem for me. 3. Having a lively imagination is important to me. 4. I am comfortable with being singled out for praise or rewards. 5. I am the same person at home that I am at school. 6. Being able to take care of myself is a primary concern for me. 7. I act the same way no matter who I am with. 8. I feel comfortable using someones first name, soon after I meet them, even when they are much older than I am. 9. I prefer to be direct and forthright when dealing with people Ive just met. 10. I enjoy being unique and different from others in many respects. 11. My personal identity independent of others is, very important to me. 12. I value being in good health above everything.
1345.4. Data Analysis and Results To assess if the manipulation was successf ul, the degree to which the participant belonged to the ingroup and the outgroup he /she entered was assessed using the standardized score of three items ( = .883). Participants regarded themselves as belonging to the ingroup, M = 5.75 significantly more than th ey felt they belonged to the outgroup, M = 1.72, F (1,243) = 39.65, p < .000. Thus the manipulation held for ingroup versus outgroup affiliations. For purposes of analysis, participants were divided into two groups based on a median split of the identity centrality measure. Individuals were cla ssified as being high in centrality if their standard ized score was greater than 3.5, while those whose score was below 3.5 were classified into the low centrality group. One-hundred and eighty-two participants rated th eir ingroup identity as high in centrality, M = 4.42; while eightyseven participants rated their ingr oup identity as low in centrality, M = 2.26, F (1,268) = 20.80, p < .0000. 5.4.1. Hypothesis 1 H1A and H1B were assessed first. These hypothese s were a direct replication of Escalas and Bettman (2005)s H1A and H1B which predicts a two-way interaction between group type and brand image match. It was e xpected that the perceived association between a reference group and a brand woul d have differential effects on self-brand connections depending on the group type. This hypothesis was tested using a univariate ANOVA with a priori contrast to see if the re sults could be replicated, and they were. A significant interaction of group type by bra nd image match on self-brand connections was found, F (1, 1064) = 369.15, p < .0000; see Figure 5.1. As predicted in Hypothesis 1A,
135 brands consistent with the image of the i ngroup resulted in higher self-brand connections Mconsistent = 62.18 than brands inconsis tent with the groups image Minconsistent = 29.60, F (1, 269) = 19.794, p < .0000. As suggested by Hypothesis 1B, brands consistent with the outgroup had less favorable self-brand connections Mconsistent = 30.43, than those that did not match the image of the outgroup Minconsistent = 48.68; F (1, 269) = -9.304, p < .0000 The results are consistent with the original authors prediction s in the directions hypothesized, thus H1A and H1B are supported. Figure 5.1 Self-Brand Connections by Group Type by Brand Image Match
136 H1C and H1D and were assessed here. These hypotheses extended the prediction of Escalas and Bettmans (2005) H1A and H1B to include the moderating effects of identity centrality. Thus, pr edicting a three-way interac tion between group type, brand image match, and identity centrality. When an individuals reference group identity is high in centrality and the brand image is cons istent with the image of the ingroup, more favorable self-brand connections were expected than when the ingroup identity is low in centrality. A significant thr ee-way interaction was found, F (1, 1064) = 22.17, p < .0000; see Figures 5.2 and 5.3. In the brand image does match condition, participants who reported their ingroup identity as being high in centrality reported higher self-brand connections Mhigh = 65.68 compared to those who reported their identity as being low in centrality, Mlow = 54.82, F (1, 268) = -4.202, p < .000. In the brand image does not match condition, participants who repor ted their ingroup identity as being high in centrality reported lower self-b rand connections Mhigh = 22.32 similar to thos e who reported their identity as being low in centrality, Mlow = 29.70, F (1, 268) = -3.693, p < .034.
137 Figure 5.2 Ingroup Self-Brand Connections by Bra nd Image Match by Identity Centrality Individuals who reported thei r ingroup identity as bei ng highly central reported lower self-brand connections for imag es consistent with the outgroup, Mhigh = 29.88 compared to those who reported their id entity as being low in centrality Mlow = 30.24; F (1, 268) = .132, p < .895. In the outgroup brand image does not match condition, individuals who rated their i ngroup identity as being high in centrality reported higher self-brand connections Mhigh = 48.97 compared to those i ndividuals who rated their ingroup identity as low in centrality Mlow = 48.09; F (1, 268) = -.267, p < .790. Thus H1C is supported, while H1D is not supported.
138 Figure 5.3 Outgroup Self-Brand Connections by Bra nd Image Match by Identity Centrality 5.4.2. Hypothesis 2 Hypothesis 2A replicates H2 from Escalas and Bettman (2005). It proposes that self-brand connections for brands with associat ions consistent with the outgroup will be lower for independent individua ls compared to interdepende nt individuals. A significant three-way interaction was found, F (1, 1064) = 4.973, p < .002; see Figures 5.4 and 5.5. The contrast comparing interdependent to independent self -construals found, Mindependent = 29.12, Minterdependent = 34.89; F (1,72) = 2.003, p < .049. This finding is consistent with the results of Escalas and Bettman (2005). As it relates to the outgroup, significant differences were found be tween an independent, Mindependent = 51.32 and interdependent Minterdependent = 42.96 individuals self-b rand connections in the brand image does not
139 match condition F (1, 241) = -2.30, p< .022. These results are shown in Figure 5.4. Thus, H2A was supported. Figure 5.4 Outgroup Self-Brand Connections by Brand Image Match by Self-Construal Type The contrast comparing interdependent to independent self-construals for the ingroup revealed interdependents and i ndependents formed similar self-brand connections, Mindependent = 62.92, Minterdependent = 63.85; F (1,241) = .329, p < .742 when the brand image was consistent with the ingroup. In the brand image does match the ingroup condition, significant differences were found in the self-brand conn ections between the two self-construal types, Mindependent = 28.06 and interdependent Minterdependent = 33.20; F (1, 241) = 2.04, p< .043. See Figure 5.5 below.
140 Figure 5.5 Ingroup Self-Brand Connections by Brand Image Match by Self-Construal Type For H2B and H2C these hypotheses extended th e prediction of Escalas and Bettmans (2005) H2 to include the moderating effect s of identity centrality. Thus, predicting a four-way interac tion between group type, brand image match, self-construal and identity centrality. This is interaction was not significant, F (1, 1064) = 0.279, p < .757. Thus, H2B and H2C were not supported. 5.4.3. Hypothesis 3 For H3A and H3B, it was hypothesized that the e ffects of ingroup and outgroup brand associations on self-brand connections wi ll be stronger when the brand was viewed as highly symbolic, a direct replica tion of Escalas and Bettmans (2005) H3. The three way interaction between group match, group t ype and brand symbo lism was significant, F (1, 1064) = 73.452, p < .000; see Figures 5.6 and 5.7. Si milar to Escalas and Bettman
141 (2005), the results are presente d after dichotomizing the brand symbolism measure. The results indicate that for brands consistent with the ingroup when the brand is highly symbolic, participants formed stronger self-b rand connections compared to brands that were low in symbolic value, MLow-symbolism = 45.67 and MHigh-symbolism = 70.93, F (1,268) = 11.44, p <.000. Opposite effects were found fo r the ingroup brand does not match condition, MLow-symbolism = 31.5 and MHigh-symbolism = 25.67, F (1,268) = 2.70, p <.007. The results are presented below in Figure 5.6 Figure 5.6 Ingroup Self-Brand Connections by Brand Image Match by Brand Symbolism In the outgroup condition, brand sy mbolism also moderated self-brand connections, significant results were found in the brand image match condition, MLowsymbolism = 32.58 and MHigh-symbolism = 27.67, F (1,268) = 2.307, p <.022; as well as in the
142 outgroup image incons istent condition, MLow-symbolism = 43.32 and MHigh-symbolism = 52.83, F (1,268) = 2.634, p <.009. The results are presented below in Figure 5.7. Thus H3A and H3B were supported. Figure 5.7 Outgroup Self-Brand Connections by Br and Image Match by Brand Symbolism For H3C and H3D, the findings in Hypotheses 3A and 3B were extended to account for differences in self-brand connections due to the degree of ident ity centrality. It was predicted that identity centrality will moderate self-brand connections when the brand was high in symbolism. The four-way in teraction between group type, brand image match, identity centrality and brand symbolism was insignificant F (1, 1064) = 1.319, p < .251. Thus H3C and H3D were not supported.
143 5.5. Discussion The overall objective of this study was to assess the moderating influence of identity centrality on the formation of self-b rand connections. Indivi duals reported higher self-brand connections for brands with associ ations congruent with an ingroup compared to associations incongruent with an ingroup (Hypothesis 1A). Brands associated with the image of an outgroup resulted in lower self-brand connections than brands with images not associated with an outgroup (Hypothesis 1B ). These findings were consistent with the findings of Escalas and Bettman (2005). The consideration of identity centrality in this study was based on the expectation that ingroup members will form differential self-brand connections for the brand image match condition. Extending these findings, identity centrality positively moderated selfbrand connections for the ingroup (H1C). This result is significant in that it reveals the significance of centrality in exam ining self-brand interactions. It is not enough to say that an identity is salient to a particular brand. As evidenced in the results, identity centrality leads to stronger self -brand connections in the brand image does match condition. This difference among ingroup members indicates there is varying levels of importance within the group. From a marketers stand point th is could mean the difference in message effectiveness. Salience may be sufficient enough to encourage an individual to attend to a particular message. But is it enough to create an emotional bond? Is it enough to get the consumer to buy the product? It is argued that centrality is true a catalyst for self-brand connections. The moderating influences of identity centrality on outgroup brand associations was insignificant (H1C). It difficult to hypothesize the directional influe nce of identity
144 centrality in outgroup effects due to various meanings associated with outgroups. Identity importance can be hypothesized to influen ce the outgroup but it wi ll vary depending on the desired degree of divergence from the outgroup. For instance, an individual may declare an outgroup but may desire to eventually become one its members (e.g. fraternities and sororities). Or perhaps an i ndividual may have some level of similarity with the outgroup. These scenarios make it diffi cult to predict how identity centrality will influence self-brand connections. Brand associations not matching the image of the outgroup were lower for individuals with an independent self-construal compared to interdependent self-construal (H2). This finding was consistent with Escal as and Bettmans (2005) results. They reasoned that outgroup brand asso ciations will have the great est effect on participants with independent self constr uals due to their strong di fferentiation needs. This was evidenced in this study as well. Identity centrality moderated these findi ngs such that the degree of self-brand connections for independent versus inte rdependent self-construal was reversed. Interdependents formed lower self-brand connections than in dependents. This is due to the importance placed on the ingroup identity. Interdependent s were thought to shift their focus from outgroup differentiation to ingroup differentiation, leading to lower self-brand connections. By doing this they are attempti ng to show ingroup favoritism and seeking distinctiveness from the outgroup. Lower self -brand connections for individuals with interdependent self-construals is more in line with the self-construal research which views interdependents are more collectivist in nature (Si ngelis 1994). The contrast in
145 findings due to identity centrality may be wo rth investigating the continued use of the self-construal to examine individual differences in self-brand connections. There was a non-significant interaction effect between group type, brand image match, brand symbolism and identity centrality It is argued that the interaction would have been significant if the scope of th e study focused on the ingroup or outgroup only. By holding this condition constant, the interac tive effect of identity centrality and brand symbolism would have been shown. Overall, the pattern of re sults supports the general id ea proposed in this essay: individuals use brands to crea te or communicate their self-concept partly in an effort to meet certain identity goals (e.g. self-verif ication, self-enhancement) and do it more so when a particular identity is central. 5.6. Managerial Implications Building brands through the development of emotional connections has been advocated in consumer markets (Pawle a nd Cooper 2006; Lindstrom 2005; Woods 2004) and identity centrality is a means to understand how consumers develop emotional connections to brands. As shown in this study, participants w ho rated their ingroup identity as high in centrality resulted in the strongest self-b rand connections. When taken in conjuction with identity commitment (Foote 1951), this finding suggests that identity centrality and identity commitment lead to stronger brand connections. This research also has implications fo r marketing communications. For marketers and retailers alike, the use of identity centrality in the cr afting of messages for targeted groups is a new opportunity. Identity centr ality may be communicated explicitly or implicitly in the advertisement. This is in line with Ashmore, Deaux, and McLaughlin-
146 Volpe (2004) who identified two distinct forms of psychological importance (e.g. centrality). Explicitly, the message should st ate the importance of an identity to the perceiver. Through an appraisal process the individual will assess th e importance of the identity and act in accordance with the appraisal (Ashmore et al. 2004). Implicitly, the message should use cues to trigger the iden tity that is highest on the individuals hierarchy of selves. Another important point for managers is th at centrality will cause the consumer to self-reference. Self-referencing is "the pr ocess of relating information to oneself" integrating communicated information with knowledge of oneself (Meyers-Levy and Peracchio 1996, p. 408). If the information comm unicated is related to an identity that sits at the top of the indi viduals hierarchy of selves and that identity is relevant to a particular good or service, individuals are mo re likely to attend to the communication and perhaps be inclined to pu rchase the good or service. 5.7. Research Limitations and Future Research The findings of this research provided impor tant managerial and theoretical insights into the role of identity centrality in the formatio n of self-brand connections. Notwithstanding these insights, several lim itations should be addressed. First caution must be exercised with the self-construal measure, Escalas and Bettman (2005) suggested their results were inconsistent with previous results on individualism-collectivism research which they say is akin to self-c onstrual. Notably, this construct may not be optimal for considering individual differen ces in the formation self-brand connections among older consumers who cannot directly relate to some of the scale items presented in the self-construal measures. This limits the interpretation of the findings of H2, because it
147 is difficult to tell if the hypothesis was not supported due to the poor measure or other theoretical arguments. Future research is warranted to address this limitation and expand the theoretical validity of the findings. One opportunity for fu ture investigation th at deserves further attention is the potential for assessing iden tity centrality within the marketing domain. For instance, this study examined the central ity of the reference group identity. Perhaps it will be more accurate to assess this measure rele vant to the brand as well. In other words, a construct that answers the question: how im portant is an identity to a specific brand? Instead of, how important is your ingroup? 5.8. Chapter Summary This chapter reported results of a qua si-experimental study investigating the moderating role of identity centrality in the formation of self-brand connections. The primary goal of the research was to test th e moderating effect of identity centrality on self-brand connections. The results indicate identity centrality was a positive moderator of self-brand connections. This is a signifi cant finding because it means future research on self-brand interactions should incorporat e centrality measure into their study designs before making conclusions on their findings. Th e next chapter concl udes this dissertation by providing a general discussion of the fi ndings and contributions across the three essays.
148 Chapter 6 Conclusions and Implications 6.1. Introduction The intent of this dissertation was to provide a theoretically grounded and consumer informed account of the various motiv ations influencing se lf-brand interactions and identity centrality. Theoretical and manage rial advances emanating from the multiple identities-multiple motive paradigm are discussed below. 6.2. Theoretical and Conceptual Contributions This dissertation advances a more robus t theory of self-referent consumption while providing an identity motive-based fram ework (Essay 1) in which identity motives are satisfied through symbolic brand associ ations. This approach draws on Identity Process Theory (Breakwell 1988 ) and leads to a more comp rehensive understanding of how multiple identity motives guide the se lf-concept in self-brand interactions. By advancing this framework, distinctions among self-brand phenomena are revealed by highlighting the multiplicity of not only an individuals motives but their identities as well. The current definition of self-brand connec tions conceptualizes th e self in terms of the overall self-concept and does little to char acterize the multidimensionality of the self. Thus one of the greatest adva ntages of the framework is that it allows one or more identities, or a context specific identity (i.e purchasing identity) to be examined. These considerations are in line with suggestions that identity is multidimensional (Markus and
149 Kunda 1986) as opposed to the unidimensional se lf (Sirgy 1982); an advance that further refines our knowledge of self-brand interactions. A second area of contribution is derived from the qualitative investigation of selfbrand interactions: including self-brand congr uence with multiple identities, life-cycle state influences, and the use of identity mo tives to manage multiple identities. Nine identity motives were put forth and hypothesi zed to influence both identity centrality and self-brand connections. Expanding the number of identity motives beyond self-esteem and self-consistency leads to new ways to study self-brand phenomena (i.e. competing identities, identity centralit y, competing motives, identity management). Most important for the purpose at hand, the participants qua litative responses demonstrated the validity of the multiple identity motives proposition put forth in this dissertation. Individuals do encounter multiple or competing identity motives. These motives take several forms, sometimes being more applicable at only the individual level, group level, or in some cases at both levels. Further, the inclusion of multiple identity motives theoretically extends the self-referent consumption literat ure, primarily in the area of how multiple identities emerge or become central. Conceptual advances offered through th e identity motive fr amework takes other forms as well. Application of the identity-motive framework increases the base knowledge of self-brand interactions beyond what has been obtai ned through dominant theories (i.e. Product Image Congruity Th eory; Extended Self-Theory) or widely accepted constructs (e.g. salience). The inclusion of identity centrality is significant in that forces marketers and researchers to c onsider a wider range of identity motives. Multiple identities operate thr ough a process of identity centrality, whereby one identity
150 becomes more central than another (Stryker and Serpe 1994; Settles 2004). This is because centrality or importance is almost always associated with some degree of motivation (Bagozzi, Bergami, Leone 2003). It was hypothesized and supported in this dissertation that as multiple identity motives are satisfied a given identity increases in centrality. This relationship suggests motiva tion is closed aligne d with importance. Therefore, if identity conflict should arise or if multiple identities have equal levels of importance, the brands ability to satisfy vari ous identity motives may ease this conflict. However in evaluating the usefulness of the identity centrality construct, one could ask a bottom-line question: Is there va lue in thinking of self-brand phenomena in terms of a central identity? The findings in this dissertation suggest the answer to this question is yes. Application of the cen trality construct compels researchers and marketers to view the self-c oncept as dynamic, an axiom that commands a more careful look at how consumer identity shapes not only brand choice, but related self-brand connections. Based on this, new research ques tions are suggested, th e answer to which can improve the accuracy in which researchers explain self-brand phenomena. For example, how does identity centrality impact brand relationships? How does identity centrality impact multiple identities simultaneously? How is centrality established? How does centrality impact consumption choices? Lastly, the identity centrality construct can serve as a meaningful starting point for the articulation of a brand-identity theory. Concepts reflective of the role of identity in brand outcomes variously labeled as self-b rand connections, consumer brand meaning, and product-image congruity represent the most studied dependent variables in the selfreferent consumption literature (Ng a nd Houston 2006; Escalas and Bettman 2005;
151 Chaplin and Roedder-John 2005). Identity centra lity related to brand interaction may hold promise for brand-related outcomes surrounding a central identity. This task is left for future research. Next prescriptive managerial guidance for effective brand management practice is offered through application of the centrality perspective. 6.3. Managerial Contributions In an effort to address the managerial imp lications of this dissertation, four central questions are considered: (1) How can manage rs determine which identity motives are worth pursuing? 2) What strategies and t actics should be employed to pursue these motives? 3) How can managers use the framew ork to attract more consumers? and (4) How can identity motives/identity centrality be evaluated and assessed? 6.3.1. Which Motives to Pursue? Identity motives determine two things: wh at consumers want to do and how much they want to do it. Brand managers would l ove to know what motiv ates individuals to buy specific goods or services. If they kne w which identity motives would lead individuals to purchase their brand, then perhaps they would develop a brand management program to satisfy relevant id entity motives among potential buyers. So why not just ask consumers about their identity motives? Motives have two key properties: direction and intensity (Bagozzi 1997). Direction is simply the valence of the motivation, being favorable or unfavorable; and intensity re presents the strength of the motive. If a particular motive is satisfied and the direction is favorab le as suggested in this dissertation this may support a specific id entity becoming central. To utilize this framework successfully, brand managers must recognize that the identity motive satisfied
152 has to be strong enough and in the right dir ection to entice individuals to purchase the offering. Successful application of the identity motive proposition wi ll further require identification of the most fertile opportuni ties for motive development. Ideally, firms should research which motives are genera lly satisfied by given product category and formulate their strategies based on those motives or a combination thereof. For instance in Essay 2, the continuity and recognition motives were more influential in an automobile brand choice context. If Toyota wanted to craft a brand campai gn around identity, they would find ways to creatively depict those motives in their mark eting communications. High potential motives include those charac terized by high social values, namely: belonging, distinctiveness, meaning, recognition, security. Identity motives beyond these mentioned can be evaluated in terms of their symbolic interaction value. In this way, consumer brand preference can be understood in the context of underlying identity motive satisfaction as mo tives are chosen with motive-congruent associations. The established link between bra nds and consumer identity motives provide opportunities in the context of marketing segmentation and targeting. Based on the framework presented in this dissertation, a heterogeneous consumer audience could be segmented into distinct clusters according to which identity motive(s) individuals are looking to satisfy by choosing a pa rticular brand. The resulti ng information could be used by marketing managers to develop appropr iate advertising campaigns and brand strategies.
153 6.3.2. How to Proceed Strategically and Tactically? Firms can take the results provided in th is dissertation to hone strategies and tactics that will evoke a consumers central identity; revealing how their brand supports the satisfaction of relevant identity motives. To demonstrate how the brand satisfies an individuals identity motives, the manager must possess an adequate repertoire of knowledge regarding the various ways in whic h motives can be pursued and how they are represented in the minds of the consumer. St rategies can then be crafted to display increased motive satisfaction and in pa rt enhance self-brand connections. Once the decision has been made to incor porate key identity mo tives into a brand strategy, the desired le vel of motive satisfaction must be determined. Since previous research suggest that motives fall into eith er an intrapersonal or interpersonal domain (Wicker, Lambert, Richardson and Kahler 1984 ), it is essential that firm develop an appropriate strategy. This means companies should consider if the motive will be satisfied on an individual or group level. T hus, the receipt of brand communications by the individual will vary across these levels in the same way that the self has been shown to vary across situations (e.g. situational self malleable self) and across social categories (e.g. social self, social identity). The mana ger may also want to devise programs to facilitate life cycle state changes by crafti ng identity-based nostalgia campaigns. This particular tactic lends itself to the satisf action of the continuity motive, in which individuals desire certain bra nds that have a level of signi ficance across an individuals identity. Beyond intrapersonal or interpersonal mo tive satisfaction, identity motives can also be incorporated at th e level of the corporation or the individual brand. The
154 preeminence of the identity motive idea suggests that any company putting its name on a line of products engages in corporate identity activities. The fact that corporate branding is where the motive process begins has not gained full consideration in the marketing literature. Moreover, the reality of todays marketplace support s the value of dealing with consumers on a more personalized corporate level. The value of a strong corporate identity strategy inclusive of a multiple identi ty paradigm lies in its ability to transcend the irritations encountered with a homogenous corporate identity stra tegy. In this sense, the corporate identity can serve as a safety net supporting multiple identity motives (on a corporate level) cultivated at both the firm and th e individual brand level. A mastery of the factors that encourag e motive satisfaction among consumers will be demanded at this stage. The barriers that normally separate large companies from their consumers must also be removed if motive satisfaction is to occur (i.e. consumer mistrust). Advertising and public relationship activities that reveal intimate knowledge of the company-as-a-person should come strongl y into play. Program execution stressing one or more of the nine motives will estab lish an identity link between the individual and the brand and in some cases the firm, fu rther supporting the psyc hological benefit of identity centrality. 6.3.3. How Can the Framework Help Firms Attract More Customers? How can firms attract more consumers using this framework? The answer to this question is based on the premise that consumers prefer brands that allow them to express their (desired or actual) identity. Firms a ttempting to connect with consumers on a deep and meaningful level will have to identify areas of similarity and synergy with their consumers. Efforts to attract more customer s are better received when aligned with the
155 brands core identity. Since brands have th e potential to characterize individuals in a split-second (Bucholz and Wordemann 2000) co nsumers are well aware of what their brand decisions tell others a bout themselves. In essence the brand becomes an identifier, which is not always separate from the functionality or quality. Consumers will selectively choose brands to send the correct message to relevant others. From the perspective of the firm, the message should be build around the satisfaction of the identity motives driving the purchase, and the identity that is central to the purchase context. By proving that the asso ciations of the brand can or will satisfy a consumers identity motives, the brand will become more appealing to consumers. Improper use of symbolic values, associations and meanings in the message will lead to the decline in brand preference leaving a seri ous impact on the brands image. Therefore market research is strongly recommended befo re an identity motive strategy is advanced. Developing a successful strategy that makes the brand (more attractive) can be achieved if the brand can demonstrate motive satisfac tion and cater to a sp ecific central identity. The identity motive(s) firms choose to pur sue should be a direct reflection of the brand core identity, or perhaps a key selli ng point of the brand. More importantly, the motive(s) must be desired by the consumer, be it a current motive or an aspired motive. Current motives are related to the actual self whereas aspired motives are related to an aspired self. From a communica tions standpoint, the brand must convey the motive that it is attempting to satisfy. It is not enough for a firm to state that the brand will satisfy particular motives, consumers must believe it. This means ma rketing communication should be genuine and based on the authenti c image of the brand. Any indication of manipulation would be a costly mistake for the firm especially across the long term
156 considering that consumers are becoming increasingly informed and demanding. The more distinctive the motive is portrayed in comparison to competing brands, the more clearly the brand can be differe ntiated from the competition. Boulding (1965) was one of the earliest re searchers to recognize the commercial importance of image. He found that individual s do not respond to reality, instead they respond to their perception of reality. In th is case, consumers respond to their desired image that the brand will help them to portr ay. Image is posited to be a combination of the brand associations and th eir perceived reality of their identity. The relationship between a consumers identity and brand identity, defined as a set of brand associations which the marketer is aiming to create and obtain (Reed 2004) is potentially where selfbrand connections began to de velop. Brand identity supports the relationship between the customer and the brand by generating a va lue proposition that includes functional, emotional, and self-expressive benefits (Aaker 1996). These self expressive benefits will attract more consumers who are looking to project a particular image through symbolic brand associations. 6.3.4. How Can Identity Motives/Identity Centrality Efforts be Evaluated and Assessed? To assess the success of the identity-mo tive based strategy there is a need for evaluation and assessment. Measures of eff ectiveness should consider two factors: (1) how well did the strategy achieve the bra nding objective? And (2) how well did the strategy contribute to attaining the overal l marketing objectives? If the strategy is successful, it should be used in future bra nding plans. Otherwise, the flaws should be corrected. A successful brand strategy should enjoy the consequences of enhanced selfbrand connections, increased brand loyalty, and increased positive brand attitudes.
157 The satisfaction of identity motives can be monitored over time to determine the overall success of the identity motive-based program. Assessment of the identity-motive strategy can be used to gauge whether or not mo tive strategies should be continued at all, and to what levels those efforts should be pursued. For example, a brand that enjoys a high level of loyalty without the benefit of identity motive-based strategy may be sustained at a lower investment than one de livering only moderate but promising quality levels with an identity motive-based strategy. 6.4. Limitations of the Research Despite the contributions mentioned above several limitations of the current research exist. While these limitations restri ct the conclusions that can be drawn here, they do not preclude the value of the ident ity centrality construct and multiple motivemultiple identities paradigm. By acknowledging the limitations, areas for improvement in future research are suggested, maximizing the research potential of identity centrality and the applicability of multiple identity motives in theory and practice. Methodologically, the dissertation employs a mixed method approach which has its advantages. However, there are a few disa dvantages to the met hods used. First, the qualitative study conducted to find evidence of id entity motives tested in this dissertation could have benefited follow-up interviewing. Ea ch participant spent a total of 45 to 90 minutes in a single meeting. Perhaps greater insight could have been gleaned from conducting follow-up interviews to confirm the motives identified so that participants could elaborate on their responses. This follo w-up did not occur in the first several interviews but after becoming acclimated w ith qualitative techniques, follow-up was done inside of the interview.
158 Conceptually, the list of identity motives is restricted to those originally identified in exploratory phase of this dissertation. Futu re researchers should seek to add to the list of identity motives tested here. It would be unwise to assume that the nine motives examined here comprise an exhaustive lis t of motivational influences on identity centrality and self-brand interactions. 6.5. Future Research Four of the most promising, research ideas are discussed here briefly. The intent is to demonstrate the potential contributory va lue of the multiple motive-multiple identity paradigm as a whole, and its ability to motivate the marketing managers agenda. The first research idea addresses the re lationship between the various motives. Although multiple motives were found to pred ict identity centrality, the relationship among the multiple motives still remains unknow n. Future research should provide an understanding of the relationships between the various motives. Specifically, future research should answer questions such as: 1) to what degree is the operation of each motive dependent upon the other motives? 2) Does a motive hierarchy exist among the motives? And 3) Is there a conflict or interference among the motives? The findings of this dissertation cannot answer these questions, as the rela tionship between the motives was not directly tested (be yond their unique contributions). A second research idea lies in the cross sectional nature of data collected. The ability of identity motives to predict identi ty centrality was inferred. The survey design did not allow the identity processes in acti on. A longitudinal study is necessary to assess identity processes across time. Thus these da ta cannot show to what extent the observed relationships were caused by processes shaping identity centrality or the meanings of the
159 identities themselves. Both may be guided by identity motives (E thier and Deaux 1994). Hence, these findings might be complemented by longitudinal research into processes shaping both structure and content of identit y. This can be achieved by assessing motive measures at two points in time and anal ysis the shift in motive importance. Comparing identity motives across different dimensions of the self is a third area ripe for future research. It may be valuable to examine in greater detail the relationship between individual and group levels of iden tity. Individuals woul d provide different aspects of their identities and sort them into group or indi vidual level categories. Once completed, motive ratings can be assessed and checked for differences between the two categories. In this dissertation, multiple id entity motives were conceptualized on the individual level. Lyons (1996) has suggested transposing these pr inciples to the group level, referring to group self-esteem group distinctiven ess, and so on. The last area of future research involves the motiva tional influence of identity centrality among reference group members. For example, brands can be used to meet self-presentation goals (e.g. meaning), serve as devices of social integration (e.g. belonging), connect consumers to the past (e .g. continuity); become symbols of personal accomplishment (e.g. recognition), provide self-e steem, allow one to differentiate oneself and by expressing individuality (e.g. distinctiveness), and may protect us from threats to identity (e.g. security). Given these multiple motives which can lead to social identity activation as well as ce ntrality, a future study that focu ses on reference group motivations should provide evidence of how the operation of multiple identity motives leads to the importance of a reference group identity.
160 6.6. A Closing Note Many conceptual and empirical tools have be en brought to bear in this dissertation. Through them, the present study has tried to illu strate identity centrality is a mechanism by which consumers manage multiple identities in the marketplace through the assignment of different levels of importance to each identity relative to various product categories. Using this perspective lens provi des a deeper understan ding of self-referent consumption and provides a more detailed understanding for marketing theory and practice. In framing identity motive-based consumption, new ideas for the study of selfbrand interactions have been advanced; yet ma ny more are left to the agenda of future researchers.
161 References Aaker, David A. (1991), Managing Brand Equity: Capitalizi ng On the Value Of a Brand Name, New York: The Free Press. Aaker, David (1996), Measuring Brand Equity Across Products and Markets, California Management Review 38(3), 102-120. Aaker, Jennifer (1999), The Malleable Self: Th e Role of Self-Expression in Persuasion, Journal of Marketing Research, 36 (February), 45-57. Aaker, Jennifer L. and Angela Y. Lee (2001), I Seek Pleasures and We Avoid Pains: The Role of Self-Regulatory Goals in Information Processing and Persuasion, Journal of Consumer Research 28 (June), 33-49. Adkins, Natalie Ross and Julie L. Ozanne (2005), The Low Literate Consumer, Journal of Consumer Research 32 (June), 93105. Agrawal, Nidhi and Durairaj Maheswaran ( 2005), The Effects of Self-Construal and Commitment on Persuasion, Journal of Consumer Research 31 (March), 841849. Ahuvia, Aaron (2005), Beyond the Extende d Self: Loved Objects and Consumers Identity Narratives, Journal of Consumer Research 32 (June), 171-184. Aiken, Leona S. and Stephen G. West (1991), Multiple Regression: Testing and Interpreting Interactions Thousand Oakes: Sage. Ajzen, Icek, and Martin Fishbein (1980), Understanding Attitudes and Predicting Social Behavior Englewood Cliff, New Jersey: Prentice-Hall. Argo, Jennifer J., Darren W. Dahl, and Rajesh V. Manchanda (2005), The Influence of a Mere Social Presence in a Retail Context, Journal of Consumer Research 32 (September), 207-212. Argo, Jennifer J., Katherine White and Darren W. Dahl (2006), Social Comparison Theory and Deception in the Interpersonal Exchange of Consumption, Journal of Consumer Research 33 (January), 99108. Armstrong, J. Scott and Terry S. Overton ( 1977), Estimating Nonresponse Bias in Mail Surveys, Journal of Marketing Research 14 (August), 396-402.
162 Arnett, Dennis B., Steve D. German, Shelby D. Hunt (2003), The Identity Salience Model of Relationship Marketing Succe ss: The Case of Nonprofit Marketing, Journal of Marketing 67 (April), 89-105. Ashforth, Blake E. and Fred Mael (1989), Social Identity Theory and the Organization, Academy of Management Review 14 (1), 20-39. Ashmore, Richard, D., Kay Deaux and T. McLaughlin-Volpe ( 2004), An Organizing Framework for Collective Identity: Articulation and Significance of Multidimensionality, Psychological Bulletin 130, 80-114. Bagozzi, Richard P. (1997), Goal-Directed Behaviors in Marketing: Cognitive and Emotional Perspectives, Psychology and Marketing 14, 539-543. Bagozzi, Richard P., Massimo Bergami, and Luigi Leone (2003), Hierarchical Representation of Motives in Goal-Setting, Journal of Applied Psychology 88 (5) 915-943. Bandura, Albert (1977), Self-Efficacy: To ward a Unifying Theory of Behavioral Change, Psychological Review 84,191-215. Baumeister, Roy F. (1991), Meanings of Life New York: Guilford Press. Baumeister, Richard F. and Mark R. Leary (1995), The Need to Belong: Desire For Interpersonal Attachments as a Fundamental Human Motivation, Psychological Bulletin (117), 497-529. Baumgartner, Hans (2002), Towar d a Personology of the Consumer, Journal of Consumer Research 29 (September), 286292. Bearden, William O. and Michael J. Et zel (1982), Reference Group Influence on Product and Brand Purchase Decisions, Journal of Consumer Research 9 (September), 183-194. Bearden, William O., David M. Hardesty a nd Randall L. Rose (2001), Consumer SelfConfidence: Refinements in Con ceptualization and Measurement, Journal of Consumer Research 28 (June), 121-134. Bearden, William, Richard G. Netemeyer and Jessie E. Teel (1989), Measurement of Consumer Susceptibility to Interpersonal Influence, Journal of Consumer Research 15 (March), 473-481. Belch, George E. and E. Laird Landon (1977), Discriminant Validity of A ProductAnchored Self-Concept Measure, Journal of Marketing Research 14, 252-256.
163 Belk, Russell W. (1988), Possessions and the Extended Self, Journal of Consumer Research 15(September), 139-168. Belk, Russell W., Kenneth D. Bahn and R obert N. Mayer (1982), Developmental Recognition of Consumption Symbolism, Journal of Consumer Research 9 (June), 4-17. Belk, Russell, Gliz Ger and Sren Askegaard (2003), The Fire of Desire: A Multi-sited Inquiry into Consumer Passion, Journal of Consumer Research 30 (December), 326-351. Belk, Russell W., Robert Mayer and Amy Dr iscoll (1984), Children's Recognition of Consumption Symbolism in Children's Products, Journal of Consumer Research 10 (March), 386-397. Berry, William Dale and Stanley Feldman (1985), Multiple Regression in Practice, Sage University Paper Series on Quantitative Applications in the Social Sciences, Series no. 07-050, Newbury Park, California: Sage. Birdwell, Al E. (1968), A Study of Influence of Image Congruence on Consumer Choice, Journal of Business (41), 76-88. Boulding, Kenneth E. (1965), The Economics of Human Conflict, in The Nature of Human Conflict ed. E.B. McNeil, Englewood Cliffs: Prentice-Hall. Boyatzis, Richard E. (1998), Transforming Qualitative Inform ation: Thematic Analysis and Code Development Thousand Oaks, CA: Sage Publications. Branscombe, Nyla R., Naomi Ellemers, Russel Spears and Berjan Doosje (1999), The Context and Content of Soci al Identity Threats, in Social Identity: Context, Commitment, Content, eds. Naomi Ellemers, Russel Spears and Berjan Doosje, Oxford: Blackwell, 35-58. Braun, Virgina and Victoria Clarke (2006), Using Thema tic Analysis in Psychology, Qualitative Research in Psychology 3, 77-101. Breakwell, Glynis M. (1983), Threatened Identities , London: Wiley. Breakwell, Glynis M. (1986), Coping With Threatened Identities London: Methuen. Breakwell, Glynis (1988), S trategies Adopted When Identity Is Threatened, International Review of Social Psychology (2),189-203. Breakwell, Glynis M. (1993), Integrating Para digms, Methodological Implications, in Empirical Approaches to Social Representations ed. Glynis M. Breakwell and David V. Canter, New York: Oxford University Press, 180204.
164 Brendl, C. Miguel, Amitava Chattopadhyay, Br ett W. Pelham, Mauricio Carvallo (2005), Name Letter Branding: Valence Transf ers When Product Specific Needs Are Active, Journal of Consumer Research 32 (December), 405-415. Brewer, Marilynn B. (1991), The Social Self : On Being the Same and Different At the Same Time, Personality and Social Psychology Bulletin 17, 475-482. Brewer, Marilynn B. and Wendi Gardner (1996), Who Is This We? Levels of Collective Identity and Self Representations, Journal of Personality and Social Psychology 71 (July), 83-93. Buchholz, Andreas and Wolfram Wordemann (2000), What Makes Winning Brands Different? The Hidden Method Behind the Worlds Most Successful Brands England: John Wiley and Sons. Cattell, Raymond B. (1965), The Scientific Analysis of Personality Baltimore, Maryland: Penguin. Chandran, Sucharita and Vicki G. Morwitz ( 2005), Effect of Participative Pricing on Consumers' Cognitions and Actions: A Goal Theoretic Perspective, Journal of Consumer Research 32 (September), 49-59. Chaplin, Lan Nguyen and Deborah Roedder-John (2005), The Emergence of Self-Brand Connections in Children and Adolescents, Journal of Consumer Research 32 (June), 119-125. Cocanougher, A. Benton and Grady D. Bruce, (1971), Socially Distant Reference Groups and Consumer Aspirations, Journal of Marketing Research 8(August), 379-381. Codol, Jean-Paul (1981), A Cognitive Appr oach to the Feeling of Identity, Social Science Information, 20, 111-136. Cohen, Jacob and Patricia Cohen (1983), Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences 2nd ed. Hillsdale, New Jersey: L. Erlbaum Associates Cotte, June and Stacy L. Wood (2004), Families and Innovative Behavior: A Triad Study of Influence, Journal of Consumer Research 31 (June), 78-86. Coulter, Robin A., Linda L. Price, and Lawr ence Feick, (2003), Rethinking the Origins of Involvement and Brand Commitment: Insights from Postsocialist Central Europe, Journal of Consumer Research 30 (March), 151169.
165 Crawford, Matthew (2004), The Renegotiation of Social Id entities in Response to a Threat to Self-Evaluation Maintenance, Journal of Experimental Social Psychology 1-11. Dahl, Darren W., Rajesh V. Manchanda and Jennifer J. Argo (2001), Embarrassment in Consumer Purchase: The Roles of Social Presence and Purchase Familiarity, Journal of Consumer Research 28 (September), 473-481. Daniels, Matthew (2001), Maslows's Concept of Self-actualization, Retrieved February 2004, from http://www.mdani.demon.co.uk/archive/MDMaslow.htm Dauenbeimer, Dirk G., Dagmar Stablberg, Sa ndra Spreemann and Constantine Sedikides (2002), Self-Enhancement, Self-Verifica tion, or Self Assessment: The Intricate Role of Trait Modifiability in the Self-Evaluation Process, International Review of Social Psychology (15) 3-4, 89-111. Deci, Edward L. and Richard M. Ryan (2000), The "What" and "Why" of Goal Pursuits: Human Needs and the Self-determination of Behavior, Psychological Inquiry (11), 227-268. Deshpand,Rohit and Naveen Donthu (1986), The Intensity of Ethnic Affiliation: A Study of the Sociology of Hispanic Consumption, Journal of Consumer Research ,13 (March), 214-220. Deutsch, Claudia (2006), Marketers Look Past Jelly Stains And See a Mom Who Has Needs, New York Times, September 15, 2006, Business/Financial Desk, Late Edition Final, Section C, Page 5, Column 1. Dey, Ian (1993), Qualitative Data Analysis: A User Friendly Guide for Social Scientists London: Routledge. Dimofte, Claudiu V., Mark R. Forehand and Rohit Deshpand (2003), Ad schema Incongruity as Elicitor of Ethnic Self-awareness and Differential Ad Response, Journal of Advertising 32 (4), 7-17. Dolich, Ira J. (1969), Congruence Relations hips Between Self-images and Product Brands, Journal of Marketing Research 11, 80-84. Duhachek, Adam (2005), A Multidimensi onal Hierarchical Model of Coping: Examining Cognitive and Emotional Antecedents and Consequences, Journal of Consumer Research 32 (January), 41-54. Either, Kathleen and Kay Deaux (1994), Ne gotiating Social Identity When Contexts Change: Maintaining Identifica tion and Responding to Threat, Journal of Personality and Social Psychology 67, 243251.
166 Ellemers, Naomi, Russell Spears and Berjan D oosje (2002), Self and Social Identity, Annual Review of Psychology 53, 161-186. Elliott, Richard and Kritsadarat Wattanasuwan (1998), Consumption and the Symbolic Project of the Self, European Advances in Consumer Research 3, 17-20. Emmons, Robert A. (1992), Abstract Versus Concrete Goals: Personal Striving Level, Physical Illness, and Psychological Well-being, Journal of Personality and Social Psychology 62, 292-300. Erdem, Tlin and Joffre Swait (2004), Brand Credibility and its Role in Brand Choice and Consideration, Journal of Consumer Research 31 (January), 191-199. Ericksen, Mary K. and M. Joseph Sir gy (1992), Employed Females Clothing Preference, Self-Image Congruen ce, and Career Anchorage, Journal of Applied Social Psychology 22 (5), 408-422. Escalas, Jennifer (2003), Narrative versus An alytical Self-Referencing and Persuasion, Working paper, Department of Mark eting, Eller College of Business and Administration, University of Arizona, Tucson. Escalas, Jennifer E. (2004), Narrative Pro cessing: Building Consumer Connections to Brands, Journal of Consumer Psychology 14 (1-2), 168-179. Escalas, Jennifer E. and James R. Bettman (2000), Using Narratives to Discern SelfIdentity Related Consumer Goals and Motivations, in The Why of Consumption: Perspectives on Consumer Motives, Goals, and Desires, eds. Ratti Ratneshwar, David Mick, and Cynthia Huffman, New York, NY: Routledge Press, 237-258. Escalas, Jennifer E. and James R. Bettman (2003), You Are What They Eat: The Influence of Reference Groups on C onsumer Connections to Brands, Journal of Consumer Psychology 13 (3), 339-348. Escalas, Jennifer E. and James R. Bettman (2005), Self-Construal, Reference Groups, and Brand Meaning, Journal of Consumer Research 32 (December), 378-389. Evans, Franklin B. (1968), The Ford ve rsus Chevrolet: Park Forest Revisited, Journal of Business 41(October), 445-459. Firat, A. Fuat and Clifford J. Shultz II (1997), From Segmentation to Fragmentation: Markets and Marketing Strategy in the Postmodern Era, European Journal of Marketing 31 (3-4), 183-207. Firat, A.Fuat and Clifford J. Shultz II (2001) Preliminary Metric Investigations into the Nature of the Postmodern Consumer, Marketing Letters 12 (2), 189-203.
167 Firat A. Fuat and Alladii Venkatesh (1995) Liberatory Postmodernism and the Reenchantment of Consumption, Journal of Consumer Research 22 (December), 239 267. Folkes, Valerie S. and Tina Kiesler (1991), Social Cognition: Consumers' Inferences About the Self and Others, in Handbook of Consumer Behavior eds. Thomas S. Robertson and Harold H. Kassarj ian, New Jersey: Prentice-Hall Foote, Nelson N. (1951), Identification as the Basis for a Theory of Motivation, American Sociological Review 26, 14-21. Forehand, Mark, Rohit Desphand and Americ us Reed II (2002), Identity Salience and the Influence of Differential Activation of the Social Self Schema on Advertising Response, Journal of Applied Psychology 87(6) 1086-1099. Fournier, Susan (1998), Consumers and Thei r Brands: Developing Relationship Theory in Consumer Research, Journal of Consumer Research 24 (March), 343-373. Gangestad, Steven. W. and Mark Snyder (2000), Self-Monitoring: Appraisal and Reappraisal, Psychological Bulletin 126, 530-555. Gecas, Viktor (1982), The Self-concept, Annual Review of Sociology 8, 133. Gergen, Kenneth J. (1991 ), The Saturated Self: Dilemmas of Identity in Contemporary Life New York: Basic Books. Gould, Steven .J. (1991), The Self-Manipul ation of My Pervasive, Perceived Vital Energy through Product Use: An Intr ospective-Praxis Perspective, Journal of Consumer Research 18 (September), 194-207. Graeff, Timothy R. (1996), Using Promoti onal Messages to Mana ge the Effects of Brand and Self-image on Brand Evaluations, Journal of Consumer Marketing 13(3), 4-18. Grub, Edward and Harrison Grathwohl (1967), Consumer Self-Concept, Symbolism and Market Behavior: A Theoretical Approach, Journal of Marketing 31 (October), 22-27. Grub, Edward L. and Greg Hupp (1968), Perce ption of Self, Gene ralizes Stereotypes and Brand Selection, Journal of Marketing Research 5 (February), 38-63. Grub, Edward L. and Barbara L. Stern (1971) Self-concept and Si gnificant Others, Journal of Marketing Research, (8) 382-385.
168 Hamilton, Rebecca W. and Gabriel J. Biehal (2005), Achieving Your Goals or Protecting Their Future? The Effects of Self-View on Goals and Choices, Journal of Consumer Research 32 (September), 277-283. Helgeson, James G. and Magne Supphellen ( 2004), A Conceptual and Measurement Comparison of Self-Congruity and Brand Personality, International Journal of Market Research 46 (2), 205-233. Higgins, E. Tory (1987), Self-discrepancy: A theory Relating Self and Affect, Psychological Review 94, 319-340. Higgins, E. Tory, Ronald N. Bond, Ruth Klein and Timothy Strauman (1986), Selfdiscrepancies and Emotional Vulnerabil ity: How Magnitude, Accessibility, and Type of Discrepancy Influence Affect, Journal of Personality and Social Psychology 51, 5-15. Hirschman, Elizabeth C. and Morris B. Holbrook (1982), Hedonic Consumption: Emerging Concepts, Methods and Propositions, Journal of Marketing 46 (Summer), 92-101. Hogg, Michael A. (2000), Subjective Uncertainty Reduction Through Selfcategorization: A Motivational Th eory of Social Identity, European Review of Social Psychology 11(1), 223-255. Hogg, Michael A. (2001), A Social Identity Theory of Leadership, Personality and Social Psychology Review 5 (2001), 184200. Hogg, Michael A. and Dominic Abrams (1988), Social Identifications: A Social Psychology Of Intergroup Relations And Group Processes London: Routledge. Hogg, Michael and M.H. Savolainen (1998), S ymbolic consumption and the situational self, European Advances in Consumer Research 3, 11-16. Holbrook, Morris B. and Elizabeth C. Hirsch man (1981), Symbolic Consumer Behavior: An Introduction, in Symbolic Consumer Behavior eds. Elizabeth C. Hirschman and Morris B. Holbrook, Ann Arbor: A ssociation for Consumer Research. Hong, Jae W. and George M. Zinkhan (1995), Self-Concept and Advertising Effectiveness: The Influence of Cong ruency, Conspicuousness, and Response Mode, Psychology and Marketing 12, 5377. Hoyle, Rick (1999), Selfhood: Identity, Esteem, Regulation Boulder, Colorado: Westview.
169 Huffman, Cynthia, S. Ratneshwar, and Da vid Glen Mick (2000), Consumer Goal Structures and Goal Determination Pro cesses: An Integrative Framework, in The Why of Consumption: Contemporary Persp ectives on Consumer Motives, Goals, and Desires eds. S. Ratneshwar, David Glen Mick, and Cynthia Huffman, London: Routledge, 9 35. Hyman, Herbert H. (1942), Psychology of Status, Archives of Psychology 269, 5-28. James, William (1890), Principles of Psychology, New York, NY: Henry Holt and Company. James, William, (1892), Psychology: The Briefer Course New York: Henry Holt and Company. Jones, Edward E. (1990), Interpersonal Perception New York: W.H. Freeman. Kates, Steven M. (2004), The Dynamics of Brand Legitimacy: An Interpretive Study in the Gay Mens Community, Journal of Consumer Research 31 (September), 455-464. Keller, Kevin Lane (1993), Conceptua lizing, Measuring, and Managing CustomerBased Brand Equity, Journal of Marketing 57 (1), 1-22. Keller, Punam (2006), Regulatory Focus and Efficacy of Health Me ssages, Journal of Consumer Research, 33(June), 109-114. Keller, Punam A., Issac M. Lipkus, Barbara K. Rimer (2002), Depressive Realism and Health Risk Accuracy: The Negativ e Consequences of Positive Mood, Journal of Consumer Research 29 (1), 57-69. Kelley, Harold H. (1947), The Two Functions of Reference Groups, In Readings in social psychology, eds. Guy E. Swanson, Theodore M. Newcomb and E. L. Hartley, New York: Holt, 410-414. King, Nigel (2004), Using Templates in th e Thematic Analysis of Text, in Essential Guide to Qualitative Methods in Organizational Research eds. C. Cassell and G. Symon. London: Sage. Kleine, Robert E., III, Susan Schultz Klei ne and Jerome B. Kernan (1993), Mundane Consumption and the Self: A SocialIdentity Perspective, Journal of Consumer Psychology 2 (3), 209-235. Krishnamurthy, Parthasarathy, and Mita Sujan (1999), Retrospection versus Anticipation: The Role of the Ad Under Retrospective and Anticipatory SelfReferencing, Journal of Consumer Research 26 (June), 55-69.
170 Kuhn, Manford H. and Thomas McPartland (195 4), An Empirical Inve stigation of SelfAttitude, American Sociological Review 19, 68-77 Labov, William (1972), Language in the Inner City: Studies in the Black English Vernacular, University of Pennsylva nia Press: Philadelphia. Landon, E. Lair, Jr. (1974), Self Concept, Id eal Self Concept, and Consumer Purchase Intentions, Journal of Consumer Research 1(September), 44-51. LaTour, Michael, Kathryn Braun-LaTour, George M. Zinkhan (2007), Using Childhood Memories to Gain Insigh t into Brand Meaning, Journal of Marketing 71(2), 4560. Laverie, Debra A., Robert E. Kleine, III and Susan Schultz Kleine (2002), A Reexamination and Extension of Kleine, Klei ne, and Kernan's Social Identity Model of Mundane Consumption: The Mediati ng Role of the Appraisal Process, Journal of Consumer Research 28 (March), 659-699. Leary, Michael R. and Roy F. Baumeister ( 2000), The Nature and Function of SelfEsteem: Sociometer theory, in Advances in Experimental Social Psychology ed. M.P. Zanna, San Diego, CA: Academic Press, 1-62. Leonard, Nancy, Laura Beauvais and Rich ard Scholl (1995), A Self-Concept-Based Model on Work Motivation, Paper pres ented at the Annual Meeting of the Academy of Management, August. Retrieved: December 2006, from http://www.cba.uri.edu/Scholl/Pap ers/Self_Concept_Motivation.HTM Levy, Sidney J. (1959), Symbols for Sale, Harvard Business Review 37 (July-August), 117124. Levy, Sidney J. (1986), Meanings in Advertising Stimuli, in Advertising and Consumer Psychology, (eds.) J. Olson and K. Sentis, Westport, CT: Praeger, 214226. Lindstrom, Martin (2005), Brand Sense. New York: Free Press. Linville, Patricia W. and Donald E. Carlst on (1994), Social Cogniti on of the Self, in Social Cognition: Its Impact on Social Psychology ed.Patricia G. Devine, David L Hamilton, and Thomas M. Ostrom, Academic, 143-93. Lowrey, Tina M., Cele C. Otnes and Julie A. Ruth (2004), Social Influences on Dyadic Giving over Time: A Taxonomy from the Givers Perspective, Journal of Consumer Research 30(December), 547-558.
171 Lyons, Evanthia (1996), Coping wi th Social Change: Process of Social Memory in the Reconstruction of Identities, in Changing European identities: Social Psychological Analyses of Social Change (Eds.), Glynis M. Breakwell and Evanthia Lyons, Oxford, England: Butterworth-Heinemann, 3139. Mandel, Naomi (2003), Shifting Selves a nd Decision Making: The Effects of SelfConstrual Priming on Consumer Risk-Taking, Journal of Consumer Research 30(January), 30-40. Maner, Jon K. Douglas T. Kenrick, D. Vaughn Becker, Theresa E. Robertson, Brian Hofer, Steven L. Neuberg, Andrew W. Delton, Jonathan Butner and Mark Schaller (2005), Functional Projection: How Fundamental So cial Motives Can Bias Interpersonal Perception, Journal of Personality and Social Psychology, 88, 6378. Markus, Hazel and Shinobu Kitayama (1991), C ulture and the Self: Implications for Cognition, Emotion, and Motivation, Psychological Review 98 (April), 224-253. Markus, Hazel and Paula Nurius (1986), Possible selves, American Psychologist 41, 954-969. Markus, Hazel and Ziva Kunda (1986), Stabil ity and Malleability of the Self-Concept, Journal of Personality and Social Psychology 51, 858-866. Markus, Hazel and Elisa Wurf (1987), The Dynamic Self Concept: A Social Psychological Perspective, Annual Review of Psychology 38, 299-333. Marques, Jos M. and Vincent Y. Yzerbyt (1998), The Black Sheep Effect: Judgmental Extremity Towards Ingroup Members In Inter-And Intra-Group Situations, European Journal of Social Psychology 18 (3), 287-292. Maslow, Abraham H. (1954), Motivation and Personality New York: Harper Collins. Martire, Lynn M., Mary Ann Stephens and Aloen Townsend (2000), Centrality of Women's Multiple Roles: Beneficial and Detrimental Consequences for Psychological Well-Being, Psychology and Aging 15, 148-156. McClelland, David (1951), Personality New York: Holt, Rinehart and Winston. McCracken, Grant (1986), Culture and Cons umption: A Theoretical Account of the Structure and Movement of the Cultu ral Meaning of Consumer Goods, Journal of Consumer Research 13, 7184. McCracken, Grant (1988), Culture and Consumption Bloomington, IN: Indiana University Press.
172 McQuarrie, Edward F. (1988), An Alternative to Purchase Intentions: The Role of Prior Behavior in Consumer Expenditure on Computers, Journal of the Market Research Society 30 (October), 407-437. Meyers-Levy Joan and Laura Peracchio (1996) Moderators of the Impact of SelfReference on Persuasion, Journal of Consumer Research 22(March) 408-423. Mick, David Glen (1996), Are Studies of Dark Side Variables Confounded by Socially Desirable Responding? The Case of Materialism, Journal of Consumer Research September, 23 (2), 106-121. Minton, Carmelle and Kay Pasley (1996), Fathe rs' Parenting Role Identity and Father Involvement: A Comparison of Nondivorced and Divorced Nonresident Fathers, Journal of Family Issues 17(1), 26-45. Moorman, Christine, Kristin Diehl, Da vid Brinberg and Blair Kidwell (2004), Subjective Knowledge, Search Lo cation, and Consumer Choice, Journal of Consumer Research 31 (December), 673-680. Morgan, Amy J. (1993), The Evolving Self in Consumer Behavior: Exploring Possible Selves, in Advances in Consumer Research, eds. L. McAlister and M. Rothschild, Provo, UT: Association for Consumer Research, (20) 429-432. Moschis, George P. (1985), The Role of Family Communication in Consumer Socialization of Child ren and Adolescents, Journal of Consumer Research 11 (March), 898-913. Muiz, Albert M., Jr. and Thomas C. OGuinn (2001), Brand Community, Journal of Consumer Research 27 (March), 412-432. Muiz, Albert M., Jr. and Hope Jensen Schau (2005), Religios ity in the Abandoned Apple Newton Brand Community, Journal of Consumer Research 31 (March), 737-747. Murray, Jeff B. (2002), The Politics of Consumption: A Re-Inquiry on Thompson and Haytko's (1997) Speaking of fashion, Journal of Consumer Research 29 (December), 427-440. Netemeyer, Richard G., William O. Bearden, Subhash Sharma (2003), Scaling Procedures: Issues and Applications London: Sage Publications. Ng, Sokling and Michael J. Houston (2006), E xemplars or Beliefs? The Impact of SelfView on the Nature and Relative In fluence of Brand Associations, Journal of Consumer Research 32 (March), 519-529.
173 Oakes, Penelope (1987), The Salience of Social Categories: A Self-Categorization Theory, in Rediscovering the social group. A Self-Categorization Theory eds. John Turner, Michael Hogg, Penelope Oake s, Stephen D. Reicher and M. S. Wetherell, Oxford: Basil Blackwell, 117 141. Osborne, Jason and Elaine Waters (2002), Four Assumptions of Multiple Regression That Researchers Should Always Test, Practical Assessment Research and Evaluation 8(2). Pace, Robert C. (1939), Factors Influenc ing Questionnaire Returns from Former University Students, Journal of Applied Psychology 23 (June) 388-397. Pawle, John and Peter Cooper (2006), Measuring Emotion Lovemarks, The Future Beyond Brands, Journal of Advertising Research 46 (1), 38-48. Rane, Thomas R. and Brent McBride ( 2000), Identity Theory As A Guide To Understanding Fathers' Involvement With Their Children, Journal of Family Issues 21, 347-366. Reed II, Americus (2004), Activating the Se lf-Importance of Consumer Selves: Exploring Identity Salien ce Effects on Judgments, Journal of Consumer Research 31 (2), 286-295. Rogers, Carl R. (1959), A Theory of Ther apy, Personality, and In terpersonal RelationShips, As Developed in the C lient-Centered Framework, in A Study of a Science: Study 1. Conceptual and Systematic : Formulations of the Person and the Social Context ed. S. Koch, New York: McGraw Hill, 184-256. Rook, Dennis (1987), The Buying Impulse, Journal of Consumer Research 14, 189199. Rook, Dennis (2001), Typology of Projective Techniques, in Advances in Consumer Research eds. Mary C. Gilly and Joan Myers-Levy, Valdosta: Association for Consumer Research: 28, 253. Rose, Randall and Stacy L. Wood (2005), Par adox and the Search for Authenticity through Reality Television, Journal of Consumer Research 32 (September), 284296. Rosenberg, Morris (1979), Conceiving the Self New York: Basic Books. Ross, Ivan (1971), Self-con cept and Brand Preferences, Journal of Business of the University of Chicago (44), 38-50. Rowan, John and Mick Cooper (Eds.) (1999), The Plural Self: Multiplicity in Everyday Life London: Sage.
174 Schau, Hope Jensen and Mary C. Gill y (2003), We Are What We Post? The Presentation of Self in Personal Webspace, Journal of Consumer Research 30 (December), 385-404. Schenk, Carolyn Turner and Rebecca Holman (1979), A Sociological Approach to Brand Choice: The Concept of Situational Self Image, Advances in Consumer Research 7, 610-614. Schlenker, Barry R. (1985) Ident ity and Self-identification, in The Self and Social Life, ed. B. R. Schlenker New York: McGraw Hill, 15-99. Schouten. John W. (1991), Selves in Transi tion: Symbolic Consumption in Personal Rites of Passage and Identity Reconstruction, Journal of Consumer Research 17(March), 412-425. Sedikides, Constantine and Michael. J. Str ube (1997), Self-Evaluation: To Thine Own Self Be Good, To Thine Own Self Be Sure, To Thine Own Self Be True, And To Thine Own Self Be Better, Advances in Experimental Social Psychology, 29, 209-269. Settles, Isis H. (2004), When Multiple Identi ties Interfere: The Role of Identity Centrality, Personality and Social Psychology Bulletin 30 (4), 487-500. Shavitt, Sharon (1990), The Role of Attit ude Objects in Attitude Functions, Journal of Experimental Social Psychology, 26,124-148. Simonson, Itamar and Steven M. Nowlis (2000) The Role of Explanations and Need For Uniqueness In Consumer Decision Making: Unconventional Choices Based On Reasons, Journal of Consumer Research, 27 (January), 49-68. Singelis, Theodore M. (1994), The Measurem ent of Independent and Interdependent Self-Construals, Personality and Social Psychology Bulletin 20 (October), 580591. Sirgy, M. Joseph (1981), Introducing a Se lf -theory to Consumer Personality Research, Catalog of Selected Documents in Psychology 11 (May), 33, Ms. 225. Sirgy, M. Joseph (1982), Self-Concept in C onsumer Behavior: A Critical Review, Journal of Consumer Research 9 (December), 287-300. Sirgy, M. Joseph (1983), Social Cognition and Consumer Behavior New York: Praeger Publishers. Sirgy, M. Joseph (1985), Using Self-Congruity and Ideal Congruity to Predict Purchase Motivation, Journal of Consumer Research 9 (December), 287-300.
175 Sirgy, M. Joseph, Dhruv Grewal, Tamera F. Mangleburg, Jae-Ok Park, Kye-Song Chon, C. B. Claiborne, J. S. Johar, and Harold Berkman ( 1997), Assessing the Predictive Validity of Two Methods of Measuring Self-Image Congruence, Journal of the Academy of Marketing Science 25(3), 229 241. Smeesters, Dirk and Naomi Mandel (2006), P ositive and Negative Media Image Effects on the Self, Journal of Consumer Research 32 (March), 576-582. Snyder, Mark (1974), Selfmonitoring of Expressive Behavior, Journal of Personality and Social Psychology 30(4), 526537. Snyder, C. R. and Howard L. Fromkin (1980), Uniqueness: The Human Pursuit of Difference New York: Plenum. Solomon, Michael R. (1983), The Role of Products as Social Stimuli: A Symbolic Interactionism Perspective, Journal of Consumer Research 10 (September), 319-329. Spangenberg, Eric R. and David E. Sprott ( 2006), Self-Monitoring a nd Susceptibility to the Influence of Self-Prophecy, Journal of Consumer Research (March), 550 556. Spradley, James P. (1979), The ethnographic interview, New York: Holt, Rinehart and Winston. Steele, Claude M. (1988), The Psychology of Self-affirmation: Sustaining the Integrity of the Self, in Advances in experimental social psychology, ed. L. Berkowitz New York: Academic Press, 21, 261-302. Stets, Jan E. and Peter J. Burke (2000), I dentity Theory and Soci al Identity Theory, Social Psychology Quarterly 63, 224237. Stryker, Sheldon (1980), Symbolic Interactionism: A Social Structural Version Menlo Park, CA: Benjamin/Cummings. Stryker, Sheldon and Richard T. Serpe ( 1994), Identity Salience and Psychological Centrality: Equivalent, Overlapping or Complementary concepts?, Social Psychology Quarterly 57, 1635. Swann, William B., Jr. (1983), Self-verificatio n: Bringing Social Reality Into Harmony With The Self, in Social Psychological Perspectives on the Self eds. J. Suls and A. G. Greenwald, Hillsdale, NJ: Lawrence Erlbaum Associates, 2, 33 66. Swann, William B., Jr. (1987), Identity Negotiation: Where Two Roads Meet, Journal of Personality and Social Psychology 53, 1038-1051.
176 Swann, William B., John J. Griffin, Jr., Steven Predmore and Bebe Gaines (1987), The Cognitive-Affective Crossfire: When Self-consistency Confronts Selfenhancement, Journal of Personality and Social Psychology 52, 881-889. Tajfel, Henri (1974), Social iden tity and Intergroup Behavior, Social Science Information 13, 65-93. Tajfel, Henri and John C. Turner (1979), An Integrative Theory of Social Conflict, in The social psychology of intergroup relations eds. W. Austin and S. Worchel, Monterey, California: Brooks/Cole. Tajfel, Henri and John C. Turner (1985), T he Social Identity Theory of Intergroup Behavior, in Psychology of Intergroup Relations eds. S. Worchel and W.G. Austin, Chicago: Nelson-Hail, pp. 724. Tesser, Abraham (1988), Towards a Self-e valuation Maintenance Model of Social Behavior, in Advances in experimental social psychology ed. L. Berkowitz, 21, 181227, San Diego: Academic Press. Thoits, Peggy A. (1986), Multiple Identitie s: Examining Gender And Marital Status Differences In Distress, American Sociological Review 51, 259-272. Tian, Kelly and Russell W. Belk (2005), Extended Self and Possessions in the Workplace, Journal of Consumer Research 32 (September), 297-310. Tepper-Tian, Kelly, William O. Bearden, and Gary L. Hunter (2001), Consumers Need for Uniqueness: Scale Development and Validation, Journal of Consumer Research 28 (June), 50-66. Thompson, Craig J. (1997), Interpreting C onsumers: A Hermeneutical Framework for Deriving Marketing Insights from the Text s of Consumers' Consumption Stories, Journal of Marketing Research 34 (November), 438-455. Tolman, Edward C. (1943), Iden tification and Post-War World, Journal of Abnormal and Social Psychology 38, 141148. Turner, John C. (1985), Social Categoriza tion and the Self-con cept: A Social-cognitive Theory of Group Behavior, in Advances in group processes: Theory and research ed. E.J. Lawler, Vol. 2, 47-74, Greenwich, CT: JAI. Turner, John C., Michael A. Hogg, Penelope Oakes, Stephen Reicher and Margaret Wetherell (1987), Rediscovering the Social Group: A self-categorization theory Oxford: Blackwell.
177 stner, Tuba and Douglas B. Holt (2007), Dominated Consumer Acculturation: The Social Construction of Poor Migrant Wo mens Consumer Identity Projects in a Turkish Squatter Neighborhood, Journal of Consumer Research 31(June), 4156. Vignoles, Vivian L., Xenia Chryssochoou a nd Glynis M. Breakwell (2000), The Distinctiveness Principle: Identity, Meaning and the Bounds of Cultural Relativity, Personality and Social Psychology Review 4, 337-354. Vignoles, Vivian L., Xenia Chryssochoou and Glynis M. Breakwell (2002), Sources of Distinctiveness: Position, Difference a nd Separateness in the Identities of Anglican Parish Priests, European Journal of Social Psychology 32(6), 761-780. Vignoles, Vivian L., Camillo Regalia, Claudi a Manzi, Jen Golledge, and Eugenia Scabini (2006), Beyond Self-Esteem: Influence of Multiple Motives on Identity Construction, Journal of Personality and Social Psychology 90(2), 308-333. Wicker, Frank W., Frank B. Lambert, Fra nk C. Richardson and Jo seph Kahler (1984), Categorical Goal Hierarchies and Classification of Human Motives, Journal of Personality 52, 285-305. Wicklund, Robert A. and Peter M. Gollwitzer (1982), Symbolic self-completion Hillsdale, NJ: Erlbaum. Woods, Richard (2004), Exploring the Emotional Territory for Brands, Journal of Consumer Behavior 3 (September), 388-403. Wooten, David B. and Americus Reed II ( 2004), Playing it Safe: Susceptibility to Normative Influence and Prot ective Self-Presentation, Journal of Consumer Research 31(September) 551-556. www.abercrombie.com\ourbrands .html. Accessed on 6/25/07 Zinkhan, George M. and Jae W. Hong (1991), Self-concept and Advertising Effectiveness: A Conceptual Model of Congruency, Conspicuousness, and Response Mode, In Advances in Consumer Research eds. R. H. Holman and M. R. Solomon, 18:348-354. Prove: UT: A ssociation for Consumer Research.
179 Appendix A: Interview Questions Brand Preferences Can you name your favorite brands ? (in any particular category) What are your thoughts on others who purchase things that are really expensive? Name a brand that you would absolutely not wear. What does a particular brand say about you? When you think of Brand X, what comes to mind? What does Brand X mean to you? What does the (specific feature) of Brand X mean to you? Doe Brand X help communicate your identity to others (whether you know them or not)? Does Brand X help you become the person you want to be? Consumer Identity Can you list four identities that you currently hold in life right now? How much do you think about any of those identities when you shop? Which identity do you think is most important for you right now? Can you relate your identity to the actual brand? Are you the typical user of Brand X? If not, how would you describe the typical user of Brand X? Do others in your reference group also consume Brand X? Why?
180 Appendix B Coded Transcriptions Data Extracted Coded for: But if they were really cute, I would never tell any of my friends, and I would never let them see the inside label cause it will say Payless. And if we had to go somewhere to take off our shoes, I wouldnt take them off, Id be too embarrassed. I cant have my friends thinking I shop at Payless. Brand prestige Conspicuous brand use Reference groups Belonging motive Security motive For where I am in my life right now, those cars are more reflective of the fact that I have reached a certain level. So I wouldnt buy an Acura because its looks just like a Toyota and everybody has a Toyota. I wouldnt buy a Cadillac, cause that is my dads car. I wouldnt by an Infiniti or Jaguar, because Jaguars are an old retired mans car. Its not a girl car; its a guy car, an old man car. And I am not the SUV type, even though some of them look nice, I am not the truck type. Life cycle stage, brand congruency Distinctiveness motive Brand user imagery Family brand associations Self-consistency motive Identity threat Brand-identity congruency Because Im not. Its not that I dont want to be married with children. But Im not and I dont want to be seen that way. It says wife, kids, cat, dog, and house with the white picket fence. Its not me as of yet. They make the big bodied cars for those kinds of women, but it doesnt say me. People are often influenced by their peers, but I am not using my peers to make a decision, I dont want to be like them. Future self Self-concept perceptions Brand-image congruity Self-consistency motive Brand associations Reference group Palmers Cocoa butter, I used it as a moisturizer, and have used it ever since. With the impending stretch marks, I have been putting it on my belly twice a day. And when she (unborn daughter) is old enough, Ill use it on her. Lets see what else. Oh yes, barrettes and beads in my hair, my mother would braid our hair and make it pretty with the beads and barrettes. I cant recall any brands of them, because I was so little, but I want to do her (unborn daughter) hair in the same way. I dont really know how to braid but I will learn. I want her to experience a lot of the same things I did when I was a little girl (smiling). Continuity motive Hair braiding Childhood memories Shared experiences Well like I said I like to shop based on my smarts, and if I am going to invest 30 grand into a car, I want it to last at least until my retirement kicks in (laughing). Volvos are good cars. They last. My station wagon, I have had that car for almost 12 years now. And it runs pretty well. With the exception of a few minor repairs, I havent really had any major problems. Self-efficacy Brand loyalty Brand knowledge
181 Appendix B (Continued) Being the responsible father, I wanted Sascha (his daughter) to have a car that was going to last her a while, at least until she was able to put a dent into her student loans. She would have a reliable car, and wouldnt have to worry about buying a new one. Well not at least until she was in a position to afford a new one. If she is a smart girl like her father, shed buy a Volvo, when that time came around (smiling). My wife, well she runs a Jewelry store, and she is price conscious like me. But she wanted a nice car without us both having to spend a lot of money. So the solution to that is . I kept the old station wagon, and she drives the newer sedan. We are both sold on the quality of Volvos. I dont need to convince her much. Family purchasing decisions Purchase motivation Brand motivation Wearing what you like, not thinking about what other people are going to judge what you are wearing, and I think someone who has self-esteem doesnt have to wear something really revealing, promote a bad message. Something that maybe it doesnt look good to you, but if they like it (meaning the person who is wearing it) then they are happy wearing it. Like a shirt from Target. Self-esteem Self-consistency So you have people that dr ess very punk rockish and they are trying to be different (they wear black things and wear chains) . gothic. But then there is a whole group of them, so they are not different. For me it is more of the friends that I come into contact with on a daily basis, Id rather be distinctive among my friends versus people I didnt know. .Some people have to buy a brand and it has to have the name all over it, like BEBE in big bold letters, I cant stand that. Like the Louis Vuitton bags, the same thing, I want high fashion, but not like what everyone else has. I dont want to go around looking like them. Self-esteem Self-consistency Distinctiveness Fashion Reference groups Palmers Cocoa Butter, I used it as a moisturizer, and have used it ever since. With the impending stretch marks, I have been putting it on my belly twice a day. And when she is old enough, Ill use it on her (unborn daughter). Lets see what else. Oh yes, barrettes and beads in my hair, my mother would braid our hair and make it pretty with the beads and barrettes. I cant recall any brands of them, because I was so little, but I want to do her (unborn daughter) hair the same way. I dont really know how to braid but I will learn. I want her to experience a lot of the same things I did when I was a little girl (smiling). Continuity Family influences Childhood identity
182 Appendix B (Continued) Being the responsible father, I wanted Sascha (his daughter) to have a car that was going to last her a while, at least until she was able to put a dent into her student loans. She would have a reliable car, and wouldnt have to worry about buying a new one. Well not at least until she was in a position to afford a new one. If she is a smart girl like her father, shed buy a Volvo, when that time came around (smiling). My wife, well she runs a Jewelry store, and she is price conscious like me. But she wanted a nice car without us both having to spend a lot of money. So the solution to that is . I kept the old station wagon, and she drives the newer sedan. We are both sold on the quality of Volvos. I dont need to convince her much. Self-efficacy Family influences I associate the F150 with friends, because with friends we think of intimacy. With the regular cab, not the extended cab, but the regular cab, you can pretty much fit one person, maybe two. They have close seats which provides for a lot of intimacy between people. And whats one thing that friends ask people to do, and that is to go out. You know when you got a truck, people ask can you haul this and haul that. Also, the practicality of the truck. Belonging motive Friendship Social Influences What do I think? Well most people I see are younger girls, and I think what a brat, their mom buys them all of their clothes (laughing). But the reason that I like A and F is really stupid, but I like the moose on their shirts. Because I like moose. Significance Abercrombie and Fitch Social influences Identity threat Because its cute, and Ill buy anything with birds and moose on it. Also my mom collects Christmas moose and things like that. Meaning motive Probably because it is the highest (most expensive) out of American Eagle and all of them, and Hollister. Brand meaning If it was cute then yeah, because I like moose. I always say like for Christmas people either have a snowman or Santa Clause, my mom used to be all snowman and Santas, but I am going to be reindeer and moose. We usually have one (moose) that we decorate every year. Meaning motive Moose
183 Appendix B (Continued) Right now I have a twelve year old Mitsubishi Gallant, Its a black car, but the paint on the car has peeled off on the back of the trunk and part of the roof and it looks speckled like a black and white car. It makes the car look beat up. But its a good car. People see me in that car and make judgments that I am poor, and in an economically low social class, otherwise I wouldnt be driving in such a low end car. If you pull up at a stop light, and someone is next to you in a Mercedes, they look at you in your car and you can tell by their facial expression that they are so not digging you. I thought for a long time it was because of my skin color, but people of my own skin color do the same thing when they drive fancy cars. But the funny thing is people who are in cars more beat up than mine, they say hi to me.(Laughing) Self-image Others perception Self-consistency If I pull up at a stop sign and they look Mexican I dont care what they think because their car is more beat up than mine. But if a Mexican pulls up in a Mercedes I think they will judge me based on my car. These thoughts are real! I am not making this up. The sad part is it never really bothered me until I got into a different salary level. Up until I hit the six figure salary level, I didnt mind. I dont know. I just dont want people to .. No! I dont want people to think of me as just another statistic as a poor African, blacklooking statistic. I want them to see me an upwardly mobile successful woman. Self-image Others perception Self-consistency They (Lexus IS350) are not sporty. They look like someone who has a family of four. I want something that is sporty, that makes me look upwardly mobile and professional. Something that says Ive made it. Im single and I dont have a family. I dont want people to think that I am older person. You know, so what I do is . while I am driving, I look at cars and I look at who is in them, and how the cars look, and pretty much everything except for the IS350 I see mature looking people in them, looking like they have kids. Lexus also has an IS250 but the people that I see in the 250 are like young collegey kids. Its fast and sporty, but the young college kids are in them. But I dont want anyone to think Let me correct myself, I dont want people to assume that I have kids. Thats why I dont only want to buy that model. I also will not buy the IS250 because I dont want anyone to think that I am one of those just trying to get my hands around life college kids. Does that sound kind of vain? Self-image Others perception Self-consistency Identity signaling
184 Appendix B (Continued) Well thats more of my black male identity not really my father, if thats the case then I would get a Volvo Station wagon. (laughing). Its a little bit of both all wrapped up. You know I cant say which one it would be . Well the car will be more for status. So if I really wanted just straight up status, I would choose the Lexus, and if I was going for the responsible, competent father, then Id choose the Volvo or the Honda. Self-consistency Status Self-image Multiple identities Reference groups Hollister is like more the middle-school aged children, and A and F is for high school kids and older and Rule, well no one knows about Rule, and they only have a few stores, but no one knows about it . I can fit them, but I really dont go in there because its for high schoolers, but every now and then, Ill buy a t-shirt out of there. Self-image Self-consistency Identity threat Just straight up me! Me being a black male, umm just, just strictly frontin. But the fact that it will be a big enough vehicle for my family would be a plus, but thats not really what I am buying that for. Im buying that for status. For a big, pretty vehicle thats it. Reference groups Family influences Status So I like designer shoes, but they dont all have to be designer, they just have to look cute. Sometimes I drop the name of the brand in front of my girlfriends like Minolo Blanick or Via Spiaga. Some of my friends think theyre all that and they know fashion, so because they act that way, we always talk about what we have on. They say oh this suit is Chinese Laundry, this suit is Gucci, or I just picked up this shirt from the Armani Exchange. When I get around them, I will let them know, that it is something expensive. So my friends who arent that snooty, I will say I just got this at a bargain price, and I will tell them how much I paid for it. But with my snooty friends, I will always lie about the price and never tell the true price of it. And say that I bought it at another store, and never an outlet. Reference groups Self-image Identity threat Yes. But only if they were really cute, I would never tell any of my friends, and I would never let them see the inside label cause it will say Payless. And if we had to go somewhere to take off our shoes, I wouldnt take them off, Id be too embarrassed. I cant have my friends thinking I shop at Payless. Self-image Others perception Self-consistency Identity threat
About the Author Tracy is the youngest child of Joyce Harm on and Joe Wright, sister to Malik and Conswello, aunt to Stedman, Mundedrick, Ni gel, Nigele, Rashaad, and Jaylen. Tracy received a B.S. degree in Chemical Engineer ing from Florida A & M University, and an M.B.A. from Rollins College. She will receive a doctoral degree from the University of South Florida. Starting in the fall of 2007, Tracy will be gin a postdoctoral as signment with the Whitman School of Management at Syracuse University. Beyond her identity-related research agenda, Tracy is also pursui ng projects in urban branding and urban consumption.