Journal of Marketing Research Vol. XLVI (February 2009), 92���104 92 �� 2009, American Marketing Association ISSN: 0022-2437 (print), 1547-7193 (electronic) *David Sprott is Boeing/Scott and Linda Carson Chaired Associate Pro- fessor of Marketing (e-mail: dsprott@wsu.edu), and Eric Spangenberg is Dean and Maughmer Freedom Philosophy Chair (e-mail: ers@wsu.edu), College of Business, Washington State University. Sandor Czellar is Assis- tant Professor of Marketing, HEC Paris (e-mail: czellar@hec.fr). The authors thank Marc Vanhuele and participants in the HEC Paris Marketing Workshop Series for their constructive help with previous versions of the manuscript. They also thank the two anonymous JMR reviewers for their guidance in strengthening this article. DAVID SPROTT, SANDOR CZELLAR, and ERIC SPANGENBERG* Brand engagement in self-concept (BESC) is a generalized view of brands in relation to the self, with consumers varying in their tendency to include important brands as part of their self-concepts. The authors develop an eight-item scale to measure BESC and demonstrate that it captures a consumer���s general engagement with brands. This scale successfully predicts consumers��� differential attention to, memory of, and preference for their favorite brands. Brand engagement in self-concept is also related to differential brand loyalty, with high-BESC consumers being less price and time sensitive regarding their favorite brands than low-BESC consumers. The authors discuss the usefulness of this construct for marketing research. Keywords: self-concept, self���brand connection, brand attitudes, brand preferences, brand equity The Importance of a General Measure of Brand Engagement on Market Behavior: Development and Validation of a Scale Exploring connections between consumers and their brands, scholars have demonstrated that self���brand connec- tions exist and can lead to favorable brand attitudes (Escalas 2004 Escalas and Bettman 2003, 2005). Prior research on these self���brand connections has predomi- nantly focused on the linkage between a specific brand and a consumer���s self-concept. We propose that consumers vary in their general engagement with brands, and we examine the nature and importance of consumers��� tendencies to include important brands as part of their self-concept. We call this generalized individual difference ���brand engage- ment in self-concept��� (BESC). We develop a measure of this new construct, demonstrate its nomological relevance, and show that consumers vary substantially in the degree to which they incorporate brands as part of their self-concept. We then examine how BESC affects important aspects of brand-related knowledge, attitudes, and behavioral intentions. BESC During the past decade, branding research has given con- siderable attention to the different forms of relationships between consumers and brands (e.g., Aaker, Fournier, and Brasel 2004 Aggarwal 2004 Chaplin and John 2005 Escalas 2004 Fournier 1998 Terrasse 2006). The current research contributes to this domain by exploring differences between consumers with respect to their general engage- ment with brands. To do so, we propose the BESC con- struct, defined as an individual difference representing con- sumers��� propensity to include important brands as part of how they view themselves. Our conceptualization builds on self-schemas to investigate the role of brands in the self- concept. Self-concept can be viewed as a set of self-schemas rep- resenting stable knowledge structures about the self that organize incoming self-related information and help people make sense of themselves in their environment (Markus 1977). People vary in their tendency to possess particular self-schemas, and this variation leads to differential atti- tudes and behaviors toward objects relevant to those schemas (Markus 1983 Markus et al. 1982). For example, Sentis and Markus (1986) show that consumers with a strong masculine self-schema described fragrance brands in more accentuated gendered terms and held sharply different brand preferences than those with weaker masculine self- schemas. The underlying assumption of the BESC con- struct is that consumers vary in their tendency to possess

A General Measure of Brand Engagement 93 brand-related schemas. While some consumers may develop self-schemas about how the brands they use and like are related to the self, others may not develop such schemas. As with other self-related schemas, we expect that differences in BESC are associated with differences in brand-related cognitions, perceptions, and behaviors. Other research supports the basic notion of BESC, sug- gesting that differences exist between consumers with respect to their tendency to engage brands in their self- concepts. For example, important variation has been shown in children and adolescents with respect to the number and nature of self���brand connections (Chaplin and John 2005). Scholars have also found that posting brand references on personal Web sites serves self-extension purposes more for some consumers than for others (Schau and Gilly 2003), and researchers have suggested that people vary in their likelihood to engage in identity building and expression through brand communities (Muniz and O���Guinn 2001). Anecdotal evidence is also suggestive of individual differ- ences regarding consumers��� use of favorite brands for self- development within loyal segments, some consumers are more likely than others to use brand tattoos as a means of identifying with the brand (Lindstrom 2005). Our conceptualization of BESC broadens existing per- spectives on the link between brands and self-concept. Arguably informative to the development of BESC are self���brand connections (Escalas 2004 Escalas and Bettman 2003) and attachment to possessions (Ball and Tasaki 1992). Although respective measurement scales have been used to demonstrate that consumers form connections to particular brands (Escalas 2004 Escalas and Bettman 2003) or objects (Ball and Tasaki 1992), important distinc- tions exist between BESC and each of these constructs. First, BESC is a generalized tendency to include brands as a part of the self-concept, whereas previous research has assessed connection with a specific brand. For example, the self���brand connection scale (Escalas 2004) measures the strength of the link between the self and a particular brand, while BESC provides a more comprehensive view of the person���brand connection by accounting for the notion that multiple brands are integrated into a consumer���s self- concept. Second, attachment to possessions taps into the extent to which a specific past, present, or future possession contributes to maintaining consumer self-concept (Ball and Tasaki 1992). As such, the attachment-to-possessions scale is not a useful measure of the general links between self- concept and brands, because it focuses on a specific (not necessarily branded) object. Furthermore, attachment to possessions is a construct anchored in materialism���a para- digm distinct from and broader in scope than BESC (Richins 1994, 2004 Richins and Dawson 1992). Although conceptually related to BESC, materialism is conceptual- ized as a consumer value, ���a set of centrally held beliefs about the importance of possessions in one���s life��� (Richins and Dawson 1992, p. 308). Possessions may be unbranded products or services, and the value consumers attach to them may not be self-related. Indeed, Richins (1994) shows that though possessions may serve purposes of self- extension or self-expression, such motivations are seldom the main reasons for valuing a possession. A primary goal of this article is to demonstrate that con- sumers indeed vary in their propensity to include important brands in their self-concept. We develop a scale to measure the BESC construct and highlight the value of this individ- ual difference variable to research by demonstrating in a nomological framework that BESC affects important aspects of brand-related consumer attitudes and behavior. SCALE DEVELOPMENT AND NOMOLOGICAL VALIDITY Scale Development Following standard procedures (Nunnally and Bernstein 1994), we based item generation on a review of relevant branding and self-concept literature streams, which resulted in an initial pool of 36 scale items intended to capture vari- ous aspects of brand engagement with the self. Nine schol- ars evaluated the content validity of each item with respect to our definition of BESC, which resulted in a modified set of 32 items. We administered these items to an undergradu- ate student sample (N = 430). Item purification using con- ventional factor analytical iterations resulted in a final BESC scale comprised of eight items anchored by ���strongly disagree��� (1) and ���strongly agree��� (7) (see Table 1). Reliability of the BESC scale met conventional stan- dards of internal consistency (�� = .94) and intertemporal reliability (ranging between .62 and .78). Details of scale development and psychometric properties appear in the Web Appendix (http://www.marketingpower.com/jmrfeb09). Nomological Validity Although various conceptualizations of self-extension are proposed in the literature, of relevance to the current work is research in psychology that focuses on the social aspects of the self and the extent to which the self-concept is construed with regard to other people. Of two basic approaches to such self-construal, the first, broader approach views the self to be construed around others at a general level (e.g., others in society), giving rise to con- structs such as independent versus interdependent self- construal (Singelis 1994) and collective self-esteem (Luhta- nen and Crocker 1992). The second, more focused view holds the self to be construed around a tighter network of other people, including those who are particularly impor- tant, such as close friends and/or relatives (Cross, Bacon, and Morris 2000 Cross and Madson 1997), and is repre- sented by the construct of relational-interdependent self- construal (RISC Cross, Bacon, and Morris 2000). Overall, we expect weak linkages between BESC and social aspects of the self because brands are not people and consumers do not typically view human beings and brands in the same ways (Aaker 1997 Yoon et al. 2006). Nevertheless, BESC bears some structural similarity to RISC in that both con- Table 1 SCALE ITEMS USED TO MEASURE BESC 1. I have a special bond with the brands that I like. 2. I consider my favorite brands to be a part of myself. 3. I often feel a personal connection between my brands and me. 4. Part of me is defined by important brands in my life. 5. I feel as if I have a close personal connection with the brands I most prefer. 6. I can identify with important brands in my life. 7. There are links between the brands that I prefer and how I view myself. 8. My favorite brands are an important indication of who I am.