9.... £3. 1 1 in... 1 $qu ii: 1. I r 1. .7...» Janna... .. i ..i .uanwmnn. )0 Jul-n7 in..." 3.. ‘3 ‘1. n to ,. c. -7 a: n; '1. r11! 2»: wife . t .v. a... .3) J: V . Pr: 1...... .1393. #5... r 33 ‘31 . ; A 5.. «YA-ralb .195... u . 3. . 3&5. .. x , . 2.3.: . .5), 13.10.. c I. ,‘l ). .b :51 .9'i‘ Gill. I 5‘1: {.19. e. if... 3 3.3.1.1.: n 5?: 3:} .t .1.) ‘ ‘0 1:19...) s’fiz. 3 .rzaDst . ll»... .ltuatt . .. . I .312: tic): V213»... . rut J MICHIGAN sure umve ill I ll! III! //I Will! Hm Liammes '- Will/{W 3 1293 01410 0592 iii! This is to certify that the thesis entitled Toward a Predictive Model of Brand Preference: The Relationship of Product Involvement, Symbolic Brand Image and Self-Congruity presented by Jae—Eun Chung has been accepted towards fulfillment of the requirements for M.S. degree in Human Environment andiDesign L/Zéflvuf’fl/ J. VQMM/Wi Major professor '/ Date H “/5 “€59“ 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan State University PLACE N RETURN TO AVOID FINES BOX to roman this chockom from your record. Mum on or baton duo duo. DATE DUE DATE DUE DATE DUE TOWARD A PREDICTIVE MODEL OF BRAND PREFERENCE: THE RELATIONSHIP OF PRODUCT INVOLVEMENT, SYMBOLIC BRAND IMAGE AND SELF-CONGRUITY By J ae-Eun Chung A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Human Environment and Design 1995 ABSTRACT TOWARD A PREDICTIVE MODEL OF BRAND PREFERENCE: THE RELATIONSHIP OF PRODUCT INVOLVEMENT, SYMBOLIC BRAND IMAGE AND SELF-CONGRUITY By Jae-Eun Chung The purpose of the study was to examine the psychological origins of national manufacturer/designer-brand and private-label preferrers on the basis of product involvement, self-image/product image congruity, and symbolic brand image. The data for the final study were collected through a structured self-report survey instrument at two major midwestem universities. The final sample consisted of 171 participants. Using discriminant analysis and paired sample t-tests, the findings of the study revealed that product involvement, ideal congruities with national manufacturer/designer brands and with private labels had a significant relationship with brand preference. However, these variables explained only a small amount of variance in brand preference. Finally, the perceived symbolic images of national manufacturer/designer brands and private labels were significantly related to brand preference. To my husband and daughter iii ACKNOWLEDGEMENTS I would like to express my great appreciation to my advisor, Dr. Dawn Pysarchik, for her support and encouragement. I would like to thank Dr. Linda Good and Dr. Pat Huddleston for their willingness to serve on my committee and valuable input. I would like to thank my parents and family in Korea for their love. Finally, I would like to thank my husband No-Wook for his love and understanding. iv TABLE OF CONTENTS PAGE LIST OF TABLES ....................................................................................................... viii LIST OF FIGURES ......................................................................................................... x CHAPTER 1. INTRODUCTION .......................................................................................... 1 Statement of the Problem and Research Objective .................................. 2 Justification ............................................................................................. 3 11. REVIEW OF LITERATURE ........................................................................ 5 Name-Brand Preferrers Versus Non-Name Brand Preferrers ................. 5 Self-Concept in Consumer Behavior ....................................................... 6 Self-Esteem and Self-Consistency Motivations ..................................... 10 Involvement ............................................................................................ 12 Symbolic Images of Name-Brand and Private-Label Product Users ..... l4 Conceptual Framework .......................................................................... 16 Sirgy’s Actual Congruity/Ideal Congruity Theory ........................... 16 The Proposed Model of Consumers’ Preferences for National Manufacturer/Designer-Brand and Private-Label Products .......... 18 Operationalization of the Proposed Model and Hypotheses ............ 24 III. RESEARCH METHODS AND PROCEDURES ........................................ 27 Research Objective ................................................................................ 27 Research Design ..................................................................................... 27 Research Methods .................................................................................. 28 Pretest ............................................................................................... 28 Selection of High and Low Involvement Products .................... 28 Selection of the Most Popular National Manufacturer/Designer Brands and Private Labels ........................................................ 3O Selection of Brand Image Descriptors ....................................... 32 Final Study ....................................................................................... 35 Sample Selection ........................................................................ 35 Data Collection ........................................................................... 36 Instrument ................................................................................... 36 Theoretical and Operational Definitions ................................................ 43 Hypotheses and Statistical Analyses ....................................................... 44 IV. RESULTS AND DISCUSSION .................................................................. 46 Demographic Characteristics of Sample ................................................ 46 Preliminary Data Preparation ................................................................. 47 Classification of National Manufacturer/Designer-Brand Preferrers and Private-Label Preferrers ........................................................... 47 Computation of Self-Image/Product Image Congruity .................... 49 Computation of Symbolic Images of National Manufacturer/ Designer Brands and Private Labels .............................................. 52 Statistical Results of Hypotheses Test .................................................. 52 Two-Group Discriminant Analysis .................................................. 53 Hypothesis 1 .............................................................................. 58 Hypothesis 2 .............................................................................. 59 Hypothesis 3 .............................................................................. 6O Hypothesis 4 .............................................................................. 61 Hypothesis 5 .............................................................................. 61 Paired Sample T-Test ....................................................................... 62 Hypothesis 6 .............................................................................. 63 Hypothesis 7 .............................................................................. 64 Discussion ....... 65 V. SUMMARY, CONCLUSION, AND RECOMMENDATIONS ................. 68 Summary of Research Methods and Data Analyses .............................. 68 Summary of Findings ............................................................................. 69 Relationship between Involvement and Brand Preference .............. 69 vi Relationship between Four Types of Self-Image/Product Image Congruities and Brand Preference .................................................. 71 Relationship between Brand Preference and Perceptions of Symbolic Images of National Manufacturer/Designer Brands and Private Labels .......................................................................... 71 Conclusions ............................................................................................ 72 Limitations ............................................................................................. 74 Implications and Recommendations ...................................................... 75 APPENDICES ................................................................................................................ 78 A. PRETEST QUESTIONNAIRE .................................................................... 78 B. FINAL STUDY QUESTIONNAIRE ........................................................... 83 REFERENCE LIST ........................................................................................................ 91 vii LIST OF TABLES TABLE PAGE 2.1 The Image of Typical National Manufacturer/Designer- and Store-Brand Owners ................................................................................................... 15 2.2 Sirgy’s (1985) Proposed Relationship between Actual Congruity/Ideal Congruity and Purchase Motivation as Theoretically Explained by Self-Consistency and Self-Esteem Motivation ....................................... 17 3.1 Pretest Results to Determine the Level of Product Involvement with Selected Products ................................................................................... 30 3.2 Frequency Distributions of Brand Popularity Ratings of National Manufacturer/Designer Brands for Blue Jeans, Oxford Cloth Shirts, and Socks: Pretest Results ..................................................................... 31 3.3 Familiarity and Image Assessments of Private Labels: Pretest Results ...... 32 3.4 Pretest Results of Brands Selected for Three Test Products (Blue Jeans, Socks, and Oxford Cloth Shirts) for Final Study .................................. 33 3.5 Wearing Lists to Measure Images of National Manufacturer/Designer Brands and Private Labels in Pretest ...................................................... 33 3.6 Paired T-Tests of Seven Descriptors to Select Attributes to Measure Images of National Manufacturer/Designer Brands and Private Labels ............ 35 3.7 The Descriptors Used to Measure Actual and Ideal Self-Image .................. 37 3.8 The Twenty Attributes of Involvement Measurement ................................ 38 3.9 The Six Wearing Lists for Brand-Image Measurement ............................... 4O viii 3.10 3.11 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 Statements for Brand Preference Measurement ........................................... 41 Cronbach’s Alpha Analysis of National Manufacturer/Designer Brand Preference Measurement for Blue Jeans, Socks and Oxford Cloth Shirts ...................................................................................................... 42 Sample Profile ............................................................................................. 46 The Frequency Distribution of High and Low Involvement Groups by Product Type .......................................................................................... 49 The Results of Canonical Discriminant Functions for National Manufacturer/ Designer-Brand Preferrers and Private-Label Preferrers ....................... 55 The Group Centroids of National Manufacturer/Designer—Brand Preferrers and Private-Label Preferrers for Analysis Group ................................... 55 Classification Matrices for Two-Group Discriminant Analysis of National Manufacturer/Designer-Brand Preferrers and Private-Label Preferrers for Both Analysis and Holdout Samples ................................................ 57 The Results of Press’s Q Statistics to Check Classification Accuracy Relative to Chance .................................................................................. 58 Univariate Analyses of Variance to Test Equality of Group Means between National Manufacturer/Designer-Brand Preferrers and Private-Label Preferrers ................................................................... 59 Discriminant Analysis of National Manufacturer/Designer-Brand Preferrers Versus Private-Label Preferrers ............................................ 60 The Results of Paired T-tests: Perception of Symbolic Images between NM/D and PL for National Manufacturer/Designer-Brand Preferrers... 63 The Results of Paired T-tests: Perception of Symbolic Images between NM/D and PL for Private-Label Preferrers ........................................... 64 ix 5.1 Summary of Hypotheses Testing ................................................................. 70 LIST OF FIGURES FIGURE PAGE 2.1 The Proposed Model of Consumers’ Preferences for National Manufacturer/Designer Brands and Private-Label Brands .................... 19 CHAPTER I INTRODUCTION Consumer behavior is an evolving phenomenon that changes with the development of a society (Assael, 1992). A review of consumer behavior studies over the past few years reveals that consumers are more knowledgeable and thus they react in more complex ways to marketing processes (Assael, 1992; Mowen, 1990). An example would be the complex responses of consumers to the recent product brand proliferation both within and outside of the United States. This brand explosion has yielded a plethora of national manufacturer/designer and private-label brands1 from which consumers may choose. Thus, a growing area of interest in consumer behavior is the psychological antecedents underlying consumers’ preferences for national manufacturer/designer and private-label brands of products. Consumer attitudes toward name-brand and private-label products have changed significantly. Although earlier studies indicated that consumers prefer name-brand products because their images are more appealing than those of private-label products (Belk, Mayer & Driscoll, 1984; Calder & Burnkrant, 1977; Jolson, Anderson & Leber, 1981), recent evidence indicates that private-label brands are growing in popularity since consumers' confidence in the quality of private-label products has improved (Deveny, 1992; Salmon & Cmar, 1987; Underwood, 1992; Wilensky, 1994). Well known brand 1 Private label brands are defined as “store names, retailers’ own brand names, designer exclusives, exclusive licensed names, or generic labels” (Klokis, 1987, p. 22). 1 2 names such as designer brand names, however, still appear to be a significant part of the retailing mix (Baugh & Davis, 1989) since the prestigious symbolic images of these brand names are appealing to consumers (Abend, 1991; Huddleston & Cassill, 1990; Kwon, 1990; Liebeck, 1994; Miller, 1992). These trends, therefore, support the notion that consumers purchase national manufacturer/designer brands or private-label products not because of functional factors (such as price or quality) but because of other factors such as the product’s symbolic attributes (for example, the enhancement of a consumer's self-concept) or a consumer’s involvement with a specific product. Further research, therefore, is needed to better understand the complex relationship of elements that affect consumers' preferences for apparel with national manufacturer/designer or private-label brands. Statement of the Problem and Research Objective The previous research regarding national manufacturer/designer-brand and private-label preferrers has been limited to profiling consumers' life-styles (Jolson et al., 1981) and to investigating product quality and demographics as predictors of apparel brand selection (Huddleston and Cassill, 1990). Thus, these studies have not investigated the effect of psychological factors or the symbolic aspects of brands on brand preference. The research problem of this study, therefore, is to identify the psychological antecedents of consumers’ preferences for national manufacturer/designer brands and private labels. Self-concept was the underlying theory used in this study. This theory proposes that the degree to which a brand's image matches a consumer’s self-image affects her or his brand preferences (Sirgy, 1982; 1985). For example, if the images of national 3 manufacturer/designer-brand products are congruent with what consumers would like to be (their ideal self-images) or their perceptions of themselves (their actual self-images), they will prefer national manufacturer/designer brands. The literature of psychology and consumer behavior suggests several factors related to self-concept theory that might affect consumers’ preferences for national manufacturer/designer brands or for private labels. These include perceptional differen- ces of the symbolic images of these two kinds of brands and product involvement. The objective of the study, therefore, is to examine the psychological origins of national manufacturer/designer—brand and private-label preferrers on the basis of product involvement, self-image/product image congruity, and symbolic brand image. Justification Consumer researchers have identified the importance of the symbolic meanings underlying product preference and purchase intention (Behling & Wilch, 1988; Belk, 1988; Birdwell, 1968; Dolich, 1969; Grubb & Grathwohl, 1967; Landon, 1974; Leigh & Gabel, 1992; Malhotra, 1988; Onkvisit & Shaw, 1987; Sirgy, 1982; 1985; Solomon, 1983). Solomon (1983) suggested that the symbolic qualities of highly valued products are often determinants of product evaluation and adoption. Several researchers have pointed out that self-concept has been rarely studied in consumer-behavior research in spite of its relevance and significance in psychological and behavioral research (Claiborne & Sirgy, 1990; Onkvisit & Shaw, 1987; Sirgy, 1982). Onkvisit & Shaw (1987) indicated that an understanding of consumers' self-concepts and self-images can provide the means for developing more effective marketing programs, 4 since self-concept may explain how and why consumers purchase a particular product or brand. The proposed study, therefore, will be the first to use self-concept theory as the basis for investigating the effect of product symbolism on purchases of national manufacturer/designer-brand and private-label clothing. CHAPTER 11 REVIEW OF LITERATURE Name-Brand Preferrers Versus Non-Name Brand Preferrers Jolson et al. (1981) conducted the first study to contrast name-brand preferrers with name-brand avoiders. This study found that name-brand preferrers were highly aggressive; that is, they had a desire for strong interpersonal relationships and a need to compete and be noticed and admired. Name-brand preferrers were more likely to be female, black, and well-educated. On the other hand, name-brand avoiders aspired to self-sufficiency, privacy, and a high degree of social anonymity. Older people, males, and Caucasians were typical of this group. Additionally, Kwon (1990) reported that brand-conscious consumers had a high interest in clothing and were usually adolescents and younger adults. To identify the predictors of female consumers' name-brand orientation, Huddleston and Cassill (1990) investigated the quality of products and studied the demographic variables within this group. The results showed that name-brand oriented consumers were more likely to be concerned with quality, to spend more money on apparel, and to shop at specialty stores. Further, this study found that apparel-selection criteria, such as "made in USA" labels, were significant predictors of name-brand orientation (Huddleston, Cassill & Hamilton, 1993). 6 Self-Concept in Consumer Behavior Self-concept, which has many dimensions, has been defined as "the totality of the individual's thoughts and feelings having reference to himself as an object" (Rosenberg, 1979, p. 7). The term "actual self“ refers to how a person perceives herself/himself; "ideal self" refers to how a person would like to perceive herself/himself; "social self" refers to how a person presents herself/himself to others; and "ideal-social self" refers to how a person would like to present herself/himself to others. The influence of self-concept on consumers’ consumption behavior is suggested by Levy's (1959) proposition that consumer behavior is significantly affected by the symbols identified with goods in the marketplace. After Levy's study was published, several other consumer researchers attempted to establish theoretical frameworks for self- concept and consumer behavior (Claiborne & Sirgy, 1990; Dolich, 1969; Grubb and Grathwohl, 1967; Landon, 1974; Sirgy, 1982; 1985; Sirgy, Johar, Samli, & Claiborne, 1991). Grubb and Grathwohl (1967) proposed a model to clarify the systematic relationships between self-concept theory, goods as symbols, and consumer behavior (p. 26). The model included the following propositions: 1. Self-concept is of value to the individual, and behavior will be directed toward the protection and enhancement of self-concept. 2. The purchase, display, and use of goods communicate symbolic meaning to the individual and to others. 3. The consuming behavior of an individual will be directed toward enhancing self-concept through the consumption of goods as symbols. Dolich (1969) argued that products and brands have images that some individuals perceive as congruent with their self-concepts. He then hypothesized the effect of 7 symbolic images of products and brands on consumer behavior by using Rogers' (1965) proposition, "individual behavior is regulated by each person's perceived similarities (or dissimilarities) of environmental conditions to the self-image" (p. 80). That is, products and brands are parts of the environment, and only those whose symbolic images are similar to the consumers' self-concept will be perceived as maintaining or enhancing the self. Accordingly, Dolich (1969) proposed that individuals will accept brands with images similar to their self-concepts and reject brands with images dissimilar to their self- concepts. Several empirical studies have found that consumer behavior is a function of self- concept and product image congruity (self-congruitY); that is, consumers preferred products (brands) with images that were consistent with their self-concepts (Bellenger, Steinberg, & Stanton, 1976; Birdwell, 1968; Dolich, 1969; Ericksen & Sirgy, 1989; Evans, 1988; Grubb & Grathwohl, 1967; Grubb & Hupp, 1968; Holman & Schenk, 1980; Landon, 1974; Malhotra, 1988; Sirgy, 1985). Self-image/product image congruity has also been treated multidimensionally. Product image has been found to be congruent or incongruent with the actual2 , ideal, social, and ideal social selves as previously defined. Actual congruity is consistency between the actual self-image and the product image; ideal congruity is agreement between the ideal self-image and product image; social congruity is correspondence between the social self-image and product image; and ideal social congruity is conformity between ideal social self-image and product image. 2 Sirgy (1985) refers to consistency between the actual self-image and the product image as “self- congruity.” 8 Landon (1974) clarified the relative importance of actual self-concept and ideal self-concept. The results showed that some subjects preferred products with images that were congruent with their actual self-image, while others chose products with images that were consistent with their ideal self-image. The study revealed that factors of the products (i.e., public or private use or social desirability) and the individuals (i.e., discretionary income) were significant determinants of the relative importance of actual self-image and ideal self-image to purchase intentions. In a study of consumers' housing, Malhotra (1988) found that a majority of the respondents (60.1%) preferred houses congruent with their ideal self-concepts. However, 22.3% preferred houses similar to their actual self-concepts and 17.6% selected houses more akin to their social selves. Therefore, this study suggested that the ideal self-image has a dominant role in consumer decision making. Some researchers have investigated moderating variables influencing the relative effect of ideal congruity and actual congruity. Dolich (1969) studied the relative effect of actual self-image and ideal self-image on most- and least-preferred brands within four product categories of socially and privately consumed products. Beer and cigarettes were used as examples of socially consumed products, and bar soap and toothpaste were used as examples of privately consumed products. The results indicated that both the actual self- and ideal self-images were consistent with the images of the most-preferred brands, regardless of whether the product was socially or privately consumed. For the least- preferred brands, however, the ideal self-image had more discrepancy with brand image than the actual self-image for only one of the privately consumed products, bar soap. 9 Ross (1971) employed a public (social and ideal social) versus private (actual and ideal) self-concept model to explain the relative effects of actual, ideal, social, and ideal social congruities on product preference. He hypothesized that private-self models would predict brand preferences for less conspicuous brands. Conversely, public-self models would predict preferences for conspicuous brands. Ross, however, did not find these relationships. Munson (1974) extended the study of self-congruity and product preference to the examination of the moderating effect of product conspicuousness and social class interaction. The results showed that preference for conspicuous products was related to ideal self-concept among upper-social class respondents, whereas among lower-class respondents, preference was not related to either actual or ideal self-concept for either conspicuous or inconspicuous products. Belch (1978) examined the moderating effect of personality on the relationship between self-concept/product image congruity and purchase intention. His findings suggested that purchase intentions of subjects who had high social needs were more closely related to ideal self-concept than to actual self-concept. In addition, Sirgy (1980) found that attitudes toward a product and attitudes toward a purchase moderated actual congruity and ideal congruity. More specifically, findings indicated that ideal and ideal- social congruity were more predictive of product preference (attitude toward product) than of purchase intention (attitude toward purchase); and conversely, actual and social congruity were more predictive of purchase intention than product preference. 10 Sirgy (1985) integrated actual congruity/ideal congruity models to explain the differential role of these congruities in consumer behavior. He found that purchase intention was a positive function of the linear combination of both actual and ideal congruities. He therefore proposed that purchase motivation is affected by actual congruity/ideal congruity mediated by self-esteem and self-consistency needs. Finally, researchers examined the relative influences of functional congruity and self-congruity on consumer brand attitude (Sirgy et al., 1991). Functional congruity refers to the degree to which a brand meets consumers' performance-related evaluative criteria, while self-congruity refers to the degree to which a brand's image matches a person's self- image. Four studies were conducted to test the hypotheses that functional congruity is a stronger predictor of consumer behavior than self-congruity and that functional congruity is significantly influenced by self-congruity. The results supported these relationships. Self-Esteem and Self-Consistency Motivations According to self-concept theory, the individual is influenced by two motives in building her/his self-concept: self-esteem and self-consistency (Sirgy, 1982; 1985; Assael, 1992). Therefore, this section examines the nature of self-esteem and self- consistency as well as the underlying motivations. Self-esteem is "a positive or negative attitude toward a particular object, namely, the self" (Rosenberg, 1965, p. 7). Psychologists have stressed the importance of self- esteem and identified. it as one of our deepest emotional needs (Hilgard, 1962, p. 10), the 11 most important element in our motivation (Brandon, 1969, p. 12), and our controlling emotion (McDougall, 1932, p. 9). Self-esteem was found to affect self-consistency (Kaplan, 1980; Rosenberg, 1965). That is, persons with low self-esteem generally were not satisfied with themselves. They sought to change their deficient self-images in order to improve their self—perceptions, and thus had less consistent self-concepts. Those with high self-esteem, in contrast, were satisfied with who they were and had no need to adjust their self- concepts. They sought to protect and maintain their high self-esteem, and they had high self-consistency. Self-esteem and self-consistency have different motivational forces. The self- esteem motivation is the tendency to seek experiences that enhance the self-concept, and the self-consistency motivation is the tendency to behave consistently with the view one has of oneself in order to avoid dissonance (Sirgy, 1982). The self-esteem motivation has primacy with respect to the self-consistency need (Elliot, 1986; Jones, 1973; Kaplan, 1980). In short, people have a stronger desire to protect and enhance their self-esteem than to act in a manner that is congruent with their self-concepts (Elliot, 1986). Although researchers have not specifically analyzed the effect of self-esteem on consumers' brand preference, some studies have investigated the discrepancy between ideal and actual self-concept as a measure of self-esteem. White (1964) defined three groups based on the discrepancy between actual and ideal self-concept. His study indicated that those with high discrepancy (the lowest self-esteem respondents) were dissatisfied with their self-image and wished for great and unrealizable changes; those 12 who fell in the middle range were somewhat dissatisfied and wanted to improve themselves in a realistic way; and those with low discrepancy (the highest self—esteem respondents) were satisfied with their situations, and their real and ideal selves were quite similar. White found that those in the middle range were the most predictable in their product purchases; that is, they were likely to select products with images similar to their ideal self-images. Involvement Consumer involvement has been a central concept in the understanding of information processing (Mowen, 1990), and researchers have distinguished between various kinds of involvement. Situational involvement has been defined as "the level of perceived importance and/or interest evoked by a stimulus (or stimuli) within a specific situation" (Mowen, 1990, p. 782). Such involvement occurs over a short period of time and is associated with a specific situation, such as a purchase. In contrast, enduring involvement entails a longer commitment and concern with a product class on a day-to- day basis (Mowen, 1990). Some researchers have related self-concept to these kinds of involvement. Beatty, Kahle & Homer (1988) referred to enduring and situational involvement as ego and purchase involvement. They defined the former as "the importance of the product to the individual and to the individual's self-concept, values and ego," and the latter as "the level of concern for or interest in the purchase process triggered by the need to consider a particular purchase" (p. 150). This research indicated that self-concept was an antecedent of enduring involvement. 13 Bloch (1982) investigated the relationship between self-concept and consumers' enduring involvement with a product. He found that the magnitude of enduring involvement was positively related to the extent of self-image/product image congruity for automobiles and clothing. Assael (1992) presented a comparison of low-involvement and high-involvement decision making. A low-involvement decision-making model suggested that personality and life-style characteristics were not related to purchase behavior because the product was not closely tied to a consumer’s identity and belief systems. Conversely, a high- involvement decision-making model suggested that such characteristics affected consumers' behavior because the product was highly related to both their identities and their belief systems. Although there has been no research specifically investigating the relationship between consumer involvement and attitude toward national manufacturer/designer—brand and private-label products, some studies have suggested that this relationship exists (Holstius & Paltschik, 1983; Kwon, 1990). Holstius and Paltschick (1983) studied the effect of brands on consumers’ perceptions of the price and quality of neckties. Their results indicated that individuals who were interested in fashion were willing to pay more for designer labels. It appeared from this study that consumers who were highly involved with fashion had more interest in name-brand goods than those who were not. Similarly, name-brand preferrers had more interest in clothing, higher brand-name awareness, and higher clothing expenditures than did those who preferred non-name brands (Kwon, 1990) SL YESU l4 Symbolic Images of Name-Brand and Private-Label Product Users The results of several studies in which the symbolism of products was investigated suggested that individuals were likely to ascribe personal characteristics to others based on the image of the products (brands) they purchased (Calder & Bumkrant, 1977; Belk, 1988; Belk, Bahn, & Mayer, 1982). Belk (1988) concluded that inferences about a product user were determined, in part, by the image of the product relative to style, color, uniqueness, condition or brand name. The inferences about product users were largely dependent on the product when little prior information about the user was available. Belk et al. (1982) investigated the ability of consumers from different age groups to recognize the product symbolism of automobiles and houses. These researchers suggested that the ability to recognize the social implications of product choices was minimal in preschool, significant by second grade, and almost fully developed by sixth grade. While college students were most likely to recognize product symbolism, this perception decreased in adulthood. Calder and Bumkrant (1977) investigated how people inferred consumers’ personality traits from their choices of manufacturer's brand and private—label cosmetic products. The respondents were presented with scenarios describing a typical consumer’s purchase situation. After reading the scenarios, they were asked to rate twenty-seven personality traits of the hypothetical consumers. Through factor analysis, the traits were reduced to two factors, "social evaluation" and "personal effectiveness" (Table 2.1). The results showed that the subjects‘ social evaluation of the consumer was more positive and Til-1 Chi pro pm 15 their personal effectiveness evaluation more negative when the shopper was associated with the manufacturer's brand. Table 2.1. The Image of Typical National Manufacturer/Designer— and Store-Brand Owners. Researcher Image of National Manufacturer/Designer-Brand Owners Calder & Bumkrant Social Evaluation: attractive, high status, popular, (1977) sociable, successful, superior, neat. . Personal Effectiveness: extravagant, immature, conforrnist, uninformed. Belk et al. (1984) Young, female, attractive, successful, extravagant, more conforming. Image of Store-Brand Owner. Workman (1988) Less active, less fashionable, negative personality, less popular, less sociable, less sophisticated, heavier, enjoy shopping less, more economical. Several researchers have also investigated how consumers' choices of clothing labels affected the impression they made on others (Belk et al., 1984; Workman, 1988; Kwon, 1990). When elementary school children viewed color slides of three different types of jeans -- designer, national, and private brand -- Belk et al. (1984) found that the children exhibited a brand-name bias when inferring owner characteristics from these products. Name-brand owners received more favorable evaluations than owners of private brands (Table 2.1). 51! co €01 CU! theo l6 Workman (1988) investigated inferences based on a person's choice of jeans. Participants were given a list of four jeans brands: designer, national brand-name, store- brand, and unlabeled jeans (used as a control). She then asked participants to evaluate the owners' personalities and shopping traits. The results indicated that inferences about the person varied significantly depending on the type of jeans. Store-brand jeans produced significantly more negative inferences than designer or name-brand jeans when compared to the control jeans (Table 2.1). More recently, Kwon (1990) investigated brand-name awareness and image of nationally known women's brand-name apparel. Brand-name awareness was influenced by the subjects' occupational status and level of clothing expenditures. She found that some brands (such as Calvin Klein and Ralph Lauren) showed more positive image characteristics than twelve other national apparel brands. Conceptual Framework This study will utilize Sirgy’s (1985) actual congruity/ideal congruity theory, since this theory synthesizes the relative impact of actual self-concept and ideal self- concept on consumer behavior. Thus, in the following sections, Sirgy’s actual congruity/ideal congruity theory will be examined in detail, and then a model for the current study will be proposed based on this theory. Sirgy's Actual Conggig/Ideal Congmig Theog Sirgy (1985) established an actual congruity/ideal congruity theory based on a theory of self-consistency and self-esteem needs relative to four conditions: high actual “l ex; COD DE€( l7 congruity/high ideal congruity (ACh/ICh), high actual congruity/low ideal congruity (ACh/ICI), low actual congruity/high ideal congruity (ACl/ICh), and low actual congruity/low ideal congruity (AC1/IC1). (See Table 2.2.) Table 2.2. Sirgy's (1985) Proposed Relationship Between Actual Congruity/Ideal Congruity and Purchase Motivation as Theoretically Explained by Self- Consistency and Self-Esteem Motivations. Theoretical Intervening Constructs Congruity Observed Purchase Condition Self-Consistency Motivation Self-Esteem Motivation Motivation ACh/IC}l Self-consistency motivation Self-esteem motivation Approach purchase (approach) (approach) motivation ACh/IC1 Self-consistency mot. Self-abasement mot. Conflict purchase (approach) (avoidance) motivation AC1/IC1-l Self-inconsistency mot. Self-esteem mot. Conflict purchase (avoidance) (approach) motivation AC1/IC1 Self-inconsistency mot. Self-abasement mot. Avoidance purchase (avoidance) (avoidance) motivation AC = actual congruity; IC = ideal congruity. The self-consistency motivation and the self-esteem motivation are used to explain the effect of actual congruity and ideal congruity on purchase motivation. Actual congruity (AC) affects purchase motivation through the operation of the self-consistency need. In other words, consumers are likely to purchase products or brands that match m C0 C02 sin (3" ideal man: 18 their actual self-image and satisfy their need for self-consistency. On the other hand, ideal congruity (1C) affects purchase motivation through activation of the self-esteem need; that is, consumers are likely to purchase products or brands that match their ideal self-image and enhance their self-esteem. Therefore, under conditions creating high actual congruity/high ideal congruity (ACh/ICh), the consumer would be motivated to buy the product or brand, because the purchase would support both her/his self-esteem and self-consistency needs (approach). Under conditions involving high actual congruity/low ideal congruity (ACh/Icl), the consumer would experience a conflict regarding the product or brand, since buying it would fi'ustrate the self-esteem need (avoidance), yet satisfy the self-consistency need (approach). When low actual congruity/high ideal congruity (ACI/ICh) existed, the consumer would also experience a conflict between the self-esteem (approach) and self- consistency motives (avoidance). If she or he experienced low actual congruity/low ideal congruity (AC1 /IC1), the consumer would be motivated to avoid the product or brand, since purchasing it would not meet her/his self-esteem and self-consistency needs (avoidance). The Promsed Model of Consumers' Preferences for National Manufacturer/Designer- Brand and Private-Label Products Antecedent Va_ria_1bles in the Model In the proposed study, an expanded version of Sirgy's (1985) actual congruity/ ideal congruity theory is applied to explain consumers' attitudes toward national manufacturer/designer and private-label apparel (Figure 2.1). The antecedent variables l9 Product Involvement High Low Symbolic Brand Image of INWD> IPL INM/D= IPL NM/D and PL ' Self-Esteem "iiigiifi a..." "an; "a..." Image Congruity J l of NM/D ACh/ICh AC1/1Ch AC1/IC11 AC1/IC1 .................. r___--—-——— _----—-—--- _-___---_-_ _-—----——-— of PL ACl/ICl ACh/ICI AC1/IC1 AC1/IC1 Brand i l v .. Preference NM/D NM/D PL PL NM/D = national manufacturer/designer brand, PL = private label I m = symbolic image of NM/D, I n. = symbolic image of PL AC = actual self-Image/product image congruity, IC = ideal self-Image/product image congruity Figure 2.1. The Proposed Model of Consumers' Preferences for National Manufacturer/Designer Brands and Private-label Brands. 20 of consumers’ involvement with the product, their perceptions of the symbolic images of national manufacturer/designer brands and private labels, and their self-esteem have been added as factors that may predict their preferences for national manufacturer/designer or private-label products. The social self and ideal-social self are excluded from this study, since social self-images are closely related to actual self-images and ideal-social self- images overlap with ideal self-images. Furthermore, Malhotra's (1988) study showed that these variables were less relevant to consumers' preferences than were actual and ideal self-concept. The Influence of Product Involvement In the proposed model, one of the predictors of consumers’ preferences for national manufacturer/designer or private-label products will be their involvement with the product. Rothschild (1979) proposed that highly involved consumers express interest in both the product class and the brand. They actively seek information and consider a large set of product attributes in their decision making. As explained above, some researchers reported that those consumers who had a greater interest in clothing were more likely to prefer name-brand goods (Holstius & Paltschick, 1983; Kwon, 1990). Thus, highly involved consumers will be national manufacturer/designer-brand preferrers. Rothschild (1979) also suggested that less involved consumers are interested in the general product class, but not in any specific brand. They are not so actively engaged, and they consider only a small number of attributes. Since these less involved consumers will have less intention to spend large sums of money on apparel products (Kwon, 1990), 21 they will consider price to be one of the most significant product attributes. Thus, these consumers will be private-label preferrers. The Influence of Brand Im_ag§ Consumers' perceptions of the symbolic images of national manufacturer/designer and private-label products will influence their brand preferences and will be affected by their involvement with the product. Because a product is related to the identities and belief systems of highly involved consumers (Assael, 1992), it is likely that these consumers will distinguish the symbolic images of national manufacturer/designer brands from those of private-label products. As previously discussed, past research has found that consumers ascribe more favorable socioeconomic and personality characteristics to users of national manufacturer/designer—brand products than they do to users of private- label products (Belk et al., 1984; Calder & Bumkrant, 1977; Workman, 1988). Because high-involvement consumers will have a more positive image of national manufacturer/ designer brands than of private labels, they will, therefore, be national manufacturer/ designer-brand preferrers. Conversely, the product is not important to the identity or belief systems of private-label preferrers, whose involvement is assumed to be low (Assael, 1992). These consumers are unlikely to make distinctions between the symbolic images of national manufacturer/designer brands and those of private-label products and, therefore, will not perceive these symbolic image differences. Research has indicated that consumers have more favorable images of individuals who use national manufacturer/designer—brand products than of those who use private- label products (Belk et al., 1984; Calder & Bumkrant, 1977; Workman, 1988). That is, 22 users of national manufacturer/designer-brand products are perceived as successfiil and sociable, and as having high socioeconomic status. In this model, therefore, consumers who are highly involved with products are assumed to select brands that are consistent with their ideal self-images and reinforce these perceptions. I_h_e Influence of Self-Esteem Self-esteem (Rosenberg, 1965, p. 7) is defined as "a positive or negative attitude toward a particular object, namely, the self. " The consumer’s level of self-esteem will determine the relative magnitude of actual congruity and ideal congruity. Consumers with high self-esteem will have actual self-images that are close to their ideal self-images, while consumers with low self-esteem will experience a gap between their ideal and actual self-images (Rosenberg, 1967). Integrating the findings of previous research, the current study presumes that the image of the national manufacturer/designer-brand products is close to the ideal self-concepts of consumers highly involved with products. Therefore, among these consumers, those with high self-esteem will have high actual congruity/high ideal congruity of national manufacturer/designer-brand products and low actual congruity/low ideal congruity of private-label products. Consumers with low self- esteem, on the other hand, will have low actual congruity/high ideal congruity of national manufacturer/designer products and high actual congruity/low ideal congruity of private- label products. Based on the above arguments, it is suggested that consumers with high involvement and high self-esteem will have ACh/ICh with national manufacturer/ 23 designer brands and ACVIC} with private-label products. Purchasing the national manufacturer/designer-brand products will meet their self-esteem and self-consistency needs, while purchasing private-label products will not satisfy these needs. It is predicted, therefore, that these consumers will prefer national manufacturer/designer brands. Consumers with high involvement and low self-esteem will have AC1/1Ch with national manufacturer/designer brands and ACh/IC1 with private labels. In this conflict situation, the self-esteem need is stronger than the self-consistency need (Elliot, 1986; Jones, 1973; Kaplan, 1980). Therefore, these consumers will prefer national manufacturer/designer brands rather than private label products. Finally, consumers with low involvement and high self-esteem and consumers with low involvement and low self-esteem will have ACI/IC, with both national manufacturer/designer and private-label products. Since they feel little involvement, these consumers do not relate their self-concepts to the product, and they purchase products without comparing their self-images (or ideal images) to the image of the product’s brand. Thus, the extent of their actual congruity and ideal congruity will be low regardless of their levels of self-esteem and the product’s brand. These consumers will prefer private labels for functional reasons, such as low prices, good quality, or good design. 24 Opgrationalization of the Promsed Model and Hmtheses The proposed model suggests causal relationships among variables. It is, however, beyond the scope of this study to test them. Thus, in order to operationalize these relationships, the focus of the study is to identify which variables will be able to discriminate the national manufacturer/designer-brand preferrers from the private-label preferrers on the basis of the proposed model. As shown in Figure 2.1, product involvement is proposed as a discriminating factor in brand preference. The national manufacturer/designer—brand preferrers will have higher product involvement than the private-label preferrers. Thus, hypothesis 1 is proposed as follows: H, : The national manufacturer/designer—brand preferrers will have higher product involvement than the private-label preferrers. The model proposes that self-esteem will not discriminate the national manufacturer/designer-brand preferrers from the private-label preferrers, since there is no distinctive pattern of self-esteem in these two groups. Self-esteem is an explanatory factor in describing the effect of image congruity on brand preference. Self-esteem, therefore, is excluded as a factor for discriminating the national manufacturer/designer- brand preferrers from the private-label preferrers. This model proposes that: (1) the level of actual congruity of national manufacturer/designer brands and actual congruity of private labels varies with the level of self-esteem (high/low) for the national manufacturer/designer—brand preferrers, and (2) the actual congruity of these brands (NMD/ and PL) will be low, regardless of the self- 25 esteem for the private-label preferrers. That is, the national manufacturer/designer—brand preferrers may have high or low actual congruity of national manufacturer/designer brands, depending upon their level of self-esteem. Therefore, the proposed model suggests that the actual congruity of national manufacturer/designer brands and actual congruity of private labels will not have predictive power for brand preference. Hypotheses 2 and 3 are proposed as follows: H2 : There will be no difference in the actual congruity of national manufacturer/designer brands between the national manufacturer/designer- brand preferrers and the private-label preferrers. H3 : There will be no difference in the actual congruity of private labels between the national manufacnirer/designer-brand preferrers and the private label-preferrers. As shown in Figure 2.1, the ideal congruity of the national manufacturer/designer brands is proposed as a discriminating factor, since the national manufacturer/designer- brand preferrers will have higher ideal congruity of the national manufacturer/designer brands than will the private-label preferrers, regardless of their level of self-esteem. Thus, hypothesis 4 is proposed as follows: H4 : The national manufacturer/designer-brand preferrers will have higher ideal congruity of national manufacturer/designer brands than will the private- label preferrers. The proposed model suggests that both the national manufacturer/designer-brand preferrers and the private-label preferrers will have low ideal congruity of private labels, regardless of their level of self-esteem or product involvement. Thus, it is proposed that 26 ideal congruity of private labels will have no affect on brand preference, and hypothesis 5 is proposed as follows: H5 : There will be no difference in the ideal congruity of private labels between the private-label preferrers and national manufacturer/designer-brand preferrers. It is proposed that the perception of the symbolic brand images of national manufacturer/designer brands and private labels will affect brand preference. The national manufacturer/designer-brand preferrers will perceive the images of national manufacturer/designer brands more favorably than those of private labels, while the private-label preferrers will perceive little, if any, symbolic image difference between these two kinds of brands. To examine these relationships, the following hypotheses have been formulated: H6 : The national manufacturer/designer-brand preferrers will have a more favorable symbolic image of national manufacturer/designer brands than of private-label brands. H7 : There will be no difference in the perceived symbolic images of national manufacturer/designer brands and private labels among the private-label preferrers. CHAPTER III RESEARCH METHODS AND PROCEDURES Research Objective The purpose of the study was to examine the psychological origins of consumers' preferences for national manufacturer/designer-brand and private-label apparel. The research is based on self-concept theory, and its objective was to identify national manufacturer/designer-brand and private-label preferrers on the basis of self-image/ product image congruity, the symbolic images of brands, and consumers' product involvement. Research Design This study was cross sectional and used survey research design to collect data on college students' self-concepts, product involvement, self-esteem, and perception of the symbolism of national manufacturer/designer—brand and private-label apparel products. Data were collected through a structured, self-report survey instrument that was specifically designed for the study. The full study consisted of a pretest, which will be presented first, and a final study, which will be discussed after the pretest results. 27 28 Research Methods Ms: A pretest was performed to identify high and low involvement apparel products, to select the national manufacturer/designer and private-label apparel brands that are most relevant to college students, and to identify brand-image descriptors for the final study. As shown in Appendix A, the pretest included three sections: selection of high and low involvement products, selection of the most popular national manufacturer/designer brands and private labels, and selection of brand image descriptors. A convenience sample of fifty undergraduate and graduate students majoring in retailing was selected for the pretest, since college students would be used as the sample for the final study. The pretest was administered in selected graduate and undergraduate classes at a major midwestem university. Selection of High+and Low Involvement Products From the literature, it has been indicated that the level of a consumer’s involvement with a product varies with the type of product. Some products are generally perceived as high or low involvement, while others may vary by the individual. For example, a previous study found that generally an automobile is a high-involvement product and toothpaste is a low-involvement product (Mowen, 1990). In order to examine the varying effect of product involvement on consumers’ brand preferences of specific apparel products in the present study, it was necessary to include a range of products from those eliciting a high level of involvement to those eliciting a low level of 29 involvement. In general, apparel is a high involvement product class (Kaiser, 1990), however, involvement will vary by specific apparel products. An objective of the pretest was to identify apparel products that could be generally viewed as high and low involvement-apparel products for inclusion in the final study. In the pretest study, respondents were asked to score three selected dimensions3 of involvement (important-unimportant, interested-uninterested, and needed-not needed) for ten apparel items (blue jeans, wool sweater, oxford cloth shirts, blazer, raincoat, T-shirt, jacket, socks, tennis shoes, and sweatshirt) on a five-point semantic differential scale (Appendix A, question 1). In the pretest study, only three dimensions of involvement were selected from the twenty attributes listed in Zaichkowsky’s (1985) involvement measurement, since it was felt that these represent three diverse aspects of involvement and would provide a global assessment of involvement for each apparel product. The mean for the three aspects of involvement was calculated and used to determine the level of involvement for each of the ten apparel items. On the basis of this analysis, blue jeans and socks were selected as products representing high-involvement, and oxford cloth shirts were selected as low-involvement products (Table 3.1). Although the mean determined for blue jeans did not represent the highest level of involvement among selected products, it was chosen based on the study of Zaichkowsky and Sood (1988), who identified blue jeans as a high-involvement product. Oxford cloth shirts were selected as low-involvement products, since they represented the lowest level of involvement among such products in the pretest. Socks were selected to complete the list 3 The three dimensions of involvement (importance, interest, and need) were selected to reflect the three areas of involvement (personal, physical, and situational) identified by Zaichkowsky (1985). 30 of basic outfit items provided for respondents to help them visualize a user wearing a complete outfit when they are asked to evaluate brand images in the final study. Table 3.1. Pretest Results to Determine the Level of Product Involvement with Selected Products (N = 50). Product Mean Std Dev Product Mean Std Dev Jacket 1.4 .67 Tennis shoes 2.0 1.1 Blue jeans 1.5 .79 Blazer 2.3 1.1 T-shirt 1.6 .87 Wool sweater 2.7 1 .1 Sweatshirt 1.7 .82 Rain coat 2.9 1.2 Socks 1.8 .98 Oxford shirts 3.1 1.3 involvement = measures of importance, interest, and need. 1 = high involvement; 5 = low involvement. Selection of the Most Popular National Manufacturer/Desiggr Brands and Private Labels In the pretest study, the most popular national manufacturer/designer brands among the ten apparel items (blue jeans, wool sweater, oxford cloth shirts, blazer, raincoat, T-shirt, jacket, socks, tennis shoes, and sweatshirt) were identified by open- ended questioning: “What are most popular national manufacturer/designer brands among college students for each of the following ten clothing items?” (Appendix A, questidn 2). For blue jeans, Guess was selected, and for socks, Hanes was selected, because respondents listed these brands most fi'equently (Table 3.2). For the oxford cloth shirts, Guess was selected for consistency with Guess blue jeans, although Polo was the most popular national manufacturer/designer brand. 31 Table 3.2. Frequency Distributions of Brand Popularity Ratings of National Manufacturer/Designer Brands for Blue Jeans, Oxford Cloth Shirts, and Socks: Pretest Results. Blue Jeans Oxford Cloth Shirts Socks Guess (43.6 %) Polo (61.9 %) Hanes (38.5 %) Levi’s (40.2 %) Guess (25.4 %) Polo (23.1 %) Jockey (12.5 %) The sum of the percentages will not be 100, since the brands with percentages below ten were not included in the table. The preselection of private brands was limited to regional store brands, since the researchers believed that some private labels have come to be perceived as national manufacturer/designer brands (e.g., GAP). In addition, it was necessary to determine the most familiar and positively perceived private labels among the five alternatives in order to avoid selecting a private label with a negatively perceived image. The researcher, therefore, identified five private labels, Mervyn, Target, JCPenney, Kmart, and Meijer. A familiarity rating was included on the pretest questionnaire, and image questions measured by status, popularity, and fashionability were adapted from previous research (Calder & Bumkrant, 1977; Workman, 1988). Respondents were asked to score the above four dimensions (familiarity, status, popularity, and fashionability) for each of the five private labels on a five-point semantic differential scale (Appendix A, question 3). Familiarity was obtained based on the attributes, in a one-word pair, unfamiliar-familiar. The image of each of the five private labels was determined based on the mean of three paired attributes (low status - high status, unpopular - popular, and unfashionable - 32 fashionable). The results of the familiarity and the average image assessment (status, popularity, and fashionability) for each of the five private labels are represented in Table 3.3. Arizona, the JCPenney brand, was finally selected as the private label for the three items (blue jeans, the oxford cloth shirts, and socks), since this brand had a high level of familiarity, and its brand image was the closest to neutral (the score three), which indicated a moderately favorable image among the five private labels. The brands selected for each of the three test products (blue jeans, socks, and oxford cloth shirts) for the final study are listed in Table 3.4. Table 3.3. Familiarity and Image Assessments of Private Labels: Pretest Results. Meiier nget JCPennev Mervvn Kmfl Familiarity(F) Image(I) F I F I F I F I Mean 1.5 1.6 2.3 2.2 2.5 2.6 2.6 2.4 2.8 2.1 Std Dev .97 .70 1.5 .96 1.4 .99 1.5 .88 1.6 .98 N 50 48 50 49 50 49 50 49 50 48 Meijer brands: Falls Creek and Comer Stone; Target brands: Basic Trends and Sostanza; JCPenney brand: Arizona; Mervyn brands: Cambridge Classic and Partners; and Kmart brands: Brittina Jean and Jaclyn Smith. 1 = not familiar; 5 = familiar. 1 = negatively perceived image (status, popularity, and fashionability); 5 = positively perceived image. Selection of Brand Image Descriptors In the final section of the pretest instrument, a projective technique was used to identify brand-image descriptors for measuring the images of national manufacturer/ 33 Table 3.4. Pretest Results of Brands Selected for Three Test Products (Blue Jeans, Socks, and Oxford Cloth Shirts) for Final Study. Product Blue Jeans Socks Oxford Cloth Shirts NM/D Guess Hanes Guess PL Arizonal Arizonal Arizonal NM/D = national manufacturer/designer brands; PL = private label. ' JCPenney’s brand. designer brands and private labels for the final study. Respondents were asked to imagine the following scenario: he/she is walking down the street and encounters two unknown individuals wearing one of each of the clothing items included in Wearing List 1 (national manufacturer/ designer brands) and Wearing List 11 (private labels) (Appendix A, questions 4 & 5) (Table 3.5). Respondents were asked to rate their perceptions of each person based upon how he/she was dressed. Seven brand-image descriptors were selected Table 3.5. Wearing Lists to Measure Images of National Manufacturer/Designer Brands and Private Labels in Pretest. Wearing List I National manufacturer/designer brand blue jeans (e.g., Levi’s, Guess, or Calvin Klein) National manufacturer/designer brand oxford cloth shirts National manufacturer/designer brand wool sweater National manufacturer/designer brand blazer Wearing List 11 Private-label blue jeans (e.g., Falls Creek, Meijer; Arizona, JCPenney) Private-label oxford cloth shirts Private-label wool sweater Private-label blazer 34 from previous studies and were presented in the format of a five-point semantic differential scale (Belk et al., 1984; Calder & Bumkrant, 1977; Workman, 1988): unattractive-attractive, unsuccessful-successfirl, low status-high status, unpopular- popular, unsociable-sociable, unfashionable-fashionable, economical-affluent. The paired t-test was performed to determine whether or not each of the seven descriptors adequately differentiated the images of national manufacturer/designer brands and private labels. The results show a significantly higher mean for the national manufacturer/ designer brand than for the private label for each of the seven descriptors, as shown in Table 3.6. Therefore, all of the seven brand-image descriptors were retained for the final study. Respondents were also asked whether there were other words/phrases not included among the seven brand-image descriptors, that they would use to describe the persons who wear either national manufacturer/designer brands or private labels. Only six respondents, however, gave answers, and they were not relevant to this study. Thus, the previously selected seven descriptors were used in the final study. 35 Table 3.6. Paired T-Tests of Seven Descriptors to Select Attributes to Measure Images of National Manufacturer/Designer Brands and Private Labels. Variable Mg t-value df NM/D PL Attractiveness 3.86 3.00 4.95*** 49 Status 4.12 2.70 839*“ 49 Popularity 3.94 2.80 7.54*** 49 Success 3.84 3.00 4.81 *"‘* 49 Sociability 3.74 3.22 3.49*** 49 Fashionability 4.28 2.84 6.85*** 49 Extravagance 2.62 3.72 -3.99*** 49 NM/D = national manufacturer/designer brand; PL = private label. semantic differential scale (1-5): 1 = unattractive, low status, unpopular, unsuccessful, unsociable, unfashionable, and extravagant; 5 = attractive, high status, popular, successful, sociable, fashionable, and economical. *** p < .001. Final Study Sample Selection The data for the final study were collected from undergraduate and graduate students at two4 major midwestem universities who were diverse in ethnic and socioeconomic background, gender, and age. College-students were selected as the target 4 T-test analyses confirmed that there were no differences in the demographic characteristics of the two university samples. 36 population, since this age group was found to be brand conscious in previous research (Jolson et al., 1981; Kwon, 1990). Data Collection Prior to initiating the project, approval was obtained from the University Committee on Research Involving Human Subjects. In order to insure a diverse student sample, the researcher distributed questionnaires to undergraduate students from selected general education courses offered in the summer of 1994. Participation was voluntary, and the questionnaires were disseminated prior to and following each class so as not to disrupt the scheduled class time. A brief introduction to the study was provided by the researcher prior to administering questionnaires. The researcher also approached every fifth person at libraries, unions, and student lounges and requested their participation in the study. Data were collected from 300 students over a two-month period, and 256 usable questionnaires were initially analyzed in the study. Subsequent analysis of the level of involvement and brand preference yielded 171 questionnaires with 224 cases for final analysis. Further explanation will be provided in Chapter IV. Instrument A self-report questionnaire included a statement identifying the researchers' university affiliation and an assurance of confidentiality and anonymity. The questionnaire consisted of six major sections: self-esteem, actual self-image and ideal self-image, product involvement, images of national manufacturer/designer-brand and private-label products, brand preference, and demographics. A copy of the survey instrument is in Appendix B. 37 Self-Esteem Measurement The self-esteem measurement for this study was Rosenberg's (1965) ten-item scale, since it evaluates a global sense of self-esteem (Pelharn & Swarm 1989, coefficient alpha = .82). (See Appendix B, questions 1 to 10.) An example of a self-esteem criterion is "On the whole, I am satisfied with myself. " This construct, however, was not analyzed in the study, but was included in the proposed model as an explanatory variable. Measurement of Actual Self-Image_and1deal Self-Imgge Actual self-image and ideal self-image were measured to obtain self- image/product image congruity. A five-point semantic differential scale was used to measure actual self-image and ideal self-image. As shown in Table 3.7, the seven adjectives identified by the pretest were used to describe the actual self- and ideal self- images. To assess actual self-image, each respondent was asked to indicate how she/he Table 3.7. The Descriptors Used to Measure Actual and Ideal Self-Image. unattractive - attractive unsociable - sociable unsuccessful - successful unfashionable - fashionable low status - high status economical - affluent unpopular - popular actually sees herself/himself in regard to these attributes on a five-point semantic differential scale (Appendix B, question 11). The question (Ericksen & Sirgy, 1989) was as follows: "How do you actually see yourself in regard to being ........... ?" While in the 38 case of ideal self-image, each respondent was asked to state how he/she would like to see herself/himself with respect to each of the attributes on a five-point semantic differential scale (Appendix B, question 12). The question (Ericksen & Sirgy 1989) was as follows: "Would you like to see yourself as .............. ?" Involvement Measurement Zaichkowsky's involvement measure (1985) was used (coefficient alpha = .90, in Goldsmith & Emmert, 1991) to measure the three pretest-selected apparel products, blue jeans, socks, and oxford cloth shirts. Zaichkowsky's instrument consisted of 20 word pairs that include three dimensions of consumer involvement: personal, physical, and situational (Table 3.8). The instrument was adjusted from a seven-point scale to a five- point semantic differential scale for consistency in the measurements. Table 3.8. The Twenty Attributes of Involvement Measurement. unimportant - important insignificant - significant of no concern - of concern to me superfluous - vital irrelevant - relevant boring - interesting means nothing - means a lot to me unexciting - exciting useless - useful unappealing - appealing worthless - valuable mundane - fascinating trivial - fundamental nonessential - essential not beneficial - beneficial undesirable - desirable doesn’t matter - matters to me unwanted - wanted uninterested in - interested in not needed - needed 39 Meaflrement of the Images of National Manufacturer/Designer-Brand anid Private-Label Products This research used a projective technique known as the "wearing list5" to assess the images of the national manufacturer/designer-brand and private-label products. Such an indirect approach is recommended by some researchers (Haire, 1950; Holbrook & Hughes, 1978; Workman, 1988) as an appropriate technique for identifying the symbolic value of products, since it is difficult to obtain accurate data on subconscious motivations, unrecognized cultural values, or feelings by direct methods. In this section of the study, six different wearing lists were presented to respondents. Each wearing list (for person 1 through person 6) included three products with national manufacturer/designer or private-label brands. The products included in the final study instrument were selected on the basis of the pretest results (Table 3.9). The seven adjectives identified by the pretest and used for measuring actual and ideal self- irnage were also utilized to describe the national manufacturer/designer-brand and private-label users’ images (Table 3.7). Using a five-point semantic differential scale, respondents were asked to indicate their impressions with regard to these attributes, of each of six persons wearing different brands. The following question (Bloch, 1982) was given to respondents: "Indicate your impressions of person X who is wearing the 3, following 3 items with regard to being ........ 5 In the present study, a projective technique, the shopping-list method (Holbrook & Hughes, 1978), was revised to a “wearing list” in order to measure the images of national manufacturer/designer—brand and private-label wearers. 40 Table 3.9. The Six Wearing Lists for Brand-Image Measurement. Person I Guess Blue Jeans Person 11 Arizona Blue Jeans Oxford Cloth Shirts Oxford Cloth Shirts Socks Socks Person 111 Blue Jeans Person IV Blue jeans Guess Oxford Cloth Shirts Arizona Oxford Cloth Shirts Socks Socks Person V Blue Jeans Person VI Blue Jeans Oxford Cloth Shirts Oxford Cloth Shirts Hanes Socks Arizona Socks _B_r_and Preference Measurement To measure the respondents’ preference for the national manufacturer/designer or the private-label brand for each of three products (blue jeans, socks and the oxford cloth shirts), a slightly modified version of Huddleston & Cassill's (1990) brand orientation measure was adopted. The original measure consisted of three statements that assessed a consumer’s awareness of name-brand products and her/his willingness to purchase these products. The statements were to be rated on a six-point Likert-type scale ranging from "strongly disagree" to "strongly agree.” This study, however, adopted two of these three statements‘5 (questions 20 and 21, as shown in Table 3.10) and included one item measuring consumers’ private-label preference (question 22). The scale was changed to a five-point Likert-type scale ranging from “strongly disagree” to “strongly agree.” 6 The original version of the measurement was revised to fit the present study. The term “brand name” was changed to “national manufacturer/designer brand” and the term “private label” was substituted for “brands with which I am not familiar” or “another brand.” 41 Table 3.10. Statements for Brand Preference Measurement. 20. I buy well-known national manufacturer/designer-brand garments rather than private label-brands. Strongly Dis- Neu- Agree Strongly disagree agree tral agree Blue jeans 1 2 3 4 5 Socks 1 2 3 4 5 Oxford Cloth Shirts 1 2 3 4 5 21. I would rather pay more for well-known national manufacturer/designer brands than to purchase private labels that look similar but cost less. Strongly Dis- Neu- Agree Strongly disagree agree tral agree Blue jeans 1 2 3 4 5 Socks 1 2 3 4 5 Oxford Cloth Shirts 1 2 3 4 5 22. I buy private label-brands rather than well-known national manufacturer/designer- brand garments. Strongly Dis- Neu- Agree Strongly disagree agree tral agree Blue jeans 1 2 3 4 5 Socks 1 2 3 4 5 Oxford Cloth Shirts 1 2 3 4 5 Cronbach’s alpha was utilized to examine the reliability of the two national manufacturer/designer-brand preference items for blue jeans, socks, and oxford cloth shirts. As shown in Table 3.11, the coefficients for each of the three products represent a high level of reliability (blue jeans, .91; socks, .85; and oxford cloth shirts, .90). Therefore, the average of the two national manufacturer/designer brand preference items was calculated and utilized to classify national manufacturer/designer-brand preferrers. 42 The single item measuring private label preference was used to classify the private-label preferrers. Table 3.11. Cronbach’s Alpha Analysis of National Manufacturer/Designer Brand Preference Measurement for Blue Jeans, Socks and Oxford Cloth Shirts. Blue Jeans Socks Oxford Cloth Shirts Reliability Coefficients .91 .85 .90 (Alpha) 43 Theoretical and Operational Definitions Product involvement -- "A person's perceived relevance of the object based on inherent interests, needs, and values" (Zaichkowsky, 1985, p.342). Self-concth -- "The totality of the individual's thoughts and feelings having reference to himself as an object" (Rosenberg, 1979, p. 7). Actual self -- "How a person perceives herself/himself" (Assael, 1992, p. 290). Ideal self -- "How a person would like to perceive herself/himself“ (Assael, 1992, p. 290) Self-esteem motivation -- "Tendency to seek experiences that enhance self- concept" (Sirgy, 1985, p. 197). Self-consistency motivation -- "Tendency for an individual to behave consistently with the view of herself/himself in order to avoid dissonance" (Sirgy, 1985, p.197) Self-esteem -- "A positive or negative attitude toward a particular object, namely, the self" (Rosenberg, 1965, p. 7). Self-image/product image congr_uity (Self-congruity) -- Correspondence between her/his self-concept and the image of a product (brand) (Sirgy, 1982; 1985); specifically, actual congruity or ideal congruity. Actual congr_uig -- "Consistency between the actual self-image and the product (brand) image" (Sirgy, 1985, p. 195). Ideal congruity -- "Agreement between the ideal self-image and product (brand) image" (Sirgy, 1985, p. 195). Brand orientation -- "A consumer's familiarity with branded products and willingness to purchase these products" (Huddleston & Cassill, 1990, p. 256). National manufacturer/designer bra_nd -- "Nationally acknowledgeable manufacturer-owned and prestigious designer brand" ( Diamond & Pintel, 1985, p. 277). 44 Private label -- “Store names, retailers’ own brand names, designer exclusives, exclusive licensed names, or generic labels” (Klokis, 1987, p. 22). Hypotheses and Statistical Analyses : The national manufacturer/designer-brand preferrers will have higher product involvement than the private-label preferrers. : There will be no difference in the actual congruity of national manufacturer/designer brands between the national manufacturer/designer- brand preferrers and the private-label preferrers. : There will be no difference in the actual congruity of private labels between the national manufacturer/designer—brand preferrers and the private label-preferrers. : The national manufacturer/designer-brand preferrers will have higher ideal congruity of national manufacturer/designer brands than will the private- label preferrers. : There will be no difference in the ideal congruity of private labels between the private-label preferrers and national manufacturer/designer-brand preferrers. : The national manufacturer/designer-brand preferrers will have a more favorable symbolic image of national manufacturer/designer brands than of private-label brands. : There will be no difference in the perceived symbolic images of national manufacturer/designer brands and private labels among the private-label preferrers. Data analyses of the instrument responses were conducted using the Statistical Package of the Social Science (SPSS). Two-group discriminant analysis was performed to distinguish national manufacturer/designer-brand preferrers and private-label preferrers 45 on the basis of their product involvement, and the four types of self-image/product image congruity (i.e., actual congruity of national/manufacturer designer brands, actual congruity of private labels, ideal congruity of national manufacturer/designer brands, and ideal congruity of private labels). Thus, H1, H2, H3, H4, and H5 were tested by using two- group discriminant analysis. The effect of symbolic image differences between the national manufacturer/ designer brand and the private label on brand preference could not be determined by discriminant analysis. Because the symbolic image differences between the two brands must be entered as INWD - IPL in the discriminant analysis, differences between the images become indistinguishable when the image score of the private label is greater than that of the national manufacturer/designer brand. Since the outcome would be negative, it would be interpreted as a small difference when, in fact, a large difference exists. If the absolute value of the image differences is used in the analysis, it is unclear which brand is perceived more favorably than the other. Thus, paired sample t-tests were used to test H6 and H7. CHAPTER IV RESULTS AND DISCUSSION Demographic Characteristics of Sample Respondents were asked about the following demographic characteristics: age, gender, ethnic origin, income, and marital status. The profile of the final sample is shown in Table 4.1. The ages in the sample ranged from 18 to 45, with an average age of 22. Females constituted 65% of the sample and males 35%. Most respondents (94%) were single, but 6% were married. The levels of family income were reasonably balanced. The majority of the respondents in the sample were white (61%). Table 4.1. Sample Profile (n = 171). Age Gender Marital n (%) n (%) Status 11 (%) 18 - 23 132 (78) Male 59 (35) Single 158 (94) 24 - 29 34 (20) Female 111 (65) Married 10 (6) 30 - 45 3 (2) Family’s Annual Ethnic Origin Income 11 (%) n (%) $ 30,000 and under 14 (12) Asian 34 (20) 30,001 - 60,000 35 (30) Black 20 (12) 60,001 - 90,000 29 (24) White 103 (61) 90,001 and over 40 (34) Others 13 (7) 46 47 Preliminary Data Preparation To test the hypotheses, the data were prepared using a two-step process. The first step was to classify national manufacturer/designer-brand and private-label preferrers, and the second step was to compute the self-image/product image congruity. Classification of National Manufacturer/Desigger—Brand Preferrers and Private-Label Preferrers Although overall assessments of involvement with specific products were identified through the pretest, product involvement may vary with each individual. Therefore, the product involvement mean (measured by twenty attributes), rated on a five-point semantic-differential scale, was obtained for each of the three products (i.e., the blue jeans, oxford cloth shirts, and socks) for each subject (Appendix B, questionnaire Part III). Among the three products, the product with the highest mean score was identified as the high-involvement product and the product with the lowest mean score was identified as the low-involvement product for each respondent. Only those respondents with mean scores above three for the high-involvement product and below three for the low-involvement product were analyzed in the final study. Thus, for the final study, only the products with the above involvement characteristics were analyzed. Given these research controls, a maximum of two products (cases) per respondent could be included in the analysis. After identifying high and low involvement products for each respondent, national manufacturer/designer—brand preferrers and private-label preferrers were determined. The classification of national manufacturer/designer-brand (NM/D) preferrers was based on 48 the average score from questions 20 and 21 in Appendix B, while classification of private-label (PL) preferrers was based on the score of question 22. Subjects answered each question with regard to the three products, blue jeans, socks, and an oxford cloth shirts (Table 3.10). In the first step, each respondent’s highest scoring product was identified as her/his high-involvement product and, conversely the lowest scoring product was identified as her/his low-involvement product. For example, if a respondent’s high- involvement product was blue jeans, her/his preference for the national manufacturer/ designer brand was determined based on responses concerning blue jeans for questions 20 and 21 and her/his preference for the private label was obtained from the response concerning blue jeans for question 22. In order to assure that only national manufacturer/designer-brand preferrers and private-label preferrers were included in the data analysis, a strict interpretation of their brand preference was utilized. If the NM/D score was above 3 (on a five-point scale), and the score for PL was below 3, the respondent was classified as a national manufacturer/designer-brand preferrer. If the reverse was true, the respondent was classified as a private-label preferrer. The cases were deleted if both the scores of NM/D and PL were below, above, or equal to three; or equal to each other. Thus, a total of 224 cases from 171 respondents were finally classified with 152 high-involvement cases and 72 low-involvement cases (Table 4.2). The frequency distribution for each of the three products for the high- and low- involvement groups for 224 cases is shown in Table 4.2. Blue jeans yielded the highest 49 frequency in the high involvement group, while the socks had the highest frequency in the low involvement group. Table 4.2. The Frequency Distribution of High and Low Involvement Groups by Product Type. High Involvement Group Low Involvement Group (N = 152) (N = 72) Product Percent (%) Percent (%) Blue Jeans 77 4 Oxford Shirt 18 43 Socks 5 53 Computation of Self-Image/Product Impge Congfly To clarify the process of computing self-image/product image congruity (actual congruity and ideal congruity), this section is divided into three parts: first, the images of national manufacturer/designer brands and private labels, second, the actual and ideal self-images, and third, the equations for self-irnage/product image congruity. The Images of National Manufacturer/Designer Brands and Private Labels The image of the national manufacturer/designer brand for each of the three products (blue jeans, socks, and oxford cloth shirts) was measured by questions 14, 16, and 18, while the image of the private label for each of the three products was measured by questions 15, 17, and 19. Questions 14 and 15 were about blue jeans, questions 16 and 17 about an oxford cloth shirts, and questions 18 and 19 about socks. These 50 questions were in a format presenting seven items on a five-point semantic differential scale. For each respondent, the symbolic images of national manufacturer/designer brands and private labels were obtained based on the apparel products identified as the high-involvement and the low-involvement product. For example, if the high involvement product of a respondent was blue jeans, question 14 was used to obtain the image of the national manufacturer/designer brand, and question 15 was used to obtain the image of the private label. If the low involvement product of a respondent was oxford cloth shirts, question 16 was used to obtain the image of the national manufacturer/ designer brand, and question 17 was used to obtain the image of the private label. Thus, the images of national manufacturer/designer brands and private labels were obtained for only two of the three products (one high and one low involvement product for each respondent), as previously identified. _EQIEtions for Actual Congruity and Ideal Congrpitv Actual self-image was measured by question 11 for actual congruity and ideal self-image was measured by question 12 for ideal congruity. The format for questions 11 and 12 consisted of a five-point semantic differential scale. Each respondent’s actual congruity and ideal congruity scores for the national manufacturer/designer—brand products and the private-label products were calculated by following the method used in Ericksen & Sirgy's research (1989). The congruence between the actual self-image (Appendix B, question 11) and the image of the national manufacturer/designer—brand (questions 14, 16, or 18) or the private-label product 51 (questions 15, 17, or 19) was obtained by computing the average of the absolute arithmetical difference between the actual self-image and the image of the national manufacturer/designer—brand or private-label product for all seven image attributes. Mathematically stated, AC1: = [ZIBIr‘k — AS111: fl/n i=1 Where: ACk = Actual congruity score for individual (k) BIik = Brand image (i) of individual (k) ASIik = Actual self-image (i) of individual (k) n = number of image attributes (i) (i = 7). Similarly, an ideal congruity score was computed by using the average of the absolute arithmetical difference between the ideal self-image and the brand image for all seven image attributes. [CI = [flan — 151,-. |]/ n i=1 Where: ICk = Ideal congruity score for individual (k) Blik = Brand image (i) of individual (k) ISIik = Ideal self-image (i) of individual (k) n = number of image attributes (i) (i = 7). In this study, the scores of actual and ideal congruity ranged from zero to four, and as the score decreased, the level of congruity increased. For consistency with the other measurements in the discriminant analysis, the obtained scores of the actual and ideal congruity were reversed and recoded, so a score of five represented the highest congruity, and a score of one represented the lowest congruity. 52 Computation of Symbolic Images of Natiopd Manufacturer/Desigper Brands and Private Labels The symbolic images of national manufacturer/designer brands and private labels were obtained by calculating the average of the seven descriptors (unattractive-attractive, low status-high status, unpopular-popular, unsuccessful-successful, unsociable-sociable, unfashionable-fashionable, extravagant-economical), which were measured on a five- point semantic differential scale. As explained previously, if a respondent identified her/his high or low involvement with a product such as blue jeans, questions 14 and 15 were used to measure the image of the national manufacturer/designer brand and the image of the private label, respectively. In the case of oxford cloth shirts, questions 16 and 17 were utilized to measure the image of the national manufacturer/designer brand and the image of the private label, respectively; and in the case of socks, questions 18 and 19 were employed to measure the image of the national manufacturer/designer brand and the image of the private label, respectively. Statistical Results of Hypotheses Test H, : The national manufacturer/designer—brand preferrers will have higher product involvement than the private-label preferrers. H2 : There will be no difference in the actual congruity of national manufacturer/designer brands between the national manufacturer/designer- brand preferrers and the private-label preferrers. H3 : There will be no difference in the actual congruity of private labels between the national manufacturer/designer-brand preferrers and the private label-preferrers. 53 H4 : The national manufacturer/designer-brand preferrers will have higher ideal congruity of national manufacturer/designer brands than will the private- label preferrers. ' H5 : There will be no difference in the ideal congruity of private labels between the private-label preferrers and national manufacturer/designer-brand preferrers. H6 : The national manufacturer/designer—brand preferrers will have a more favorable symbolic image of national manufacturer/designer brands than of private-label brands. H7 : There will be no difference in the perceived symbolic images of national manufacturer/designer brands and private labels among the private-label preferrers. Two-Group Discriminant Analysis Two-group discriminant analysis was performed to distinguish national manufacturer/designer—brand preferrers and private-label preferrers on the basis of their product involvement and the four types of self-image/product image congruities (i.e., actual congruity of national/manufacturer designer brands, actual congruity of private labels, ideal congruity of national manufacturer/designer brands, and ideal congruity of private labels). After checking the assumptions of normality and equality of group covariance matrices7, the simultaneous method was used, since the objective of the study was to determine the discriminating capabilities of the entire set of independent variables, 7 The assumptions for discriminant analysis are 1) each group must be a sample from “a multivariate normal population”; and 2) “the population covariance matrices” must all be equal (Norusis, 1993). The normal probability plots for each variable showed that the data were approximately normally distributed. To check the population covariance matrices, Box’s M was performed. The results showed that the group covariance matrices were equal. 54 rather than to consider intermediate results based only on the most discriminating variables. The cross-validation approach was used in this study to validate the discriminant function (Hair, Anderson, Tatham, & Black, 1992). That is, the total group was divided so that approximately two-thirds of the respondents were placed in the analysis sample, and one-third in the holdout sample. In the holdout group, the proportion of national manufacturer/designer-brand preferrers to private-label preferrers was the same as for the total sample distribution. Thus, 130 cases were randomly selected for the analysis sample, and 94 different cases were chosen for the holdout sample. The multivariate aspects of the model are reported under the heading “Canonical Discriminant Functions,” as shown in Table 4.3. The result of the chi-square shows that the discriminant function is highly significant (p < .0001) and indicates a canonical correlation of .4337. According to Hair et al. (1992), this correlation can be squared (.4337)2, which means that 18.81% of the variance in the dependent variable can be explained by this model. This relatively low percentage indicates that the psychological characteristics of consumers, i.e., product involvement, and the four types of self- image/product image congruities (actual congruity of national manufacturer/designer brands, actual congruity of private labels, ideal congruity of national manufacturer/ designer brands, and ideal congruity of private labels), explain only a small amount of variance in their brand preferences. 55 Table 4.3. The Results of Canonical Discriminant Functions for National Manufacturer/ Designer-Brand Preferrers and Private-Label Preferrers. Eigenvalue Canonical Wilk’s Chi-Square DF Correlation Lambda Enter Method .2317 .4337 .8119 26.157**** 5 **** p< .0001. The group centroids, which are the mean values for the discriminant scores for each of the national manufacturer/designer preferrers and private-label preferrers, are shown in Table 4.4. The group centroids are important in that the test for the statistical significance of the discriminant function is a generalized measure of the distance between the group centroids (Hair et al., 1992). Table 4.4. The Group Centroids of National Manufacturer/Designer-Brand Preferrers and Private-Label Preferrers for Analysis Group. Group Centroid NM/D Preferrers .3963 PL Preferrers - .5758 NM/D = national manufacturer/designer brand; PL = private label. The cutting score is the criterion against which each individual’s discriminant score is judged to determine into which group the individual should be classified (Hair et al., 1992). Because the groups are not of equal size in this study, a weighted average of 56 the group centroides provided an optimal cutting score. Thus, the cutting score for this study was zero. Table 4.5 provides evidence of the predictive validity of the model. The classification analysis compares actual with predicted brand preference behavior for the respondents in this study. As shown, 70.0 % of the analysis sample is correctly classified by the discriminant model, and 69.2 % of the holdout sample is correctly classified. To check the classification accuracy relative to chance, Press’s Q statistic was analyzed. As shown in Table 4.6, the results of Press’s Q statistic indicated that the predictions for both the analysis and holdout samples were significantly better than chance, which would be a correct classification rate of 50 % (p < .01 in each case). 57 Table 4.5. Classification Matrices of Two-Group Discriminant Analysis of National Manufacturer/Designer-Brand Preferrers and Private-Label Preferrers for Both Analysis and Holdout Samples“. Classification Results: Analysis Sampleb No. of Predicted Group Membership Actual Group Cases 1 2 Group 1 77 48 29 62.3 % 37.7 % Group 2 53 10 43 18.9 % 81.1 % Classification Results: Holdout Samplec No. of Predicted Group Membership Actual Group Cases 1 2 Group 1 60 38 22 63.3 % 36.7 % Group 2 34 7 27 20.6 % 79.4% Group 1: national manufacturer/designer-brand preferrers. Group 2: private-label preferrers. a. n == 224 b. percent of “correctly classified grouped” cases: 70.0 %. 0. percent of “correctly classified grouped” cases: 69.2 %. 58 Table 4.6. The Results of Press’s Q Statistic to Check Classification Accuracy Relative to Chance. Press’s Q Analysis Sample 208’ ** Holdout Sample 13.8b ** computation: Press’s Q = [N - (n * K)]2/N (K - 1), where N = total sample size, n = number of observations correctly classified, K = number of groups. a Press’s Q = [130 - (91 * 2)]2/130 (2 -1). b Press’s Q = [94 - (65 * 2)]2/94 (2 - 1). ** p < .01; critical value at a significance level of .01 = 6.63 (Hair et al., 1992, p. 106). The results concerning the discriminating capabilities of independent variables are summarized in Tables 4.7 and 4.8. The independent variables can be ranked in terms of both their weights and loadings to represent their discriminating power. Since the loadings are considered more valid than the weights (Hair et al., 1992), the independent variables were ranked on the basis of the loadings. These results are discussed separately following each of the relevant hypotheses. Hypothesis 1 The national manufacturer/designer-brand preferrers will have higher product involvement than the private-label preferrers. The results of the univariate analysis of variance indicate that product involvement is a significant discriminating variable of brand preference (p < .0001 in Table 4.7). As shown in Table 4.8, product involvement is the most influential variable in discriminating the national manufacturer/designer-brand preferrers and the private- 59 label preferrers (with a loading of .7203). The national manufacturer/designer-brand preferrers (with a mean of 3.8) have higher product involvement than private-label preferrers (with a mean of 3.2) (Table 4.8). Therefore, the null hypothesis of no difference was rejected. Table 4.7. Univariate Analyses of Variance to Test Equality of Group Means between National Manufacturer/Designer-Brand Preferrers and Private-Label Preferrers. Independent Wilk’s Univariate Variables Lambda F Ratio Product Involvement .8927 15.3 898”" Ideal Congruity of NM/D .9211 10.9667“ Ideal Congruity of PL .9617 5.0964* Actual Congruity of PL .9792 2.7201 Actual Congruity of NM/D .9978 .2852 NM/D = national manufacturer/designer brand; PL = private label. * p < .05. ** p < .01. **** p < .0001. Hypothesis 2 There will be no difference in the actual congruity of national manufacturer/ designer brands between the national manufacturer/designer—brand preferrers and the private-label preferrers. The results of the univariate analysis of variance indicate that there is no significant difference between the actual congruity of national manufacturer/designer brands for the two groups as shown in Table 4.7. Thus, actual congruity of national 60 manufacturer/designer brands was found to have little discriminating capability in brand preference, therefore, hypothesis 2 was not rejected. Table 4.8. Discriminant Analysis of National Manufacturer/Designer-Brand Preferrers Versus Private-Label Preferrers. Mean Responses by Group Variables Discriminant“ Standardizedb Rank NM/D PL Loadings Weights Preferrers Preferrers (n = 77) (n = 53) Product InvolvementI .7203 .4506 1 3.8 3.2 Ideal Congruity2 .6808 .7070 2 4.0 3.7 of NM/D Ideal Congruity2 -4145 -.5911 3 3.3 3.6 of PL Actual Congruityz -.3028 -.0094 4 4.0 4.2 of PL Actual Congruityz -.O981 .0236 5 4.1 4.1 of NM/D NM/D = national manufacturer/designer brand; PL = private label. a structure matrix: pooled within-group correlation between discriminating variables and canonical discriminant functions. b standardized canonical discriminant function coefficients. 1 = low involvement; 5 = high involvement. 1 = low congruity; 5 = high congruity. Hypothesis 3 There will be no difference in the actual congruity of private labels between the national manufacturer/designer—brand preferrers and the private label-preferrers. The results of the univariate analysis of variance indicate that there is no significant difference in the actual congruity of private labels for the two groups as shown 61 in Table 4.7. From the results, it appears that the actual congruity of private labels has no significant distinguishing ability in brand preference. Thus, hypothesis 3 was not rejected. Hypothesis 4 The national manufacturer/designer—brand preferrers will have higher ideal congruity of national manufacturer/designer brands than will the private- label preferrers. The results of the univariate analysis of variance indicate that there is a significant difference between the ideal congruity of national manufacturer/designer brands for the two groups (p < .01) as shown in Table 4.7. In Table 4.8, it is indicated that the ideal congruity of national manufacturer/designer brands was the second most discriminating variable (with a loading of .6808). The ideal congruity of national manufacturer/ designer brands for the national manufacturer/designer—brand preferrers was 4.0, while the mean for the private-label preferrers was 3.7. Thus, the national manufacturer/ designer-brand preferrers have higher ideal congruity of national manufacturer/designer brands than the private-label preferrers. Therefore, the null hypothesis of no difference was rejected. Hypothesis 5 There will be no difference in the ideal congruity of private labels between the private-label preferrers and national manufacturer/designer-brand preferrers. The results of the univariate analysis of variance indicate that there is a significant difference between the ideal congruity of private labels for the two groups (p < .05) as 62 shown in Table 4.7. The Table 4.8 shows that ideal congruity of private labels was the third most discriminating factor (with a loading of -.4145). Since the mean of the ideal congruity of private labels for the national manufacturer/designer-brand preferrers was smaller (3.3) than the mean for the private-label preferrers (3.6), the loading was negative. Thus, private-label preferrers have significantly higher ideal congruity of private labels than the national manufacturer/designer—brand preferrers. Therefore, hypothesis 5 was rejected. The finding indicates that the ideal congruity of private label is a predictor of brand preference, which is not consistent with the model. That is, the model proposed that both the national manufacturer/designer—brand preferrers and the private-label preferrers would have low ideal congruity of the private label. It should be noted that the means of the ideal congruity of private labels for both groups (N M/D, PL) were relatively lower than the actual congruity of the brands (NM/D, PL) (Table 4.8). Paired Sample T-Test Paired sample t-tests were used to examine the effects on brand preference of the perceived symbolic image differences between the national manufacturer/designer brands and the private labels. This is tested in H6 and H7. The images of national manufacturer/designer brands and private brands were obtained by calculating the mean of seven descriptors measured on a five-point semantic differential scale (unattractive- attractive, unsuccessful-successful, low status-high status, unpopular-popular, unsociable- sociable, unfashionable-fashionable, economical-affluent) (Appendix B, questions 14 - 19). Thus, a score of one represents an unfavorable image and five represents a favorable 63 image. The tests were performed after checking the assumptions of normality and homogeneity of variance8 for each of the national manufacturer/designer-brand preferrers and private-label preferrers. HVDOthesis 6 The national manufacturer/designer-brand preferrers will have a more favorable symbolic image of national manufacturer/designer brands than of private-label brands. The results of a paired t-test are shown in Table 4.9. One hundred and thirty- ' seven cases of the national manufacturer/designer—brand preferrers were analyzed, and there was a significant difference in the respondents’ perceptions of the symbolic images of national manufacturer/designer brands and private brands (p < .001 ). The national Table 4.9. The Results of Paired T-tests: Perception of Symbolic Images between NM/D and PL for National Manufacturer/Designer-Brand Preferrers. National Manufacturer/Designer-Brand Preferrers Variable Number of Mean t-value df Pairs Image NM/D 137 3.7 11.45*** 136 PL 2.9 NM/D = national manufacturer/designer brand; PL = private label. 1 = not favorable; 5 = favorable. *** p < .001, one-tailed. 3 The normal probability plots for each variable showed the data were approximately normally distributed. To check the homogeneity of variance, Levene’s test was performed. The results showed that the variances of each variable was equal. 64 manufacturer/designer—brand preferrers (with a mean of 3.7) perceived the image of the national manufacturer/designer brand more favorably than that of the private label (with a mean of 2.9). Thus, the null hypothesis of no difference was rejected. Hymthesis 7 There will be no difference in the perceived symbolic images of national manufacturer/designer brands and private labels among the private-label preferrers. The results of a paired t-test are presented in Table 4.10. When eighty-seven cases among the private-label preferrers were analyzed, a significant difference was found in their perception of the symbolic images of the two types of brands at p < .01. That is, private-label preferrers (with a mean of 3.3) perceived the image of the national manufacturer/designer brand more favorably than that of the private label (3.2). Thus, hypothesis 7 was rejected. Table 4.10. The Results of Paired T-tests: Perception of Symbolic Images between NM/D and PL for Private-Label Preferrers. Private-Label Preferrers Variable Number of Mean t-value df Pairs Image NM/D 87 3.3 3.27" 136 PL 3.2 NM/D = national manufacturer/designer brand; PL = private label. 1 = not favorable; 5 = favorable. ** p < .01, two-tailed. 65 Discussion The results of discriminant analysis revealed that product involvement influenced consumers’ preference for national manufacturer/designer brands or private labels. These results are consistent with previous research (Holstius & Paltschik, 1983; Kwon, 1990). From the results, it appears that consumers who are highly involved with a product are more likely to prefer national manufacturer/designer brands, and those who are less involved with a product are more likely to prefer private labels. The results of discriminant analysis indicated that ideal congruity was a better predictor than actual congruity for both the national manufacturer/designer brands and the private labels among the four types of self-image/product image congruities. That is, individuals whose ideal self-images were highly congruent with the image of the national manufacturer/designer brands were identified as national manufacturer/designer-brand preferrers, and those whose ideal self-images were congruent with the image of the private labels were identified as private-label preferrers. This result may reflect the fact that self-esteem is a more fundamental need than self-consistency (Elliot, 1986; Jones, 1973; Kaplan, 1980), and that the self-esteem motivation is reflected in ideal congruity. That is, people may select either national manufacturer/designer brands or private-label brands whose images are congruent with their ideal self-concepts to protect and enhance their self-esteem. The canonical correlation indicated that the psychological factors of involvement and self-image/product image congruities explained only a small amount of the variance of brand preference. This finding may indicate consistency with the previous researchers 66 who found that functional congruity - the degree to which a brand meets consumers' performance-related evaluative criteria - is a better predictor of consumer behavior than self-congruity (Sirgy & Samli, 1985; Sirgy et al., 1991). In view of the previous findings related to functional congruity, it may help to explain why the hypothesized psychological factors accounted for only a small amount of the variance of brand preference. The results of paired sample t-tests indicated that the perceived symbolic images of national manufacturer/designer brands and private labels influenced a consumer’s brand preference. The national manufacturer/designer-brand preferrers perceived greater differences in the symbolic image between these two brands than did the private-label preferrers. This finding supports Solomon’s (1983) notion; the symbolism of prestigious products often affects consumers’ product preference and adoption. This suggests that some consumers preferred to purchase national manufacturer/designer-brands because the symbolic images of these brands were more favorable than those of private labels. In the case of the private-label preferrers, although the results of paired sample t- tests indicated a statistically significant difference in their perceptions of the symbolic images of the national manufacturer/designer brands and the private labels, the difference between the means of these two groups was very small. In fact, with means of 3.3 and 3.2, respectively, these represent neutral responses. Thus, the statistical significance may result from the large sample size, and the private-label preferrers may not be as aware of symbolic image differences between national manufacturer/designer brands and private labels as are the national manufacturer/designer-brand preferrers. Therefore, from the 67 results it appears that the private-label preferrers are relatively insensitive to the symbolism of national manufacturer/designer—brand and private-label products. These findings suggest that the symbolic characteristics of apparel products were not important reasons for purchasing the product. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS This chapter is divided into four sections. Section one contains a summary of the research methodology, data analysis, and findings of the research. The second section presents the conclusions of this research study. The third section discusses limitations associated with the research. Lastly, section four contains the implications of the findings and the recommendations for the national manufacturer/designer-brand and private-label apparel industries and future research. Summary of Research Methods and Data Analyses The overall intent of this study was to investigate the psychological origins of consumers' preference for national manufacturer/designer-brand and private-label apparel. The research was based on self-concept theory, and its objective was to distinguish national manufacturer/designer-brand preferrers from private-label preferrers through an investigation of self-image/product image congruity, product involvement, and the symbolic images of national manufacturer/designer brands and private labels. The full study consisted of a pretest and a subsequent study. In the pretest, data were collected fi'om fifty college students to identify the sample products and brand names. Based on the results, the questionnaire was developed by adapting established questionnaires (Rosenberg, 1965; Ericksen and Sirgy, 1989; Zaichkowsky, 1985; Huddleston & Cassill, 1990). The data were collected through a structured self-report 68 69 survey instrument consisting of six sections. These sections utilized five-point Likert- type scales and semantic differential scales. In the final study, data were collected at two major midwestem universities from 300 undergraduate and graduate students who were diverse in ethnic and socioeconomic background, gender, and age. The researcher distributed questionnaires to undergraduate students from selected general education classes and also distributed questionnaires outside of classrooms during the summer semester of 1994. Finally, 171 usable questionnaires were analyzed in the study. Data analyses of the self-report instrument were conducted using the Statistical Package of the Social Science (SPSS). Descriptive analyses were used to identify demographic characteristics of samples. Discriminant analysis and paired sample t-tests were employed to test the hypotheses. Results of the statistical analyses are presented in Table 5.1. Summary of Findings Relationship between Involvement and Brand Preference Product involvement had a significant relationship with brand preference. Consumers who were highly involved with a product were more likely to prefer national manufacturer/designer brands, and those who were less involved with a product were more likely to prefer private labels. Thus, the finding suggests involvement as an important antecedent of brand preference. 70 Table 5.1. Summary of Hypotheses Testing. Hypothesis Test Result H1 Discriminant Analysis Dependent Variable = Brand Preference Significant Independent Variable = Product Involvement H2 Discriminant Analysis Not Dependent Variable = Brand Preference Significant Independent Variable = Actual Congruity of NM/D H3 Discriminant Analysis Not Dependent Variable = Brand Preference Significant Independent Variable = Actual Congruity of PL H4 Discriminant Analysis Dependent Variable = Brand Preference Significant Independent Variable = Ideal Congruity of NM/D H5 Discriminant Analysis Dependent Variable = Brand Preference Significant Independent Variable = Ideal Congruity of PL H6 Paired Sample T-Test Significant Among National Manufacturer/designer-brand preferrers Dependent Variable = Perceived Symbolic Images of National Manufacturer/Designer Brands and Private Labels H7 Paired Sample T-Test Significant Among Private-Label Preferrers Dependent Variable = Perceived Symbolic Images of National Manufacturer/Designer Brands and Private Labels NM/D = national manufacturer/designer brand; PL = private label. 71 Relationship between Four Types of Self-Ingge/Product Image Congrg'ties and Brand Preference Among the four types of self-image/product image congruities (actual congruity of national manufacturer/designer brands, actual congruity of private labels, ideal congruity of national manufacturer/designer brands, and ideal congruity of private labels) ideal congruity of national manufacturer/designer brands and ideal congruity of private labels were found to be predictors of brand preference. The ideal self-images of the national manufacturer/designer—brand preferrers were more congruent with the image of the national manufacturer/designer brands than were those of private-label preferrers. On the other hand, the ideal self-images of the private-label preferrers were more consistent with the images of the private labels than were those of the national manufacturer/ designer-brand preferrers. Relationship between Brand Preference fid Perceptions of Symbolic Images of Na_ti9_rgl_ Manufacturer/Designer Bragls_ and Private Labels Both national manufacturer/designer—brand preferrers and private-label preferrers perceived users of national manufacturer/designer brands more positively than they perceived users of private labels. Although the results for the private-label preferrers indicated a statistically significant difference in their perceptions of the symbolic images of national manufacturer/designer brands and private labels, the difference between the mean of the national manufacturer/designer-brand images and those of the private-label images were very small. This suggests that the private-label preferrers may not have perceived the product symbolism of these two types of brands as differently as did the national manufacturer/designer-brand preferrers. 72 Conclusions Results of this study provide insights into the psychological origins of consumers’ preference for national manufacturer/designer brands or private labels. Several conclusions can be drawn from an analysis of these findings. First, the findings revealed that the national manufacturer/designer—brand preferrers had higher product involvement than did the private-label preferrers. Both national manufacturer/designer—brand preferrers and private-label preferrers perceived the image of the national manufacturer/designer brand more favorably. Therefore, it can be concluded that product involvement and perception of the symbolic images of these two brands are significant psychological predictors of consumers’ preference for national manufacturer/designer brands or private labels. Second, the findings suggest that the ideal congruity of national manufacturer/ designer brands and the ideal congruity of private labels had discriminating power for brand preference. That is, national manufacturer/designer-brand preferrers are indivi- duals whose ideal self-images are highly congruent with the images of the national manufacturer/designer brands; private-label preferrers are those whose ideal self-images are congruent with the images of private labels. Therefore, it can be concluded that idea] congruity is a better predictor than actual congruity for both national manufacturer/ designer brands and private labels among the four types of self-image/product image congruities. That is, ideal self-image (how a person would like to perceive herself/himself) has a dominant role in consumer decision making. Enhancing self- esteem through identifying with the symbolic image of products appears to be more 73 important than strengthening self-consistency through the actual self-concept (how a person perceives herself/himself). Lastly, the findings indicated that psychological factors -- involvement and self- image/product image congruity -- explained only a small amount of the variance of brand preference. Thus, it is concluded that the proposed model, including product involvement and self-image/product image congruity, has limited relevancy in predicting consumers’ preference for national manufacturer/designer brands and private labels. Given these conclusions, the proposed model (Figure 2.1) was partially supported in that product involvement and the symbolic brand images of national manufacturer/designer brands and private labels were significant predictors of brand preference. However, the model proposed that only ideal congruity with national manufacturer/designer brands would be a significant predictor of brand preference among the four types of the self-image/product image congruities, yet the ideal congruity with private labels was also found to be a significant predictor. The model proposed that among national manufacturer/designer-brand preferrers, the actual congruities with national manufacturer/designer brands and with private labels would vary according to the level of self-esteem. It also proposed that these consumers would have a high level of ideal congruity with national manufacturer/designer brands and a low level of ideal congruity with private labels. In addition, the model proposed that the private-label preferrers would have a low level of actual and ideal congruities with national manufacturer/designer brands and private labels regardless of their level of self-esteem. Although this study did not test these relationships (self-esteem, image 74 congruity, and brand preference), it is possible to make inferences based on the findings of the study. First, the national manufacturer/designer-brand preferrers had a high level of ideal congruity with national manufacturer/designer brands and had a low level (relative to the other types of image congruities) of ideal congruity with private labels. Secondly, the private-label preferrers had a high level of actual congruities with national manufacturer/designer brands and private labels and a relatively high level of ideal congruities with national manufacturer/designer brands and private labels. In conclusion, therefore, the findings of this study partially supported the proposed model. Limitations This study involved the use of survey research methodology. Although this is a frequently used research mode, several weaknesses can be associated with it (Abrahamson, 1983; Babbie, 1992). Specifically, Abraharnson (1983) discussed the difficulty of creating a measurement that deals with sensitive issues. This study examined the respondents’ actual self-images and ideal self-images. The validity of their responses may be questionable, since self-concept is a sensitive topic. In addition, the subjects’ responses to questions about their perceptions of the symbolic images of national manufacturer/designer brands and private labels may not have been candid, since many people are taught not to judge others by appearances. Thus, even though a person perceived the images of the users of national manufacturer/designer brands and the users of private labels differently, he/she may have been reluctant to give forthright answers. Secondly, this study obtained a relatively high rate of neutral responses (a score of three on a five-point scale) on the rating of brand-image perception among the national 75 manufacturer/designer-brand preferrers and the private-label preferrers. The reasons for these results are unclear. The respondents may not in actuality have perceived the symbolic image differences between national manufacturer/designer brands and private labels, or they may have been indifferent to the product and the brand. If the respondents did not perceive image differences, it may have been true that the functional attributes of private labels have been significantly improved and, thus, on balance the importance of the symbolism of national manufacturer/designer-brand products has been diminished. It is also possible that the descriptors selected in the study did not accurately reflect the perceived image differences among these brands. Therefore, more refined measurements are required for future studies. Thirdly, given the modes of statistical analysis used in the study (discriminant analysis and paired sample t-tests), it was not possible to examine the proposed causal relationships among variables. Therefore, further study is needed to investigate the proposed model through causal modeling. Lastly, in considering the results of the present study, generalization should be cautious, since the assessed sample was limited to college students. It should be recognized that there are inherent limitations associated with generalizing the findings of this study beyond its geographic, demographic limits, or applying the findings to products and brand names other than those surveyed. Implications and Recommendations The results of this study revealed that the ideal self-image/product image congruity and the perceived symbolic image differences between national manufacturer/ 76 designer brands and private labels were significant predictors of consumers’ brand preference. This may reflect the importance of self-esteem rather than self-consistency in a consumer’s decision making. These results, however, do not present a definitive explanation of the role of the antecedent variables in brand preference. Thus, the findings of this study recommend that promotional strategies for the national manufacturer/ designer-brand apparel industry balance the self-esteem attributes of a brand with the product’s functional aspects, such as product quality and price. The results of the study suggested that national manufacturer/designer-brand preferrers were those who had high involvement with the product, and private-label preferrers had low involvement with the product. In previous research, highly involved consumers were characterized as 1) actively seeking information, 2) making decisions based on multiattributes of brands, 3) being influenced in their brand choice by reference groups, such as friends and relatives, 4) not being significantly influenced by advertisement, and 5) relating their personality and life-style characteristics to their behavior (Assael, 1992; Urban & Star, 1991). Therefore, this study recommends that marketing strategies for the national manufacturer/designer-brand apparel industry be focused on both flmctional and symbolic attributes of brands. In addition, this industry should identify the source of information as well as the role of advertising and personal selling (Urban & Star, 1991). On the other hand, less involved consumers are characterized as 1) passively gathering information, 2) making decisions based on a few brand attributes, 3) being strongly influenced in brand choice by familiarity with brands, 4) not being significantly 77 influenced in brand choice by reference groups, 5) being influenced by advertisements, and 6) not relating their personality and life style characteristics to their behavior (Assael, 1992; Urban & Star, 1991). Therefore, marketing strategies for the private-label apparel industry should be focused on the functional attributes of brands, especially price, since it is an important attribute for low-involvement consumers. In addition, the effect of advertising is important to these consumers. Thus, repetition of advertisements is important, and advertisements should focus on a few key points, rather than general attributes of brands. Television, rather than print media, may be a good means of communicating with these consumers (Assael, 1992). Some researchers suggest that the functional congruity and self-image/product image congruity should be examined simultaneously to establish a more complete model for determining consumers’ brand choice (Malhotra, 1988; Sirgy et al., 1991). The results of this study appear to support this notion, therefore, it is recommended that further research be designed to establish a model of consumers’ preferences for national manufacturer/designer brands and private labels that includes functional congruity as well as self-image/product image congruity. The present study involved only apparel products and college students, who tend to have homogeneous characteristics. Thus, a final recommendation is that further study build upon this effort by assessing more diverse population, and more products and brands. APPENDICES APPENDIX A. PRETEST QUESTIONNAIRE 78 APPENDIX A. PRETEST QUESTIONNAIRE Part I. A. Apparel Product Selection 1. The purpose of this study is to measure a person’s involvement or interest in different clothing items. In order to measure this, we need you to judge each of the following ten clothing items against three descriptive scales according to how YOU perceive the products presented below. (Blue Jeans) important _________ unimportant interested in _:_:_:_:_ uninterested in needed _:_:_:_:_ not needed (Wool Sweater) important _ : _ : _ : _ : _ unimportant interested in _ : _ : _ : _ : _ uninterested in needed _:_:_:_:_ not needed (Oxford Cloth Shirt) important _ : _ : _ : _ : _ unimportant interested in _ : _ : _ : _ : _ uninterested in needed _:_ : _:_ : _ not needed (Blazer) important _ . _: _____ unimportant interested in _ : _ : _ : _ : _ uninterested in needed _ : _ : _ : _ : _ not needed (Raincoat) important _ : _ : _ : _ . _ unimportant interested in _ : _ : _ : _ : _ uninterested in needed _ : _ : _ : _ : _ not needed (T-shirt) important __ : _ : _ : _ . _ unimportant interested in : : : : uninterested in needed : : : : notneeded 79 (Jacket) important _ _:_:_._ unimportant interested in _ : _ : _ : _ : _ uninterested in needed _:_:_:_:_ not needed (Socks) important _________ unimportant interested in _ : _ : _ : _ : _ uninterested in needed _:_:_:_:_ not needed (Tennis Shoes) important _ : _ : _ : _ : _ unimportant interested in _ : _ : __ : _ : _ uninterested in needed _ : _ : _:_ : _ not needed (Sweatshirt) important _ : _ : _ : _ : _ unimportant interested in _ : _ : _ : __ : __ uninterested in needed : : : : not needed 2. What are the most popular national manufacturer/designer brands among college students for each or the following ten clothing items? (National manufacturer/designer brands mean nationally acknowledgeable manufacturer - owned and prestigious designer brands.) (Private labels mean retailers” or wholesalers’ own brands.) National Manufacturer/Designer Brand Blue Jeans Wool Sweater Oxford Cloth Shirts Blazer Raincoat T-shirt Jacket 80 Socks Tennis Shoes Sweatshirt 3. We are interested in your perception of the following private-label brands. For each brand, indicate your perception of familiarity, status, popularity, and fashionability. (Cambridge Classic, Partners: Mervyn’s Brand) unfamiliar _: _: _: _: _ familiar low status _: _: _: _: _ high status unpopular ______ popular unfashionable : : : : fashionable (Basic Trends, Sostanza: Target’s Brand) unfamiliar _: _: _: _: _ familiar low status _: _: _: _: _ high status unpopular ______ popular unfashionable : : : : fashionable (Arizona: JCPenney's Brand) unfamiliar _: _: _: _: _ familiar low status _: _: _: _: _ high status unpopular ______ popular unfashionable : : : : fashionable (Falls Creek, Corner Stone: Meijer’s Brand) unfamiliar _: _: _: _: _ familiar low status _: _: _: _: _ high status unpopular _______ popular unfashionable : : : : fashionable (Brittina Jean, Jaclyn Smith: Kmart’s Brand) unfamiliar _: _: _: _: _ familiar low status _: _: _: _: _ high status unpopular ________ popular unfashionable : : : : fashionable 81 Realizing that you may buy private-label brands other than those listed above (Question 3), please list two private labels that you most frequently buy when buying any of the ten apparel items (Blue Jeans, Wool Sweater, Oxford Cloth Shirt, Blazer, Raincoat, T-shirt, Jacket, Socks, Tennis Shoes, and Sweatshirt). Please write in the private-label brand on the line below and indicate the type of product for which you would buy this brand. Also, indicate your perception of each private label with regard to status, popularity, and fashionability. (Private-Label Brand) (Type of Product) low status _: _z _: _: _ high status unpopular ________ popular unfashionable _: _: _: _: _ fashionable (Private-Label Brand) (Type of Product) low status _: _: _: _: _ high status unpopular _________ popular unfashionable : : : : fashionable Part I. B. Brand Image Descriptors Imagine that you are walking down a street and encounter two individuals you do not know, wearing one of each of the following clothing items included in Wearing List I and Wearing List II. The Wearing List I and Wearing List II have the same clothing items except that they are either national manufacturer/designer brands or private labels. Please read the following statements and indicate your perceptions of each person based upon how helshe is dressed. Wearing list I national manufacturerldesigner-brand jeans (e.g., Levi’s, Guess, or Calvin Klein), national manufacturerldesigner-brand oxford cloth shirt, national manufacturerldesigner-brand wool sweater, national manufacturerldesignor-brand blazer. 82 4. Indicate your impressions of the person to whom such clothing belonged with regard to being ....... (as listed in Wearing List I) ? unattractive _: _: _: _2 _ attractive unsociable _: _: _: _: _ sociable unsuccessful _:_: _:_: _ successful unfashionable _: _: _: _: _ fashionable low status _:_:_:_:_ high status economical _:_: _: _: __ affluent unpopular _: _: _: _: _ popular Wearing list II private-label jeans (e.g., Falls Creek: Meijer’s Brand, Arizona: JCPenney’s Brand), private-label oxford cloth shirt, private-label wool sweater, private-label blazer. 5. Indicate your impressions of the person to whom such clothing belonged with regard to being ....... (as listed in Wearing List ll) 7 unattractive _: _: _: _: _ attractive unsociable _: _: _: _: __ sociable unsuccessful _: _: _: _: _ successful unfashionable _: _: _: _: _ fashionable low status _:_:_:_:_ high status economical _:_: _: _:_ affluent unpopular _: _: _: _: _ popular 6. Are there other wordslphrases that you would use to describe the person who wears national manufacturerldesignor-brand or private-label apparel, that are not included in Questions #4 and 5? If so, please write the additional adjectives below. National Manufacturer/Designer- Private-Label-Brand Wearers Brand Wearers APPENDIX B. FINAL STUDY QUESTIONNAIRE 83 APPENDIX B. FINAL STUDY QUESTIONNAIRE [_ntroflctorv Letter: First University COLLEGE OF HUMAN ECOLOGY Dopartrnont ol Human Environment and Design Michigan State UlllVClSlly 204 Human Ecology Bolldlnq East Lansing, Michigan 48824-1010 51713559771? FAX 511/336 1058 MICHIGAN STATE UNIVLRSIIY Dear M.S.U. student: I am a graduate student in a master’s program at Michigan State University. Currently, I am working on my thesis which is consumer-behavior based and focuses on the relationship between self-perception and brand loyalty of consumer products. You have been selected to answer the questionnaire for this study. The questionnaire consists of 27 questions and it will take approximately 15 minutes to complete. You will be asked about your perceptions of and preferences for selected apparel products. There are no riqht or wrong armwers to any questions. All of your answers will be treated with §t_r_iat confidence and responses will be reported together so that you will not be identified. The answers you give to questions will never be linked to you in any way. Your participation in this study is voluntary and you may choose not to answer any question or to discontinue participation at any time. If you have any questions about the questionnaire, please feel free to ask. I would greatly appreciate it if you take a few minutes to answer these questions. Should you have any questions regarding this study, please contact Dr. Dawn Pysarchik (353- 0650). Sincerely, %.; ........{f-~—¥ Jae-Eon Chung ( l M. S. Graduate Stude ' . , 1;»: K9 L Dawn I. Pysarchik, Ph D Associate Professor 84 IntrodJuctorv Letter: Second University COLLEGE or HUMAN ECOLOGY Department of Human Environment and Design Michigan State Unwersdy 204 Human Emlogy Burlnlng Easl lanslnq Mulligan 48824 1030 517555 7N? fAX FILMS.“ ll')‘;»8 MICHIGAN STATE UNIVERSITY Dear University of Michigan student; I am a graduate student in a master's program at Michigan State University. Currently, I am working on my thesis which is consumer-behavior based and focuses on the relationship between self-perception and brand loyalty of consumer products. You have been selected to answer the questionnaire for this study. The questionnaire consists of 27 questions and it will take approximately 15 minutes to complete. You will be asked about your perceptions of and preferences for selected apparel products. There are no riqht or wrong was to any questions. All of your answers will be treated with s_t_r§t confidence and responses will be reported together so that you will not be identified. The answers you give to questions will never be linked to you in any way. Your participation in this study is voluntary and you may choose not to answer any question or to discontinue participation at any time. If you have any questions about the questionnaire, please feel free to ask. I would greatly appreciate it if you take a few minutes to answer these questions. Should you have any questions regarding this study, please contact Dr. Dawn Pysarchik (517- 353—0650). Sincerely, f"- /r . fie {*11.-L’,/\“-2-f3 Jéé -Eun Chung C" M.S. Graduate Student I / g “/2; " ”‘ 21x2. )3 #7 that rc’ A4. 1 Dawn l. Pysarchik- Associate Professor 85 PART I. In this section, we are interested in your feelings about yourself. Please circle your degree of agreementldisagreement with each statement where 1 = strongly disagree to 5 = strongly agree. Strongly Dis- Neu- Agree Strongly disagree agree tral agree 1) On the whole, l am satisfied 1 2 3 4 5 with myself. 2) At times, I think I am no good at all. 1 2 3 4 5 3) Ifeel that I have a 1 2 3 4 5 number of good qualities. 4) I am able to do things 1 2 3 4 S as well as most other people. 5) Ifeel I do not have 1 2 3 4 5 much to be proud of. 6) lcertainly feel useless at times. 1 2 3 4 5 7) I feel that I am a person of worth, 1 2 3 4 5 at least on an equal plane with others. 8) Iwish I could have 1 2 3 4 5 more respect for myself. 9) All in all, lam inclined to 1 2 3 4 5 feel that I am a failure. 10) I take a positive attitude 1 2 3 4 5 toward myself. PART II. In this segment of the survey you are asked how you actually see yourself, based on seven pairs of adjectives. You should note that you are m; asked how you would like to see yourself. In some cases how you see yourself may be the same as how you would like to see yourself; in other cases, they may not be the same. Place an x above the line on the five-point scale which most accurately describes how you see yourself for each set of adjectives. Some examples are provided below. If you actually see yourself as being closely related to one end of the scale, you should place your check mark as follows: conservative £1 _: _: _._ liberal or conservative _: _: _: _; )1 liberal If you actually see yourself as being somewhat related to one end of the scale, you should place your check mark as follows: conservative _: X; _' __ __ liberal or conservative _. _: _i )5 _ liberal 86 If you actually see yourself as not being related to either end of the scale (that is, neutral), you should place your check mark as follows: conservative_: _: 1g: _: _ liberal 1 1. How do you actually see yourself in regard to being ........... ? unattractive _: _: _: _: _ attractive unsociable ______ sociable unsuccessful _: _: _: _: __ successful unfashionable _: _: _: _; _ fashionable low status _: _: _: _: _ high status economical _____ affluent unpopular _: _: _: _: _ popular 12. Now, you are asked about how you would like to see yourself. Note that you are n_ot asked about how you actually see yourself. Would you like to see yourself as .............. ? unattractive _: _: _: _: _ attractive unsociable _: _; _: _: __ sociable unsuccessful _: _: _: _: __ successful unfashionable _: _: _: _: _ fashionable low status _: _: _: _: __ high status economical ______ affluent unpopular _: _: _: _: __ popular PART III. In this part of the study we are interested in measuring your involvement and interest in three different clothing items. In order to assess this, we need you to judge each of the three clothing items listed below (blue jeans, socks, and oxford cloth shirt) using the twenty descriptive scales, according to how YOU perceive each product. Some examples are provided below. If you feel that blue jeans are not at all practical, you should place your check mark as follows : not practical X: _: _: _: _ practical If you feel that socks are somewhat practical, you should place your check mark as follows: not practical _3_3_3.K_ practical If you feel that an oxford cloth shirt is not related to either end of the scale (that is neutral), you should place your check mark as follows: not practical _: _: L: _; _ practical 13. Please place an x above the line on the five—point scale which most accurately indicates your interest and involvement with each product, based upon each set of adjectives. unimportant of no concern irrelevant means nothing useless worthless trivial not beneficial doesn’t matter uninterested in unimportant of no concern irrelevant means nothing useless worthless trivial not beneficial doesn't matter uninterested in unimportant of no concern irrelevant means nothing useless worthless trivial not beneficial . . . . — — — — .— . . _ — — — — .——.————-— doesn‘t matter uninterested in 87 My}: important insignificant of concern to me superfluous relevant boring means a lot to me unexciting useful unappealing valuable mundane fundamental nonessential beneficial undesirable matters to me unwanted interested in not needed Socks important insignificant of concern to me superfluous relevant boring means a lot to me unexciting useful unappealing valuable mundane fundamental nonessential beneficial undesirable matters to me unwanted interested in not needed Oxford Cloth Shi_rt important insignificant of concern to me superfluous relevant boring means a lot to me unexciting useful unappealing valuable mundane fundamental nonessential beneficial undesrrable matters to me unwanted interested in not needed ——_—————-.—_ . . . . —— — —— — _— 1 o . . — — — — _ significant vital interesting exciting appealing fascinating essential desirable wanted needed significant vital interesting exciting appeahng fascinating essential deSirable wanted needed Significant vital interesting excrting appeanng fascrnating essenhal desrrable wanted needed 88 PART IV. In this portion of the survey we are interested in your perceptions of the images of selected products. Imagine that you are shopping at J. C. Penny and observe six individuals you do not know. They are in a fitting room trying on different clothing items. These six persons are wearing the same 3 items (blue jeans, oxford cloth shirt, and socks), the only difference is in the brand of one of the 3 items. The items they have decided to purchase are listed below, by person. Give your perceptions of each of the six people based upon the apparel items they are wearing. Place an X above the line on the five-point scale. 14. Indicate your impressions of person I who is wearing the following 3 items with regard to being ...... ? , Person I: Guess Blue Jeans Oxford Cloth Shirt Socks unattractive _:_:_:_:_ attractive unsociable _:_: _:_:_ sociable unsuccessful _:_:_:_:_ successful unfashionable _:_:_:_:_ fashionable low status _:_:_:_:_ high status economical _;_;_:_:_ affluent unpopular _:_:_:_:_ popular 15. Indicate your impressions of person II who is wearing the following 3 items with regard to being ....... 7 Person ll: Arizona Blue Jeans Oxford Cloth Shirt Socks unattractive _: _: _: _: __ attractive unsociable _: _ _: _: __ sociable unsuccessful _: _: _: _: __ successful unfashionable _: _: _: _' _ fashionable low status _: _: _: _: _ high status economical _: _: _: _: _ affluent unpopular _: _: _: _: _ popular 16. Indicate your impressions of person III who is wearing the following 3 items with regard to being ....... 7 Person Ill: Blue Jeans Guess Oxford Cloth Shirt Socks unattractive _: _: _: _: _ attractive unsociable _. _ _. _. _ sociable unsuccessful _: _: ___: _; __ successful unfashionable __ _. __ _. _ fashionable low status _: _: _: _: _ high status economical _’ __ __' ___. _ affluent unpopular ________ popular 89 17. Indicate your impressions of person IV who is wearing the following 3 items with regard to being ....... ? Person IV: Blue Jeans Arizona Oxford Cloth Shirt Socks unattractive _:_:_:_:_ attractive unsociable _:_:_‘_:___ sociable unsuccessful _: _: _: __: _ successful unfashionable _: _: _: _: _ fashionable low status _:_: _: _:_ high status economical _: _: _: _: _ affluent unpopular _:_:_:_:_ popular 18. Indicate your impressions of person V who is wearing the following 3 items with regard to being ....... ? Person V: Blue Jeans Oxford Cloth Shirt Hanes Socks unattractive _:_:_:_:_ attractive unsociable _:_:_'__:_ sociable unsuccessful _:_:_:_:_ successful unfashionable _:_:_:_:_ fashionable low status _:_:_._:_ high status economical _:_:_:_:_ affluent unpopular _:_:_;_:_ popular 19. Indicate your impressions of person VI who is wearing the following 3 items with regard to being ....... ? Person VI: Blue Jeans Oxford Cloth Shirt Arizona Socks unattractive _: _: _: _: _ attractive unsociable _: _: _: _: _ sociable unsuccessful _: _: _: _i _ successful unfashionable _: _: _.' _' _ fashionable low status _: _: _: _: __ high status economical _: _: _i _g _ affluent unpopular _: _:_: _: _ popular PART V. This section includes statements about apparel buying decisions regarding national manufacturerldesigner brands and private labels. National manufacturerldesigner brands mean nationally recognized manufacturer-owned or prestigious designer brands (e.g., Lee or Calvin Klein). Private labels mean retailers' or wholesalers’- own brands (e.g., Mervyn’s - Partners label or Hudson’s - Boundarywater label). Keep in mind each of the blue jeans, socks, and oxford cloth shirt purchases that you have made for yourself as you answer each of these questions. Circle the appropriate number (1 to 5) which indicates your agreement with each of the following statements for each of the 3 products (blue jeans, socks, and oxford cloth shirt). 9O 20. I buy well-known national manufacturerldesigner-brand garments rather than private label- brands. Strongly Dis- Neu- Agree Strongly Disagree agree tral agree Blue Jeans: 1 2 3 4 5 Socks: 1 2 3 4 5 Oxford Cloth Shirt: 1 2 3 4 5 21. I would rather pay more for well-known national manufacturerldesigner brands than to purchase private labels that look similar but cost less. Strongly Dis- Neu- Agree Strongly Disagree agree tral agree Blue Jeans: 1 2 3 4 5 Socks: 1 2 3 4 5 Oxford Cloth Shirt: 1 2 3 4 5 22. I buy private label-brands rather than well-known national manufacturerldesigner-brand garments. Strongly Dis- Neu- Agree Strongly Disagree agree tral agree Blue Jeans: 1 2 3 4 5 Socks: 1 2 3 4 5 Oxford Cloth Shirt: 1 2 3 4 5 PART VI. This section is for statistical purposes. Circle the appropriate response for each. 23. What is your gender? 1. Female 2. Male 24. What is your ethnic origin? 1. Asran or Pacific Islander 5. American Indian 2. Black 6. Others 3. White (Please complete.) 4 Hispanic or Spanish 25. What is your marital status? 1. Now married 4. Separated 2. Widowed 5. Never married 3 Divorced Please fill in your response on the line provided. 26. Your Age 27. Annual Household Income of Your Parents (3) REFERENCE LIST 91 REFERENCE LIST Abrahamson, M. (1983). Social Research Method. Englewood Cliffs, N.J.: Prentice- Hall, INC. Abend, J. (1991, May). License to Profit. Bobbin 3_2 (9), 58-66. Assael, H. (1992). Consumer Behavior and Marketing Action. (4th ed.). Boston: PWS- Kent Publishing. Babbie, E. (1992). he Practice of Social Research. (6th ed.). Belmont, California: Wadsworth Publishing Company. Baugh, D. F., & Davis, L. L. (1989). The Effect of Store Image on Consumers’ Perception of Designer and Private Label Clothing. Clothing and Textiles Research Joumal, 7_(3), 15-21. Beatty, S. E., Kahle, L. R., & Homer, P. (1988), The Involvement-Commitment Model: Theory and Implications. Journal of Business Research 1_6_(2), 149-167. Behling, D. U., & Wilch, J. (1988). Perceptions of Branded Clothing by Mall Consumers. Clothing and Textiles Research Journfl, Q, 43-47. Belch, G. E. (1978). Belief Systems and the Differential Role of the Self-Concept. In K. H. Hunt (Eds), Advances in Consumer Research, Q, (pp. 320-325). Ann Arbor, MI: Association for Consumer Research. Belk, R. W., Bahn, K. D., & Mayer, R. N. (1982). Developmental Recognition of Consumption Symbolism. Joum_al of Con_sumer Research, 2, 5-17. Belk, R. W. (1988, September). Possessions and the Extended Self. Journal of Consumer Research _1_§_, 139-168. Belk, R. W. , R. Mayer, & Driscoll, A. (1984). Children's Recognition of Consumption Symbolism in Children's Products. Journal of Consumer Research, 1_Q, 386-396. Bellenger, D. N., Steinberg, E., & Stanton, W. W. (1976). The Congruence of Store Image and Self-Image. Journal of Retailing, fl, 17-32. Birdwell, A. E. (1968). A Study of the Influence of Image Congruence on Consumer Choice. The Journal of Business fl, 76-88. 92 Bloch, P. H. (1982). Involvement Beyond The Purchase Process: Conceptual Issues and Empirical Investigation. In A. Mitchell (Eds), Advances in Consumer Research 2, (pp. 413-417). Ann Arbor, MI: Association for Consumer Research. Brandon, N. (1969). The Psvc_hology of Self-Esteem. Los Angeles: Nash Publishing Company. Calder, B. J ., & Bumkrant, R. E. (1977, June). Interpersonal Influence on Consumer Behavior: an Attribution Theory Approach. Journal of Con_sumer Research, 4, 29-38. Claiborne, C. B., & Sirgy, M. J. (1990). Self-Image Congruence as a Model of Consumer Attitude Formation and Behavior: A Conceptual Review and Guide for Future Research. Developments, in Marketing Science, _13, 3-7. Deveny, K. (1992, October 20). Store Shoppers Bypass Big-Name Brands and Steer Carts to Private-Label Products. Wall Street Journal p. B1. Diamond, J., & Pintel, G. (1985). Retail Buying (2nd ed.). Englewood Cliffs, NJ: Prentice-Hall, Inc., (pp. 245-256). Dolich, I. J. (1969). Congruence Relationships Between Self-Images and Product Brands. Journal of Marketig Research 6, 80-84. Elliot, G. C. (1986). Self-Esteem and Self-Consistency: A Theoretical and Empirical Link Between Primary Motivation. Social Psychology Quarterly, fl, 218-297. Ericksen M. K., & Sirgy, M. J. (1989). Achievement Motivation and Clothing Behavior: A Self-Image Congruence Analysis. Journal of Social Behavior and Personalig, 5(4), 307-326. Evans, M. (1988). Consumer Behavior towards Fashion. Europ_ean Journal of Marketing, 23, 7-16. Goldsmith, R. E., & Emmert, J. (1991). Measuring Product Category Involvement: A Multitrait-Multimethod Study. Journal of Business Research 23, 363-371 . Grubb, E. L., & Grathwohl, H. L. (1967, October). Consumer Self-concept, Symbolism and Market Behavior: A Theoretical Approach. Jourfll of Marketing, 3 , 22-27. Grubb E. L., & Hupp, G. (1968, February). Perception of Self, Generalized Stereotypes, and Brand Selection. Journal of Marketing, i, 58-63. Hair, J. F. Jr., Anderson, R. E., Tatham, R. L., & Black, W. C. (1992). Multivariate Data Analysis with Readings. New York: Macmillan Publishing Company. 93 Haire, M. (1950). Projective Techniques in Marketing Research. Journi of Marketing, l_8_, 314-321. Hilgard, E. R. (1962). Introduction to Psychology. U.S.A.: Harcourt, Brace & World, Inc.. Holbrook, M. B., & Hughes, N. C. (1978). Product Images: How Structured Ratings Scales Facilitate Using A Projective Technique in Hypothesis Testing. Journal of Psychology, 1%, 323-328. Holman, R. H., & Schenk, C. T. (1980). A Sociological Approach to Brand Choice: The Concept of Self-Image. Advaaces in Consumer Research, 1, 610-614. Holstius, K., & Paltschik, M. (1983, October). Brand Names and Perceived Value. Europaan Research , 1_1_, 151-155. Huddleston, P., & Cassill, N. L. (1990). Female Consumers' Brand Orientation: The Influence of Quality and Demographics. Home Economics Research Journal, _l_8(3), 255-262. Huddleston, P., Cassill, N. L., & Hamilton, L. K. (1993, Fall). Apparel Selection Criteria as Predictors of Brand Orientation. Clothing and Textiles Research Journal, _12 (1), 51-56. Jones, S. (1973). Self and Interpersonal Evaluations: Esteem Theories Versus Consistency Theories. Psychological Bulletin, 22, 185-199. Jolson, M. A., Anderson, R. E., & Leber, N. J. (1981). Profiles of Signature Goods Consumers and Avoiders. Journal of Retailing, 5_7_ (4), 19-38. Kaiser, S. B. (1990). The Social Psychology of Clothing: Syt_nbolic Appaarances in Context. New York: Macmillan Publishing Company. Kaplan, H. B. (1980). Deviant Behavior in Defense of Self. New York: Academic Press. Klokis, H. (1987, February). Private Brands Game: Chains Ante Up. Chain Store Age Executive, _6_3 (2), 21-23. Kwon, Y. H. (1990). Brand Name Awareness and Image Perception of Women's Daytime Apparel. Perceptual and Motor Skills, 11, 743-752. Landon, E. L. Jr. (1974, September). Self-Concept, Ideal Self-Concept, and Consumer Purchase Intentions. Journal of Con_sumer Resear_ch, ;, 44-51. 94 Leigh, J. H., & Gabel, T. G. (1992, Winter). Symbolic Interactionsim: Its Effects on Consumer Behavior and Implications For Marketing Strategy. Journal of Consumer Marketing, 2(1), 27-38. Levy, S. J. (1959). Symbols For Sales. Harvard Buainess Review. 3_7(4), 117-124. Liebeck, L. (1994, October 3). 1994 Power Brands - Part 1. Discount Store News, £09), 25-61. McDougall, W. (1932). The Energies of Men. London: Methuen. Malhotra, N. K. (1988). Self-Concept and Product Choice: An Integrated Perspective. Journal of Economic Psychology, 2, 1-28. Miller, C. (1992, June). Brands Are Extremely Important: Parents Make Fashion Statement by Dressing Kids in ‘Status’. Marketing News, _2£(12), 1,14-15. Mowen, J. C. (1990). Con_sumer Behavior. New York: Macmillian Publishing Company. Munson, J. M. (1974). Typological Investigation of Self Concept Congr_uig and Brand Preference: Toward a Predictive Model. Ann Arbor: University Microm International. Norusis, M. J. (1993). SPSS for Windows: Professional Statistics. Relea_se 6.0. US: SPSS Inc.. Onkvisit, S., & Shaw, J. (1987, Winter). Self-Concept and Image Congruence: Some Research and Managerial Implications. Jouraal of Consumer Marketing, 4_ (1), 13-23. Pelham, B. W., & Swann, W. B. Jr. (1989). From Self-Conceptions to Self-Worth: On the Sources and Structure of Global Self-Esteem. Journal of Personaligy and Social Psychology, fl(4), 672-680. Rogers, C. R. (1965). Client-Centered Therapy. Boston: Hougton Miffilin Company. Rosenberg, M. (1965). Socieg and the Adolescents' Self-Esteem. Princeton, NJ: Princeton University Press. (1967). Psychological Selectivity In Self-Esteem Formation. In C. Sherif & M. Sherif (Eds), Attitude, Ego-Involvement And Change, New York: John Wiley. (1979). Conceiving the Self, New York: Basic Books. 95 Ross, 1. (1971). Self-Concept and Brand Preference. Journal of Business of the Univers_itv of Chicago, fl, 38-50. Rothschild, M. L. (1979). Advertising Strategies for High and Low Involvement Situations. In J. C. Maloney & B. Silverman, Attitude Research Plays for High Stakes (pp. 74-93), Chicago, IL: American Marketing Association. Salmon, W. J., & Cmar, K. A. (1987, June). Private Labels Are Back in Fashion. Harvard Business Review 6_5, pp. 99-106. Sirgy, M. J. (1980). Self-Concept In Relation To Product Preference and Purchase Intention. Developments in Marketing Science, 2, 12-23. (1982, December). Self-Concept in Consumer Behavior: A Critical Review. Journal of Consumer Research 2, 287-300. (1985). Using Self-Congruity and Ideal Congruity to Predict Purchase Motivation. Journal of Bu_siness Research, fl, 195-206. Sirgy, M. J., Johar, J. S., Samli, A. C., & Claiborne, C. B. (1991). Self-Congruity versus Functional Congruity: Predictors of Consumer Behavior. Journal of the Academy of Marketing Science, QM), 363-375. Sirgy, M. J., & Samli, A. C. (1985, Summer). A Path Analytic Model of Store Loyalty Involving Self-Concept, Store Image, Socioeconomic Status, and Geographic Loyalty. Journal of the Academy of Marketing Science, 13, 265-291. Solomon, M. R. (1983, December). The Role of Products as Social Stimuli: A Symbolic Interactionism Perspective. Journal of Consumer Research fl, 319-329. Underwood, E. (1992, September). Something For Everyone. Brandweek pp. 19-20, 24. Urban, G. L., & Star, S. H. (1991). Advanced Marketing Strategy: Phenomena, Analysis, and Decisions. Prentice-Hall, Inc: New Jersey. White, I. S. (1964). Perception of Value in Products. In Newman, J. W. (1966). 9a Knowing Consumers (pp. 103-104). John Whiley: New York. Wilensky, D. (1994, May 16). Store Brands Take Paramount Importance as Retailers Scale Back National Brands. Discount Store News £00), 24-27. 96 Workman, J. E. (1988). Traits Inferences Based on Perceived Ownership of Designer, Brand Name, or Store Brand Jeans. Clothing and Textiles Research Journfl, 6 (2), 23-29. Zaichkowsky, J. L. (1985, December). Measuring The Involvement Construct. Journal of Consumer Research Q, 341-352. Zaichkowsky, J. L. & Sood, J. H. (1988, February). A Global Look at Consumer Involvement and Use of Products. International Marketing Review, 6 (1), 20 - 34. IE5 "‘i(tilllllllllll