Toward an Integrated Model of Self-Congruity and Functional Congruity

M. Joseph Sirgy, Virginia Tech, U.S.A.
J. S. Johar, California State University, U.S.A.
ABSTRACT - Self-congruity is defined as the match between the brand image and the consumer’s self-concept, while functional congruity refers to the match between the perceived functional or performance characteristics and the consumer’s desired or important functional characteristics. It was hypothesized that the attitude predictiveness of self-congruity and functional congruity is moderate by brand conspicuousness, uniqueness, differentiation, and involvement. A study was conducted involving eight different product categories and 429 subjects. The overall pattern of results provided some support for the moderating effects of conspicuousness, uniqueness, involvement, and differentiation on attitude predictiveness of self-congruity and functional congruity.
[ to cite ]:
M. Joseph Sirgy and J. S. Johar (1999) ,"Toward an Integrated Model of Self-Congruity and Functional Congruity", in E - European Advances in Consumer Research Volume 4, eds. Bernard Dubois, Tina M. Lowrey, and L. J. Shrum, Marc Vanhuele, Provo, UT : Association for Consumer Research, Pages: 252-256.

European Advances in Consumer Research Volume 4, 1999      Pages 252-256

TOWARD AN INTEGRATED MODEL OF SELF-CONGRUITY AND FUNCTIONAL CONGRUITY

M. Joseph Sirgy, Virginia Tech, U.S.A.

J. S. Johar, California State University, U.S.A.

ABSTRACT -

Self-congruity is defined as the match between the brand image and the consumer’s self-concept, while functional congruity refers to the match between the perceived functional or performance characteristics and the consumer’s desired or important functional characteristics. It was hypothesized that the attitude predictiveness of self-congruity and functional congruity is moderate by brand conspicuousness, uniqueness, differentiation, and involvement. A study was conducted involving eight different product categories and 429 subjects. The overall pattern of results provided some support for the moderating effects of conspicuousness, uniqueness, involvement, and differentiation on attitude predictiveness of self-congruity and functional congruity.

INTRODUCTION

Sirgy, Johar, Samli, and Claiborne (1991) conducted four studies testing the hypotheses that (1) consumer behavior (e.g., brand attitude, store patronage) can be predicted by both self-congruity and functional congruity, (2) that functional congruity may be a stronger predictor of consumer behavior than self-congruity, and (3) that self-congruity may influence consumer behavior through the mediating effects of functional congruity. The results provided support for these expectations. Johar and Sirgy (1991) proposed that attitude predictiveness of self-congruity may be enhanced under conditions of high brand conspicuousness and uncommon usage. Furthermore, the predictiveness of functional congruity may be enhanced under conditions of high brand involvement and brand differentiation (cf. Shavitt 1992; Sirgy and Johar 1992). The purpose of the present study is to develop and test an integrated model of self-congruity and functional congruity. The model is designed to integrate theoretical notions from Sirgy et al. (1991) and Johar and Sirgy (1991) and subject the integrated model to an empirical test.

The marketing researcher uses concepts of functional congruity and self-congruity (through the use of multiattribute attitude models and self-image congruence models) for brand positioning (Johar and Sirgy 1989). Hence, it would be quite useful for marketing researchers to know under what circumstance self-congruity and functional congruity models should be used in brand positioning research. To use self-congruity and functional congruity in consumer research, we need to understand the conditions under which these constructs explain and predict consumer behavior.

CONCEPTUAL DEVELOPMENT

We refer to self-congruity as the match/mismatch between the brand user image and the consumer actual self-image, ideal self-image, social self-image, and/or ideal social self-image (Johar and Sirgy 1991). An ideal self-image incorporates attributes one aspires to have. A social self-image involves beliefs about how one is viewed by others, and the ideal social self-image is the image one aspires others to have of him or herself. Hence, the self-congruity process is viewed as a psychological process in which the consumer focuses on the brand user image and compares this image to his/her self-concept. The greater the match of the brand user image, the more positive and attitude toward the brand, and vice versa (Sirgy 1986).

Functional congruity, on the other hand, is defined as the match between the consumer’s beliefs about brand utilitarian attributes (performance) and the consumer’s referent attributes. The referent (e.g., ideal) attributes are the criteria used to evaluate the actual performance characteristics of the brand. For example, a specific brand of toothpaste may be evaluated along a set of utilitarian attributes such as preventing cavities, freshening breath, whitening teeth, and tasting good. The consumer may have referent points or standards (e.g., an ideal) used to judge the relative goodness of the perceived attributes. The greater the congruence between the consumer’s utilitarian beliefs about the actual brand the more positive the attitude (Sirgy, Johar, Samli, and Claiborne 1991). Expectancy-value researchers explain that a positive attitude is likely if the object of evluation is perceived to have characteristics that are positively valued.

The relationship between self-congruity and functional congruity has been addressed by several studies (Samli and Sirgy, 1982; Sirgy and Samli, 1985; Sirgy et al. 1991). Samli and Sirgy (1981) conducted a study to test the differential impact of self-congruity and functional congruity on store loyalty. Specifically, store loyalty was regressed on self-congruity (social and ideal social congruities) and functional congruity (functional evaluation of store-image) in addition to socioeconomic status, area loyalty, and shopping-complex loyalty. The results indicated that although self-congruity failed to significantly predict store loyalty, the self-congruity variables (social congruity and ideal social congruity) were significantly correlated with functional congruity (functional store-image evaluation or multiattribute attitude). In a follow-up study, Sirgy and Samli (1985) demonstrated through causal path analysis that store loyalty may be primarily influenced by functional congruity (functional store-image evaluation), and that functional congruity is influenced by self-congruity (social and ideal social congruity). That is, the study demonstrated a "biasing effect" of self-congruity on functional congruity. This effect suggests that, although functional congruity is more closely related to consumer behavior than self-congruity, functional congruity is significantly influenced by self-congruity.

Sirgy et al. (1991) addressed these issues in some detail. Four studies were conducted to test the hypotheses that (1) consumer behavior is more strongly predicted by functional congruity than by self-congruity, and (2) functional congruity is influenced by self-congruity. The model posits that the consumer experiences a match or congruity between a particular product/store’s symbolic image and his/her self-image resulting in self-congruity. Such congruity will bias the consumer’s perception and evaluation of the product/store’s functional image (functional congruity). In contrast, a consumer who experiences self-incongruity is likely to form an initial unfavorable attitude, which in turn biases her evaluation of the store’s functional characteristics in a negative way. Perception and evaluation of functional attributes operate at a more-conscious level to influence consumer behavior. In contrast, self-congruity is argued to operate at a less-conscious level to influence the consumer’s perception and evaluation of the product/store’s functional attributes. Also, self-congruity was hypothesized to influence consumer behavior directly but less strongly than functional congruity. That is, the motivational tendency generated as a result of self-congruity not only biases functional congruity, but also impacts consumer behavior directly.

Johar and Sirgy (1991) developed a conceptual model addressing the psychological dynamics involved in value-expressive and utilitarian appeals. The model helped clarify the distinctions between these two types of advertising appeals. The model showed how value-expressive appeals involve a psychological process in which the brand user image is compared with the audience self-image (i.e., self-congruity), while utilitarian appeals involve a psychological process in which the brand performance, concrete, or functional features are compared with the features of a referent (i.e. functional congruity). The model further articulated a set of hypotheses concerning moderator effects, such as the conditions which would enhance the effectiveness of value-expressive appeals, and similarly the conditions which moderate the effectiveness of utilitarian appeals. It was hypothesized that a value-expressive advertising appeal may be effective when the product (brand) is uncommonly used (or unique), and when the product (brand) is conspicuously consumed. Conversely, utilitarian advertising appeal may be effective when the product (brand) is highly differentiated from its competition, and when the consumer is highly involved with the product (brand).

The Self-Congruity Effect

Self-congruity models are based on the cognitive matching between value-expressive attributes of a given brand and consumer self-concept. ne models are designed to predict consumer behavior variables, such as product (brand or store) attitude, intention, behavior, and loyalty (see Sirgy 1982, 1985a, 1985b, 1986, Sirgy et al. 1991; and Claiborne and Sirgy 1990; Johar and Sirgy 1991, for comprehensive reviews of the consumer research literature in self-concept). Among the commonly used self-congruity models arc the actual self-congruity model, the ideal self-congruity model, the social self-congruity model, the ideal social self-congruity model, and the affective self-(social) congruity models.

If consumers experience actual self-congruity (ideal self-congruity, social self-congruity, and/or ideal social self-congruity), they will be motivated to purchase that product, because the use of the product will satisfy self-consistency (self-esteem, social consistency, and/or social approval) needs. By using that product, consumers would be able to fulfill their needs of self-esteem, self-consistency, social consistency, and social approval. This situation is expected to maximize brand preference and use. Based on the above discussion, the following hypothesis was deduced:

Hypothesis 1: Self-congruity (actual ideal, social, and ideal social-self-congruity) is predictive of brand attitude.

The Moderating Effects of Brand Conspicuousness

Brands that are socially visible (conspicuously consumed) reveal the personal characteristics of their users more than brands that are consumed inconspicuously. The social visibility of the brand helps establish consensual beliefs regarding the stereotyped image of the brand user. Sirgy, Johar, and Wood (1986) have demonstrated that brand social conspicuousness is positively related to brand value-expressiveness (which in turn may evoke a self-congruity process). They explained that socially visible brands can be associated with the personal characteristics of their users more than brands can be associated with the personal characteristics of their users more than brands that are consumed privately. Based on this discussion, we hypothesize the following:

Hypothesis 2: Self-congruity is likely to be more predictive of brand attitude when brand conspicuousness is high rather than low.

The Moderating Effects of Brand Uniqueness

From attribution theory in social psychology we know that unique or uncommon events are used by people to make causal attributions: i.e., the informational weight of unique characteristics is augmented, whereas common characteristics are discounted in perceptual inferences. That is, the use of uncommon or unique brands may lead to inferences about the brand user (person attributions). Based on this discussion, we hypothesize the following:

Hypothesis 3: Self-congruity is likely to be more predictive of brand attitude when brand uniqueness is high rather than low.

The Functional Congruity Effect

The functional congruity effect, as previously discussed, involves the match between the beliefs about the brand utilitarian attributes (performance-related) ad the consumer’s referent attributes. The greater the congruence between the consumer’s utilitarian beliefs about the actual brand and the referent beliefs, the more positive the attitude toward the brand. Therefore, the following hypothesis is deduced.

Hypothesis 4: Functional congruity is predictive of brand attitude.

The Moderating Effects of Brand Involvement

Consumer involvement may be defined in terms of such observable actions as time and /or intensity of effort expended in consumer behavior. Consumer behavior may include overt information search, in-store shopping, price and brand comparison, and product use. Under high involvement conditions, consumers tend to engage in more information processing and alternative evaluation than under low involvement conditions. Brand involvement is similarly defined in terms of cognitive effort in decision-making. Johar and Sirgy (1991) argued that brands that induce high emotional involvement are likely to involve a functional congruity route to persuasion. The consumer is likely to allocate greater cognitive processing capacity to focus on the concrete, performance-like, or functional aspects of the brand. Based on the above argument, we hypothesize the following:

Hypothesis 5: Functional congruity is likely to be more brand predictive of attitude when brand involvement is high rather than low.

The Moderating Effects of Brand Differentiation

Brand differentiation may cause consumers to invest considerable cognitive effort searching for differences among brands. Under conditions of high brand differentiation, consumers may acquire more knowledge about the benefits of the brand. Low involvement processing often may occur when the differences between brands are not perceived (low brand differentiation). If we accept the notion that brand differentiation may be related to product involvement, we can logically conclude that brand differentiation may play a similar role in moderating the relationship between functional-congruity and attitude toward the product. Based on the above argument, we hypothesize the following:

Hypothesis 6: Functional congruity is likely to be more predictive of brand attitude when brand differentiation is high rather than low.

Self-Congruity versus Functional Congruity

Sirgy et al. (1991) theorized that although both self-congruity and functional congruity impact brand attitude directly self-congruity may play a more subtle role. That is, self-congruity operates at a less conscious level biasing functional evaluations. Thus, the direct impact of self-congruity on brand attitude is not as strong as functional congruity. By the same token, functional congruity is influenced by self-congruity. Therefore, the following hypotheses can be tested:

Hypothesis 7: Functional congruity is likely to be more predictive of brand attitude than self-congruity.

Hypothesis 8: Functional congruity is a direct function of self-congruity, i.e., the self-congruity effect is partially mediated through functional congruity.

METHOD

roduct and Brand Selection

Eight brands pertaining to eight different products were selected for the study. The selection criterion dictated that the products and brands would vary in involvement, differentiation, conspicuousness, and uniqueness. It should be noted that the extent to which a specific brand is judged as differentiated, conspicuous, or commonly used is dependent on the consumer’s frame of reference. For example, a brand of automobile such as a Porsche can be judged as high in conspicuousness relative to a Ford Escort, but low in conspicuousness relative to a Ferrari. We decided to control possible confounds from different frames of references by specifying a referent brand for each of the selected (focal) brands. Note that we did not measure involvement in relation to a referent-brand but at the product level because we felt that a consumer’s level of involvement is determined mostly by the product category and not his or her perception of the brand relative to a referent brand. The selection of the eight brands was based on a pretest using 39 college students and involving 20 brands. Our objective electing two was to work with brands that would produce the greatest variability in the moderator variables.

Sampling

Eight different samples were obtained, each corresponding to a different product (the eight brands used in the study). The goal was set to obtain 70 respondents for each of the eight samples (eight brands), a total of 560 respondents. Marketing students were used as consumer subjects for this study. A total of 429 students completed the questionnaire, producing a response rate of 76%. More specifically, the responses were broken down as follows: auto (n=54), camera (n=39), tires (n=36), watch (n=47), soft drink (n=70), TV (n=51), beer (n=64), and headache remedy (n=68).

Attitude Measures

Four self-report items using 5-point Likert-type scales were used to measure attitude for each of the eight products. These were: I like Focal Brand better than Referent Brand. I would use Focal Brand more than I would Referent Brand. Focal Brand is my preferred brand over Referent Brand. I would be inclined to buy Focal Brand over Referent Brand. The Cronbach Alpha coefficient across the eight brands was .915. The mean (and standard deviation) of the average composite was 3.111 (1.075).

The Self-Congruity Measure

The self-congruity measure involved the summation of scores from four self-congruity constructs (for a comprehensive review of self-congruity measures, see Sirgy et al. 1997). These are: actual self-congruity, ideal self-congruity, social self-congruity, and ideal social self-congruity.

Actual Self-Congruity. This construct was measured using 10 self-report items along 5-point Likert-type scales. Reliability analysis produced a Cronbach Alpha of .82 (pooled across all eight products) after deleting five items. The retained items were: (+)The image of the (user of focal brand) is highly consistent with how I see myself, more so than the image of (referent brand). (-) I can’t relate to those people who (use focal brand) rather than (referent brand). (-) I can’t identify with those people who prefer (focal brand) over (referent brand). (-) People who are very different from me prefer (focal brand) over (referent brand). (+) I am very much like the typical person who prefers to (use) (focal brand) rather than (referent brand).

Ideal Self-Congruity. This construct was measured using 10 self-report items along 5-point Likert-type scales. Reliability analysis produced a Cronbach Alpha of .84 (pooled all eight products) after deleting four items. The retained items are: (+) I may like myself better if I were to (use) (focal brand) rather than (referent brand). (-) (Using focal bran) may make me less special than (using referent brand). (-) I hate the image of (focal brand) (user) compared to the image of (referent brand). (+) I prefer the image of (focal brand) (user) than the image of (referent brand). (-) I may not think highly of myself if I were to (use) (focal brand) rather than (referent brand). (+) I like the kind of person who (uses) (focal brand) better than the kind of person who (uses) (referent brand).

Social Self-Congruity. This construct was measured using 10 self-report items along 5-point Likert-type scales. Reliability analysis produced a Cronbach Alpha of .90 (pooled across all eight products) after deleting four items. The retained items are: (-) People who are close to me have a hard time seeing me as (using) a (focal brand) over (referent brand). (-) People who know me think that I’m very different from those who (use focal brand) instead of (referent brand). (-) People that I know think of me as the kind of person who (uses) (focal brand) and I’m not the kind who (uses) (referent brand). (+) The image of the (user) of (referent brand) is highly consistent with how I’m seen by the people who are close to me. (+) People who know me think of me as the kind of person who is more likely to (use) (focal brand) than (referent brand). (+) I am usually viewed by my relatives and friends like the typical person who prefers to (use) (focal brand) rather the kind of person who prefers to (use) (referent brand).

Ideal Social Self-Congruity. This construct was measured using 10 self-report items along 5-point Likert-type scales. Reliability analysis produced a Cronbach Alpha of .87 (pooled across all eight products) after deleting four items. The retained items are: (-) My friends and associates don’t like to see me as a (user) of a (focal brand) compared to (referent brand). (-) People that I associate with do not have much regard for the image of the (focal brand user) compared to the image of the (referent brand user). (+) (Using) (focal brand) may make people think more special of me than if I were to (use) (referent brand). (-) (Using) (focal brand) may make my friends and associates have less regard for me than if I were to use (referent brand). (+) People around me may like me more if I were to (use) (focal brand) than (referent brand). (+) My friends and associates prefer the image of (focal brand) (user) than the image of (referent brand) (user).

The Functional Congruity Measure

From the open-ended responses in the preliminary study, a set of functional (utilitarian) evaluative criteria were content-analyzed. From the content analysis results, a set of functional attributes were formulated specific to each of the eight brands. These attributes were used to measure the component variables involved in a belief-importance multiattribute attitude. The model involves the sum of the product of belief strength and importance weights across a set of utilitarian attributes. The belief strength variable was measured using the following response cue: "Listed below are possible attributes of Product X. For each of these attributes, please indicate by circling the appropriate number how likely or unlikely it is that the Focal Brand [Brand X] would possess each attribute compared to the Referent Brand" [Brand Y]. Responses to this question were formulated in terms of a set of utilitarian attributes measured using a 5-point likelihood scale (from "very unlikely" to "very likely"). A "don’t know" category was also concluded, and responses in this category were treated as missing values. The importance variable was measured using the following response cue: "If you were considering Brand X, in general, how important or unimportant are the following characteristics to you? For most people, some things are more important than others. Please circle the number which is closest to the degree of importance you would attach to that characteristic when shopping for Brand X." The same utilitarian attributesused to tap the belief strength variable also were used to tap the importance variable. The response scale was a 5-point scale ranging from "very unimportant" to "very important." The mean (standard deviation) of the functional congruity measures was 10.923 (5.589).

The Involvement Measure

Zaickowsky’s (1985) involvement scale was used for this study. The measure involved 7-point semantic differential scales consisting of 20 bipolar adjectives. Reliability and factor analyses showed the following items to interrelate highly: important/unimportant, of no concern/of concern to me, irrelevant/relevant, means a lot to me/means nothing to me, useless/useful, valuable/worthless, trivial/fundamental, beneficial/not beneficial, matters to me/doesn’t matter, uninterested/interested, significant/insignificant, and vital/superfluous. The mean and standard deviation of the overall measure were 3.317 and 1.306, respectively.

The Differentiation Measure

Ten Likert 5-point scales were designed to measure the brand differentiation construct. This measure was then modified from a measure used by Sirgy, Johar, and Wood (1986). Based on factor and reliability analyses, the following items were retained: (-) I can hardly notice the difference between "X" and most other brands I know; however, there is definitely a difference between "Y" and the other. (+) "X" is much more differentiated from other brands than "Y." (-) It is harder to distinguish "X" from its competition than "Y." (-) "X" is very similar to the competitor brands. The mean and standard deviation of the overall measure were 3.710 and .978, respectively.

The Conspicuousness Measure

Ten Likert 5-point scales were designed to measure the brand conspicuousness construct. This measure was modified from a measure used by Sirgy, Johar, and Wood (1986) in which product conspicuousness was hypothesized and found to be positively related to product value-expressiveness. The findings provided construct validation support for the measure. Based on factor and reliability analyses, the following items were retained: (+) The user of "X" is more of an attention-seeker than the user of "Y." (+) The user of "X" is more noticeable when using it than when using "Y." (+) People who use "X" show off; people who use "Y" don’t. (+) The use of "X" draws attention from other people more than the use of "Y." (-) The use of "X" is much more private than "Y." (-) The use of "X" is more inconspicuous than the use of "Y." (+) The use of "X" is more attention-getting than the use of "Y." (+) You can’t avoid people not seeing you when one uses "X." This is not the case when one uses "Y." The mean and standard deviation of the overall measure were 2.651 and .744, respectively.

The Uniqueness Measure

Ten Likert 5-point scales were designed to measure the brand uniqueness construct. This measure was modified from a measure used by Sirgy, Johar, and Wood (1986) in which product common usage was hypothesized and found to be negatively related to product value-expressiveness. Based on factor and reliability analyses, the following items were retained: (+) "X" is directed to a highly select market, whereas "Y" isn’t. (-) The majority of consumers buy "X." This is not the case for "Y." (+) Only a very select few use "X." Everyone else seems to use "Y." (+) Not a lot of people use "X." This is not the case with regard to "Y." (+) There is a much smaller minority of people who use "X" than use "Y." The mean and standard deviation of the overall measure were 3.556 and .828, respectively.

RESULTS

The results are reported by hypothesis. Hypothesis 1 states that self-congruity (actual-, ideal-, social-, and ideal social self-congruity) is predictive of brand attitude. Two regressions functional congruity treated as a covariate were run. Self-congruity was found to be significantly (p<.01) related to brand attitude both with (.271) and without functional congruity (.522). These results provide support for the notion that self-congruity does play a significant role in predicting brand attitude (Hypothesis 1).

Hypothesis 2 states that self-congruity is likely to be more predictive of brand attitude under conditions of high than low brand conspicuousness. A median split was used to create two groups, one set with functional congruity as a covariate and another set without the covariate. The results show that self-congruity was slightly more predictive of brand attitude in the high conspicuousness condition (.481) than the low condition (.442); however, the difference was not statistically significant at the .05 alpha level. However, the same pattern was more evident when the effects of functional congruity was partialled out (high brand conspicuousness=.348; low brand conspicuousness=.219).

Hypothesis 3 states that self-congruity is likely to be more predictive of brand attitude under conditions of high than low uniqueness. A median split was used to create two groups of uniqueness (high and low). Regression analyses were run on these two groups, one set with functional congruity as a covariate and the other without. The results show that self-congruity was not more predictive of brand attitude in the high uniqueness condition (.588) than the low uniqueness condition (.577). This was also evident when the effects of the covariate was partialled out (.311 vs. .334).

Hypothesis 4 states that functional congruity is predictive of brand attitude. Two regressions (one with self-congruity as a covariate and one without the covariate) were run. Functional congruity was found to be significantly related to brand attitude both with the covariate (.462) and without the covariate (.612). These results provide support that functional congruity does play a significant role in predicting brand attitude.

Hypothesis 5 states that functional congruity is likely to be more predictive of brand attitude under conditions of high than low brand involvement A median split was used to create two groups of brand involvement: high and low. Regression analyses were run on these two groups, one set with self-congruity as a covariate (.495 vs. .380) and another set without the covariate (.575 vs. .537). The results show that functional congruity was indeed slightly more predictive under high than low involvement conditions, with and without the covariate.

Hypothesis 6 states that functional congruity is likely to be more predictive of brand attitude under conditions of high rather than low brand differentiation. A median split was used to create two groups of brand differentiation: high and low. Regression analyses were run on these two groups, one set with a self-congruity as covariate (.441 vs. .377) and another without the covariate (.603 vs. .585). The pattern shows that functional congruity was somewhat more predictive under high rather than low differentiation conditions, with and without the covariate.

Hypothesis 7 states that functional congruity is likely to be more predictive of brand attitude than self-congruity. The results indicate that indeed functional congruity is more predictive (.462) than self-congruity (.271), accounting for 42 percent of the variance in brand attitude. These results provide support for Hypothesis 7.

Hypothesis 8 states that functional congruity is a direct function of self-congruity, i.e., the self-congruity effect is partially mediated through functional congruity. The regression esults show that functional congruity is significantly predicted by self-congruity (.180) with the covariates (brand attitude, brand differentiation, brand conspicuousness, brand uniqueness, and brand involvement) and without the covariates (.554).

REFERENCES

Claiborne, C. B. and Sirgy, M. J., Self-Congruity as a Model of Attitude Formation and Change: Conceptual Review and Guide for Future Research, in Developments in Marketing Science, B. J. Dunlap, ed., Academy of Marketing Science, Cullowhee, NC, (1990): 1-7.

Johar, J. S. and Sirgy, M. J. Positioning Models in Marketing: Toward a Normative-Integrated Model. Journal of Business and Psychology 3 (1989): 475-485.

Johar, J. S. and Sirgy, M. J. Value-Expressive Versus Utilitarian Advertising Appeals: When and Why to Use Which Appeal, Journal of Advertising 20 (September 1991): 23-35.

Samli, A. C. and Sirgy, M. J. A Multidimensional Approach to Analyzing Store Loyalty: A Predictive Model. In The Changing Marketing Environment: New Theories and Applications, K. Bernhardt and B. Kehoe (eds)., Chicago: American Marketing Association (1982): 113-116.

Shavitt, S. Evidence for Predicting the Effectiveness of Value. Expressive versus Utilitarian Appeals: A Reply to Johar and Sirgy. Journal of Advertising, 21 (June 1992): 47-52.

Sirgy, M. J. Self-Concept in Consumer Behavior: A Critical Review. Journal of Consumer Research 9 (December 1982): 287-300.

Sirgy, M. J. Self-Image/Product-Image Congruity and Consumer Decision-Making. The International Journal of Management 2 (December 1985a): 49-63.

Sirgy, M. J. Using Self-Congruity and Ideal Congruity to Predict Purchase Motivation. Journal of Business Research 13 (June 1985b): 195-206.

Sirgy, M. J. Self-Congruity: Toward a Theory of Personality and Cybernetics. New York: Praeger Publishers (1986).

Sirgy, M. J., Dhruv, G., Mangleburg, T., Park, J. O., Chon, K. S., Claiborne, C. B., Johar, J. S., and Berkman, H. Assessing the Predictive Validity of Two Methods of Self-Image Congruence. Journal of the Academy of Marketing Science 25 (Summer 1997): 229-241.

Sirgy, M. J., Johar, J. S., Samli, A. C., and Claiborne, C. B., Self-Congruity: Predictors of Consumer Behavior. Journal of the Academy of Marketing Science 19 (Fall 1991): 363-375.

Sirgy, M. J., and Johar, J. S. Value-Expressive Versus Utilitarian Advertising Appeals: A Reply to Shavitt. Journal of Advertising: 21 (June 1992): 53-54.

Sirgy, M. J., Johar, J. S. and Wood, M. Determinants of Product Value-Expressiveness: Another Look at Conspicuousness, Differentiation, and Common Usage. In Developments in Marketing Science, N. Malhotra, (ed.), Atlanta, GA: Academy of Marketing Science, Atlanta, GA (1986).

Sirgy, M. J., Johar, J. S., Samli, A. C., and Claiborne, C. B. Self-Congruity versus Functional Congruity: Predictors of Consumer Behavior. Journal of the Academy of Marketing Science 19 (Fall 1991): 363-375.

Sirgy, M. J. and Samli, A. C. A Path Analytic Model of Store Loyalty Involving Self Concept, Store Image, Socioeconomic Status, and Geographic Loyalty. The Journal of the Academy of Marketing Science 13 (Summer 1985): 265-291.

Zaichowsky, J. Measuring the Involvement Construct. Journal of Consumer Research 12 (December 1985): 341-352.

----------------------------------------