Some Observations on Self-Concept in Consumer Behavior Research

J. Paul Peter, Washington University
[ to cite ]:
J. Paul Peter (1980) ,"Some Observations on Self-Concept in Consumer Behavior Research", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 615-616.

Advances in Consumer Research Volume 7, 1980     Pages 615-616


J. Paul Peter, Washington University


The study of self-concept has recently been categorized as being in the decline stage of its life cycle (Bettman, et. al., 1978). This should not lead one to believe that self-concept ever received a major research effort in marketing/consumer behavior since there are probably less than twenty studies of the construct in the area. Several reasons could be offered for the failure of self-concept to become a major consumer behavior construct. First, the construct is so broad conceptually that it is difficult to operationalize in a meaningful way. For example, most definitions include a variety of other constructs such as attitudes, feelings, perceptions, evaluations, etc., which an individual holds about himself/ herself and some view self-concept as everything mediating between stimuli and responses. Clearly, such broad conceptual views are not readily operationalizable in a battery of valid measures.

Second, the construct is not as directly related to the purchase/consumption process as some other notions, such as behavioral intentions or attitudes. Thus, self-concept may not be as useful as other constructs for market segmentation or predicting consumer behavior, two of our more heavily researched issues. However, the construct is intuitively appealing and will probably continue to receive a modicum of research effort. It will also continue to be part of the conventional wisdom in consumer behavior.

The three papers here are all concerned to some degree with the self-concept, yet their focus is quite different. Thus, each paper will be discussed individually in the order of their presentation.

Allison, et. al. - Sex Typed Product Images. . .

This paper is commendable for providing a good review of sex role research in consumer behavior and extending such research. However, there are three issues which need to be considered.

First, the paper could be improved by providing the reader a conceptual definition of what the term "sex-role self-concept" means and how the notion relates to the self-concept literature. It would seem from the operational definition that the term relates to how a person perceives his/her sexuality. If so, then a discussion of how sexuality relates to sex-roles and self-concept would seem warranted.

The second issue deals with the interpretation of the repeated measures analysis of variance results. In all the analyses the A x C interaction was significant and the A x B x C interaction was .062 and .078 for two of the analyses. Naturally, this makes interpretation of the results somewhat difficult.

The third issue deals with the addition of two (at best) interval items purported to measure masculinity and femininity. To interpret the sum of the two scales equaling 10 as an indication of unidimensionality rests on two strong assumptions: (1) the scales have the same origin and (2) there is no measurement error in the scales. The same is true for the interpretation of multidimensionality. Thus, it is not clear that this procedure is appropriate. Further, it is not clear how many persons responded "unidimensionally" or "multidimensionally" so the results may be totally random.

In terms of future research, it would seem that usage of the product could well be an important variable which effects responses. For example, if a man perceives himself as masculine and actually uses hairspray, he may well perceive the product as more masculine. In such cases, we may well be measuring cognitive consistency or behavior justification rather than sex-role or self-concept notions.

Munson and Spivey - Assessing Self Concept

This paper is commendable for trying to develop some new notions in the area of self-concept and investigate their applicability in consumer research. However, the reader immediately becomes concerned with the question of whether SP and OSP are in any way isomorphic with cognitive functioning. In other words, the two notions imply that a preference is formed for a specific product and then self perception and other perceptions of self are determined. It seems more tenable that if self perception and others perceptions of self are important aspects of consumer decision making, they would seem to be inputs into forming preferences rather than post-preference evaluations.

A closely related question deals with whether SP and OSP are conceptually distinct. It would seem that self perceptions and others perceptions of self should be highly related and it should not be surprising that people did not distinguish between the two. The authors suggestion for future research concerning sensitizing respondents about the need to distinguish among the different constructs would not seem to be an appropriate approach for discovering how valid self-concept notions really are.

The final issue is operational and concerns the use of D2 in making profile comparisons. Although the authors correctly point out that D2 has some advantages over correlational procedures for comparing profiles, they ignore the fact that D2 exaggerates the larger difference between persons. It is for this reason that Cronbach and Gleser (1953, p. 459) suggest that in most cases it is preferable to obtain D rather than D2. This point may be important in the comparisons of means with and without the actual measure since if the differences were not squared, they may well not be significant.

Schenk and Holman - A Sociological Approach. . .

This paper is commendable for attempting to develop a self-concept notion new to consumer behavior, viz., the "situational self-image." The construct is defined as "the meaning of self that the individual wishes others to have of him/herself." It includes attitudes, perceptions and feelings the individual wishes other individuals in the situation to formulate about his/her character, and appropriate behaviors. The paper is also commendable for its attempt to integrate self and situational influences in terms of symbolic interactionism.

While much of the terminology used in the paper is "new" to consumer behavior, there is some questions as to whether the underlying logic is in any way different from the conventional view. At the risk of oversimplification, it would seem that the paper discusses and illustrates that:

(1) People may play different roles in different situations (presumably) to obtain social rewards.

(2) People may select conspicuous products to enhance their own self-image, i.e., the way they feel about themselves. A portion of how they feel about themselves depends on how they perceive others feel about them.

(3) Situational influences may be an important dimension in determining the appropriate role to play.

Somehow, while the attempt to relate the concepts may he useful, it is not clear that there are really any new insights here. Perhaps the real contribution of the paper is that it illustrates that we can get to the same place through different doors.


Although the major focus of this paper was on suggesting some issues in self-concept research, these suggestions are intended to be constructive. In other words, if the life cycle for self-concept is to be extended, it is hoped that attention to these issues may be helpful to researchers interested in the area.


Bettman, James R., Harold H. Kassarjian and Richard J. Lutz (1978), "Consumer Behavior," in Review of Marketing 1978, Chicago: American Marketing Association, 194-239.

Cronbach, Lee J. and G. C. Gleser, (1953), "Assessing Similarity Between Profiles," Psychological Bulletin, 50, 465-474.