Connie Wolfe,       Jennifer Crocker,  and  Janetta Lun

   Hanover College               University of Michigan

 

 

Note -  This poster is largely based on a manuscript by Crocker & Wolfe (2000) that has been accepted for publication pending revisions in Psychological Review.

Address correspondence to Connie Wolfe (wolfec@hanover.edu) or Jennifer Crocker (jcrocker@umich.edu).

ABSTRACT

The confusion surrounding the nature and functioning of self-esteem (SE) may be resolved by examining the contingencies on which SE is based, and the match between those contingencies and the events in one's life.  We contend that SE will be influenced by positive and negative life outcomes only when those outcomes are relevant to a person's contingencies of worth, and that people differ in their chronically held contingencies of worth.  Studying contingencies of self-esteem can help to answer some existing questions in the literature. For example, our self-report data suggests that African-Americans may have equal or higher levels of SE as European-Americans in part because they do not base their self-esteem on reflected appraisals (commonly assumed to be a universal and central source of SE). The contingencies perspective also raises new and interesting questions about the development and functioning self-esteem.

    

     We propose that global judgments of self-worth depend on outcomes in domains upon which self-esteem (SE) is contingent. This theoretical approach rests on three basic assumptions (Crocker & Wolfe, 2000):

(1) People differ in their contingencies of self-esteem. For example, one person may stake their self-worth on reflected appraisals while another person's SE may rise and fall with success and failure in school. Others may base their self-worth on religion (God's love), power, self-reliance, competency, love from family, physical strength or appearance, virtue, or social identities. Existing literature supports the notion that self-esteem may come from one or more of several sources (e.g., Beck, 1983; Coopersmith, 1967; Franks & Morolla, 1976; Kernis & Waschull, 1995). This literature, however, tends to focus on various competencies and or reflected appraisals as primary sources of self-worth whereas our perspective allows greater individual variation in contingencies of worth.

(2) Self-esteem is a judgment made in a context and, thus, is not necessarily a stable trait. While SE may be stable over time, we suggest global self-esteem will fluctuate when outcomes in domains in which SE is contingent fluctuate. Past research suggests that SE is more unstable for some people than others (Kernis & Waschull, 1995), and that SE will fluctuate as a result of the standards salient in a context and how one is doing relative to those standards (Quinn & Crocker, 1999, Study 2).

(3) As stated above, we also propose that global self-esteem will rise or fall as a function of positive or negative feedback in domains on which SE is contingent. There is a surprising lack of experimental evidence in which a manipulation successfully changes SE as a dependent variable. We suggest that this may be due to the difficulty of creating laboratory manipulations with real impact, but also because these manipulations were likely not targeted at participants' particular contingencies of worth.  Despite these difficulties, past work by Kernis and his colleagues (Kernis, Cornell, Sun, Berry & Harlow, 1993) supports our assertion by showing a relationship between fluctuations in self-esteem and fluctuations in perceived competence, physical attractiveness and social approval (as a function of importance of those domains).

 

    

     Crocker, Luhtanen, Wolfe & Bouvrette (1999) have developed a self-report scale designed to measure people's perceptions of the extent to which their self-esteem is based on several possible contingencies including: God's love, family support, school competency, outperforming others, and virtue.  Wolfe, Crocker, Coon & Luhtanen (1999) have developed a separate scale to measure reflected appraisals as a contingency of worth.  As self-report measures, these scales are of course imperfect, but should, nevertheless, provide a reasonable estimate of the extent to which each domain is a contingency of worth. In several samples, each scale and subscale has been shown to be reliable (alphas range from .97 to .78) and to have anticipated convergent and divergent validity.

     Using these scales, we are beginning to accumulate direct evidence for the utility of the contingencies perspective. Examination of the standard deviations of each subscale among a large sample of college students (N=543) suggests that, even among a relatively homogeneous group, there are individual differences in the extent to which each domain is regarded as a contingency of worth (Crocker, et al., 1999).

     Perhaps more compellingly, we have also found group differences in the endorsement of several contingencies (Crocker, et al., 1999; Wolfe, et al., 1999). For example, African-Americans are less likely than European-Americans to endorse reflected appraisals as a basis of SE, and more likely, on average, to endorse God's love as a contingency of worth. Preliminary findings revealed a gender difference such that women are more likely to report basing self-esteem on appearance and family support relative to men.

     Important evidence for our perspective is found in research demonstrating that self-reported contingencies of worth do predict reactions to actual outcomes. Crocker, Sommers & Luhtanen (2000) found just such evidence: school competency as a basis of SE predicted whether success or failure in a related domain would impact on global self-esteem.  See Sam Sommers’ poster (the second in this "poster symposium") for details on their intriguing research.

     Finally, preliminary research by Lun & Wolfe (1999) indirectly suggests that self-reported contingencies of worth may also impact on behavioral responses. Lun and Wolfe conducted a partial replication of Fein & Spencer's (1997) work on stereotyping as a response to a self-esteem threat. Lun and Wolfe exposed all participants to a self-esteem threat and were asked, subsequently, to evaluate a target. Results found that people who more strongly endorsed power over others as a basis of SE evaluated a target more harshly than did those who were less likely to endorse power as a contingency of worth. Post hoc analyses suggest an interaction trend such that those basing SE on power were particularly likely to derogate a target who was a member of a stereotyped group. Several other contingencies of worth (competition, reflected appraisals, and school competency) were found to be unrelated to evaluations of the targets.     

     While this research is preliminary, as a whole it provides encouraging empirical support for the contingencies of SE theoretical perspective. There are individual and group differences in contingencies of worth; and, those contingencies have been shown to predict responses to actual outcomes and, perhaps, to influence behavioral responses.

 

    

     This theoretical perspective sheds light on many of the controversies surrounding the nature and functioning of self-esteem.  For example, it has long been assumed that African-Americans would have lower self-esteem than members of non-stigmatized groups-- this assertion rested on the widely held assumption that reflected appraisals are a central source of self-esteem for all people. As mentioned earlier, Wolfe et al. (1999) found, however, that African-Americans were less likely than European-Americans to endorse reflected appraisals as a basis of self-worth. This finding may explain, in part, the substantial empirical findings indicating that African-Americans have as high or higher self-esteem than European-Americans (Twenge & Crocker, 1998).  Another debate in the SE literature is whether self-esteem is best thought of as a stable trait or temporary state. This theoretical approach allows for both possibilities - depending on the stability of the outcomes in domains on which SE is contingent. This theoretical perspective can shed light on many issues in the SE literature, as well as having important implications for work on self-esteem and depression, stigma and aging.

     A contingencies approach to self-esteem also opens up new questions for future research. For example, how many contingencies are adaptive to hold? How do these contingencies develop and change? Is non-contingent SE possible?

     In sum we feel that a contingencies approach to self-esteem can help "rescue" self-esteem as a useful and important construct.

References

 

Beck, A.T. (1983). Cognitive therapy of depression: New perspectives. In P.J. Clayton & J.E. Barrett (Eds.), Treatment of depression: Old controversies and new approaches (pp. 265-284). New York: Raven Press.

 

Coopersmith, S. (1967).  The antecedents of self-esteem. San Francisco: W.H. Freeman.

 

Crocker, J. & Wolfe, C. (2000).  Rescuing self-esteem: A contingencies of worth perspective. Psychological Review.

 

Crocker, J., Luhtanen, R., Wolfe, C. & Bouvrette, S. (1999). Determining the sources of self-esteem: The contingencies of self-esteem scale.  Manuscript in preparation, University of Michigan.

 

Crocker, J., Sommers, S.R., & Luhtanen, R.K. (2000).  Hopes dashed and dreams fulfilled: contingencies of self-esteem and the graduate school admissions process.  Manuscript in preparation, University of Michigan, Ann Arbor, MI.

 

Franks, D. D.,  & Morolla, J.  (1976). Efficacious action and social approval as interaction dimensions of self-esteem: A tentative formulation through construct validation. Sociometry, 39, 324-341.

 

Kernis, M.H. & Waschull, S.B. (1995). The interactive roles of stability and level of self-esteem: Research and theory. In M.P. Zanna (Ed.), Advances in Experimental Social Psychology, Vol. 27 (pp. 93-141). San Diego: Academic Press.

 

Kernis, M.H., Cornell, D.P., Sun, C.R., Berry A.J., & Harlow, T. (1993). There's more to self-esteem than whether it is high or low: The importance of stability of self-esteem. Journal of Personality and Social Psychology, 65, 1190-1204.

 

Lun, J. & Wolfe, C. (1999). Power as a basis of self-esteem: The effects on stereotyping and self-affirmation.  Unpublished senior thesis, University of Michigan.

 

Quinn, D.M. & Crocker, J. (1999). When ideology hurts: Effects of feeling fat and the Protestant Ethic on the psychological well-being of women. Journal of Personality and Social Psychology, 77, 402-414.

 

Fein, S. & Spencer, S.J. (1997). Prejudice as self-image maintenance: affirming the self though derogating others. Journal of Personality and Social Psychology, 73, 31-44.

 

Twenge, J. & Crocker, J. (1998).  Race differences in self-esteem: A meta-analysis.  Manuscript in preparation, University of Michigan.

 

Wolfe, C. Crocker, J., Coon, H., & Luhtanen, R.K. (1999). Reflected and deflected appraisals: Basing self-esteem on others' regard. Manuscript in preparation, University of Michigan.