This longitudinal study tested three theoretical models of the relationship between contingency of self-worth on social approval and trait self-esteem. These included (1) a selective-determinants model positing that self-reports of social approval contingency moderate effects of perceived regard from others on trait self-esteem; (2) an autonomy model positing that approval contingency is a cause of low self-esteem; and (3) a sociometer model positing that approval contingency is a consequence of self-esteem. Perceived regard from others predicted concurrent trait self-esteem and longitudinal change in self-esteem, irrespective of social approval contingency. Approval contingency did not predict longitudinal change in self-esteem, although self-esteem predicted longitudinal change in approval contingency. These results, which support the sociometer model, were consistent across two somewhat different measures of social approval contingency and four indices of perceived regard and social inclusion.
“Properly speaking, a man has as many social selves as there are individuals who recognize him and carry an image of him in their mind. To wound any one of these his [sic] images is to wound him” ( James, 1890/1950, p. 294).
James’ thoughts regarding the social self have profound implications for the understanding of self-esteem. James suggested that self-esteem is linked to the regard accorded by others, a perspective shared by early symbolic interactionists (Cooley, 1902/1922 and Mead, 1934) and sustained by contemporary interpersonal theories positing that perceived acceptance from others is a determinant of, if not interchangeable with, feelings of self-worth (e.g., Harter, 1993, Leary and Baumeister, 2000, Leary and Downs, 1995, Leary et al., 1995, Murray et al., 2002 and Rosenberg, 1979).
Yet, lay persons’ beliefs regarding their contingencies of self-esteem are often at odds with an interpersonal perspective. For instance, Harter, Stocker, and Robinson (1996) asked middle-school children to indicate whether their self-esteem was a cause of, a consequence of, or unrelated to the approval they received from others. Of those who endorsed only one of the three perspectives, only about a third reported that social approval determined their self-esteem. In the development of the Contingencies of Self-Worth Scale, Crocker and colleagues (Crocker, Luhtanen, Cooper, & Bouvrette, 2003) found that college students’ reports of contingency of self-worth on social approval were only moderate (4.49 on a 1–7 scale), widely variable, and lower than their reports of most other contingencies (i.e., appearance, competition, academic competence, family support, and virtue; the only exception was contingency of self-worth on “God’s love”). As these studies indicate, many people deny that their self-esteem is a function of social approval, in contrast to the interpersonal perspective advocated by many self-esteem theorists.
What are we to make of these disclaimers? In the current research, we test three theoretically derived answers. According to a selective-determinants model, self-reports of social approval contingency reflect actual differences in contingencies of self-worth and thus self-reports of approval contingency should moderate effects of perceived regard from others on trait self-esteem. From an autonomy perspective, contingency of self-worth on social approval reflects an inauthentic, fragile, and conditional sense of self, and consequently has negative consequences for psychological well-being. Thus, we might expect that social approval contingency is a direct cause of low trait self-esteem. The sociometer model posits that perceived regard affects self-esteem for nearly everyone and self-reports of social approval contingency do not moderate effects of perceived regard on self-esteem. Furthermore, the sociometer model construes beliefs regarding contingency of self-worth on social approval as consequence, rather than cause, of trait self-esteem. In its relation to trait self-esteem, then, self-reports of social approval contingency have been construed as moderator, cause, and consequence. Before we describe the current tests of these alternatives, we briefly discuss each of these theoretical perspectives.
Approval-contingent self-esteem appears to be a psychological reality for most people, at least for those in our college student sample; although many deny that their self-esteem is contingent on approval from others, trait self-esteem, even for them, appears strongly related to others’ evaluations and will change over time as a result of approval and disapproval. These results converge with and extend other research suggesting that disclaimers about the effect of social approval on self-esteem are invalid (Leary et al., 2003). Apparently, people cannot truthfully disavow the sociality of the self. Perhaps this explains why James (1890) presented his ideas about the social self and the selective-determinants model in distinct sections of his well-known treatise.
An awareness of the approval-contingent nature of self-esteem does not appear to have detrimental effects on self-esteem. In fact, low trait self-esteem may be a cause, rather than a consequence, of this knowledge. People with low trait self-esteem, with their history of tenuous social bonds, know quite well the effects of rejection on feelings of self-worth. Just as most of us overlook the ground that bears us up, people with high self-esteem may take for granted the social approval that forms the foundation of their self-esteem.
