Journal ol"Personality and Social Psychology 1995. Vol. 68. No. 3. 518-530 Copyright 1995 by the American Psychological Association, Inc. 0O22-3514/95/$3.0O Self-Esteem as an Interpersonal Monitor: The Sociometer Hypothesis Mark R. Leary Wake Forest University Ellen S. Tambor Johns Hopkins University Sonja K. Terdal Northwestern Michigan College Deborah L. Downs Ohio State University Five studies tested hypotheses derived from the sociometer model of self-esteem according to which the self-esteem system monitors others' reactions and alerts the individual to the possibility of social exclusion. Study 1 showed that the effects of events on participants' state self-esteem paralleled their assumptions about whether such events would lead others to accept or reject them. In Study 2, participants' ratings of how included they felt in a real social situation correlated highly with their self-esteem feelings. In Studies 3 and 4, social exclusion caused decreases in self-esteem when re- spondents were excluded from a group for personal reasons, but not when exclusion was random, but this effect was not mediated by self-presentation. Study 5 showed that trait self-esteem correlated highly with the degree to which respondents generally felt included versus excluded by other people. Overall, results provided converging evidence for the sociometer model. The proposition that people have a fundamental need to maintain their self-esteem has provided the cornerstone for a great deal ofwork in personality, social, developmental, clinical, and counseling psychology. In the century since William James (1890) first referred to self-esteem as an "elementary endow- ment of human nature," many classic theories of personality have addressed the importance of self-esteem needs, many emotional and behavioral problems have been attributed to un- fulfilled needs for self-esteem, and many psychotherapeutic ap- proaches have focused in one way or another on the client's feel- ings about himself or herself (Adler, 1930 Allport, 1937 Bed- nar. Wells, & Peterson, 1989 Horney, 1937 Maslow, 1968 Rogers, 1959). Among social psychologists, the self-esteem mo- tive has been offered as an explanation of a wide array of phe- nomena, including self-serving attributions (Blaine & Crocker, 1993), reactions to evaluation (S. C. Jones, 1973), self-handi- capping (E. E. Jones & Berglas, 1978), downward social com- parison (Wills, 1981), attitude change (Steele, 1988), and in- group/but-group perceptions (Crocker, Thompson, McGraw, &Ingerman, 1987). Despite the fact that the self-esteem motive has been invoked to explain so many phenomena, little attention has been paid to the source or functions ofthe self-esteem motive itself. The field has taken it for granted that people have a motive to protect Mark R. Leary, Department of Psychology, Wake Forest University Ellen S. Tambor, Genetics and Public Policy Studies, Johns Hopkins University Sonja K. Terdal, Department of Sociology, Northwestern Michigan College Deborah L. Downs, Department of Psychology, Ohio State University. We thank Robin Kowalski, Cathy Seta, and James Shepperd for their comments on earlier versions of this article. Correspondence concerning this article should be sent to Mark R. Leary, Department of Psychology, Wake Forest University, Winston-Sa- lem. North Carolina 27109. their self-esteem without adequately addressing the question of why they should have such a motive or what function it might serve. In five studies we evaluated the hypothesis that the self- esteem system functions as a sociometer that monitors the de- gree to which the individual is being included versus excluded by other people and that motivates the person to behave in ways that minimize the probability of rejection or exclusion. Explanations of the Self-Esteem Motive Although few efforts have been made to systematically ad- dress the functions of the self-esteem motive, at least three gen- eral explanations of the motive can be gleaned from the literature. The most widely acknowledged explanation is that people strive for self-esteem because high self-esteem promotes posi- tive affect by buffering the person against stress and other nega- tive emotions and by enhancing personal adjustment, whereas low self-esteem is associated with depression, anxiety, and mal- adjustment. Research findings attest that people with low self- esteem experience virtually every negative emotion more com- monly than those with high self-esteem (e.g., Cutrona, 1982 Goswick & Jones, 1981 Leary, 1983 Taylor & Brown, 1988 White, 1981). Furthermore, high self-esteem appears to buffer people against feelings ofanxiety, enhance coping, and promote physical health (Baumeister, 1993 Greenberg et al., 1992 Tay- lor & Brown, 1988). Although the link between self-esteem, affect, adjustment, and health is undisputed, it is less clear why self-esteem should produce these effects. One possibility is that, because self-es- teem is associated with confidence and high expectations ofsuc- cess, high self-esteem is associated with optimism and lowered anxiety. In a variation on this theme, Greenberg, Pyszczynski, and Solomon (1986) suggested that high self-esteem serves as a buffer against the existential anxiety people experience when they contemplate their own fragility and mortality. However, it 518

SELF-ESTEEM FEELINGS 519 is unclear why such a psychological system for buffering people against anxiety and uncertainty should have developed. Indeed, from an evolutionary perspective, we might expect that people who worried about possible misfortunes (including death) would have been more likely to survive and reproduce. A second set of explanations emphasizes the role of high self- esteem in promoting goal achievement. The motive to seek self- esteem may have developed because high self-esteem enhances people's willingness to strive toward desired goals and to persist in the face of obstacles and setbacks (Bandura, 1977 Greenwald, 1980 Kernis, 1995). In a related vein, Tedeschi and Norman (1985) suggested that people seek self-esteem be- cause self-esteem is associated with feelings of control over one's environment. In support of explanations that implicate goal accomplish- ment, people with high self-esteem often work harder and per- form better after an initial failure than people with low self- esteem (Perez, 1973 Shrauger & Sorman, 1977). However, it is also true that high self-esteem may lead to nonproductive per- sistence when tasks prove to be insurmountable (McFarlin, Baumeister, & Blascovich, 1984). Although self-efficacious cog- nitions can undoubtedly facilitate achievement, this explana- tion cannot explain why self-esteem is inherently affectively laden (Brown, 1993). Although we concur that it may often be useful for people to think that they possess certain favorable attributes or abilities (i.e., to have high self-efficacy), this does not explain why self-esteem is intimately linked to strong self- relevant emotions. A third set of explanations of the self-esteem motive involves the possibility that people seek self-esteem for its own sake. Many writers have implicitly assumed the existence of a self- system that maintains a sense of integrity or adequacy (e.g., Epstein, 1973 James, 1890 Steele, 1988). One difficulty with this assumption is that it fails to explain why a motive to behave in ways that promote self-esteem should exist at all. It also does not adequately explain why certain events pose threats to self- integrity and others do not. In fact, to the extent that, over the long run, rewarding encounters and the attainment of one's goals depends on accurate knowledge of oneself, a self-system that functioned purely to elevate one's sense ofselfmay actually be less adaptive than one that sought accurate self-knowledge (Heatherton & Ambady, 1993). Properties of the Self-Esteem System Although each of these explanations can explain why people prefer to evaluate themselves positively under certain circum- stances, none clearly and fully explains why people appear gen- erally to need self-esteem and regularly behave in ways to main- tain and enhance it. Before offering an alternative explanation of the self-esteem motive that we believe parsimoniously ex- plains the properties of the self-esteem system, we must clarify precisely what we mean by the term self-esteem. Self-esteem has often been described as an attitude, specifi- cally an attitude toward oneself (Coopersmith, 1967 Rosen- berg, 1965). Like all attitudes, self-esteem has cognitive and affective components. A distinction can be drawn between the self-concept (beliefs about the self) and self-esteem (evaluation ofoneself in light ofthose beliefs). Although self-esteem is often based on self-relevant cognitions, not all cognitions about the self, even evaluatively laden ones, are relevant to a person's self- esteem. Each person has many self-beliefs that have no affective quality. People may believe firmly that they are very good or very bad at certain mundane tasks, for example, yet experience no corresponding increase or decrease in their self-esteem. Self-esteem includes an essential affective quality that "cold" cognitions about the self do not. Brown (1993) persuasively ar- gued that self-esteem is fundamentally based in affective pro- cesses, specifically positive and negative feelings about oneself. People do not simply think favorable or unfavorable self-rele- vant thoughts theyfeel good or bad about themselves. Further- more, they fiercely desire to feel good rather than bad. Most previous explanations of the self-esteem motive have difficulty explaining the inherent emotional and motivational qualities of self-esteem. We return to this point shortly. Although people can be characterized as having some average level of self-esteem over situations and time (trait self-esteem), self-esteem inevitably fluctuates as people move about their daily lives (state self-esteem). To our knowledge, researchers have not previously addressed the question of whether people are motivated to maintain state self-esteem, trait self-esteem, or both. However, we think it is reasonable to assume that people want both to feel good about themselves in the present moment as well as maintain positive self-feelings over time. As will be- come clear, stale self-esteem is of paramount importance in the explanation of the self-esteem system we describe. Self-Esteem System as a Sociometer The hypothesis to be considered in this article is that the self- esteem system is a sociometer that is involved in the mainte- nance of interpersonal relations (Leary, 1990 Leary & Downs, 1995). Specifically, a person's feelings of state self-esteem are an internal, subjective index or marker of the degree to which the individual is being included versus excluded by other people (the person's inclusionary status) and the motive to maintain self-esteem functions to protect the person against social rejec- tion and exclusion. We believe that this perspective on self-es- teem more parsimoniously explains the emotional and motiva- tional aspects of self-esteem than other explanations. Many writers have observed that human beings possess a fun- damental motive to seek inclusion and to avoid exclusion from important social groups and that such a motive to promote gre- gariousness and social bonding may have evolved because of its survival value (Ainsworth, 1989 Barash, 1977 Baumeister & Leary, in press Baumeister & Tice, 1990 Bowlby, 1969 Hogan, 1982 Hogan, Jones, & Cheek, 1985). Because solitary human beings in a primitive state are unlikely to survive and reproduce, psychological systems evolved that motivated people to develop and maintain some minimum level of inclusion in social rela- tionships and groups. Successfully maintaining one's connections to other people requires a system for monitoring others' reactions, specifically the degree to which other people are likely to reject or exclude the individual. Such a system must monitor one's inclusionary status more or less continuously for cues that connote disap- proval, rejection, or exclusion (i.e., it must be capable of func- tioning preconsciously), it must alert the individual to changes