Better Than Better-Than-Average (or Not): Elevated and Depressed Self-evaluations Following Unfavorable Social Comparisons

Two experiments were designed to investigate perceivers’ self-evaluations when they received objectively positive above-average performance feedback but were told about another coactor who performed either moderately or much better than the participant. Results indicated that participants responded negatively to this comparison information even though they received better-than-average performance feedback. Participants were given the opportunity to evaluate themselves relative to another coactor who was described as performing at an average level. When the negative implications of the unfavorable social comparisons were relatively mild, both low and high self-esteem participants raised their self-evaluations vis-a`-vis the inferior coactor who performed at an average level on the task. However, when the upward comparisons were especially unfavorable (i.e., when there was a large discrepancy between the performance level of the participant and the coactor—the comparison target), only high self-esteem participants raised their self-evaluations. Results provided evidence for active compensation and relatively passive spreading activation, supporting a schema-maintenance through compensation model (e.g., Seta & Seta, 1982, 1993; Seta, Seta, & McElroy, 2003). As noted by Festinger over fifty years ago, the social environment provides us with a rich source of social comparison information that is helpful for self-knowledge in the absence of more objective forms of information (Festinger, 1954). Although this literature is too vast to adequately review in this paper, the research that was generated from his classic social comparison theory also has made it clear that social comparisons can be painful and can represent a threat to our ego or self-concept (see Collins, 1996; Suls & Wills, 1991). In this paper, we explored how individuals responded to information that they had performed at a level that was above average (i.e., received objectively positive feedback) but also were told that another participant performed either one rank (mild threat) or two ranks (severe threat) above them. A competitive context was induced in which the participants expected to meet with the experimenter to discuss differences in participants’ performance levels. These participants also were given information about a performer who had performed at an average level, putting this performer in a relatively inferior social position vis-a` -vis the participant. Thus, our procedures created a social hierarchy of performance comparison information in a competitive context. Control conditions also were included in which participants received either no information about superior others (Experiment 1) or received information about another performer who performed similarly to the participant (Experiment 2). We asked participants to make self-evaluations of their ability levels relative to the average-other. Based upon the view that participants would respond to these upward comparisons in a relatively negative way, we expected them to be motivated to compensate for these social comparison threats by raising their self-evaluations in relation to the average performer. We also included measurements of participants’ self-esteem levels in this research. We expected levels of self-esteem to interact with the social comparison threat conditions such that both high and low self-esteem participants would be able to compensate for mild social comparison threats but that low self-esteem participants would be unable to compensate for relatively severe social comparison threats. In contrast, we expected high self-esteem participants to be able to compensate for relatively severe levels of threat by raising their selfevaluations. Our reasoning for these predictions is discussed below. We used a schema-maintenance through compensation model (e.g., Seta & Seta, 1982; Seta, Seta, & Erber, 1993; Seta, Seta, & McElroy, 2003) as a theoretical backdrop for this research. Although this model has not been applied to research in social comparison, it is applicable to this domain because unfavorable social comparisons can represent events that are inconsistent with self-related schema. For example, unfavorable upward social comparisons can provide information that one is not meeting goals for self-improvement, or can create negative affective reactions that are not commensurate with maintaining a positive social position. Other research using the schema-maintenance through compensation model has found that individuals respond to threats to important schemata (e.g., stereotypes) by generating information that at least attempts to compensate for, or balances-out, the threatening information (e.g., Seta et al., 1993, 2003).1 It also may be the case that individuals respond similarly to self-threatening information and generate compensatory self-relevant information. Thus, individuals also may attempt to balance the negative implications of social comparison threats. According to our view, compensatory effects occur when individuals are both motivated and able to generate (or access) information that has the potential to balance-out inconsistent and negative experiences. However, when individuals are not motivated or capable, the inconsistent event can have a direct influence and produces responses that are driven by the inconsistent experience. Although there has been research that supports these assumptions (e.g., Alter & Seta, 2005; Hughes & Seta, 2003; Seta et al., 1993; Seta, Seta, & Goodman, 1998; Seta, Hundt, & Seta, 1995; Seta & Seta, 1982, 1993; Seta, Seta, & McElroy, 2003), research has not tested the role of these processes for the self in a context in which inconsistent experiences were induced by upward social comparisons. The purpose of the following research is to extend this model into the domain of social comparison threat. Compensation and spreading of activation. It is worthwhile to consider why individuals might be motivated to compensate for upward social comparisons when they have just received objective information that they performed better than average. Why not just accept the objectively positive feedback and ignore the social comparison information that others have performed better than they in the setting? In many cases, this may indeed occur. However, in competitive contexts upward social comparisons can signal an inferior social position that is associated with the loss of social status and the potential for rejection. This is especially likely in a context in which differences between persons’ performance levels are salient and consequential. Being outperformed by another can represent an important loss and comparers may attempt to compensate for this loss by providing themselves with an especially positive social position relative to another group member. To successfully accomplish this goal, a suitable target must be available—one that allows performers to access relatively positive self-attributes. Typically, a person holding a relatively inferior social position relative to performers can serve this role. Thus, because an above-average performers’ accomplishments are superior in comparison to those of the average person, performers may be able to access self-attributes that are relatively more positive than those of this inferior other. In addition to motivating performers to compensate for the unfavorable social comparison, however, the negative implications associated with this type of upward comparison can serve as a predictive cue for performers’ self-worth by activating negatively valenced self-knowledge information, which spreads outside of awareness and relatively automatically to associatively linked self-concepts. The result of this passive spreading of activation process (e.g., Collins & Loftus, 1975; Higgins, 1996), is an increase in the availability of negative self-information which, in turn, can constrain attempts to access information that compensates for the unfavorable comparison. When the negative implications of an unfavorable social comparison are relatively mild, a performer should be capable of accessing information that at least partly balances the unfavorable experience. However, when the negative implications are severe, they may reduce performers’ ability to access positive self-attributes and, thus, their self-evaluations may be driven by the negative implications of the unfavorable social comparison resulting in deflated self-evaluations. Compensation and self-esteem. If a lack of ability to access positive self-attributes constrains compensation following a failure experience, then person variables related to this factor also should affect this capability. Several studies have examined person factors, such as self-esteem, on perceivers’ self-evaluations following an experienced failure (e.g., Baumeister, 1982; Brown & Gallagher, 1992; Brown & Smart, 1991). In general, these studies have found patterns of elevated self-evaluation following an experienced threat for high, but not low, self-esteem persons (for reviews concerning self-esteem, see Hoyle, Kernis, Leary, & Baldwin, 1999; Kernis, 1995). This may be because low self-esteem persons believe that they have relatively few resources or positive attributes (e.g., Brockner & Elkind, 1984; Brockner et al., 1998), or because they may not be confident that they can defend their positive self-views—either to themselves or to others (e.g., Baumeister, Tice, & Hutton, 1989; Brown & Smart, 1991; Crocker & Wolfe, 2001). If so, then the inability of low self-esteem individuals to self-enhance would be due to deficiencies in cognitive abilities rather than lack of motivation (e.g., Seta, Donaldson, & Seta, 1999; Swann & Schroeder, 1995).2 If deficiencies in cognitive abilities are responsible for the failure of low self-esteem performers to self-enhance, then less ‘‘cognitively demanding’’ conditions may reveal their self-enhancing motivations. With few exceptions (see Brown & Gallagher, 1992), the vast majority of studies that have found especially positive self-evaluations following a negative evaluation have compared the responses of persons who have received negative versus positive evaluations (e.g., Baumeister, 1982; Baumeister & Jones, 1978; Brown & Smart, 1991; Dunning, Luenberger, & Sherman., 1995; Tesser, Crepaz, Collins, Cornell, & Beach, 2000). Although these studies are informative about individuals’ selfevaluations following negative versus positive feedback, these studies do not provide information concerning their responses relative to a comparatively neutral control group in a social comparison context. Therefore, in the present studies, we compared the self-evaluations of performers to those who were not exposed to a threatening upward comparison—a control group. We also provided high and low self-esteem performers with upward social comparisons that represented different levels of negativity and threat, and then measured their self-evaluations relative to an inferior comparison target. A competitive environment was created. In the experimental comparison conditions, relative evaluations were made salient; a relatively high status evaluator was purportedly scheduled to objectively compare differences in the participants’ performance relative to those of another person who was always described as performing at a relatively superior level. Therefore, there was competition established between performers and the superior, and there were clear consequences, such as loss of public face, associated with this situation. Thus, it would be difficult for performers to alter the fact that they were different from—and inferior to—the superior. Our expectations were that both low and high self-esteem performers would be motivated to balance this negative experience by positively differentiating themselves from the inferior. However, low self-esteem performers— because of their limited ability to access positive self-attributes—should be especially likely to be constrained in their ability to compensate. Thus, although they may be capable of compensating when the unfavorable social comparison is relatively mild, they should be less likely to compensate when it becomes more severe. In fact, when exposed to a very unfavorable social comparison, the selfevaluations of low self-esteem persons may be driven primarily by the inferior experience itself. If so, then their self-evaluations would be deflated relative to those of low self-esteem performers who were not exposed to the unfavorable social comparison—a control group. Similarity and elevated self-evaluations. Research has shown that upward social comparisons may be interpreted in a positive light leading to positive affect and elevated self-evaluations, such as when individuals focus on the similarities between themselves and the superior person (e.g., Buunk, Collins, Taylor, Van Yperen, & Dakof, 1990; Buunk & Ybema, 1997; Collins, 1996; Lockwood & Kunda, 1997). Thus, performers also may access positive self-attributes and raise their selfevaluations because they believe they are similar to the superior person. According to the selective accessibility model (e.g., Mussweiler, 2001, 2003; Strack & Mussweiler, 1997), when perceivers believe they belong in the same category as the standard, they engage in similarity testing and assimilate themselves to the standard; they engage in dissimilarity testing, however, and contrast themselves away from the standard when they perceive that they and the superior are in different categories (e.g., Mussweiler & Bodenhausen, 2002). In the present studies, we created a competitive situation and placed performers into a different achievement category relative to the superior; they were either one or two ranks below the superior. Thus, we created a situation in which performers would see themselves as distinct from the superior (e.g., Brewer & Weber, 1994; Brown, Novick, Lord, & Richards, 1992). Differences between performers and the superior also had external consequences in that performers expected an important evaluator to compare their accomplishments to those of the superior. Thus, this type of setting should not induce performers to see their accomplishments as being interdependent or similar to those of the superior. Rather, they should see their accomplishments as discrepant and inferior to the superior, and they should be concerned about the negative consequences associated with differences in their achievements relative to those of the superior. Overview of Experiments 1 and 2. Two studies were designed to determine whether performers would elevate their self-views as a compensatory response to an unfavorable social comparison. In addition, we tested the hypothesis that high and low self-esteem participants would differ in their ability to compensate for this comparison threat. Specifically, we expected both low and high self-esteem performers to elevate their self-evaluations when the social comparison threat was relatively mild and expected only high self-esteem participants to be capable of this form of compensation when the threat was relatively severe. In fact, when faced with a relatively severe threat, the self-evaluations of low self-esteem performers may be deflated relative to comparatively neutral control group performers. In Experiment 2, we measured performers’ negative affective reactions as well as their self-evaluations. Comparisons to superior others were expected to produce negative affect. Furthermore, levels of negative affective reactions should be most intense when the difference between the participant and the comparison target is relatively large. Therefore, we expected to observe the most intense negative affective reactions in the condition in which participants were confronted with the accomplishments of a very superior person, less intense negative affect reactions when the comparison target was only somewhat superior, and the least intense reactions when the accomplishments of performers were similar to those of the target. If elevated social comparisons are compensatory in nature, then we should find relationships between experienced negative affect and self-ratings.