Biases in social comparisons: Optimism or pessimism? ¬リニ

Social comparisons typically lead to two kinds of biases: A comparative optimism bias (i.e., a tendency for people to view themselves as more likely than others to be the beneficiaries of positive outcomes) or a comparative pessimism bias (i.e., a tendency for people to view themselves as less likely than others to be such beneficiaries); rarely are people fully calibrated in terms of how they compare to others. However, there is little systematic research on the factors that determine when a comparative optimism versus pessimism bias will occur, how they can be attenuated and whether such attenuation is always desirable. In this paper, we report four studies which demonstrate the following key results: First, we show that perceived level of control over the outcome drives whether a comparative optimism or pessimism bias will occur (Study 1). Second, an increase in perceived similarity between self and a comparison target person attenuates the comparative optimism bias in domains that people view as highly controllable (Study 2a) and attenuates the comparative pessimism bias in domains that people view as less controllable (Study 2b). Finally, we show that people are willing to work harder when they experience more comparative optimism in higher control scenarios and when they experience less comparative pessimism in lower control scenarios, illustrating that motivating people to strive harder for positive outcomes can result from exacerbated or attenuated bias, depending on the context (Study 3). 2008 Elsevier Inc. All rights reserved. The tendency for people to be comparatively optimistic about themselves, to the point of being unrealistic, has been well-documented. People tend to believe that good things are more likely to happen to oneself relative to the average person (e.g., having a gifted child; Weinstein, 1980) and that bad things are less likely to happen to oneself relative to the average person (e.g., the likelihood of being the victim of a crime or contracting a disease; Taylor & Brown, 1988; Taylor, Lerner, Sherman, Sage, & McDowell, 2003; Weinstein, 1980). This bias has been demonstrated across a variety of judgments including susceptibility to health risk (Keller, Lipkus, & Rimer, 2002; Lin, Lin, & Raghubir, 2003; Luce & Kahn, 1999; Menon, Block, & Ramanathan, 2002; Perloff & Fetzer, 1986; Raghubir & Menon, 1998), evaluation of personal characteristics (Alicke, Klotz, Breitenbecher, Yurak, & Vredenburg, 1995; Heine & Lehman, 1997; Kruger & Dunning, 1999; Messick, Bloom, Boldizar, & Samuelson, 1985; Weinstein, 1980), and odds of winning in competitive situations (Windschitl, Kruger, & Simms, 2003). Recent research has also demonstrated a comparative pessimism bias—the tendency for people to believe that certain positive outcomes are more likely to happen for others than for themselves and that certain negative outcomes are more likely to happen for themselves than for others (Higgins, 1987; Lin et al., 2003). While understandably demonstrated in the case of depressives who tend to see themselves in a more negative light (Keller et al., 2002; Lin et al., 2003), this pessimism bias has also been demonstrated in the realms of skill assessment (Kruger, 1999), competitive situations (Windschitl et al., 2003), and games of chance (Lin, Lin, & Raghubir, 2004). But under what circumstances does the comparative optimism versus pessimism bias occur? The literature is relatively silent on this issue and has focused on demonstrating only one of these biases at a time; that is, the evidence supports optimism bias in some situations (e.g., Weinstein, 1980) and pessimism bias in others (e.g., Keller et al., 2002), but there is little research that explicitly examines both biases in the same context or domain. In fact, in a review article, Dunning, Heath, and Suls (2004) highlight the importance of such research. The focus of our paper is to determine conditions under which one might expect to observe a comparative optimism or pessimism bias with the goal of gaining a better understanding of the antecedents of such biases in social comparison. In a series of four studies, we show the following: (a) Comparative optimism is likely to occur 0749-5978/$ see front matter 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.obhdp.2008.05.001 q Order of authorship is reverse-alphabetical and reflects equal contributions by each of the authors. We thank Justin Kruger for his insightful comments on an earlier version of this manuscript, Marie L. Palumbo for her editorial comments, and Andrew C. Park for his graphic assistance. We also gratefully acknowledge the constructive comments from the associate editor, Dr. William P. Bottom, and three anonymous reviewers. * Corresponding author. Fax: +1 201 221 8155. E-mail addresses: [email protected] (G. Menon), [email protected] (E.J. Kyung), [email protected] (N. Agrawal). 1 Fax: +1 212 995 4006. 2 Fax: +1 847 491 2498. Organizational Behavior and Human Decision Processes 108 (2009) 39–52