Direct-comparison judgments: When and why above- and below-average effects reverse.

Abstract

Above-average and below-average effects appear to be common and consistent across a variety of judgment domains. For example, several studies show that individual items from a high- (low-) quality set tend to be rated as better (worse) than the other items in the set (e.g., E. E. Giladi & Y. Klar, 2002). Experiments in this article demonstrate reversals of these effects. A novel account is supported, which describes how the timing of the denotation of the to-be-judged item influences attention and ultimately affects the size or direction of comparative biases. The authors discuss how this timing account is relevant for many types of referent-dependent judgments (e.g., probability judgments, resource allocations) and how it intersects with various accounts of comparative bias (focalism, generalized-group, compromise between local and general standards [LOGE]).

Losing sight of oneself in the above-average effect: When egocentrism, focalism, and group diffuseness collide

“…This motivation clearly plays a role (e.g., Alicke & Govorun, 2005), but there are several reasons to question whether it is a necessary condition (see Chambers & Windschitl, 2004). As noted above, AAEs are observed when other persons, such as acquaintances (Klar, 2002), serve as comparison targets, but also when nonsocial objects (e.g., Giladi & Klar, 2002;Windschitl, Conybeare, & Krizan, 2008) are targets. In addition, people sometimes report that they are below-average on certain desirable attributes (e.g., Kruger, 1999).…”

Are people excessive or judicious in their egocentrism? A modeling approach to understanding bias and accuracy in people's optimism.

“…However, when the evidential support for the two hypotheses is strong, the relevant sum might systematically exceed 1.0. These instances of noncomplementarity (and closely related findings, e.g., see Eiser, Pahl, & Prins, 2001;Klar & Giladi, 1997;Lehman, Krosnick, West, & Li, 1992;Moore & Kim, 2003;Windschitl, Conybeare, & Krizan, 2008;Windschitl et al, 2003) can be generally attributed to focalism: a tendency to base a probability judgment primarily on the evidence directly relevant to the focal hypothesis (i.e., the hypothesis for which a probability estimate is sought) rather than the alternative hypothesis. There are actually several versions of focalism accounts (see Chambers & Windschitl, 2004), but each one ultimately stipulates that probability judgments will reflect assessments of evidence for the focal hypothesis more so than assessments of evidence for the alternative hypothesis.…”

“…K would be a default or general expectancy for the types of items being considered. EST would predict that individual focal items drawn from an attractive or otherwise superior set of items would tend to be rated as comparatively better than the other items in the set because s(A) would be greater than K. Focal items from an inferior set would be rated as comparatively worse because s(A) < K (see e.g., Giladi & Klar, 2002;Suls et al, 2007;Windschitl et al, 2008).…”

“…A second and related reason for why EST does not replace specific egocentrism and related accounts is that the EST model is agnostic about the elements of the judgment context that are the preconditions for differential weighting (see Windschitl et al, 2008). That is, the structure of the model would be identical for representing the differential weighting that is due to generic focalism (i.e., the fact that one entity was denoted as focal in the question that solicited a response), some form of egocentrism (i.e., the fact that the question asked about the self), or some difficulty in assessing support for a group of competitors (i.e., the fact that respondent faced more than one competitor).…”

“…However, EST has broad potential and can be applied to understanding bias in many types of referent-dependent judgments (i.e., judgments that require evidence for a target to be compared with evidence for a specific referent or set of referents; see Windschitl et al, 2008). These referent-dependent judgments include comparative ability and trait judgments (e.g., Alicke & Govorun, 2005;Alicke, Klotz, Breitenbecher, Yurak, & Vredenburg, 1995;Dun-ning, Meyerowitz, & Holzberg, 1989;Goethals, Messick, & Alli-son, 1991;Hoorens, 1995;Kruger, 1999;Pahl & Eiser, 2005), comparative optimism judgments (e.g., Blanton et al, 2001;Burger & Burns, 1988;Chambers et al, 2003;Eiser et al, 2001;Heine & Lehman, 1995;HelwegLarsen & Shepperd, 2001;Klar, Medding, & Sarel, 1996;Klein & Weinstein, 1997;Price, 2001;Price, Smith, & Lench, 2006;Rose, Endo, Windschitl, & Suls, in press;Weinstein, 1980Weinstein, , 1984, and generic direct-comparison judgments (e.g., Giladi & Klar, 2002;Klar, 2002;Posavac, Brakus, Jain, & Cronley, 2006;Suls, Krizan, Chambers, & Mortensen, 2007;Windschitl et al, 2008).…”