Do people really believe they are above average?
Abstract (Via Booth)
A question that has plagued self-enhancement research is whether participants truly believe the overly positive self-assessments they report, or whether better-than-average effects reflect mere hopes or self-presentation. In a test of people’s belief in the accuracy of their self-enhancing trait ratings, participants made a series of bets, each time choosing between betting that they had scored at least as high on a personality test as a random other participant, or betting on a random drawing in which the probability of success was matched to their self-assigned percentile rank on the test. They also reported the point at which they would switch their bet from their self-rating to the drawing, or vice versa. Participants were indifferent between betting on themselves or on the drawing, and it took only a slight change in the drawing’s probability for them to switch their bet, indicating that people truly believe their self-enhancing self-assessments.
Introduction (Via Booth)
People are unlikely to know precisely where they stand among their peers on such traits as warmth, wisdom, or wastefulness, or on such abilities as wit, whistling, or woodworking. That has not stopped psychologists from asking people precisely where they think they stand on such dimensions. Numerous investigators, including both authors of this paper, have asked participants to assign themselves a single percentile value specifying their standing among their peers (e.g., Dunning, Meyerowitz, & Holzberg, 1989; Gilovich, Kerr, & Medvec, 1993; Kruger, 1999; Kruger & Dunning, 1999; Williams & Gilovich, in preparation). But what are we to make of these ratings? Imagine someone who is uncertain of her standing on the trait of intelligence, thinking she might rank as high as the 70th percentile and no lower than the 40th. Is she likely to report the midpoint of these high and low values (55th) as her percentile estimate? Or is she likely to give herself the benefit of the doubt and report an estimate close to her subjective ceiling?
Implicit in these questions is the broader issue of whether people stand by their percentile estimates. The issue is important in light of the frequently-observed tendency for people’s percentile estimates to yield an ‘‘above average effect” (e.g., Dunning, Heath, & Suls, 2004). On most positive traits and abilities, the average response is well above average—roughly at the 60th or 65th percentile across positive traits (e.g., Dunning et al., 1989; Kruger & Dunning, 1999). Do people really mean it when they say they are, on average, above average?