Describing behaviours of self and others: self-enhancing beliefs and language abstraction level

Abstract

Abstractness and frequency of descriptions of positive and negative behaviours were examined as a function of target (self versus others), valence (positive versus negative), and type of behaviour (publicly verifiable behaviours versus not publicly verifiable behaviours). Positive behaviours of self were expected to be reported more often and more abstractly than positive behaviours of others. For negative behaviours, this pattern of effects should be reversed. Further, publicly verifiable behaviours were expected to offer less opportunity for self-enhancement than private and ambiguous behaviours. ANOVA partially supported these expectations. However, the magnitude of self-enhancing beliefs did not depend on the degree to which the behaviours were publicly verifiable. Implications are discussed. Copyright © 1999 John Wiley & Sons, Ltd.