Elusive specific variance: A marginal effect on the accuracy of personality judgment

Abstract

It has become popular to demonstrate that the specific variance can make a substantial contribution to accurate personality judgements. This study challenges the main assumption that transformation of manifest scores into residual scores is capable of separating the specific from the common variance. Based on two large samples (N = 11,086), in which participants and their well-acquainted observers completed either the NEO-PI-3 or the HEXACO-PI-R personality inventory, we demonstrated that the residual scores still contain a substantial amount of the common variance, which inflates correlations with criterion variables. An alternative method using multiple regression showed that different forms of residual scores have a modest incremental improvement of explained variance beyond the effects of the common variance contained in the manifest scores. On average, the specific variance improves the self-other agreement less than 4 % consequently having a marginal effect on accurate personality judgments.