Self- versus peer ratings of specific emotional traits: Evidence of convergent and discriminant validity

David Watson, University of Iowa
Lee A. Clark

Abstract

This study tested a hierarchical model of emotional experience by examining relations between self- and peer ratings on 8 factor-analytically derived affect scales. 150 Ss were assessed in 5-person groups, so that each self-rater was also judged by 4 peers. The raters were generally well-acquainted, but a minority of the judges knew their targets less well. The data showed both higher and lower order factors and thus supported a hierarchical view of rated affect. Specifically, (1) general Negative Affect and Positive Affect factors emerged in both the self- and peer ratings; (2) significant self-peer agreement was found for all 8 specific affect scales; and (3) despite the presence of substantial general factor variance, most of the scales also showed significant discriminant validity. Other analyses demonstrated that self-peer agreement increased with the addition of more peer raters and with greater judge-target acquaintance. Implications of the results for the validity of self- and peer ratings of affect are discussed. (PsycINFO Database Record (c) 2008 APA, all rights reserved)