Confounding of Causal Attribution and Trait Consistency in Research on Determinants of Trait Inference Patterns

Abstract

Listen

Translate article

Evaluative and descriptive consistency have been proposed as principles accounting for trait inference patterns (Schneider, 1973). This study raises the possibility of a confounding in earlier research of these two principles and another plausible but previously unrecognized trait inference principle, causal attribution. Subjects rated 96 pairs of personality traits (derived fron Peabody, 1967) for probability of cooc currence and probability of causal relationship. These trait pairs were divided into four categories: evaluatively consistent--deseriptively con sistent, evaluatively consistent--descriptively inconsistent, evalua tively inconsistent--descriptively consistent, and evaluatively incon sistent--descriptively inconsistent. In an analysis of variance using evaluative and descriptive consistency as independent variables and prob ability of causal relationship as dependent variable, both types of con sistency were related significantly to probability of causal relation ship. In addition, correlations between probability of causal relation ship and probability of cooccurrence indicate that causal attribution might have accounted for trait inference patterns previously explained in terms of trait consistency.