Confounding Constructs and Levels of Constructs in Affectivity Measurement: An Empirical Investigation

This study examined the potential for confounding constructs with levels of constructs in the measurement of positive and negative affectivity. Evidence for this confound was discovered in connection with measures grounded in a well-researched theoretical structure of affectivity. Self-report responses to affect-related descriptors were collected from a sample of managers in a major manufacturing organization. The results of a usefulness analysis suggested that when negative affectivity and positive affectivity are assessed by using only high-pole descriptors, a reduction in the amount of variance explained by affectivity measures may occur.