Effects of Response Style on Polarity and Validity of Two-Dimensional Mood Models
NAVAL HEALTH RESEARCH CENTER SAN DIEGO CA
Pagination or Media Count:
Unipolar and bipolar two-dimensional models have been proposed to represent mood. This study showed that a given data set will produce both a unipolar model and a bipolar model when a response-style adjustment is employed. The two models provided mathematically equivalent descriptions of a single- factor space and were equally valid for describing mood differences between successful and unsuccessful military recruits and between recruits in different basic training platoons. Both models discriminated between these groups as well as a six-factor model. The findings confirmed the existence of a reliable two- dimensional representation of self-reported mood but indicated that any choice between unipolar and bipolar models must be based on criteria other than the structure of self-reported mood. Further comparisons to more complex mood models are needed to determine whether two dimensions adequately represent self- reported mood when a wider range of situation factors and behavioral correlates are considered. If so, a simple, reliable consensual model for mood would be available to researchers interested in studying mood determinants and effects. Keywords Emotions, Attrition, Factor analysis, Criterion validity, Reprints.