Reliability and scope in personality assessment: A comparison of the Cattell and Eysenck inventories

Abstract

One questionnaire designed to measure primary personality traits (the 16 PF) and another designed to measure second-order traits (the EPI) were compared in terms of reliability and comprehensiveness. Reliability differences between the two tests appeared to be due to their relative length rather than any inherent characteristics of the underlying traits. Two approaches for evaluating the overlap between the two scales were explored. The theoretical conclusion appears to be that the reliable portions of the EPI scale scores can be entirely reconstructed from a knowledge of 16 PF scores. In contrast, the EPI is far narrower in its scope and able to explain less than a third of a reliable variance in the 16 PF.