The Relative Validity of Scales Prepared By Naive Item Writers and Those Based On Empirical Methods of Personality Scale Construction1

Abstract

In an effort to evaluate alternative strategies of personality scale construction, the extent to which relatively naive item writers could produce valid personality scales was investigated. Each of 22 undergraduate psychology students was asked individually to prepare 16 items for one of three scales: Social Participation, Tolerance, and Self-Esteem. These items were administered to a sample of 116 females, comprising pairs of roommates, together with like-named scales from the empirically-derived California Psychological Inventory (CPI) and from the Jackson Psychological Inventory (JPI). Self-ratings and roommate ratings on these trait dimensions which served as the two criterion measures were also obtained. Student scale validities, which were much higher than those obtained for the CPI, were almost comparable to those for the JPI. Student scale scores were less free from desirability variance than were JPI scale scores. Like the earlier Ashton-Goldberg study, the results were interpreted as supporting a construct-oriented over an external-empirical strategy of personality scale construction.