A Cross Validation of the Z Scale and a Further Exploration

Abstract

PSYCHOLOGISTS working with college stu dents who are highly selected have long attempted to refine their efforts at predicting academic suc cess. At University of California campuses where students in the top 12V2 % of the graduating high school class are considered for entrance, the dis tribution of scores on standard measures of academic promise (such as the ACE and the Ter man Concept Mastery Test) becomes truncated, and in order to measure success in terms of grade point average it becomes necessary to investigate variables importantly related to but not identical with scholastic ability. The search for dimensions which would aid in the predictive formula has naturally turned to personality variables. Gough (4) put forth the hypothesis that persons' emotional, personal, and social adjustment relate directly to academic per formance when ability is held more or less com parable. This thesis was supported by his re search (5) which resulted in a scale having a mean correlation of .38 with course grades, a correlation which seemed to hold for a large number of cases and over eleven cross-validating samples. Earlier, Altus (1) on the basis of an item analysis of the MMPI found 60 non-intellective items which correlated .39 with psychology term grades, and using the upper-lower quartile method found 26 items correlating .39 with the honor point ratio. More recently, Hackett (7), based on his