Abstract
15 items from a background information questionnaire, administered to 7,929 students at the beginning of an occupational training program, were used as predictors of performance in 9 Navy “A” schools. The sample of students within each specialty was divided into validation and cross-validation subsamples; basis for this division was the criterion of 8 successive levels of final grade average or disenrollment from school. Although the multiple regression composite consisted of different variables for each specialty, the most significant predictors included: level of schooling prior to enlistment, choosing the specialty by oneself, being knowledgeable about work done in that rating group, and responding that the assignment was congruent with one's abilities and interests. In addition to increasing the multiple correlations in these subgrouping analyses, these non-cognitive variables would probably be quite useful for classification and recruiting personnel.
Get full access to this article
View all access options for this article.
