Predictive Validity of Person Matching at the Scale and Discriminant Analysis Levels

Abstract

Validity comparisons (including hit rates, kappa coefficients, and chance expectancy rates) were made between person matching at the scale and discriminant analysis levels with 5,143 medical students who had taken an interest inventory and had entered their medical residencies. Person matching using the 30 discriminant analysis items demonstrated the greatest ability to assign test takers to the highest number of occupations at a rate greater than chance.