Language Aptitude as Predictor of Performance

Abstract

West Point attempts to enroll cadets in their required core language courses based on their top three choices/preferences. Because seats are limited in the most popular languages, it was necessary to devise an objective method for reassigning cadets to courses with available seats, most of which are in Cat III and IV languages, like Arabic, Chinese, Russian, and Persian. West Point has used the Modern Language Aptitude Test (MLAT) and Pre-Defense Language Aptitude Battery (Pre-DLAB) as tools for making this decision. Analyzing the records of 17,000 cadets, this study uses logistic regression to assess the effectiveness of these aptitude tests in predicting expected performance. The analysis finds that aptitude as measured by the MLAT and Pre-DLAB modestly predicts performance.