Linking to a Large-Scale Assessment: An Empirical Evaluation

Abstract

This article develops and evaluates a linkage of a routinely administered measure, the Armed Services Vocational Aptitude Battery (ASVAB), to a large-scale assessment, the mathematics proficiency scale of the National Assessment of Educational Progress (NAEP). Analyses of both simulated and real data indicated that the statistical accuracy of the proficiency distribution estimated by projection (i.e., by regression on the ASVAB sub-scales) was very close to the accuracy obtained by the use of the NAEP methodology. However, for both methods, proficiency may have been systematically underestimated because of motivational factors in the administration of the NAEP measures to examinees in this study.