No abstract

THE Miller Analogies Test (MAT) continues to be one of the most widely used instruments for graduate school selection. A review of MAT research reveals a wide range of validity coefficients. Values from .00 (Travers, 1948) to .69 (Gustad, 1950) are noted. Such discrepancies can in part be attributed to the considerable variation in candidate qualifications and institutional differences. The lack of reliable success criteria is also a significant contributor. The usual criterion of grades is generally conceded to lack sufficient consistency to allow for stable prediction. Very few validity studies can be located which involve the use of objective criteria, such as standardized instruments or end-of-program comprehensive examinations. The purpose of this paper is to report data which compare the predictability of the MAT relative to the criteria of grades and comprehensive examination scores. VariablesThe course work leading to a Masters Degree in Education at a large urban university requires, in addition to the usual electives and prescribed major instruction, a minimum of one course in each of the following areas, Educational Psychology (EP), Meas-1 Data analysis was completed

No abstract