The graduate record examination as a predictor of graduate student performance: The case of criminal justice*

Abstract

Research on the Graduate Record Examination (GRE) as a predictor of graduate student performance has been marked by much debate. Thornell and McCoy (1985) have found that the relationship between student performance and the GRE may be sensitive to the discipline being studied. While unexplained to date, this disciplinary variation may be related to the differences in means and/or relative dispersions of GRE scores among disciplines. Further, there has been only one study (McKee, Mallory, and Campbell, in press) for criminal justice, the field with the lowest mean GRE score. This paper focuses on this neglected field. Data were collected from the files of 70 criminal justice majors in a master's program at a large midwestern urban university. The results indicate that GRE scores are largely unrelated to indicators of graduate student performance (GPA, grades in specific classes, and the completion of the M.A. degree). The one exception is a strong relationship between verbal GRE and graduate GPA. This relationship may indicate a nonlinear ceiling effect wherein verbal GRE scores affect GGPA, mainly in disciplines with means at the low end of the verbal GRE score distribution.