A study of relationships between scores on the aptitude test of the GRE and graduate quality point averages made by students in a large university in the South



Journal Title

Journal ISSN

Volume Title



The deans of graduate schools have for years sought to discover the most effective methods and techniques for evaluating applicants. Various types of information concerning the qualifications of the applicant and rather readily available at the time of application have been used and experimented with over the years. Among the standardized tests in use, the Graduate Record Examination has been used most expensively. The purpose of this study was to determine the degree of relationship, if any, between scores made on the Aptitude Test of the Graduate Record Examination and grades made in the Graduate School of a large university of the South. In effect, the study involved a somewhat limited evaluation of the GRE Aptitude Test as a predictor of success in academic graduate courses. Stated in another way: What value or significance could be attached to the GRE-AT as an instrument for screening applicants for admission to this particular graduate division? The sample group consisted of 113 students who were admitted and enrolled in the Graduate Division of the large university in the South in the Fall Semester, 1964, in the College of Arts and Sciences. The basic data for the study was obtained from copies of the academic transcripts in the university and from summary sheets provided by the dean of the graduate division. These basic data consisted of (1) Name of student (as per code number of TABLE I), (2) Major area of study, (3) Sex, (4) Age, (5) Undergraduate Grade Point Average (A is 4.00), (7) GRE Quantitative Score, (8) GRE Total Score (sum of V and Q), and (9) Graduate Grade Point Average. These data were punched into cards, the design programmed, and run through a computer. Pearson Product-Moment coefficients of correlation were obtained among the variables involved, but particularly the coefficients between Graduate Grade Point Averages and each of the three scores on the Graduate Record, Verbal, Quantitative, and Total Scores. Expectancy Tables were also set up for these same variables by means of which one might determine the odds, based on appropriate GRE score for the student to attain a particular level of graduate grade point average. In brief, these expectancy tables revealed the percent of students in the sample group with particular levels of GRE scores that attained a Graduate GPA of 3.0, or above. The coefficients of correlation between GRE scores and graduate GPA's on the whole were relatively low, except for several sub groups. Major conclusions are as follows: 1. Coefficients of correlation between GRE Scores and graduate grade point averages for all students (113) ranged from a .30 for Verbal Score to a .01 for the Quantitative Score. 2. When r's were computed between GRE Scores and graduate GPA's separately for the 67 males and 46 females of the total sample group, the r's for the males ranged from .185 for Verbal Scores and .080 for Total Scores, but for females these r's were much higher, from .519 for Verbal to .041 for Quantitative Score. 3. The r's between GRE Scores and graduate GPA's for sub groups were largest for a group of fifteen English Majors ranging from .765 for the Total Score to a .612 for Quantitative Score with a multiple R of .835. 4. In the expectancy tables 85 out of 113 students, or 75% of all students, attained a graduate GPA of 3.0, or above; 47 of 113, or 41%, a GPA of 3.5, or above. 5. From the GRE expectancy tables for all 113 students, 73 of the 85 students with graduate GPA's of 3.0 and above made 400 or above on the GRE-Q; 81 of the 85 with 3.0 or above scored 400 or above on the GRE-V; and 79 of the 85 scored 800 or above on the GRE-T. There was some evidence that quite a few of the students of the sample group may not have done their best on the GRE, for those with an undergraduate GPA of 3.0, or above, usually took the GRE after they had already been admitted to graduate school. For this reason the study may not represent a fair evaluation of the GRE for the purpose of screening applicants for graduate work.



Graduate Record Examination, Prediction of scholastic success