Usefulness Of High School Average And Act Scores In Making College Admission Decisions

Ample correlational evidence indicates that high school GPA is usually better than admission test scores in predicting first-year college GPA, although test scores have incremental predictive validity. Many people conclude that this correlational evidence translates directly to usefulness in making admission decisions. The issue of usefulness is more complex than is implied by correlations or by other regression statistics, however. This paper considers two common goals in college admission: maximizing academic success and accurately identifying potentially successful applicants. The usefulness of selection variables in achieving these goals depends not only on the predictive strength of the selection variables (such as measured by correlations), but also on other factors, including the distribution of the selection variables in the applicant population, institutions' selectivity, and their criteria for what constitutes success. This paper considers indicators of usefulness in achieving admission goals, and presents estimates of the indicators based on data from a large sample of four-year institutions. The results suggest that high school GPA is more useful than admission test scores in situations involving low selectivity in admissions and minimal to average academic performance in college. In contrast, test scores are more useful than high school GPA in situations involving high selectivity and high academic performance. In nearly all contexts, test scores have incremental usefulness beyond high school GPA. The paper also presents evidence for two other interesting phenomena: Students use their high school GPAs and test scores to select the institutions they want to attend, and this self selection may be more important than institutions' selection in admissions. Moreover, high school GPA by test score interactions are important in predicting academic success. Relationship of conditional probability of success and its marginal distribution with Success Rate and Accuracy Rate is appended. (Contains 11 tables and 10 figures.).