Role of Admissions Test Scores, Socioeconomic Status, and High School Grades in Predicting College Achievement

Abstract

Although test scores are widely used in college admissions in the United States, their use is the subject of ongoing debate, partly because of the association between test performance and socioeconomic status (SES). Although test critics have argued that this association is due to the particular content of admissions tests or to the differential availability of coaching, large socioeconomic effects are also found in assessments that are tied to school achievement and for which coaching is not available, such as the National Assessment of Educational Progress and in other academic measures. Some commentators have argued, however, that high school grade-point average has a smaller correlation with SES than admissions test scores and is therefore a superior admissions criterion. In this paper I examine the association between SES and test scores, as well as the association between SES and high school grades, and discuss the relevance of this complex web of associations to college admissions research. While the perennial finding that socioeconomic inequities manifest themselves as educational inequities is disheartening, the analysis of performance differences can point the way toward possible remedies.

Keywords

• admissions
• standardized tests
• admissions testing
• predictive validity