An Attempt to Improve Prediction of College Success of Minority Students by Adjusting for High School Characteristics

• Published in: Educational and psychological measurement 1976-12, Vol.36 (4), p.953-957
• Main Authors: Hedges, Larry V., Majer, Kenneth
• Format: Article
• Language: English
• ISSN: 0013-1644; 1552-3888
• DOI: 10.1177/001316447603600420

A validity study was conducted to determine whether the addition of a high school rating factor would increase the predictivity of freshman grade point average (GPA) for Educational Opportunity Program (EOP) students at the University of California, San Diego (UCSD). The rating factor for each high school was the mean difference between the high school GPA and the UC freshman GPA of the previous year's graduates from the same high school who attended the University of California (UC). The subjects were the 161 EOP students who entered UCSD in fall quarter, 1972, and persisted for at least one year. All students had taken the College Board Scholastic Aptitude Test (SAT) prior to college admission. A stepwise multiple regression analysis yielded no significant increase in the multiple correlation when the high school factor was added to high school GPA, SAT verbal, and SAT mathematics scores as predictors of freshman GPA. Implications of these results for recruitment and counseling of EOP students are discussed.