How well does high school grade point average predict college performance?

This report examines how well high school GPA and college entrance exams predict college grades for particular subgroups of students who enrolled directly in college math and English in the University of Alaska system over a four-year period. The report builds on a previous Regional Educational Laboratory Northwest study and examines whether high school GPA is less predictive for certain groups of students, such as students who come from different parts of the state or recent high school graduates versus older students.

This study used regression analysis to assess the extent to which high school GPA and test scores predict college grades. Regressions were estimated separately for English and math course grades and within each subject area for students who took the SAT, students who took the ACT, and students who took ACCUPLACER.

 Overall, high school GPA surpassed test scores in explaining variance in college course grades regardless of where students were from in Alaska. High school GPA explained 9–18 percentage of variance in course grades for urban students, while test scores explained 1–5 percentage of variance. Similarly, high school GPA explained 7–21 percentage of variance in course grades for rural students, while test scores explain 0–3 percentage of variance in course grades. High school GPA was also more predictive of college course performance for students who directly entered college from high school compared to those who delayed entry.

These findings provide evidence of the predictive power of high school GPA in explaining the readiness of college students for college English and math across different groups of students. Secondary and postsecondary stakeholders can use these findings to engage in conversations regarding whether and how to use high school grade point average as part of the placement process.