ACT Scores Are Better Than High School GPA

September 25, 2002

ACT Scores Are Better Than High School GPA in Predicting Superior Performance in First Year of College

IOWA CITY, Iowa—ACT scores do a better job than high school grade point average (GPA) in predicting high levels of student success during the first year of college, according to the findings of a recent study conducted by ACT. The results suggest high grades in high school do not necessarily translate into high grades in college, another potential sign of grade inflation in America's high schools.

College admissions officials typically use both high school GPA and scores on college entrance exams, among other criteria, to predict, formally or informally, an applicant's probability of academic success in the first year of college. Academic success is usually measured by first-year college GPA.

According to the study, both ACT scores and high school GPA are effective at predicting academic success in the first year of college when success is defined as a "C" average or higher, with high school GPA being the slightly better predictor. However, ACT scores are also effective at predicting higher levels of success—a "B+" average or higher in college—while high school GPA is not. This is true across all gender, race, and ethnic groups.

"If a student earns high grades in high school, we can't accurately predict that he or she will earn high grades in their first year of college," said ACT Principal Research Associate Julie Noble, a co-author of the study. "If a student earns a high composite score on the ACT Assessment, however, we can effectively make that prediction."

High school GPA is not exclusively a measure of students' academic achievement, according to Noble. Factors such as attendance, completion of assignments, participation in class, and other non-cognitive factors can affect the grades earned by high school students as much as can actual learning.

"It's possible for a student to receive high grades in high school without necessarily mastering all of the content," said Noble, "but this is less likely to occur in college. As a result, some students who earned high grades in high school and expect to perform at the same level in college may find this to be more difficult than they assumed."

While non-cognitive factors such as effort and motivation also affect college grades, noted Noble, they are not likely alone to cause a student to attain superior grade levels unless the academic achievement is there as well.

"Previous research has shown that the difference between getting a 'C' and getting a 'B' in a college course can be a combination of academic achievement and other factors such as attendance, participation in class, diligence, etc.," said Noble. "The difference between getting a 'B' and getting an 'A,' however is more likely to be academic achievement, period."

The ACT study examined the use of high school GPA and ACT composite scores for predicting the first-year college GPA of nearly 700,000 students enrolled at 216 colleges across the nation. The students were enrolled in college during the 1996-1997 school year. The results from the 1996-1997 data were also checked against data from students who enrolled in 1997-1998.

In contrast to high school GPA, the ACT Assessment measures actual student learning without regard to non-cognitive factors. ACT scores are as effective at predicting college GPAs of 3.50 or higher as they are at predicting GPAs of 2.00 or higher.

Nevertheless, the study's findings suggest the best predictor of college success—when success is defined as earning a "C" (2.00) average or higher—remains a combination of ACT scores and high school GPA.

ACT recommends that colleges use scores on the ACT Assessment as one of several factors, not as the sole criterion, when making admission decisions regarding student applicants.