Colleges have their own ways to view SAT and ACT scores when they’re comparing applicants for admission, and that’s particularly true when applicants have taken those tests more than once.
Some schools want to see only the scores from the latest test, while others want to see the scores from all of the tests taken (and might well average them), while still others “superscore,” which involves taking the highest score achieved on each of a test’s subsections across multiple tests. Somewhat bewilderingly, there are also a few colleges that will superscore either the SAT or the ACT – but not both! (Please see our blog at this link for a more detailed description of superscoring and a listing of colleges that superscore the ACT.)
Because of a recent study by the ACT, we’re going to focus on that test and the surprising findings of its study.
The ACT noted that increasing numbers of applicants were taking the ACT multiple times and wondered which set of data – composite score from the last test taken, highest single composite score achieved, average composite score across multiple tests, or superscore – was the best predictor of students’ subsequent first-year-of-college GPA (FYGPA).
The ACT “found that students who retest[ed] on the ACT perform[ed] better than expected in college based on their test scores for all four scoring methods; however, the prediction error was minimized when superscores were used compared to the other three scoring methods.” So, of those four ways in which scores might be examined, superscores – even though they under-predicted FYGPA – were the best predictor of FYGPA.
One concern about superscoring was that it might present an unrealistically high picture of a student’s ability to achieve (as measured by test scores), and that the more times a student takes the test, the more unrealistically high it might be. The expectation, therefore, was that superscoring would result in an over-prediction of FYGPA. According to the ACT, “The results of the study, however, showed exactly the opposite.”
Among its conclusions, the ACT states that “…we empirically evaluated the validity and fairness of different score-use policies. Based on the findings, ACT now supports the use of superscoring in making college admissions decisions.”
While the trend toward superscoring is undeniable, it will remain up to individual colleges whether or not they’ll superscore the ACT, and that they might not do so now doesn’t mean that they won’t in the future – and vice versa. For that reason, prospective applicants should view as nearly mandatory contacting admissions at colleges of interest to get confirmation of their current superscoring policies.
To students who are contemplating taking the ACT multiple times, unless none of their schools of possible interest superscore the ACT and will never do so (the latter conclusion requiring unlikely prescience), we offer our enthusiastic encouragement: It’s a virtually no-lose situation now, and as increasing numbers of admission offices become of aware of the ACT study’s conclusions, which they surely will, superscoring is almost certain to continue to gain acceptance.