High school GPA turns out to be a superior predictor of academic success compared to standardized test scores for students at a public university, finds new study.gettyHigh school grade point average (HSGPA) is a superior predictor of college student success compared to standardized test scores, finds a new study that assessed the power of different admission criteria in predicting academic success for students admitted to a large public, urban university system.That’s a main takeaway from a recent National Bureau of Economic Research working paper entitled “Standardized Test Scores and Academic Performance at a Public University System,” authored by City University of New York researchers Theodore J. Joyce, Mina Afrouzi Khosroshahi, Sarah Truelsch, Kerstin Gentsch, and Kyle Du.The results add a new wrinkle to the ongoing debate about the relative effectiveness of standardized tests scores like the ACT or SAT versus high school GPA in predicting various dimensions of college student success like first-year GPA, retention, and graduation. BackgroundDuring the COVID-19 pandemic, most colleges and universities suspended the requirement that students submit SAT/ACT scores as part of their application packages. After the pandemic, while many institutions continued to make submission of test scores optional, a number of elite private colleges and prominent public universities re-instated their SAT/ACT requirement, citing empirical evidence that standardized tests were better predictors of first- year academic outcomes than high school grade point average. In some cases, institutions additionally claimed that requiring ACT/SAT scores actually increased the likelihood that high-achieving students from disadvantaged backgrounds would be admitted.Most of those studies were conducted at highly selective institutions, however, increasing the likelihood that the restricted range of GPA scores that would be typical of applicants to such institutions would attenuate correlations with subsequent college outcomes. Prior research involving public institutions rather than elite colleges has found high school GPA to be more predictive of six-year graduation rates at public universities than ACT/SAT scores.MORE FOR YOUThe New StudyThe researchers analyzed enrollment data from the fall cohorts of first-time students from 2010 to 2024 at 11 four-year colleges that were part of a large public university system. They then linked three different predictors — high school GPA, standardized admission test scores, and scores on a statewide exam in English Language Arts meant to assess high school learning— with three different colleges outcomes — first-year college grade point average, first-to-second-year retention, and four- and six-year graduation rates.They conducted two empirical analyses. First, using data from the 2010- 2019 fall cohorts, they compared the relative importance of high school GPA and SAT scores in predicting student outcomes for the overall sample and for different groups of students. And second, they simulated the task of a college admission officer by predicting first-year outcomes for the 2019-2024 incoming cohorts using only high school grades or HSGPA and standardized test scores to evaluate applicants. They found that high school GPA was the strongest predictor of both first-year college GPA and first-year retention — “a one-standard deviation increase in HSGPA raises first-year GPA by between 0.36 to 0.41 points over a mean of 2.86.” A similar result was found for first-to-second year retention. A one-standard deviation increase in HSGPA increased retention by 6 percentage points in prediction models that included both SAT scores and the quality of the high school students had attended.While both high school GPA and SAT predicted four-year graduation rates, a one-standard deviation increase in HSGPA raised the probability of graduation in four years by between 12 and 14 percentage points over a mean of 35 percent. SAT scores were also a significant predictor of four-year graduation rates, but their effect was less than 25 percent of the effect of HSGPA when the effects of the high school attended were also considered. The high school GPA advantage was magnified with six-year graduation rates – the effect of SAT scores and the ELA exam were only about a tenth of the effect of HSGPA.The researchers also tested whether the predictive strength of high school GPA held up for two subgroups of students: under-represented minorities and those awarded a full Pell grant compared to those with partial Pell support or no Pell support. They again found that HSGPA had stronger correlations with all the academic outcomes compared to SAT or ELA scores.Finally, the researchers estimated how well different models would perform in predicting academic outcomes during the post-COVID period (2020–2024). They found that test-optional models that relied only on HSGPA saw relatively little loss in predicting college success compared to models that included standardized test scores.What might account for the strong predictive power of HSGPA? The researchers speculate that “HSGPA captures a set of skills that are arguably more relevant for college graduation than the academic skills measured by standardized tests.” High school grades provide information about students’ ability to resist distractions, regulate their emotions, and sustain their motivation and work habits across long periods of time as they strive for important goals. “These are the kinds of skills required to persist in the pursuit of a college degree, especially for students facing more obstacles and enjoying fewer resources than those at elite colleges,” write the researchers.