Q&A: Grades & Test Scores

Key Takeaways

• If we look at the overall population of college-going students, students’ course grades are consistently the strongest indicators of students’ general academic preparation.
  • If you were just going to focus on one for assessing academic readiness, grades provide the most comprehensive information. Tests like the SAT and ACT provide some additional information.
• Some studies may find different conclusions, often because they’re looking at a restricted sample (e.g. students with high GPAs going to very selective schools) and/or using weighted or self-reported GPAs, which are less accurate.
  • How a study is done, and what data researchers use, always matters. It matters a lot here, too.
• Standardized tests provide one type of information about students’ skills and content knowledge in the subject being tested. But they aren’t taken in all subjects, and even within a, subject they can’t measure everything that teachers expect students to learn
  • Standardized tests are like measuring sprint times for soccer players. They can tell you a bit about the player, but not everything you need to know.
  • Grades are more like the full scouting report. Grades are a much more comprehensive measure of learning. Seeing how a player performs on a field might not be as systematic as seeing how fast they run 100 meters, and it might be influenced by who they’re playing and the conditions of the field, but it’s going to provide you more information than just how fast they run.
• There are differences in grading practices and standards across teachers and schools.
  • But since the research body shows that, across all students, grades are very predictive, differences across teachers are small relative to students’ actual academic preparation.
    • Additionally, GPAs average out differences across teachers, so GPAs are particularly good indicators of students’ preparation for further academic work beyond their grades for an individual class.
  • There is also subjectivity in terms of what test makers decide to include on standardized tests, as well as different versions of the test each time students take it, which can lead to different scores, too.

Related Research

• High school GPAs and ACT scores as predictors of college completion: Examining assumptions about consistency across high schools (Allensworth & Clark, 2020)
  • Directly compares high school GPAs and ACT scores as predictors of college graduation.
• Looking forward to high school and college: Middle grade indicators of readiness in Chicago Public Schools (Allensworth, Gwynne, Moore, & de la Torre, 2014)
  • Looks at grades and standardized test scores in the middle grades as predictors of high school performance.
• The predictive power of ninth-grade GPA (Easton, Johnson, & Sartain, 2017)
  • Considers ninth-grade GPA, standardized test scores, and later outcomes.
• Contextualized high school performance: Evidence to inform equitable holistic, test-optional, and test-free admissions policies (Bastedo, Umbricht, Bauch, Byun, & Bai, 2023)
  • Analyzes measures of “contextualized” high school performance (as opposed to “raw” performance)—for example, by considering an applicant’s GPA and standardized test scores in relation to others at their high school. The study found that contextualized high school GPA had a stronger relationship with success than contextualized standardized test scores.
• Crossing the finish line: Completing college at America's public universities (Bowen, Chingos, & McPherson, 2009)
  • Probes graduation rates at 21 flagship public universities and four statewide systems of public higher education and shows the predictivity of GPA for college freshmen’s success.
• Is the sky falling? Grade inflation and the signally power of grades (Pattison, Grodsky, & Muller, 2013)
  • Analyzes GPA data from four nationally representative, longitudinal samples and "do not observe any statistically significant attenuation in the signaling power of grades over this time period. In fact, we find some evidence of an increase in the signaling power of grades."
• Attendance playbook: Smart solutions for reducing student absenteeism post-pandemic (Jordan, 2023)
  • Analyzes more than two dozen effective, readily scalable approaches covering topics ranging from family engagement to the value of attendance incentives, as well as students’ social and emotional well-being, a high priority for educators post-pandemic.

Related Webinar

On January 25, 2024 the To&Through Project hosted a Research Spotlight to discuss research by Dr. Michael Bastedo from the University of Michigan Marsal Family School of Education and Dr. Mark Umbricht, Director of Data and Research for the University of North Carolina System. 

Their study, Contextualized High School Performance: Evidence to Inform Equitable Holistic, Test-Optional, and Test-Free Admissions Policies, looks at measures of “contextualized” high school performance as opposed to “raw” performance—for example, by considering an applicant’s grade-point average (GPA) and standardized test scores in relation to others at their high school.

The study found that contextualized high school GPA had a stronger relationship with success than contextualized standardized test scores. The research spotlight's presentation and discussion addressed the question: how are folks re-thinking what equitable admissions policies can look like?  Watch the presentation here.