Exploring the Causal Effect of Test-Optional Policies on Enrollment Rates for Underrepresented Minority Students
Faculty Advisor Name
Joseph Kush
Department
Department of Graduate Psychology
Description
Test-optional policies have become increasingly prevalent among institutions of higher education over the past decade. As of fall 2023, more than 80% of U.S. bachelor-degree granting institutions no longer require SAT or ACT scores of applicants (FairTest.org). While the SAT and ACT were designed with the intention of promoting access to college, some have argued that standardized tests have perpetuated socioeconomic and racial divides in higher education admissions (e.g., Grodsky et al., 2008). Historically, student scores on standardized admissions tests like the SAT and ACT have had significant weight in admissions decisions. Thus, as the number of universities adopting test-optional policies continues to grow, a notable shift is occurring in college admissions. This has sparked a broader conversation about the impact of admissions policy changes on equity and access in higher education.
Researchers have recently begun investigating the impacts of changes in test-optional policies on various outcomes such as enrollment rates, graduation rates, and academic achievement (i.e., GPA). In particular, researchers have explored potential differential impacts of these policies across various aspects of student identity (e.g., race, gender, socioeconomic status). In general, there have been conflicting findings; some have found that test-optional policies increase enrollment rates for underrepresented groups, while others have claimed test-optional policies stifle diversity efforts. This lack of agreement may be attributed to different samples (e.g., public vs. private, R1 vs. R2 vs. R3, more/less selective) and different methodology (e.g., pre-post t-tests, difference-in-differences).
The current study aims to include as many institutions with available data as possible, to attempt to generalize the effects of test-optional policies on American institutions as a whole. Several prior studies implemented DiD analysis, which estimates the effect of an intervention or treatment by comparing changes in outcomes measured at two single time points (i.e., before intervention, after intervention) for a treatment group and a control group. This method is limited as it only includes observations at two individual time-points, thus making it difficult to meet the parallel trends assumption (Bernal et al., 2019). The current study implements comparative interrupted time series (CITS) which uses multiple consecutive observations pre- and post- intervention. Results yield opportunities for more generalizable discussions regarding the impact of test-optional policies on underrepresented minority enrollment over time.
Exploring the Causal Effect of Test-Optional Policies on Enrollment Rates for Underrepresented Minority Students
Test-optional policies have become increasingly prevalent among institutions of higher education over the past decade. As of fall 2023, more than 80% of U.S. bachelor-degree granting institutions no longer require SAT or ACT scores of applicants (FairTest.org). While the SAT and ACT were designed with the intention of promoting access to college, some have argued that standardized tests have perpetuated socioeconomic and racial divides in higher education admissions (e.g., Grodsky et al., 2008). Historically, student scores on standardized admissions tests like the SAT and ACT have had significant weight in admissions decisions. Thus, as the number of universities adopting test-optional policies continues to grow, a notable shift is occurring in college admissions. This has sparked a broader conversation about the impact of admissions policy changes on equity and access in higher education.
Researchers have recently begun investigating the impacts of changes in test-optional policies on various outcomes such as enrollment rates, graduation rates, and academic achievement (i.e., GPA). In particular, researchers have explored potential differential impacts of these policies across various aspects of student identity (e.g., race, gender, socioeconomic status). In general, there have been conflicting findings; some have found that test-optional policies increase enrollment rates for underrepresented groups, while others have claimed test-optional policies stifle diversity efforts. This lack of agreement may be attributed to different samples (e.g., public vs. private, R1 vs. R2 vs. R3, more/less selective) and different methodology (e.g., pre-post t-tests, difference-in-differences).
The current study aims to include as many institutions with available data as possible, to attempt to generalize the effects of test-optional policies on American institutions as a whole. Several prior studies implemented DiD analysis, which estimates the effect of an intervention or treatment by comparing changes in outcomes measured at two single time points (i.e., before intervention, after intervention) for a treatment group and a control group. This method is limited as it only includes observations at two individual time-points, thus making it difficult to meet the parallel trends assumption (Bernal et al., 2019). The current study implements comparative interrupted time series (CITS) which uses multiple consecutive observations pre- and post- intervention. Results yield opportunities for more generalizable discussions regarding the impact of test-optional policies on underrepresented minority enrollment over time.
