Changes to standardized college admission exams that decrease the exam’s ability to predict college preparedness can result in reduced earnings for both high- and low-income students, according to new research by Assistant Professor Evan Riehl.

In the paper, “Do less informative college admission exams reduce earnings inequality? Evidence from Colombia,” Riehl examines the national college admission exam in Colombia, which was redesigned in 2000 to address concerns that the test was biased in favor of high-income students.

The exam overhaul successfully reduced test score gaps between high- and low-income students, but it did so primarily by becoming a worse measure of abilities that are important for college success. As a result, the reform caused students to attend colleges where they were more likely to drop out, thereby reducing labor market earnings for both high- and low-income students.

“The redesign of the Colombian exam included a number of changes that made the exam easier,” Riehl said. “For example, the new exam had fewer multiple-choice options per question and fewer questions overall. As a result, the exam became less informative as a measure of students’ potential to succeed in college. This caused low-income students to attend more selective colleges for which they were academically unprepared, and it displaced high-income students to less-selective schools, which led to a decline in graduation rates for both sets of students.”

The findings show that students have higher graduation rates and earnings when their academic preparation is close to that of their classmates. Well-designed college admission exams can improve the match between students and colleges based on academic preparation.

Read the full story on the ILR website.
Julie Greco is a Senior Communications Specialist in the ILR School.