Across the U.S., policies that require children to repeat a grade in elementary school—known as retention policies—have become a popular approach to improve reading proficiency. As of 2025, 18 states have enacted or are set to implement third-grade retention laws.
In a recent article published in the Journal of Public Economics, University of Michigan professors Brian Jacob, Christina Weiland, and Jordan Berne (who earned his doctorate in economics from U-M), along with Katharine Strunk, dean of the University of Pennsylvania's Graduate School of Education, argue that traditional research methods for evaluating these policies are insufficient.
Researchers have often relied on regression discontinuity approaches to estimate the impact of grade retention. However, these methods do not account for the additional support services most students receive when they are identified for possible retention. Furthermore, many modern retention policies include "good cause" exemptions, which reduce the number of students actually held back. The authors contend that these factors can lead to overstated estimates of retention impact.
Their conclusions stem from an analysis of Michigan's Public Act 306, the "Read by Grade Three" law. This law previously mandated retention for third-grade students who scored more than one grade level below on state standardized tests, while also requiring districts to provide extra literacy support to all students eligible for retention—even those promoted to fourth grade.
The researchers found that children who scored just below the retention threshold showed modest but educationally meaningful gains in reading scores one year later, compared to those who barely passed. However, using the standard assumptions of regression discontinuity, the effect size appeared unrealistically large. They also discovered that students' reading skills improved even in districts that did not retain any students.
The authors concluded that positive outcomes resulted not only from retention, but also from other interventions.
"These findings shed light on two neglected issues in the literature: the potential bias in previous estimates of retention effects and the critical importance of implementation data in evaluations of multifaceted literacy laws."
Read "Assessing the impact of grade retention: A cautionary tale of exclusion restrict violations," authored by Jordan S. Berne, Brian A. Jacob, Christina Weiland, and Katharine O. Strunk, in Journal of Public Economics.
More news from the Ford School