Overview

Ohio’s near-cutoff study finds a local response to retention threat, while a 2025 dissertation synthesizing 45 selected U.S. studies and 801 effect sizes finds no reliable average benefit from actual retention. The contrast is compatible because treatments, comparators, timing, supports, selection, costs, and scope differ; retention-plus leaves the repeated year’s value unresolved.

Learning outcomes

Ohio measures a local threat effect; the synthesis measures actual grade repetition, so their different results are not contradictory.
Across 45 selected U.S. studies and 801 effect sizes, the synthesis found weighted Hedges’ g = 0.00, with no reliable average academic or social-emotional benefit.
Retention-plus bundles can include valuable support, but available evidence cannot assign improvement to repetition alone.

Keywords

Third Grade Reading GuaranteeRetentionMeta-analysisRetention-plusEquityCosts

What Happens When the Threat Becomes Repetition

The hardest judgment in the retention debate comes when two credible-looking findings point in different directions. Ohio students near a fall cutoff improved when they faced a risk of being held back, yet a broader synthesis finds no reliable average benefit for students who actually repeat a grade. This chapter compares those results without forcing them into a contradiction. The key is to ask what treatment each source measures, who receives it, what supports are bundled with it, and what costs the decision creates.

A positive effect of a warning does not automatically validate the sanction that makes the warning credible. Conversely, a weak average effect of repeating a grade does not prove that no student ever benefits from a carefully designed additional year. The evidence is strongest when those two statements are held together rather than simplified into a verdict about “retention” in the abstract.

The broader synthesis finds no reliable average benefit

The contemporary synthesis is a 2025 dissertation that reviews U.S. research published from 2000 through 2023. It includes studies that reported the effects of grade retention, used a promoted comparison group, and provided enough information to calculate a standardized effect. After searching and screening the literature, the author included 45 studies and calculated 801 effect sizes. Each effect size describes the difference between retained students and a comparison group on a common scale; the estimates were weighted so that results with greater statistical precision contributed more to the pooled average.

Across all 801 estimates, the weighted average Hedges’ g was 0.00, reported as not statistically significant. The academic-achievement category averaged 0.01 across 514 analyses, and social-emotional adjustment averaged −0.01 across 287 analyses. In practical terms, the synthesis does not find a dependable average advantage for retained students over the promoted comparison groups in the selected contemporary literature.

A result near zero needs to be interpreted carefully. It is not evidence that every retained student has exactly the same outcome as a promoted student. Some students may benefit from a repeated year, some may be harmed, and some may experience no meaningful change. A pooled average can conceal that variation. Nor is the estimate a test of the Ohio threat. The retained students in the synthesis received a later and different treatment, and many of the underlying studies examine students selected for retention through ordinary school decisions rather than through a near-cutoff comparison.

The outcome categories also matter. The synthesis does not ask only whether scores rise during the repeated year. It includes academic achievement and social-emotional adjustment, with specific categories such as reading, mathematics, attendance or enrollment, behavior, self-concept, and other forms of adjustment. The overall conclusion is consequently about the absence of a reliable average benefit across a wide set of outcomes, not merely about one reading test at one point in time.

The distinction between traditional retention and retention plus makes the average more informative. Traditional retention means that a student repeats the grade without the added package of intensive services that defines the other category.

Published by DeepMox

FAQ

What did the 2025 synthesis find about actual retention?

It found no statistically reliable average benefit from actual retention across 45 selected U.S. studies published from 2000 through 2023 and 801 effect sizes, including academic and social-emotional outcomes.

Why can Ohio’s threat result and the retention synthesis differ?

They study different treatments: Ohio compares near-cutoff threat exposure, while the synthesis compares students who repeated grades with promoted groups; timing, selection, supports, and outcomes also differ.

Does retention-plus prove that extra support works?

No. Retention-plus combines repetition with services, so available studies cannot separate the support’s contribution from the repeated year or their interaction.

Previous11 minOhio’s Third Grade Reading Guarantee: Support and Capacity

Next10 minOhio’s Third Grade Reading Guarantee: Conditional Answer

Ohio’s Third Grade Reading Guarantee: Repetition Evidence

What Happens When the Threat Becomes Repetition

The broader synthesis finds no reliable average benefit

FAQ