In recent years, a wealth of studies have been published, showing that tutoring, especially high-dosage tutoring, is one of the most promising interventions for closing learning gaps. I’ve also researched this. Everything consistently pointed to substantial and often encouraging effects.
That’s why the latest interim report from the Personalized Learning Initiative (UChicago Education Lab and MDRC ) is quite sobering. Not because tutoring doesn’t work—it remains an effective approach on average—but because the results at scale are far less spectacular. We’d come to expect better based on previous studies. As is so often the case in education, something works in a small setting. However, scaling it up is a different story.
What did the researchers find? Across all sites combined, we observe effects of approximately 0.06 to 0.09 standard deviations. This corresponds to roughly one to two months of additional learning. This is statistically significant, but much less than the 0.25 or even 0.3 SD found in previous studies. The reason is simple but troubling. In practice, students receive far fewer minutes of tutoring than in those earlier projects. Whereas the original target was 34 to 82 hours per year, students in this large-scale experiment often received no more than 10 to 38 hours.
The pattern is clear: per-minute tutoring gains remain comparable to previous studies, but the dosage is simply too low. Schools proved unable to free up the time, no matter how significant the investments. Conflicting priorities, insufficient time in the schedule, and the multitude of needs after the pandemic often meant that tutoring remained limited in practice.
That’s what makes this report both interesting and confronting. Yes, tutoring remains one of the best tools we have. Yes, cheaper models with larger groups or the use of technology proved to be about as effective. More expensive forms, however, are still necessary. The bottom line remains: without sufficient intensity, much of the promise is lost.
For those like me who have had high hopes and conducted research on the power of tutoring, this is a reality check. The problem isn’t that tutoring doesn’t work, but that we’re barely reaching the necessary level in schools. The challenge, then, is less about “tutoring or not?” and more about “how do we ensure that students actually receive it sufficiently?”
Image: https://commons.wikimedia.org/wiki/File:Tutoring_Center,_Tulane_University_2009.jpg
