Chalkbeat Ideas is a section featuring reported columns on the big ideas and debates shaping American schools. Sign up for the Ideas newsletter to follow our work.
In 2020, a wonky but buzzy new phrase entered the common educational lexicon: “high-dosage tutoring.”
Academics, policymakers, and journalists, myself included, began describing the idea as a potential solution to the moment’s unfolding learning loss crisis. After all, an average student might vault to the 65th percentile of performance when regularly tutored during the school day, according to widely circulated research summaries.
Education is usually beset with intense political divides — not here. “Tutoring was a shockingly broadly supported policy,” says Matthew Kraft, a Brown University professor who has studied and touted the benefits of tutoring.
At first the barrier seemed to be money — school budgets were tight at that moment — but then in early 2021 schools received a windfall of federal funds.
Among many others, then Education Secretary Miguel Cardona began evangelizing the new strategy. Students who are behind should get “at least 30 minutes per day, three times a week, with a well-trained tutor,” he said in January 2022. If learning loss was the problem, high-dosage tutoring was the research-backed, expert-sanctioned, federally funded answer.
Did it live up to the hopes and hype?
Complications soon emerged as schools sought to build out tutoring. Third-party companies rushed to fill demand, branding themselves as “high-dosage” and “high-impact.” In reality, many companies offered only remote homework help for students who sought it — nothing like the vision of a tutor regularly sitting with a student throughout the school year. These companies often had trouble engaging students, according to reporting from Chalkbeat, other news outlets, and external researchers.
Even when schools kept tutoring in-house, they faced their own issues. In Nashville, Tennessee, district officials initially looked for volunteers, retired teachers, college students, and others to serve as tutors. But these people proved difficult to find and train, according to a study coauthored by Kraft. Nashville schools ended up largely relying on their own teachers and staff. Yet teachers had little free time during the school day, so many tutoring sessions were done before or after school — rather than during the school day, a key tenet of the high-dosage tutoring model.
Denver schools contracted with two external companies to conduct tutoring. One company struggled to find qualified tutors. The other conducted tutoring virtually, which made staffing easier, but students often missed tutoring sessions. Many students in the district didn’t get the recommended hours of tutoring. “It’s great if you can do it with fidelity and if you have qualified tutors,” a district official previously told Chalkbeat. “There are just a lot of components that go into it that make it effective or ineffective.”
The timing was tough in many ways. Schools were still reeling from the pandemic and its consequences, making it tough to stand up a complex new initiative. And a hot labor market made it hard to find would-be tutors.
Meanwhile, the $190 billion in federal money for schools was dwindling quickly. Tutoring was competing with a variety of other needs and demands. Some districts were wary of using one-time funds for staffing so poured much of it into building upgrades. Others used part of the money to lower property tax rates rather than expand education programs.
One widely missed complication was with the underlying research base. When Kraft and colleagues recently re-analyzed prior tutoring studies they found a striking pattern: Larger programs tended to produce meaningful but notably smaller gains. Kraft theorized that as tutoring efforts scaled they skimped on costs and quality. This implied that widely touted effect sizes were unrealistic.
Indeed, studies began to emerge showing only modest gains from tutoring — far smaller than initial research suggested. In Nashville, tutoring produced moderate improvements in reading scores, but null results in math, according to the study by Kraft and colleagues.
A paper released by the University of Chicago Education Lab also found modest gains from tutoring and described the results as both encouraging and “sobering.” These findings kicked off a post-mortem among policy wonks, including in a piece in the Hechinger Report and an extended discussion on the Substack Schooled. Some even suggested the money going into tutoring had largely been wasted.
To Monica Bhatt, coauthor of the University of Chicago research, there’s a ready explanation for the smaller effects of tutoring seen in recent studies. Prior research had focused on deluxe programs that cost multiple thousands of dollars per student to ensure they received regular tutoring.
In contrast, the pandemic-era tutoring was often spread thin. “You spend less, you get less dosage, you get less learning gains,” Bhatt says. In other words, despite its name, high-dosage tutoring was often not high-dosage. Bhatt has been studying ways to keep tutoring costs down while maintaining dosage and effects.
Like many hyped initiatives before it, tutoring produced some gains but fell short of initial hopes. That is disappointing but in some ways was inevitable. The real shame will be if the idea is simply cast aside as another failure, rather than a complicated story with both setbacks and successes that we can continue to learn from.
“An effort to increase personalized and individualized instruction in school, in my view, is following the right North Star,” Kraft says. “But if I had to go back and make the pitch, I would be more conservative about the effects that we might expect and far more vocal about the challenges.”
Matt Barnum is Chalkbeat’s ideas editor. Reach him at mbarnum@chalkbeat.org.
