How Much Faith Should Educators Have in High-Dosage Tutoring?
When you’ve been around as long as I have, one gets all manner of intriguing questions. While I usually respond to such queries in private, some seem likely to be of broader interest. So, in “Ask Rick,” I occasionally take up reader queries. If you’d like to send one along, just send it to me, care of Caitlyn Aversman, at caitlyn.aversman@aei.org.
I’m wondering how you’re thinking about high-dosage tutoring. I know the U.S. Secretary of Education has urged districts to embrace it, and lots of communities are investing heavily on it. In my system, we’re thinking hard about this. I read your interview with the head of Varsity Tutors, and it seemed like you’re supportive. But you also frequently seem skeptical of improvement strategies, so I’m curious what your take is on this one or what cautions you might have?
In theory, the promise of tutoring is obvious. It means more individualized attention, geared to a student’s real-time needs. And there’s evidence that tutoring can deliver. A systematic survey of gold-standard studies found that a year of tutoring consistently adds months of academic progress—especially in the early grades. Another review, of nearly 200 high-quality studies, found that high-dosage tutoring (basically, 90 minutes of tutoring a week) is one of the few school-based interventions with big benefits for both reading and math.
So, the appeal is clear. However, as powerful as tutoring may be in theory, it can be expensive and logistically difficult in practice. A number of years ago, the Houston school district launched Apollo 20, an ambitious tutoring experiment for students in grades 5 and 9 in a targeted set of middle and high schools.
The effort was backed by millions in start-up funding and intensive hands-on support from a research team at Harvard University. Despite all these advantages, the practical challenges of recruiting, training, and retaining enough part-time tutors proved daunting. The idea was a good one, but ultimately too difficult and too costly to operate at the desired scale.
And I think you see a lot of this cropping up as states scramble to recruit and train rosters of tutors, districts to create programs or contract with vendors, and schools to coordinate delivery. For starters, it’s critical to get clear on the problem tutoring is intended to solve. Is it providing more demonstration, practice, and feedback at particular tasks (like math operations or essay writing)? Is it coaching on behavior and study skills? Is it providing a connection to a caring adult who will offer support and encouragement?
The answer frequently gets lost in the mix, but it really matters. Providing students with real-time assistance on math operations is different from cultivating a personal relationship with a mentor. An extended face-to-face session is very different from a brief, text-based virtual interaction.
So, what’s my take? I think there’s some real value here, but that it depends on how tutoring is designed and employed.
This should start, I think, with matching student needs to available tools. If students need assistance with early literacy or math operations, there are powerful online tutorials that can help. While schools may have trouble arranging (or affording) 90 minutes a week of in-person tutoring, these resources can help ease that problem. Computer-assisted tutoring may not be as effective as the best human tutors, but it’s accessible, reliable, and pretty effective at teaching basic skills.
It’s no great surprise that out-of-school tutoring is less effective than tutoring during the school day, especially when we consider the logistical challenges and potential distractions. But finding 90 minutes a week during school can be an impossible task, especially for schools struggling to find enough staff. Rather than proceeding haphazardly, with teachers expected to “just make it work,” there’s a need to reexamine the school schedule and the duties of staff. (If this makes things slower or more difficult, that that’s the difference between a slapdash approach and one that might deliver).
Research has pointed to the promise of intensive vacation-academy programs in which groups of struggling students devote a week-long break to a single subject. With student-teacher ratios of roughly 10 to 1, these programs are relatively inexpensive but they’re also more akin to classroom teaching. That means they require experienced teachers. Again, incorporating these is feasible but requires rethinking school calendars, schedules, and teacher roles and compensation.
As a one-off, tutoring is pretty simple. But doing it as part of a large-scale strategy requires leveraging part-time tutors, trained staff, and digital tools in ways that’s not natural for school systems. I fear that the easiest thing in the world is for high-dosage tutoring to wind up meaning, “We’ll tutor as best we can with who we can find, how and when it’s convenient.” That strikes me as a recipe for disappointment. Doing it effectively means figuring out what needs to change in terms of calendars, contracts, coordination, and connectivity.
Tutoring has real promise. But my bottom line is that, as is so often the case, what matters is less whether schools or systems spend on high-dosage tutoring than how they do it.
