Major support for MindShift comes from
Landmark College
upper waypoint

What we know about tutoring research and how schools are using tutoring in pandemic recovery

Save ArticleSave Article
Failed to save article

Please try again

Teacher helping three high school or college students at a table with a laptop.
 (Eduard Figueres/iStock)

Ever since the pandemic shut down schools almost three years ago, I’ve been writing about tutoring as the most promising way to help kids catch up academically. I often get questions about research on tutoring. How effective is tutoring? How many schools are doing it? How is it going so far? In this column, I’m recapping the evidence for tutoring and what we know now about pandemic tutoring. For those who want to learn more, there are links to sources throughout and at the end, a list of Hechinger Report stories on tutoring. 

Well before the pandemic, researchers were zeroing in on tutoring as a way to help children who were significantly behind grade level. Remedial classes had generally been a failure, and researchers often saw disappointing results from after-school and summer school programs because students didn’t show up or didn’t want to go to school during vacation. 

But evidence for tutoring has been building for more than 30 years, as tutoring organizations designed reading and math programs, partnered with schools and invited in researchers. The results have been striking. In almost 100 randomized controlled trials, where students were randomly assigned to receive tutoring, the average gains were equivalent to moving an average child from the 50th percentile to the 66th percentile. In education, that’s a giant jump. One estimate equated the jump from tutoring to five months of learning beyond a student’s ordinary progress in a school year. There are no magic bullets in education, but tutoring comes as close to one as you get.

What researchers mean when they say “tutoring,” however, is not what many people might imagine. It’s not provided by the kind of tutors that well-to-do families might hire for their children at home. Studies have found that sessions once or twice a week haven’t boosted achievement much, nor has frequent after-school homework help. Instead, tutoring produces outsized gains in reading and math when it takes place daily, using paid, well-trained tutors who are following a proven curriculum or lesson plans that are linked to what the student is learning in class. Effective tutoring sessions are scheduled during the school day, when attendance is mandatory, not after school. Researchers call it “high-dosage” or “high-impact” tutoring. 

Think of it as the difference between outpatient visits and intensive care at a hospital. High-dosage tutoring is more like the latter. It’s expensive to hire and train tutors and this type of tutoring can cost schools $4,000 or more per student annually. (Surprisingly, the tutoring doesn’t have to be one-to-one; researchers have found that well-designed tutoring programs can be very effective when tutors work with two or three students.) 


The Biden administration has urged schools to use their $122 billion in pandemic recovery funds on tutoring. But it’s been hard for schools to launch tutoring operations. For starters, it’s tough to hire tutors amid a strong labor market when there aren’t many people looking for work and “help wanted” signs are everywhere. The logistical issues are complex: tutor training, rescheduling the school day to make time for tutoring periods, finding physical space to hold tutoring sessions and figuring out how to allow a stream of adult tutors to flow in and out of school buildings all day. There are also tough decisions, such as which students should be tutored, and which curriculums to choose. Educators have to become operations experts and build a whole new organization amid everything else they’re juggling.

So far we have spotty data on how many schools have actually implemented tutoring. Among those who have, it’s unclear how many have launched good high-dosage programs and which students are getting it. 

The U.S. Department of Education estimates that more than four out of five schools were offering a version of tutoring to some of their students during the 2022-23 school year, based on a December 2022 survey of 1,000 schools. The majority said they were delivering “standard” tutoring, such as once a week extra-help sessions after school. Only 37 percent said they were delivering “high-dosage” tutoring. Even among the 37 percent of schools that said they were delivering high-dosage tutoring, only 30 percent of the students were receiving it. This translates into an estimate of 10 percent of public school students nationwide who are receiving high-dosage tutoring – far less than the need. In the same survey, school principals estimated that half of their students were behind grade level. 

Sixteen states are using $470 million of their federal pandemic recovery funds to launch large tutoring programs that will reach millions of children, according to a separate February 2023 report by the Council of Chief State School Officers, a group of public officials who head state education departments that oversee elementary, middle and high schools.  Among them are Arkansas, Colorado, Louisiana and Tennessee. Another four states are sending more $200 million directly to families to hire their own tutors. Indiana, for example, gives families up to $1,000 per qualifying student to spend on high-impact tutoring. (Local school districts are spending much more than a total of $700 million on tutoring. The school officers’ report covers only direct state spending.)

In many cases, tutoring this year is taking place virtually over screens instead of in person. Often, students are texting with tutors and not hearing or seeing one another – akin to a customer service chat session. But there are also tutoring companies that are trying to recreate an in-person tutoring experience through live video and audio. It feels more like a Zoom meeting with a shared whiteboard that both student and teacher can write on.

It remains to be seen if the outsized academic gains from in-person tutoring can be replicated online. A study of low-income middle schoolers in Chicago was disappointing. The program was riddled with problems: poor attendance, technical glitches and slow recruitment of college student volunteers to serve as tutors. Students who were assigned tutoring didn’t catch up more than those who didn’t get that extra help. But there were some signs of hope, too. Kids who started the tutoring sooner made larger academic gains. 

Another pandemic study of virtual tutoring for low-income immigrant middle schoolers in Italy yielded good results when students received four hours a week, but much worse results when they got only two hours a week. When the hours were halved, the academic gains dropped by more than half.  

Saga Education, an organization which has built an impressive track record with in-person tutoring, is currently testing whether its high-dosage model works as well in the virtual world. I am eager to see their data when it comes out. Earlier this month I observed Saga’s virtual tutoring at a New York City high school, where the students sat in a classroom and connected to their algebra tutors through laptops. I noticed how much more engaged the students were with a tutor who was physically present. Many ninth graders weren’t keen to be seen on camera and angled their laptops away. It was harder to develop an easy, friendly rapport between student and tutor.

School administrators have told me that it is hard to squeeze in three or more tutoring sessions a week, or make sure that students log in when sessions are scheduled. No-shows are common.

Many schools have purchased unlimited online tutoring from for-profit companies, such as Paper, and Varsity Tutors, where students can login anytime for homework help. Companies have marketed this voluntary 24/7 tutoring as high-dosage because, in theory, students could use it frequently. And it is much cheaper for schools; it can cost $40 per student instead of $4,000 for in-person, high-dosage tutoring. But several reports, such as this one in Fairfax County, Virginia, find that students aren’t using it very much, and the students who need tutoring the most are the least likely to use these drop-in tutoring services. 

Efforts by researchers to increase usage through text nudges convinced only 27 percent of the students at one charter school chain in California to try an online tutor even once. More than 70 percent of the students never logged into the tutoring platform. Among students who needed tutoring the most because they had failed a class with a D or an F, only 12 percent ever logged on. Just 26 of the 7,000 students in the charter network used it three times or more a week, which is what researchers are recommending.

Even though the services are marketed as one-to-one tutoring, some tutoring companies, such as Paper, have their tutors handling multiple students at once. Several tutors explained to me how challenging it is to juggle homework questions from different grades and different subjects simultaneously. Students sometimes have to wait patiently for their tutor to reply to a text while the tutor is texting with others. Relying on students’ homework questions, instead of using a structured tutoring curriculum, makes it hard to know if you’re teaching students the topics they need to catch up. Part of the magic of tutoring may be forming a long-term relationship with a caring adult. But tutors at several of these companies rarely see the same student twice. It’s no wonder that most students aren’t eager to log in.

Even though there’s good evidence for the effectiveness of intensive tutoring, districts are struggling to build functional programs. The for-profit tutoring services many schools are buying in the meantime don’t make the grade.

Previous Proof Points columns on tutoring:

High-dosage tutoring is the best known strategy for learning loss recovery. Is it actually happening?

How the life of an online tutor can resemble that of an assembly line worker

Many schools are buying on-demand tutoring but a study finds that few students are using it

Early data on 'high-dosage' tutoring shows schools are sometimes finding it tough to deliver even low doses

Uncertain evidence for online tutoring

Research evidence increases for intensive tutoring

Even older teens benefit from catch-up classes

Takeaways from research on tutoring to address coronavirus learning loss

Cheaper human tutors can be highly effective, studies show

Hechinger coverage of tutoring:

How one district went all-in on a tutoring program to catch kids up

The science of catching up

The simple intervention that could lift kids out of ‘Covid slide’

This math tutoring program gets ‘blockbuster’ results in high-poverty schools

Related Proof Points columns on pandemic learning loss:

Federal funds to combat pandemic learning loss don’t reflect need

Third graders struggling the most to recover in reading after the pandemic

Several surprises in gloomy NAEP report

Six puzzling questions from the disastrous NAEP results

Pace of learning back to normal during the 2021-22 pandemic school year but student achievement lags far behind, data shows

More studies mark the pandemic’s toll on student achievement

Three reports on student achievement during the pandemic


This story about tutoring research was written by Jill Barshay and produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for the Hechinger newsletter.

lower waypoint
next waypoint