The Washington PostDemocracy Dies in Darkness

Schools sink money into tutoring, but some programs fall short

Some methods are proved to work -- but many of the most needy students can’t access them

Retired teacher Darrell Shives leads a math tutoring session with students in Greensboro, N.C., in November. Some types of tutoring show better results than others. (Ted Richardson for The Washington Post)

Tracy Compton knew her fourth-grader needed tutoring to make up for what she missed during the pandemic. But when the Virginia mother learned about the online program offered by her well-regarded suburban school system, her heart sank.

“It’s homework help,” she lamented.

Tutoring has become a go-to solution for school districts across the country, amid an unprecedented gush of federal dollars to address pandemic setbacks. But many districts are investing in opt-in online programs that experts say have not been shown to fill in learning gaps for students who need it most.

What actually works, they say, is something else: more structured, data-driven “high-impact” tutoring, a method backed by years of research. It is typically delivered several times a week, over an extended period, using the same well-trained tutor to drill down on missed learning while staying aligned with classroom instruction. School districts identify students who need it, ideally incorporating it into their regular school day. The cost and level of difficulty are higher, but the payoff is worth it, education experts say.

“They both have the word ‘tutoring’ in them, so it seems like the same thing but it’s really not,” said Susanna Loeb, an education professor at Stanford University who has been deeply involved in research on the subject and worked with school systems nationally. Loeb said there may be benefits to opt-in tutoring for students who use it, but it is not a proven way to reach the mass of students struggling the most.

“Implemented well, with the right evidence-based features in place, targeted intensive tutoring can be a game changer for students,” said Allison Socol, a vice president at the nonprofit research organization Education Trust.

Federal data collected in June showed that 56 percent of schools reported using high-impact — also called “high-dosage” — tutoring and 36 percent of schools reported using other forms. Some school leaders say they use multiple kinds of tutoring.

With opt-in programs, a student logs into a tutoring service through their school system’s online platform — requesting help on a math problem, for instance, or feedback on an English paper. The service is typically open 24/7, often covering multiple languages, and students communicate live with a tutor via written messages or audio and sometimes video. The service is free to students and paid for by school systems.

Students are behind in math and reading. Are schools doing enough?

Virginia’s Fairfax County is a good example of the drawbacks. Less than 2 percent of the student body used an opt-in tutoring service — tutor.com — after it debuted for the final quarter of the previous school year, according to an internal analysis. Most of those who did log in used it for less than an hour — the median was 29 minutes — “an amount of time that is unlikely to yield tangible benefits to student achievement, particularly for those with greater academic need,” the analysis said. A majority of those who accessed the online tutoring were students who did not demonstrate academic need, the report said.

The report said the timing of the rollout, when students and families already had set patterns, probably contributed to low usage in the 181,000-student system. But it’s unclear whether it’s been more popular this year. School system officials did not answer questions about how many students used it. They said 23,000 sessions had been held, totaling 8,000 hours, from school’s opening to winter break.

Low participation has been an issue elsewhere, too. In Columbus, Ohio, for instance, roughly 7 percent of students used opt-in tutoring — through the company Paper — so district leaders did not renew a one-year $913,000 contract, according to city schools spokeswoman Jacqueline Bryant.

In Fairfax, critics question the cost of the service: $488,000 last spring — which translates to about $130 an hour — and an estimated $2.8 million for this academic year, according to the internal analysis.

Fairfax County school officials denied an interview request on the subject, at first saying no one was available because of the winter break and, after the break, rejecting the request. They said in written statements that each school gets funding for interventions and that targeted, in-person tutoring is given to below-grade-level students across the school system. They did not answer questions about the number of students served. They said in an email they provide high-dosage tutoring, but gave no details.

The school system “realizes that there is much work yet to be done, but we are already seeing considerable gains by our students,” they said, noting that test scores had risen considerably from the deepest pandemic lows. (Math is still well below pre-pandemic levels.)

Nearly $3.1 billion in federal covid relief funding could go to tutoring and academic coaching, according to an analysis of 5,000 school districts’ spending plans by FutureEd, a think tank at Georgetown University’s McCourt School of Public Policy.

In some districts, opt-in online tutoring was made possible by the distribution of laptops and other devices during the pandemic. All students could have access.

Neal Kellogg, director of educational technology services in Oklahoma City Public Schools, said his district hired TutorMe.com 18 months ago, at a time when the district provided an iPad or Chromebook to each of its 33,000 students. “The main thing is … to help our kids have access to learning, 24/7,” he said.

Kellogg had no data on results yet but said he felt the opt-in setup worked well for students who used it. “The biggest problem is … those kids that are reluctant to participate,” he said. “Our biggest challenge is finding a way to reach those families.”

The crisis of student mental health is much vaster than we realize

Research by Loeb and others showed just 19 percent of students ever opted into an online tutoring service offered by a California charter school system, according to a study published by EdWorking Papers, a joint project of the Annenberg Institute at Brown University and Stanford University. Students who struggle were far less likely to use it than their more engaged and higher achieving peers, the research found.

“Our results have immediate and broad implications for education policy,” the three researchers wrote. “ … If the goal is to mitigate learning gaps for students who struggled the most,” they added, “providing opt-in access … is unlikely to help.”

There is not robust independent research on the success of companies that provide opt-in tutoring.

At tutor.com, a formal efficacy study is underway, but meanwhile it gets strong reviews from students, said Sandi White, senior vice president. The company, which has been providing online tutoring for more than 20 years, holds contracts with more than 400 districts. White said her company does provide homework help, but maintains it is a form of tutoring. “We are helping them with their academic work that is being presented to them from their teachers,” she said.

In some districts, teachers or schools require their students to use the service, White said. “That’s when we see the highest participation rates,” she said. White said tutor.com expects to offer a high-impact tutoring program in the future.

Paper pointed to its own case studies as a sign of its impact, including one of the 20,000-student Val Verde school district in California. The company holds contracts with 400 school districts, in Los Angeles, Las Vegas and Florida’s Palm Beach and Hillsborough among others.

In Val Verde, Superintendent Mike McCormick said the company came across the district’s radar as officials looked for ways to support students after the pandemic began in 2020. Nearly 85 percent of students get free and reduced-cost meals, and many strapped parents were glad to get the help at no cost, he said. The district pays about $150,000 a year for its sixth- through 12th-graders.

From August to December, there were 8,500 tutoring sessions for the 10,000 students eligible to participate, McCormick said, and 200 college admissions essays were reviewed. “Even the students have been liking it,” he said — though he added that some forget it’s there and need to be nudged.

Philip Cutler, chief executive of Paper, said he tells school districts that opt-in tutoring is not a silver bullet and they will probably need other solutions, too, but that “this is obviously a very scalable way to support thousands of students — hundreds of thousands in some cases.”

Student use of the program, he said, “varies enormously from school to school and from district to district based on how well it’s introduced to the students.” He cited a high-needs school in Columbus where the rate of participation was 56 percent while other schools had no students using it; district officials said the overall rate was 7 percent.

“Our role in all of this is to be a resource to support the students from an academic perspective when they get stuck,” Cutler said.

Opt-in services often falter because some students lack confidence, motivation or clarity about what they need, so they don’t sign in, said Anthony Salcito, chief institution business officer at Varsity Tutors, a longtime tutoring company. Varsity provides opt-in tutoring but also high-dosage tutoring and a new program called “teacher-assigned tutoring.”

American students’ test scores plunge to levels unseen for decades

Salcito said teachers are in the best position to identify when a student needs help and request sessions for the student. He said that students are often selected for the high-dosage programs based on past testing, while teacher-assigned tutoring is linked to what teachers observe in real time.

In Gwinnett County, home to Georgia’s largest school system, students had access to tutor.com through a joint effort between the school system and the local library system that began before the pandemic.

But the toll of covid-19 touched off new efforts. Gwinnett worked with researchers at the National Student Support Accelerator, at Brown’s Annenberg Institute, which became a hub of tutoring research and strategy during the pandemic. The support accelerator is now at Stanford too.

High-impact tutoring in Gwinnett started nearly a year ago, with 1,000 students served by June 2022. This school year, administrators’ goal is 4,000 students. They use a string of high-impact providers, said Babak Mostaghimi, assistant superintendent of curriculum and instruction support and innovation. Depending on the provider, the cost is $800 to $1,400 per student. Tutors meet with students in person or virtually, but the format stays the same: frequent sessions, always with the same tutor, aligned with the curriculum, focused on learning gaps.

High-impact tutoring is designed to be part of the school day because attendance is better than before or after school, researchers said. For instance, Saga Education, a nonprofit that helps districts implement high-dosage tutoring, designates its own tutoring as a period of the day at the 28 schools in Chicago and supports the city’s high-dosage tutoring on 120 more campuses.

Some companies or organizations claim to do high dosage but are not adhering to all of the research-based characteristics that lead to substantial learning gains and will not achieve the same effect, said Saga’s co-founder, A.J. Gutierrez.

Gutierrez said he tells schools that if they have to prioritize, third-grade reading and ninth-grade algebra are paramount — two major markers of academic success. When students struggle at those grades, rates of failure soar.

He also sees great potential in virtual high-dosage tutoring, which could greatly expand the tutor pool and help address labor shortages that schools face. Saga has been involved in multiple randomized control trial studies showing the efficacy of high-impact tutoring, Gutierrez said.

“You’re learning two and a half times as much as a traditional student would,” he said.

Loading...