The Importance of Getting Tutoring Right
This post is part of LPI's Learning in the Time of COVID-19 blog series, which explores evidence-based and equity-focused strategies and investments to address the current crisis and build long-term systems capacity.
As we near the pandemic’s one-year mark, the majority of children in the United States may have gone a full calendar year with little or no in-person schooling. Prior to COVID-19, according to a 2018 analysis, nearly 17 million students did not have the necessary connectivity and devices to access virtual learning. Despite substantial investments, many households still lack the reliable high-speed internet that is essential to closing the digital divide. The scale of lost learning time and the impact on children’s development—particularly for students of color, students from low-income families, and other underserved students—is profound.
As we prepare for a post-pandemic world, policymakers are in search of solutions to provide more instructional time for students and more instructional support for teachers. One strategy receiving a lot of attention is tutoring, with some looking to England as inspiration for a national tutoring program of our own.
Can large-scale tutoring support students and teachers in addressing lost instructional time? The short answer is yes, but only if we pay close attention to the details of implementation to avoid the mistakes of the past. New policy proposals can draw upon the most up-to-date education research to design tutoring programs that are effective in meeting student needs.
We know a good amount about how to design effective tutoring programs, thanks to a raft of rigorous, scientific evidence on what works, and under what conditions. In particular, the last decade has brought us more than a dozen new large-scale randomized control trials (RCTs) of tutoring programs, considered the “gold standard” in research. A recent systematic review of 96 RCTs finds that tutoring yields large improvements on a variety of outcomes across all grades. These studies also highlight best practices and pitfalls to avoid.
Chief among these lessons is that a large, poorly trained, insufficiently compensated tutoring corps would be an inefficient use of precious time and money that would not do enough to accelerate student learning. The two largest evaluations of volunteer literacy tutoring programs in the United Kingdom, where tutors were neither compensated nor well-prepared, did not find improvements in reading comprehension. In the U.S., a previous attempt at a nationwide literacy program under the Clinton Administration, called "America Reads," was never able to meet the challenge of securing thousands of nonprofessional volunteer tutors, and those recruited did not transform reading outcomes. Another national attempt, funded through the No Child Left Behind Act under the George W. Bush Administration, failed to yield positive results. These supplemental education services were provided on an inconsistent basis, often in groups that were larger than five, not integrated into the school day, and delivered for only 21 hours, on average, over the course of an entire school year.
Rather than reviving past attempts—and making the same mistakes—policymakers should look to what works. The literature is clear about the characteristics of effective tutoring programs that lead to success in the classroom. Effective tutoring:
- Employs certified classroom teachers when available (whether currently teaching or not), or paraprofessional staff, such as existing paraprofessionals, teacher candidates enrolled in preparation programs, or well-trained tutors who earn a stipend, such as AmeriCorps members;
- Is provided at least 3 days per week for at least 30 minutes, as part of the regular school day, in groups of 5 or fewer;
- Invests in staff capacity building by providing quality training and ongoing support; and
- Builds relationships among students, tutors, and teachers through structured time that is well-aligned with regular classroom curriculum.
Of the 96 studies in the review discussed above, four evaluations of tutoring programs, each with large sample sizes, stand out as reporting some of the largest gains for students: Reading Recovery, Number Rockets, ROOTS, and Match Corps Tutoring. While three of these programs serve struggling learners, one offers universal tutoring, a practice that could reduce the stigma associated with remediation. Each of these programs provides key lessons for how federal, state, and local investments should be targeted.
Reading Recovery: A Teacher-Led Intervention for Struggling Readers
Developed in the 1970s and 80s by Dr. Marie Clay, a developmental psychologist, Reading Recovery has been extensively studied around the world. U.S. studies have documented the success of more than 2.3 million first graders, including students with reading disabilities and English language learners. Depending on their rate of progress, students spend 12 to 20 weeks in 30-minute daily tutoring sessions. They work one-on-one with a certified teacher trained in reading instruction, including a yearlong Reading Recovery graduate-level course. In one of the many controlled studies of this program, participation resulted in a reading growth rate that is 31% greater than the average growth rate nationally for beginning first graders.
The start-up costs of training and materials are $2,500 per teacher, which averages to about $100 per student in the first year of a teacher’s practice, with no additional costs beyond salary in the subsequent years. There are already hundreds of training sites at universities across the country, staffed by teacher leaders who could help implement the program in districts where students are behind in reading comprehension as a result of pandemic-related disruptions to their education.
Number Rockets and ROOTS: Leveraging Paraprofessionals, Substitutes, Retired Teachers, and Education Students
It may be challenging to hire teachers as tutors in districts with persistent staffing shortages (now exacerbated by the pandemic). But in addition to the existing teaching workforce, there are thousands of retired and substitute teachers, paraprofessionals, and teacher candidates enrolled in teacher-preparation programs who could help deliver quality tutoring at scale.
Number Rockets, a tutoring program originally implemented using graduate students in training to become general and special education teachers, has been found to markedly improve first graders’ foundational math skills. A replication study that employed retired teachers and substitutes instead of graduate students achieved the same positive results as the original study. Tutors participated in 10 hours of training, during which they learned how to use a scripted curriculum designed for first graders struggling in math. They worked with two or three students at a time, participating in three 40-minute sessions a week, consistently over 17 weeks. The program improved scores on a standardized math test by 0.34 standard deviations (SD), which is widely recognized as a substantial effect size. The cost of the training is $1,500, plus travel expenses, $64 for implementation manuals, and $30 for supplemental materials.
An evaluation of ROOTS, a math tutoring intervention for kindergarteners, compared the effectiveness of tutoring groups comprised of two students with groups of five students and found similar results to the Number Rockets study. After receiving 10 hours of training and two or more feedback sessions from coaches, existing district-employed paraprofessionals delivered daily 20-minute math lessons for 50 days. The larger tutoring groups of 5 students actually gained more (0.45 SD) on standardized tests than the smaller group size (0.35 SD—both moderately large effect sizes). According to the National Center on Intensive Intervention, which includes a clearinghouse for evidence-based information on tutoring programs and other interventions, ROOTS training costs $250 per teacher.
Match Corps: The Promise of Universal Tutoring
Although the programs above target young students with lower math and literacy scores, tutoring does not have to be remedial or selective in nature. In fact, providing universal tutoring can reduce the stigma associated with receiving extra help. It can also ensure that these supports are available to all students, regardless of family income. For affluent students, tutoring has long operated as what one historian calls a “shadow education system,” boosting their outcomes, a phenomenon also known as “dream hoarding.”
While there are fewer RCTs of high school tutoring, an evaluation of Match Corps Tutoring, delivered within the existing school day, demonstrated benefits for all students. Originally developed at the Match Charter Public School in Boston, the Match Corps Tutoring RCT provided more than 2,500 9th-and 10th-grade students in 12 Chicago schools with 60 minutes of 2-on-1 tutoring each day for a full school year—far more than is typical for a tutoring intervention. The program employed AmeriCorps members, who received 100 hours of training, daily supervision, and feedback for continuous improvement. Results were promising: Students’ math achievement scores significantly improved (0.19-0.31 SD, or a medium-to-large effect size), and their course failures were reduced by half. According to the study’s authors, at scale this program would cost around $2,500 per student to provide a year’s worth of daily tutoring, inclusive of standard AmeriCorps stipends.
Worth Doing Well
If tutoring is worth doing—and it is—it must be done well, based on existing knowledge and research.
To effectively address lost instructional time, policymakers can draw from effective models to adequately fund targeted, evidence-based interventions. Universal tutoring would have the added benefit of providing every student with the same individualized attention that many affluent students have long received. While much of what has been lost during this pandemic cannot be replaced, a well-designed, well-funded tutoring initiative is one way we can increase instructional time for students and provide instructional support for teachers.