In the United States, millions of children are behind grade level. Reading test scores are declining nationally, and math test scores are flat or increasing only slightly. When children fall behind early on, it’s harder to succeed later in life: third graders who are not reading proficiently are four times less likely to graduate high school than their peers. Among the many education innovations tried and tested in the United States, tutoring programs are one of the most effective.

The impacts of tutoring programs are strongest for the youngest students. A review of 96 randomized evaluations found that the effects of tutoring programs tended to be strongest among students in earlier grades, though a smaller set of programs at the secondary level were also found to be effective at improving learning outcomes.

Design of tutoring programs makes a difference. Research shows that effective tutoring programs share certain features. One-on-one or small-group tutoring led by professionals, conducted during school hours, and taking place at least three times a week is consistently more effective. The cost and design considerations below elaborate on these important design features.

Tutoring programs are more effective when led by consistent teachers or professional tutors who have been trained and are paid for their work. Professional tutors include school staff members, students in education fields, and fellows who have been trained as part of a formal program. Programs that use these types of tutors (rather than nonprofessional volunteers or parents) are consistently more impactful.

Tutoring programs held during school tend to be more effective than those held after school. A review of 96 randomized evaluations found that during-school programs had twice the impact as after-school programs. This is likely because teachers and program operators can more effectively ensure that tutoring actually happens during the scheduled time, and there may also be fewer distractions.

Students learn the most when they have at least three tutoring sessions a week. Across age groups, students who have three sessions a week consistently learned more than those with one or two sessions a week. The additional time gives tutors a chance to not only cover more material but also develop strong, supportive relationships with students.

Tutoring sessions should be informed by data. Schools and districts can collect data to help tutors identify students’ strengths and learning gaps and tailor tutoring sessions accordingly. Ongoing data collection helps tutors track progress and adjust their approach to better support student learning.