In-person and high-dosage tutoring is gaining popularity among practitioners as an evidence-based approach to accelerate learning in the post-pandemic education system. Facing the challenge of insufficient funding and lack of local tutors to provide in-person tutoring, some practitioners regard online tutoring as an alternative outlet. However, there is limited research to establish the effectiveness of online tutoring programs. To investigate this topic of increasing interest, Dr. Kraft and his colleagues conducted a pilot study and found answers align with conventional wisdom.
This pilot study recruited 230 volunteer college student tutors from 47 highly selective universities to deliver one-to-one online tutoring using Zoom. All tutors participated in a three-hour training session before the intervention and weekly peer-mentoring sessions during the intervention. The 560 participants were 6th-8th grade students from Chicago. Almost all of them come from low-income households. The study randomly assigned students to treatment condition or control condition within each grade level. Students in treatment received online tutoring while students in control participated in regular advisory period activities. The intervention was delivered in the spring of 2021 for 12 weeks with an intended dosage of two 30-minute sessions each week.
In terms of implementation, there were significant challenges. Students were meant to receive 9 hours of tutoring, but on average received 3.1 hours. Many students (18%) received no tutoring at all. To measure impacts on student achievement, researchers used scores on the Illinois Assessment of Readiness and the i-Ready tests. Analyses showed that online tutoring produced positive but statistically insignificant results on math (ES = +0.07) and reading (ES = +0.04). Compared to in-person tutoring programs’ effect sizes reported in Nickow et al.’s meta-analysis (2020), this online tutoring model only produced around a third of the effects. While the extreme low costs of such online tutoring remain highly attractive to policymakers, the authors also cautioned that free-of-charge online college tutors may not be sustainable, replicable, or scalable when universities switch from remote to in-person.
Kraft, M.A., List, J.A., Livingston, J.A., & Sadoff, S. (forthcoming). Online tutoring by college volunteers: Experimental evidence from a pilot program. American Economic Review: Papers & Proceedings.