Preliminary results from the Personalized Learning Initiative

By Cynthia Lake, Johns Hopkins University

Despite substantial federal investment through the Elementary and Secondary School Emergency Relief (ESSER) Fund, the education sector still grapples with challenges related to pandemic-induced learning loss. High dosage tutoring has been a leading strategy in accelerating academic catch-up efforts. The Personalized Learning Initiative (PLI) led by the University of Chicago Education Lab, in collaboration with MDRC and researchers from the University of Toronto and Stanford University, focuses on whether and how high dosage tutoring can be scaled to benefit more students. Their preliminary report finds that incorporating tutoring during the school day is more effective than after-school or virtual tutoring and underscores the importance of structured, consistent approaches.

The initiative worked with Chicago Public Schools (CPS), Illinois; Fulton County Schools, Georgia; the New Mexico Public Education Department; and a mid-sized urban school district in California. The New Mexico and California partners focused on out-of-school tutoring, virtual tutoring at home and after-school respectively, and both struggled with low student participation that precluded rigorous evaluations. In contrast, CPS and Fulton County Schools, who used their ESSER funding to incorporate tutoring during the school day, had more robust participation and were able to participate in randomized controlled trials during the 2022-2023 school year.  

Across sites, over 2,000 K-11 students were randomized into high dosage tutoring in math or reading or to business-as-usual control groups. In Fulton County, most of the schools began tutoring in January and 82% of students in the high dosage tutoring group participated in at least one session. Also notable was that 35% of those assigned to the business-as-usual group also received at least one tutoring session. The mean dosage for all assigned students was 9.0 sessions in the treatment group compared to 3.48 sessions in the control group. In Chicago, the mean dosage for the treatment group was 18.40 sessions for all assigned students, with 67% of them attending at least one session. The control group had a crossover rate of 18%, with a mean dosage of 4.61 sessions. Across both subjects and sites, the intent-to-treat impact was +0.04 and the treatment-on-the-treated effect size was +0.14. Despite challenges in maintaining data quality and delivery consistency, the combined outcomes from Fulton County and Chicago indicate a generally positive impact, particularly in math, where the treatment-on-the-treated (TOT) effect was 0.27 SD.

The broader lesson learned from the first year of the Personalized Learning Initiative is that high dosage tutoring can generate meaningful learning gains, provided it is structured and well-supported. The strategy that appears to work best involves a combination of smaller student-tutor ratios, structured curricula, and the use of trained tutors during the school day.