randomized trial

Tutor CoPilot improved short-term topic mastery by supporting human tutors, with the largest gain among lower-rated tutors

Rose E. Wang, Ana T. Ribeiro, Carly D. Robinson, Susanna Loeb, Dora Demszky

arXiv preprint

A tutor-level randomized trial across 782 active tutors and 1,787 Grade 3–8 students found that Tutor CoPilot access raised short-term lesson mastery by four percentage points and by nine points for lower-rated tutors. It did not significantly improve the year-end mathematics test.