The Distributional Efficacy of Collaborative Learning Recitation Sessions on Student Outcomes

• Activity Number: 478
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Education
• Author(s): James K. Self*+ and Kim P. Huynh and David T. Jacho-Chavez
• Companies: Indiana University and Indiana University and Indiana University
• Keywords: Collaborative Learning ; Quantile Treatment Effects ; Program Evaluation

Abstract

Reported here are the preliminary results found on the distributional efficacy of Collaborative Learning (CL) recitation sessions on student outcomes in two large introductory microeconomics sections. Students participating in CL (voluntary weekly one-hour session) are treated to an interactive small group deep learning experience without any loss of regular class lecture time. To account for selection bias a quasi-randomization experiment and measure the treatment effects by programme evaluation methods. The data reveals that the average treatment effect of CL is 4.6-4.9 percentage points of the final course grade or in qualitative terms, a grade change i.e. B to B+. We then focus on the distributional effects on students by computing the quantile treatment effects and find that students at the bottom 40 percentile gain the greatest improvement when they received this CL treatment.