How to Best Group Students in Math Based Courses: Evidence from an Experiment

• In Event: Poster Session: Experimental Research & Qualitative and Multi-Method Research
  In Poster Session: Poster Session: Experimental Research

Abstract

There is widespread agreement that collaborative work in math and science courses has positive effects on students outcomes. However, little is known on how to best structure such collaborative work in out-of-classroom activities, especially when dealing with students from a wide-distribution of math backgrounds. Based on a sequence of experiments where students are randomly assigned to different collaboration structures when working on homework for a course on data analysis and politics, this study will identify the way of pairing students that most improves students' performance. Preliminary results from the first experiment suggest that assigning students partners with slightly different math backgrounds, as compared to allowing students to choose their own partners, has positive effects on (a) level of collaboration, (b) engagement in the class, and (c) performance.

Author

• Elena Llaudet, Suffolk University