A key component of interteaching, as described by Boyce and Hineline (2002), is the opportunity for students to participate in "dyadic" or pair discussions. Although the rationale for pair discussions is evident, only one study involving interteaching, to date, has evaluated the relative effectiveness of student performance when group size is manipulated. The present investigation was designed to further evaluate the effect of group size during student discussions on quiz scores in an introductory psychology course with a diverse group of learners. An alternating treatments design was implemented whereby students were assigned to work in a dyad or in groups of 4-5 students to discuss a preparation guide. All of the major components of interteaching were in effect during both conditions (i.e., availability of prep guides and quality points, clarifying lectures, and frequent test probes). Results showed a small advantage for performance following discussion in dyads. A social validity measure indicated students favored discussion in larger groups. Implications of these findings and suggestions for future work will be discussed.