Two frequently-used discussion protocols were investigated as part of a program to implement teaching cases in undergraduate educational psychology classes designed for preservice teachers. One protocol involved synchronous face-to-face (FTF) discussion of teaching cases, which occurred in class after students had individually completed written case analyses as homework outside of class. The other was asynchronous computer-mediated (CM) discussion taking place outside of class, simultaneous to students' completion of their written case analyses. Six class offerings of an undergraduate child development course taught by two instructors (three classes by each instructor) were randomly assigned within instructor in a quasi-experimental design to one of the three discussion conditions: FTF, CM, or no discussion of the cases across the semester. Findings indicated that both CM and FTF discussion conditions were associated with positive outcomes relative to the control condition. Both CM and FTF discussion related to higher cognitive-affective engagement with the cases than the control condition; and the CM discussion condition was associated with higher cognitive-affective engagement than FTF discussion. In contrast, FTF discussion, but not CM discussion, was associated with higher-than-control-condition case analysis ability at the end of the semester. Potential explanations for these findings and directions for future research are discussed. (Contains 2 tables and 1 figure.)