Each of four instructors was assigned two sections of introductory psychology, one experimental and one lecture/control. Experimental sections were divided into teams of five to eight students and complied with the following schedule for each of seven two-week unites. In the first class meeting, subject matter was introduced and discussion topics assigned. In subsequent class periods, half the students assembled to present their projects while the remaining half met outside of class for preparation. Discussion was monitored for comparability to lecture material. Course effectiveness and popularity were measured at the end of the semester. Teaching method was not differentially effective as measured by either quiz or final exam scores. However, significant method-by-instructor interactions were observed on the popularity measures. The results are summarized as follows: (1) for two instructors the lecture method was rated more favorably, although students in the experimental sections reported more group discussion; (2) for a third instructor the lecture approach was more popular and produced greater discussion; and (3) for the fourth instructor increased favor and discussion were observed in the experimental team method. Preference for a given teaching method appears here to be highly dependent on the particular instructor. (Author/MSE)