A typical classroom consists of a 1:30 teacher-student ratio. In physical education, this ratio is often higher. As a result, the dominant method of instruction has been the direct style. Group strategies such as cooperative learning have been proposed as alternative approaches to the direct style of instruction. Many of these strategies are rooted in the recognition that “one size does not fit all” in complex learning environments found in educational settings. Recently, a programmatic line of research on cooperative learning in physical education has been initiated. To date, three studies have analyzed the effects of cooperative learning applied directly with students in gymnasia. In the first study, a multitreatment reversal design was used to analyze a cooperative learning strategy called Performer and Coach Earn Rewards (PACER). The subsequent two studies used an A-B-A-B reversal design to assess the effects of PACER, and Jigsaw-PE, respectively. Several dependent variables were used to assess the effects of cooperative learning. The Academic Learning Time in Physical Education (ALT-PE) instrument was used to answer how the teacher and students spent their time in class. Student performance was assessed through total trials, correct trials, and percent correct trials. In addition, student social behavior was analyzed through the frequency and duration of cooperative interaction. Critics argue that cooperative learning takes too much time to implement. This was not found in any of the studies. In fact, once cooperative learning routines were learned, time in activity increased during the cooperative learning conditions. In each study, the total number of trials, number correct, and percent correct, respectively, were all substantially higher during the cooperative learning conditions. In the second and third studies, there was no functional relationship between cooperative learning and the frequency of cooperative interaction. However, there was a functional relationship found between cooperative learning and the duration of student social behavior. The data indicate that cooperative learning in physical education shows promise as an alternative teaching strategy. First, ALT-PE data indicate similar amounts of time for student learning regardless of the method implemented. Second, trial data indicate that students perform more trials, more correct trials, and have a consistently higher percentage of correct trials. Finally, social data indicate a functional relationship between cooperative learning and student cooperative interaction. Moreover, the use of peer assessment strategies, a component of cooperative learning, reflects the importance of social development as an educational outcome.