This study was designed to test the null hypothesis which states that there will be no significant difference in mean creativity scores on the verbal or figural subtests of the Torrance Tests of Creative Thinking among subjects with no creativity training, those with creativity training under traditional, and those with creativity training under cooperative learning structures, either immediately after training, 2 months after training, or on both occasions. Subjects were 159 fifth-grade students from 8 classrooms in a small city in Mississippi. Of the 159 students, 40 students were in a control group receiving no training, 60 were in the traditional group, and 59 were in the cooperative learning group. Creativity training lasting 50 minutes a day for 10 days was presented to the 2 treatment groups. The cooperative learning structure used was Student Team Achievement Divisions. The Torrance Tests of Creative Thinking were administered immediately after the training sessions and again after 60 days. It appeared that in the short run the traditional classroom is a more effective classroom structure with which to hold such a training workshop for children, but, in the long run, results become less clear. (Contains 35 references.) (JDD)