This study investigated the possible causes of the contradiction between the results of two projects. Indiana's Prime Time project compared the achievement of students in large (averaging 26 students) and small (averaging 19 students) classrooms in grades 1 through 3. Results indicated that 3 years in smaller classes had little effect on student achievement. Tennessee's Student Teacher Achievement Ratio (STAR) project was a longitudinal study of class-size effects on student achievement in kindergarten through grade 3. The study concluded that small classes (13-17 students) had an advantage over large classes (22-26 students) in reading and mathematics. The present study examined whether students in the small classes in the STAR program really learned more than students in the large classes, and offered four hypotheses: (1) there was a relationship between the methodologies of the two projects and the contradictions in their results; (2) a Hawthorne effect occurred in the STAR program, according to which students in experimental groups tried harder than students in control groups; (3) a John Henry effect occurred in the STAR program, according to which students in control groups did not try harder than students in experimental groups; and (4) the research methodology of the STAR project was no better than that of the Prime Time project. The present study collected information about both projects' methodologies, designs, and circumstances. The study concluded that evidence did not definitively confirm a Hawthorne or John Henry effect, and that the STAR methodology was not better than the Prime Time methodology. (TM)