The relationship between class size and students' evaluations was examined in two experiments conducted at a public and a private university. Results indicated that the relationship between course enrollment and students' evaluations--with the exception of the quality of Group Interaction factor score--tended to be small, with both very small and very large classes being rated more highly than classes in between. The initial decline in students' evaluations, as course enrollment increased, was not considered large enough to warrant separate norm groups based on the first experiment. In the second experiment, the magnitude of the relationship was sufficient for the Group Interaction factor to make enrollment important in interpreting the student ratings. (Author/LBH)