The educational effects of frequent classroom testing have been studied and discussed since the early part of this century. Many possible reasons for its presumed benefits have been offered, such as increased practice, review, improved course organization, and frequent opportunities for feedback. However, reviewers of evaluations of frequent testing have been unsure about whether the benefits are real. The present review makes two contributions to this discussion. First, it distinguishes the research on frequent classroom testing from two related areas, research on adjunct questions and research on mastery testing. Second, it provides a meta-analytic review of research on frequent testing. The meta-analysis shows that tested students performed about a half standard deviation better than untested students on summative measures. However, improvement was much smaller, about one-tenth of a standared deviation, when frequently tested students were compared to other students who also received tests, only less frequently. Frequently tested students rated their classes more favorably than did students who were less frequently tested. (Author/JAZ)