This critique of two recent meta-analyses of class size research is unique among all previous studies and reports published by Educational Research Service. It was made necessary by two extraordinary reviews of class size research prepared by Gene V Glass and Mary Lee Smith and published by the Far West Laboratory for Educational Research and Development under a federal grant. Because of the many and far-reaching instructional, financial, and political implications of the two meta-analyses, this critique analyzes fully the methods, findings, and implications of these two class size reports. Full treatment is essential to provide school officials and others with important information they should have when making class size decisions. Glass and Smith stated that their techniques enabled them to make "bold generalizations" about the effects of class size on pupil achievement where other analysts could make only "timid qualifications." They declared unequivocally that their analysis "established clearly that reduced class-size can be expected to produce increased academic achievement." Glass and Smith collected 76 studies and 725 comparisons they used to justify the claimed superiority of their meta-analysis findings over all previous class size research analyses. Yet their claim that small classes increase academic achievement is actually based on only 14 "well-controlled" studies. Seventy-three percent of the comparisons used to make this claim came from only four studies. Eight of the 14 studies dealt with instructional conditions not typical to public schools (i.e., college classes or tutoring). ERS concludes that the meta-analyses only confuse the class size issue, fail to provide practical guidelines or help for making class size decisions, and do not change the conclusions it made in its 1978 report Class Size: A Summary of Research.