A recent report by Dr. Randall Ryder evaluated the use of Direct Instruction (DI) reading programs in 2 school districts in Wisconsin. In the report, Ryder claimed that students in 1st, 2nd, and 3rd grade who received Direct Instruction scored significantly lower on several standardized tests of reading than students who received more traditional reading instruction. This article examines the validity of the Ryder report. Examination of the report revealed that (a) the quality of implementation of Direct Instruction is highly suspect; (b) the group labeled "Direct Instruction" apparently included numerous students who received an undefined mix of DI and non-DI reading instruction; (c) the selection and assignment of classrooms and students to groups resulted in DI groups that performed substantially below the non-DI groups before the study began; (d) there are numerous ambiguities and contradictions regarding the number of students in various groups in each year of the study; (e) statistical reporting failed to include basic information such as degrees of freedom, means, and standard deviation for some or all analyses; (f) ANCOVA was assumed to control for systematically systematically biased assignment without consideration of the assumptions, limitations, and interpretive difficulties involved; and (g) Ryder fails to report results from subtests on which, in previous reports, the DI group outperformed the non-DI group by a statistically significant margin. As a result of these and other problems, no firm conclusions can be drawn from Ryder's report. We conclude that Ryder's report should be subjected to an independent peer review process, and the results of that process should be publicized as widely as the report has been. (Contains 7 tables.) ["A Critical Review of Randall Ryder's Report of Direct Instruction Reading in Two Wisconsin School Districts" was written with Gary L. Railsback, Scott A. Gallagher, Sarah A. McCright, Randy A. Uchytil, William W. Conlon, and James T. Davis.]