Critics of bilingual education claim that research supporting native language instruction is weak, a claim that has been echoed by some prominent supporters of bilingual education. This claim has had a damaging effect on the political fate of bilingual education in some states. This paper argues that the primary metric used to support this critique, the percentage of research studies meta-analysts consider methodologically acceptable, is a vague and not widely-accepted approach for weighing the quality of research. Data for the study come from prominent research reviews in the field of education and social sciences and from a random sample of empirical literature reviews from two major journals of research reviews. This paper suggests that the percentage of studies found methodologically acceptable in bilingual education research is not very different from similar federally funded research in education and the social sciences. It notes that there is little basis for comparison for bilingual education research and other psychology- and education-related literatures, since percentages of methodologically acceptable studies are rarely reported in research reviews. It concludes that higher quality research is necessary but should not be viewed in isolation to real-world constraints on such endeavors. (Contains 10 references.) (SM)