The findings of the Cost, Quality, and Child Outcomes (CQO) study in Child Care Centers, the largest and most visible child care research conducted in the 1990s, were widely publicized and used to promote increased spending on initiatives to improve child care quality, the redesign of subsidy systems to provide quality incentives, more stringent licensing regulations, and overall increases in teacher compensation. This paper examines the methodology and execution of the study to determine whether they were robust enough to justify confidence in the study's findings. Scrutinized in this report are the sample design, response rates, methods of measuring child care costs, methods of measuring quality, and analytic issues related to the cost and quality component and the developmental outcomes component. A limited reanalysis of the cost and quality data is presented to shed additional light on problems in the study methodology. The report asserts that the combination of problems of selection bias, low response rates, and attrition makes it difficult to identify the population to which the findings might be generalized. Also questioned was the use of the center rather than the classroom as the unit of analysis, which, it is argued, ignores differences in cost and quality of child care for children of different ages. The report ends with two sets of conclusions: (1) the findings highlighted in the CQO report summaries overstated the findings of the analyses and do not identify the impact of improving center quality on child outcomes; and (2) given the flaws in the methodology and implementation of the CQO study, the study's findings should be interpreted with caution. (KB)