Studies of racial/ethnic disparities in criminal case processing have yielded mixed results. Some differences in findings have probably resulted from analyses of different social settings, but some could be attributed to differences in analytical rigor between studies. Contextual analyses are pointless unless the research yields unbiased estimates of the true relationships between a defendant's race/ethnicity and case dispositions. This goal may be furthered by conducting analyses that simultaneously incorporate the following: corrections for sample bias, analyses of several stages of case processing, measures of prior record and offense seriousness that maximize "explained" variation in the dependent variables examined, statistical controls for extralegal variables that correlate with case dispositions, and more rigorous statistical tests for interactions. To demonstrate potential differences in findings from analyses with and without these characteristics, this study applies different analytical methods to the processing of a sample of defendants. It found that the less rigorous analysis yielded a different picture of ethnic disparities in case processing compared to the more rigorous analysis. In order to conduct more reliable studies of ethnic and racial disparities in case processing and outcome, this paper recommends that a study include corrections for selection biases to control for unmeasured influences on case outcomes. The analysis should also examine multiple stages of case processing to provide a more comprehensive picture of decisionmaking in the courts. It should also include measures of offense and prior record that maximize explained variation in the dependent variable for each race/ethnic group examined. Finally, the analysis should examine separate models for each ethnic/race group, with statistical tests for cross-group differences in equations and regression coefficients. 4 tables and 74 references