When data are fragmentary and available analytical tools are crude, there is considerable potential for inconsistent results and interpretive disagreements. Specific problems faced by civil justice researchers include matching data and questions, selecting statistical indicators, evaluating magnitudes and differences, spotting trends, and generalizing from limited data. Researchers have a responsibility to indicate the types of inferences that can and cannot be drawn from the data they collect and analyze. They also have a responsibility to educate policymakers about the uses of different statistical data. They need to be self-conscious about how they describe magnitude differences, select standards for comparisons, and define time periods for graphing trends. Researchers should be clear about definitions of what constitutes data and what should be relegated to the world of anecdotes. They need to assist policymakers in deciding when it makes sense to extrapolate from limited data and when it does not. Finally, researchers need to be sensitive to the ways in which their own political and social values affect their choices of research questions, research designs, and reporting of research data. 121 references.