Empirical analysis is central in both legal scholarship and litigation, but it is not credible. Researchers can manipulate data to arrive at any conclusion they wish to obtain. A practice known as data fishing—searching for and selectively reporting methods and results that are favorable to the researcher—entirely invalidates a study’s results by giving rise to false positives and false impressions. Nevertheless, it is prevalent in law, leading to false claims, incorrect verdicts, and destructive policy. In this article, I examine the harm that data fishing in empirical legal research causes. I then build on methods in the sciences to develop a framework for eliminating data fishing and restoring confidence in empirical analysis in legal scholarship and litigation. This framework—which I call DASS (an acronym for Design, Analyze, Scrutinize, and Substantiate)—is designed to be simple, flexible, and practical for application in legal settings. It provides a concrete method for researchers to use to safeguard against data fishing and for consumers of empirical analysis to use to evaluate a researcher’s empirical claims. Finally, after describing the DASS framework and its application in various legal settings, I consider its implications for the “hired-gun” problem and other difficulties related to the reliability of expert evidence.
Hillel J. Bavli,
Credibility in Empirical Legal Analysis,
87 Brook. L. Rev.
Available at: https://brooklynworks.brooklaw.edu/blr/vol87/iss2/2