March 16, 2008

When Do Courts Think Base Rate Statistics Are Relevant?

April 2001:

ABSTRACT
Sometimes courts regard probability evidence to be relevant and sometimes they do not. This paper identifies a set of the conditions under which appellate and supreme courts are more and less likely to treat background probabilities (i.e., base rates) as relevant evidence. Base rates identify the relative frequency with which an event occurs or an attribute is present in some reference population (e.g., “eighty to eighty-five percent of child sexual abuse is committed by a close relative”). Following an introductory section that reviews the relationship between base rates and Bayes’ theorem, the following section is a critical review of the historical mistrust courts have exhibited toward probability evidence in general and base rates in particular. Next, the probabilistic reasoning literature and high court opinions are used to identify a series of conditions under which courts are likely to view base rates as relevant. This tends to occur when base rates (a) arise in cases that appear to have a statistical structure, (b) are offered to rebut an it-happened-by-chance theory, (c) are computed using reference classes that incorporate specific features of the focal case, or (d) are offered in cases when it is difficult or impossible to obtain evidence of a more individuating sort. The final section is a conclusion. ..more.. by Jonathan J. Koehler, The University of Texas at Austin, McCombs School of Business and School of Law

No comments: