StatisticsAsEvidence

Interpretations of probability & how statistics should / do count as evidence.

  • jaynes_ch1.ps jaynes_ch2.ps: The first two chapters of ET Jaynes' book on probability, in which he presents Bayesian statistics as an extension of propositional logic that can handle inductive reasoning. This angle gets short shrift in the Stanford Encyclopedia of Philosophy for some reason.

Forensic evidence

Jury consumption of DNA evidence is the ultimate statistical decision problem. The stakes are often literally life-or-death, the options are stark, and the math isn't complicated enough to obscure the fundamental unresolved issues: how probabilities should count as evidence, and how that evidence should be presented to decision-makers.

  • Dawid_comment.pdf: See comment 2 and the rejoinder for a sampling of the actual debate.

  • davis_cold_hit_motion.rtf: The court decision Frye vs. United States sets the standard for admission of scientific evidence in court in the US: "Just when a scientific principle or discovery crosses the line between the experimental and demonstrable stages is difficult to define. Somewhere in this twilight zone the evidential force of the principle must be recognized, and while courts will go a long way in admitting expert testimony deduced from a well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs." In this motion, a defense lawyer argues that the DNA fingerprint evidence against his client fails this standard.