News for hackers. Stuff that matters.
Re:Court of Appeal bans Bayesian probability by rayiner on Monday February 25, 2013 @10:44AM (#5279447)
90% of the work of a jury trial is managing human reactions to things. There are things that judges need to know about statistics. E.g. That in modern DNA forensics, lab error totally dominates the probability of a coincidental match, limiting accuracy to 1-2%. On the other hand, they must be deeply cognizant of human responses. A mathematician, told that some test shows a person 1,000 times more likely than average to be the killer, will intuitively realize that the person is still unlikely to be the killer. A jury of ordinary people won't. We have judges precisely to mediate between these two domains.

There are no comments on this comment yet.