jpet comments on You Be the Jury: Survey on a Current Event - Less Wrong

31 Post author: komponisto 09 December 2009 04:25AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (260)

You are viewing a single comment's thread.

Comment author: jpet 09 December 2009 09:34:43PM *  10 points [-]

I was unfamiliar with the case. I came up with: 1 - 20% 2 - 20% 3 - 96% 4 - probably in the same direction, but no idea how confident you were.

From reading other comments, it seems like I put a different interpretation on the numbers than most people. Mine were based on times in the past that I've formed an opinion from secondhand sources (blogs etc.) on a controversial issue like this, and then later reversed that opinion after learning many more facts.

Thus, about 1 time in 5 when I'm convinced by a similar story of how some innocent person was falsely convicted, then later get more facts, I change my mind about their innocence. Hence the 20%.

I don't think it's correct to put any evidential weight on the jury's ruling. Conditioning on the simple fact that thier ruling is controversial screens off most of its value.

Comment author: Cyan 10 December 2009 01:16:42AM *  4 points [-]

This is a very interesting analysis -- I like your choice of reference set and your Outside View approach.

Comment author: Sebastian_Hagen 11 December 2009 11:41:02PM *  0 points [-]

I don't think it's correct to put any evidential weight on the jury's ruling. Conditioning on the simple fact that thier ruling is controversial screens off most of its value.

I disagree. Do we have specific data about the correlation between the controversy of jury rulings, and their accuracy (or some half-decent proxy, like the likelihood of the rulings being sustained in appeal)?

Most of the controversy in this specific case appears to originate from people who have significantly worse access to the factual evidence than the jury; and it's likely to be in the interest of some entities reporting about this case to play up the controversy to attract readers. I don't think there's any strong evidence to be gained from this, and consider the original ruling to still be significant evidence even after taking the controversy into account.