jimoo wrote:I never said "coincidence", statistical chance does not = coincidence IMO.
True - and that's a good point.
jimoo wrote:2nd paragraph: It is more like a heave from half court, you are missing WAY WAY too much information to get within even the three point line of making that determination. You don't know anything about the pool of these candidates, or about the qualifications desired compared to those qualification as they are distributed in the population. You're missing WAY too much information to get to the conclusion you came to. Again, this is my opinion.
Fair enough. But what I'm saying is that these questions are beside the point - if the gender bias is upstream from the process by which someone becomes 'qualified,' then analyzing the statistics solely in the context of whether someone is 'qualified' or not is a pretty obvious waste of time, no?
jimoo wrote:In general, a level playing field does not result necessarily in a completely evenly distributed outcome, especially with small samples sizes.
Agreed. But even given the small sample size and the greater statistical variance that goes with that, I honestly can't understand how anyone would look at the swing between 19% and 50+% and not think that the difference was meaningful.
I guess that when it comes down to it I'm enough of a rationalist that I believe the numbers over the warm and fuzzy stories people tell themselves.
Don't tell anyone, though. If this gets out my post-modernist cred is shot.