[eDebate] Harvard -- Octos Judges -- 8am
Tue Oct 31 06:29:41 CST 2006
RE: [eDebate] Harvard -- Octos Judges -- 8amNot a lot of time to spend on this today. First, you're right that I didn't state things accurately enough in my initial post. I should have said that the mutual preferences of the 8 quarterfinals at Harvard resulted in zero women judges.
That said, I've posted several times before attempting to tease out the various possible explanations for the lesser representation of women in outround panels at large tournaments. Others are obviously welcome to interpret the data differently, but I think it's clear from the data that:
1. On average, women judges are less preferred than men judges (Gary Larson has confirmed that with respect to a couple of tournaments he's tabbed);
2. On average, women judges are even less preferred by top teams than those at the bottom of the bracket (yes, I understand that you get lower preferences once you're eliminated; even in rounds before that takes place, teams that eventually place lower see more women judges than teams that eventually place higher).
The question then becomes why that takes place, and there are indeed multiple explanations (out-and-out discrimination, networking, experience, etc.), but we should try to answer those questions.
While your explanation (half the teams in octas prefed women highly, but they faced the half that didn't) is perhaps plausible, it certainly isn't mathematically likely (though I will say from eyeballing the data that the extreme nature of the octas distribution was probably quirky in a manner like the one you suggest; I would guess that a round robin between the 16 teams (with MPJ) would result in fewer women than random judging, but certainly more than zero). So, I think you're right that 24-0 was a quirk; I think, however, that it would be virtually impossible to happen in a pool where those 16 teams preferred women equally to men in the pool. We need to figure out why this keeps taking place (not to this extreme, admittedly).
----- Original Message -----
From: Martin Harris<mailto:mharris02 at drury.edu>
To: NEIL BERCH<mailto:berchnorto at msn.com>
Cc: edebate at ndtceda.com<mailto:edebate at ndtceda.com>
Sent: Tuesday, October 31, 2006 1:29 AM
Subject: RE: [eDebate] Harvard -- Octos Judges -- 8am
I'll try this another way, with stats of my own. I think people should be careful in over analyzing the data. I will not contest there is a statistical disparity in the representation of females in the Octos judging requirements (that is a matter of fact). I think it is hazardous to conclude from that data that NO women (your words) were preferred by the Octos representatives. Consider this, at least 4 of the 16 representatives have female Directors or Assistant Directors (Greta Stahl, Melissa Wade, Sara Partlow-Lefevre, and Heather Walters). I know in one case the prefs were filled out by one of that group of four. I watched her do part of them on the computer in my office. I will add to that list Jackie Massey because he was been a vocal critic of sexism in debate, and will hence conclude he is conscious of how he fills out pref sheets. I would also likely add to that list Matt Stannard and Matt Vega because I have heard both of them speak to the issue as well. I am also not sure that Linda's retirement was effective at the beginning of this year or the end, so the group of 4 maybe 5. This does assume that DOFs/ADOFs do prefs, but I think it is fair to assume even if they don't directly fill them out, they influence selection. I will also add Wayne State to the list of teams that I am fairly certain doesn't exclude women judges since they were half of the round 8 pairing that had only one judge, a woman, that fit the A-A model.
My point, while a narrow one, is I think your conclusion is vastly underjustified by the data you selected to justify it. I am also unsure how to take your new evidence, at least what you are attempting to argue from it. I can't speak to the way that Dallas/Harvard tab room pairs debates, but it is at least common practice at many tournaments to not honor prefs in down 4 rounds due to constraints in pool size. Hence, I would not find it at all surprising that women judge more rounds without MPJ occuring then with. This still doesn't answer the vast potential of alternative causalities to the wholesale deselection of woman as preferred critics. Is it the algorithm for judge placement that deselects women or the prefs themselves? Notice the prelim sheet commitments seem to have a disparity in assignment to women with some having 2 or 3 round commitments, and male judges being statistically more likely to have 5 and 6 round commitments. So if it tries to maximize placement by using more committed judges first, that would occur. If it tries to use the MOST prefed judges, I would aslo expect the disparity in outcome from the possibility that SOME pref males disproportionate to females. This is why I said I won't make arguments in the broader debate, but merely question your conclusion.
I think there are many possible explanations to why women were not represented in the Octos debates. The least of which is HALF the pool could highly pref women, and if the half they debated didn't pref any women at all, or different women, you would see the outcome we have now, no women, since MUTUAL makes one sexist able to take out a whole host of the other's prefs. It just seems disadvantageous to the cause to try to make broad sweeping generalizations using correlative inferences. You potentially marginalize a group of people that may agree with the "cause" but were powerless to change it with their pref sheets. What part of the process of preference is failing us is the important question if you want to be able to approach movement towards fixing the problem since I don't think you are going to get much movement towards eliminating MPJ. Baylor tried it a few years ago, and while also potentially merely correlative, it went from a tournament with 60 or so teams, and about Harvard importance in travel schedules to ... doesn't it even exist anymore?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mailman