Quote:
Originally Posted by CMDelta
I was just sure that I remembered reading somewhere that having the model with a 3 preference party format AND you maximize all your options by listing ALL that you attended then you were almost guaranteed a bid short of something highly unlikely coming up. Yet in the 2 pref party model, I thought it was certainly more probable not to match. Maybe I misunderstood and I couldn't find where I had read it but I thought the schools that had the 3 pref party format had higher match rates. No?
|
I don't think that's the case if everyone truly maximizes. Hypothetical example: 30 PNM's, 3 chapters...working quota is 30/3 = 10.
ABC has a 100% return rate (every PNM at their party lists them first), DEF has a 50% return rate, and XYZ has a 0% return rate.
If the PNM's could be ordered 1 (most desirable) to 30 (least desirable), ABC can invite 1-10, DEF can invite 1-20, and XYZ can invite all 30.
With three parties:
1-10 attend all three parties and all list ABC, then DEF, then XYZ, bid from ABC
11-20 attend DEF and XYZ and all list DEF than XYZ, bid from DEF
21-30 attend XYZ and list XYZ, bid from XYZ
All 30 PNM's are placed. Viola!
With two parties:
1-10 attend ABC and DEF, list ABC, then DEF, bid from ABC
11-20 attend DEF and XYZ, list DEF then XYZ, bid from DEF
21-30 attend XYZ and list XYZ, bid from XYZ
All 30 PNM's are placed. Viola!
The example breaks down in a lot of places: Neither PNM nor chapter preference is absolute, women don't maximize, etc. However, I would guess the real issue is this: WOMEN WHO DON'T SHOW UP ON BID DAY COUNT AS "PLACED" WHEN WE ALL KNOW THAT'S NOT REALLY THE CASE. If placement rates go down with 2 parties vs. 3, I would bet there are women dropping out before pref who would otherwise drop out on bid day.