Data to help find reach, match, safety

@merc81 #26
I think that’s right.
I doubt USNWR intends the 65-25-10 weighting to reflect the relative importance of each factor in college admission decisions. If you flipped the test score and T10% weights, you might be closer to the weights college adcoms assign, but I really don’t know if the results would reflect relative admission selectivity better.

Well, let’s try it.
Below are the T50 I get, still using the same 4-5 year old numbers, but with a 25-65-10 weighting (for T10% Standing-Scores-AdmitRate, respectively). I’ve marked some of the schools that move way up in selectivity by this new measure:
Cal Tech
MIT
Harvard
Yale
Princeton
Stanford
Harvey Mudd
Washington U
Penn
UC Berkeley (+21)
Brown
Pomona
Haverford
Northwestern
Dartmouth
Georgetown
Duke
Vanderbilt
Tufts
UCLA (+30)
Swarthmore
Notre Dame
Rice
UCSD (+38)
Williams
Amherst
UVA
Cornell
UCSB (+40)
Bowdoin
JHU
UC Davis (+44)
USC
Carnegie Mellon
Wellesley
Carleton
W&L
UC Irvine (+45)
GA Tech
Claremont McK
W&M
Davidson
UWashington (+42)
Middlebury
Emory
Barnard
Colgate
Wesleyan
Hamilton
Scripps

Many schools above stay pretty close to where they were according to the 65-25-10 weighting. However, some of the West Coast state schools (especially the UCs) move way up. I can’t say for sure which way better represents true admission selectivity. However, … UC-SD’s 2017-18 CDS reports their average entering GPA as “4.08” (even though they’re supposed to be using a 4.0 scale). UC Davis reported its entering average as 3.99; UC Irvine reported its entering average as 3.97. Stanford’s 2017-18 reported average was 3.95; Princeton’s 2016-17 reported average was 3.89. Are UC Irvine, Davis, and SD more selective than Stanford and Princeton? Even by HS academic performance alone? It may be the case that some of the super-selective private schools draw more heavily from super competitive high schools and make allowances for that.

Now if we jacked up the weight for admission rates, I suppose we’d get yet another set of results.