vanwinkle wrote:The correlation is strong with regard to student bodies as a whole, not individually. GPA/LSAT are strong enough predictors that if you take a large number of people with those stats, you can predict how many of them will succeed or fail at a certain level. It has very little to do with individual performance at all;
Sure, but all of this analysis is focused on maximizing your statistical fitness. If you don't accept that, you can't say that attending Harvard is better than attending Touro.
vanwinkle wrote:To put it another way: From HLS to GULC, the difference in medians is 3 LSAT points and 0.07 GPA. With the numbers that close, do you really believe the difference between student bodies is so great that it would substantially change how well you could do at each?
Half the class at HLS falls within 5 LSAT points and 0.18 points of GPA. The statistical correlation has been derived on data sets with very similar input variance to the difference between schools.
Obviously, it's impossible to do an empirical correlation across schools, but I would be confident in an analytical correlation. I'm not sure if the data is available, but from what has been published, it's a pretty easy deduction that one's statistical fitness is higher at GULC than HLS.