Right on. Merit is much more than SAT scores and GPA. It is, in fact, whatever the school wants it to be.

Plus, standardized test scores have a significant income bias. Affluent kids get SAT prep and can take the SAT multiple times if they're unsatisfied in their scores. And I don't have any data, but my guess is that there's a negative correlation between how many hours per week that you are employed while attending high school and your GPA.

These factors may be part of the consideration when accounting for lower quantitative admissions requirements for black and hispanic applicants, and those from lower income brackets in general.

Studies which aren't funded by Kaplan tend to show a minimal effect due to test prep.

http://online.wsj.com/article/SB124278685697537839.html

Also, bringing up income when defending racial preferences is disingenuous. If colleges wanted to discriminate by income, they would.

Fun fact: income is not the cause of racial gaps in educational performance. It plays a role, but the gap exists even holding income constant.

http://www.umich.edu/~rdytolrn/pathwaysconference/presentati...

http://www.jstor.org/pss/2963200

Surely merit does not include race?

Merit can include a consideration of race. One factor of merit is being an outlier of your group. There are many ways to analyze a given population that reveals different clusters. Being an outlier of this cluster is a factor in merit. The question is, is race a relevant dimension to consider? At face value, the answer is no. But in the US at least, race is a proxy to many other factors that are extremely hard to quantify. These hard to quantify factors are relevant in determining who "deserves" a spot at your prestigious school. Thus a consideration of race within specific contexts can be relevant, college admissions being one of them.

That isn't at all what I was saying, but the idea that two IDENTICAL students with nothing other than one being white and the other being asian would require 200 more SAT score points to get in is inaccurate.

As a class, the students who applied who happened to be asian are probably lacking in other areas as well if this BIG of a bias is being seen. Perhaps there aren't enough participation in volunteering/jobs/sports. Perhaps there isn't a good relationship in the alumni interviews, who knows. There may STILL be a racial bias on top of that, however I'm guessing not 200 points worth of pure racial bias.

It occurs to me that a place like Harvard could be as interested in what a class will look like at its 25th reunion as anything else.

Easy. Take "ethnicity" out of the application and blot out the applicant's last name when evaluating applications.

Extracurriculars are another dog-whistle for that sort of thing. A heuristic like "Deduct 150 SAT points for violin or piano" would be pretty predictive.

While this is true, I don't see why this can used as justification to keep name and ethnicity showing. There will always be ways to guess an applicant's ethnicity but since it (supposedly) has no bearing on admissions, why keep it in the application?

Not once it becomes known - I imagine Amy Chua types would stop forcing their kids to learn it, and applicants would stop mentioning it on applications (unless they are applying to Juilliard).

You're correct. Precision is important. It would have been more accurate to say "...100% academic merit..."

The material point is the same though, I really can see where the hypothetical snowboarder might have something to offer the Ivy community even if there are others more qualified academically. But again...I can't attach a number of points to that. I don't know what that number would be.

To the issue...I do think there should be more taken into account than just academics. I just can't think of a way to take those things into account AND be fair.