Upon Further Review: It's ridiculous, of course, absent any scientific evidence that adversity affects raw scores. But at least they are keeping the categories separate. (They are, aren't they? I'm not misreading this?) It would be worse if they changed the questions, measuring whether groups did equally well rather than predictive value of the test. They did that years ago, I dimly recall, culling math questions in the 70s on which males and females scored differently. I have heard that they changed back to previous objective standards, and I have heard they did no such thing. I read it years ago - I don't know the truth of it, but I can't find a thing about it now.
At the time it was considered automatic proof that if boys and girls had different rates of getting a problem right, then it was obviously a bad question. I thought so too. At the time, I only thought "Huh. Isn't that interesting. Males and females, though they are obviously the same in all things, still have mildly different styles. Cool."
Adversity scores are fine, so long as they don't bleed over. If a college wanted to specialise in accepting kids with high adversity scores I can see a defense for that. I don't think they will get the results they want, but at least it's intellectually consistent.
Update: I was wrong. The adversity score is going to be invisible and not easily separable from the final count. I suspect that this will make it worse for students with obviously Hispanic, black, or Asian names, as admissions offices will mentally correct for what they guess the adversity score is, to the detriment of some.
4 comments:
I suspect there'll be a jump in the number of mail "pass-throughs" in poor neighborhoods. If you "live" in a deprived area your score will go up a bit. Even if your school isn't one of the bad ones, every bit helps.
The college admissions machine is becoming such a rube goldberg contraption.
I'm guessing the SAT people see the writing on the wall. In 5 to 10 years the Affirmative Action position is going to be intellectually indefensible, not only due to the obvious genetic link to average IQ but also because the diversity experience rational propagated since the 1990s will collapse under the growing demand for identity unique cultural spaces. There will, however, still be legal and cultural demands for proportional representation by certain identity groups. The only way to square that circle is a value that incorporates the socioeconomic markers associated with those groups, not likely to change significantly due to the genetic component of IQ and behavior, but sufficiently opaque to provide plausible deniability that it exists primarily to allow interested parties to identify group members when they wish to do so.
Now that I think of it, I graduated from a school in Africa. What does that do to my score?
Eldest Daughter graduated from home school...
Gotta love the corner cases
Post a Comment