News for hackers. Stuff that matters.
Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by SerialOwl on Saturday February 02, 2019 @05:07PM (#19066050)
And of course it's just a coincidence that these "holistic qualities" happen to be found more frequently in some races than others?
Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by csa on Saturday February 02, 2019 @06:48PM (#19066677)
Asians at Harvard undergrad are represented at roughly 4x their proportion of the general US population. Assuming that Harvard undergrad values “holistic qualities” (I think it does), then the answer to that may be yes.

Is this the argument that the folks in the lawsuit are making? I don’t think so. The argument to me sounds more like “why can’t Harvard make admissions just based just on a test?”

Short answer: There isn’t a standardized test that does a good job of selecting for the “holistic qualities” that certain schools want. Hence the admissions system that most elite schools use.

Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by leereeves on Sunday February 03, 2019 @06:24AM (#19069011)
If these "holistic qualities" can't be detailed or measured, they may become merely a reflection of the interviewer's personal bias.

> On average, alumni give white and Asian American applicants similar ratings, but Harvard staff give whites substantially better reviews than they give Asian Americans.

Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by betterunix2 on Saturday February 02, 2019 @05:12PM (#19066082)
Actually, no, it is not a coincidence that holistic qualities are not uniformly distributed across every social group. For example, a member of a historically privileged group is less likely to be the first member of their family to attend college than a member of a historically disadvantaged group.
Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by sonnyblarney on Saturday February 02, 2019 @05:11PM (#19066072)
Maybe not even a coincidence!

Maybe extremely high SAT and grades are correlated with poor 'holistic' qualities such as social skills, communication, leadership, creativity, 'round character' (i.e. athleticism, disposition, generosity, musicianship).

I'm not saying they are - I'm saying it wouldn't really surprise me if they were, and then we have to think about it a little bit!

It's a tricky subject no doubt.

Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by ThrowawayR2 on Saturday February 02, 2019 @07:03PM (#19066762)
> Maybe extremely high SAT and grades are correlated with poor 'holistic' qualities such as social skills, communication, leadership, creativity, 'round character' (i.e. athleticism, disposition, generosity, musicianship).

If that were so, it would affect members of all ethnicities who had the same high SAT scores and grades equally.

Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by sonnyblarney on Sunday February 03, 2019 @12:02AM (#19067969)
Good point, but those 'other qualities' may not be distributed very evenly among the remaining applicants. This is an ugly problem to have. I don't envy anyone in this situation.
Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by leereeves on Sunday February 03, 2019 @06:06AM (#19068950)
Many of those "poor 'holistic' qualities" can be code words for discrimination. "Social skills" for neurotypical, "leadership" for attractive, "athleticism" for not disabled, "disposition" for obedient.

Even if you're right and extremely high SAT and grades are correlated with neurodiversity, physical problems, or rebelliousness, are those qualities colleges should be discriminating against?

Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by sonnyblarney on Sunday February 03, 2019 @01:33PM (#19071289)
Or maybe 'leadership' actually means 'leadership'?

And 'athletic' means 'athletic'?

They are unless you have a heap of evidence to suggest they are not.

And yes, Universities should definitely be considering all of those qualities, most of them do to the extent they can.

Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by leereeves on Sunday February 03, 2019 @03:32PM (#19071996)
There's certainly a heap of evidence of discrimination against neurodiversity, physical problems, and rebelliousness. It's certainly been talked about enough. If you're not aware of it that's only because you've chosen to ignore it.
Re:What Statistics Can't Tell Us in the Fight Over Affirmati... by tomrod on Saturday February 02, 2019 @05:52PM (#19066311)
I don't think it's a tricky subject in the way that fluid dynamics is tricky. I do think it is difficult to communicate openly about because systemic favor and disfavor lie so close to individual identity that those favored have a hard time acknowledging it and those disfavored have a hard time being listened to. The ego is super enmeshed.