Hacker News Reader

Comment by Veelox | Hacker News Reader

>Obviously hard objective criteria are not be enough to get into Harvard, and they need to dip into subjectivity which is rife with bias.

I think you are missing the point OP is making. The top 80% of blacks score well on the subjective stuff. Only the top 10% of Asians score that well. With that, there are a few possibilities.

1) Blacks are more interesting than Asians.

2) Personality is uncorrelated with test scores.

3) Harvard is purposely scoring to change the racial split of the class.

Of those, I think 3 is most likely. I am open to discussion but I think Harvard is to smart to be doing it on accident and the racial split is to politically acceptable to be by chance.

View on HN

I wish you were right, because if the issue was just a few racist admissions officers, then the solution would be simple: fire them and get non-racist ones, problem solved. However, there is a fourth, more likely explanation:

4) Harvard's scoring is unintentionally hurting Asian students.

From the article, it seems that Harvard focuses on establishing student diversity that roughly tracks the US population, bringing in big money donors, bringing in legacy students, and finding leaders that have traits similar to other successful leaders in the past.

None of these values have the intent to reduce Asian admission, but that is their undeniable effect.

It means that Asian students don't face a single racist admissions officer, they face a facially race-neutral system that has a discriminatory effect. To fix it, you can't just fire a few people, you need to make much larger changes. My original point was that the only fair and effective solution would need to require a larger class size.

>4) Harvard's scoring is unintentionally hurting Asian students.

It's Harvard we are taking about, not a community college. Someone in admissions knew it was happening and let it keep happening. If you are willing to say unintentional means knew it was happening and didn't care to stop it, sure it was unintentional. The issue I have with that is if the same thing was happening but you replaced Asian with African, people would be up in arms. Such a double standard offends my sense of equality.

I am not arguing with you that it's wrong, I am saying the solution isn't as easy as you describe. This isn't a tumor you can cut out, it's a veiny cancer that is hard to see and widespread. I'd bet there were a number of individuals that knew about it, felt it was wrong, and maybe some did do something about it. Firing them would probably make the system worse, not better.

It's quite easy for a bunch of non-racist individuals to create a well-intentioned system that has a racist effect. It happens in many institutional systems, this year's Pulitzer prize went to a guy who showed that D.C.'s well intentioned and black-run criminal justice system discriminated against black people in their own city.

When you have systemic problems like this, you can't just swap out some individuals, you need to fix the system. My original point is that increasing Harvard's class size will be a necessary component. It may also require a change of leadership (e.g., president or dean of admissions), but this is not a personnel problem, it 's a structural one.

We all think the claim that just policies result in just consequences, is logically true. Empirically, as this case demonstrates, that is not true.

It seems unlikely to be unintentional. Since Harvard is grading personality on a USA-centric scale, it would most likely accentuate the traits of the current USA leaders and executives who are overwhelmingly white. We would expect the white children to inherit, genetically and environmentally, these traits from their leader and executive parents and score the highest by a good margin and all other minorities to score low.

The results contradict the expectations. Conveniently, it is black > Hispanic > white > Asian which is exactly the diversity hierarchy

If their only aim was to favor "the traits of the current USA leaders and executives", then that would indeed to a preference of white males, but they have other goals as well, and when taken together, do fall in line with expectations.

What "other goals" and "expectations"?

It conveniently falls in line with racial diversity goals but it is unclear what other goals it accomplishes. The reasonable assumption is that any soft skill goals or measures will have a high preference for white males. It is very suspect that this isn't true.

Reply on news.ycombinator.com