"Race" in College Applications FAQ & Discussion 12

Applying to College
4401

" @Data10 you had posted earlier, (it seems a long time ago I know) that for Harvard “a large portion of the class is accepted without hooks”

For the 2022 class, I got from the website and Crimson surveys, about 32% URM, 23% white legacy, 17% first gen, 12% international, 12% recruited athlete. There’s overlap for sure, but there is no overlap between URM and white legacy, so already we have 40% of the class spoken for. And with 25% first gen being non-URM, that brings first-gen Asian/white at 4%, so hooked at 44% of the class, before athletes. Even if we don’t consider international a hook, I think it would be a safe to say that half of Harvard’s class have a hook, maybe a large portion have hooks.

4402
The Arcidiacono and Card analyses use the same sample, the same general regression analysis methodology, and a similar set of controls; so they come to similar conclusions. The primary differences relate to both Arcidiacono and Card making a few minor differences in choice of controls that favor their desired conclusion about Asian applicants. Card doesn't say Arcidiacono's primary regression methodology is wrong or the math is wrong. Instead the rebuttal primarily disputes the referenced choice of controls and interpretation of results.

For example, my earlier post mentioned Arcidiacono found a 0.33 gap between White and Asian applicants who were not SES disadvantaged, and a much smaller gap of 0.18 gap between White and Asian applicants who were SES disadvantaged. Therefore the gap between Asian and White applicants looks larger if you make the default group not SES disadvantaged, and separately look at low SES with a racial interaction variable. So this is what Arcidiacono does in his analysis. And if you instead do not consider SES disadvantaged by race, then the difference between White and Asian applicants looks smaller; so this is what Card does in his analysis. Card goes in to more detail about his reasons in the 2nd rebuttal https://projects.iq.harvard.edu/files/diverse-education/files/legal_-_card_rebuttal_report_revised_filing.pdf , which found an average change of 0.14 from this Asian & disadvantaged interaction variable – nearly the same as Arcidiacono. With the same influence of Asian & disadvantaged, I’d expect similar conclusions for Black & disadvantaged as well.

In my opinion both Arcidiacono and Card choose controls that slightly favor their desired conclusions to a similar extent, rather than one doing it the right way and the other trying to bias the results. As such, I expect the actual gap between White and Asian is somewhere between the two claims, likely near the threshold of statistical significance at the 95% level.

Using a different source that was done without plans of releasing for the lawsuit, the regression coefficients from a Harvard OIR internal analysis are below. This was an extremely primitive analysis that didn’t consider thing like specific application reader ratings (only high 1-2 rating, or not high 3-5 rating), LORs, a variety of hooks, SCEA/RD, etc. It also looked at income <$60k rather than getting the SES "disadvantaged " flag in the application file, so it looks at actual low income applicants rather than the applicants that admission readers make educated gueses are low income. The anlaysis found that low SES boost was far smaller for Black applicants than Hispanic, White, and Asian races. For example, the gap between Black low/high SES was 0.51,.while the gap between Hispanic low/high SES was 1.13 – more than twice as large. There was also a difference iin admit rate. Black applicants had a similar acceptance rate at both income levels, while Hispanic <$60k income applicants had a 1.4x higher acceptance rate than Hispanic >$60k income.

Harvard OIR Regression Coefficients for Chance of Admission: Primitive set of ~20 controls
Black and income <$60k: +3.02
Black and income >$60k: +2.51
Hispanic and income <$60k: +2.35
Hispanic and income >$60k +1.23
White and income <$60k: +1.00
Asian and income <$60k: +0.72
White and income >$60k: 0.00
Asian and income >$60k: -0.42

4403

I’m not sure how you are computing 23% White legacy, but note that legacy as marked in admission files means parent attended Harvard… not the additional Harvard Crimson legacy categories like aunt/uncle, cousin, or similar. I have seen no evidence that Harvard considers the latter categories a hook. Also note that the class of 2022 legacy percentages appear to be re-scaled to make the totals sum to 100% instead of the actual ~117% sum of categories that occurs in other years, due to students having more than 1 relative attending Harvard. This likely relates to the comment that appears below the graph – “A previous version of this graph incorrectly indicated the percentage of legacy respondents.” I suspect the original version was correct, then someone later incorrectly decided that the sum should be 100% and re-scaled things. The pre-2022 valid years suggest ~16% of the class is what Harvard considers a legacy (parent attended), which seems similar to the 14% legacy in the lawsuit domestic sample.

I also don’t get 32% for URM. The survey reports 10.7% Black and 6.7% Hispanic for a total of 17% between these two single-race categories. Maybe you are assuming multi-racial cannot be a mix of the 2 largest and most common racial groups – Asian and White, so all multi-racial are URM? Also note that the survey doesn’t have a separate check box for international, meaning that some international students will choose the URM categories in the survey, instead of international being a mutually exclusive category.

That said, I think your comment ~half hooked is a reasonable estimate. The specific value depends on a variety of assumptions and what you consider a hook. When I wrote the comment, “a large portion of the class is accepted without hooks,” I meant something similar to ~half, depending on definitions. From your post, it sounds like you also consider ~half to be “a large portion”.

4404

I took a few days off and had a lot to catch up on…

@tpike12 said to me…

“Are you saying that all white men think the same? I bet every person that reads this can think of multiple people that share their skin color, but thinks very differently from them.”

I’m not saying that any two people think exactly the same. But are you saying that people from different backgrounds cannot add different perspectives just based on personal experiences?

Also, as @privatebanker pointed out a few pages ago, yes, there are a lot of inequalities out there.

Why are people so obsessed with race based inequality “solutions”? To me, it seems like AA is being attacked so much because it’s race based, but there are so many other programs out there that are equally “unfair” that you don’t hear as much complaining about (but maybe that’s just me).

I could be wrong but I feel that the playing of the “race card” really gets some people going. Why is that? Why that more than the other admissions preferences?

So we do away with AA, but at the same time we are going to dictate to Harvard that they are not allowed to pick the students that they want? Are we going to force them to use SAT scores? How is THAT OK??? Because unless we get rid of every other subjective measure, that would be the only scenario that is “fair”?

By getting rid of AA, aren’t we in many ways making the government dictate how a PRIVATE college select which students to admit? I would think that many of you that are in favor of dismantling AA are against the government dictating rules to our “free” society, so how does this sit OK with you? So we might be telling Harvard, if AA goes away, that they cannot admit Bobby the Black Boy over Jimmy the White Boy, because Jimmy’s scores and grades were higher? But Bobby still had a 3.9 and a 1450 SAT and would definitely survive at Harvard. Why are we trying to take that power away from a school like Harvard that is likely in a better position to decide what is the ideal learning environment than Congress is…and with all due respect, the side of Congress that would support doing away with race based admissions is predominantly made up of white men.

And yes, @privatebanker what is the obsession with Harvard anyway? I couldn’t agree more and have made the same point over and over…it’s not like these kids that aren’t getting in to Harvard and the like are ending up on skid row. And if any kid only applies to Harvard level schools and their state flagship and they aren’t prepared to accept going to their state flagship, they need a new guidance counselor.

As for the lower SES Asians being discriminated against, I agree that is WRONG, and I don’t have stats on this but I thought (from reading posts here, mostly) that the SES factor helps applicants of all backgrounds (at the need blind schools, at least) and is now kind of a hook of it’s own (but I’m not sure to what extent)…And schools like Harvard are thoroughly reading essays and I would think they might give an Asian student whose essay describes overcoming the adversity of growing up poor a second look. But I’m really not sure how this all plays out.

As for the “life isn’t fair” concept…

My mother is a widow. Her sister also just became a widow. They are both now living off of just one social security check and really have to watch their pennies just to be able to stay in the houses they have lived in for the last 50 years. My mother was complaining that my aunt (her sister) pays such less real estate taxes than her. And that she, my aunt, makes so little money that she doesn’t even pay income taxes. She thought this was NOT FAIR!!! And I guess in a completely level playing field world, this isn’t fair. But then I pointed out to my mom that she could move to my aunt’s neighborhood and pay less taxes (but of course she wouldn’t want to leave her nicer neighborhood!). I then asked her if she would rather have a smaller social security check so that she wouldn’t have to file a tax return. Again, no.

Yes, life isn’t fair, but I think oftentimes those that aren’t being treated “fairly” still have it better off. So while I understand the argument that AA isn’t completely “fair”, would you want to swap places with those people that it helps?

And there is a lot of passion on this thread about getting rid of AA…and it seems particularly strong to me in how it helps blacks. If you go back and read about the Jim Crow laws, even just on Wikipedia, would you feel the same outrage? Or might it help explain a little why as a society some of us feel we have to help certain groups of people catch up a little??

I am not black. But part of my outrage is that so much of this seems to be about black vs. white. And the discrimination against blacks in this country that still exists (and which is far more worse than I thought, as these last few years have revealed) is extremely unsettling to me.

And not to pick on you @OhiBro but some of your generalized comments about blacks just reinforces these concerns to me. I believe you that you don’t think of yourself as a racist, but the fact that you think you can make some of the generalizations that you have made and not realize how racist they are, is part of the problem, IMO.

4405

I’m not saying anything, I’m trying to understand your statement that implied all white men think alike and different skin colors are required for diversity of thought.

4406

“So we do away with AA, but at the same time we are going to dictate to Harvard that they are not allowed to pick the students that they want?”

It’s not doing away with AA, but using race in admissions would be deemed discriminatory, and therefore illegal for colleges to use race. Harvard can pick who they want, I agree, but they can’t break laws right, like discrimination based on race and quotas?

“And yes, privatebanker what is the obsession with Harvard anyway?”

Harvard sets the tone for college admissions, if they are forced to not consider race, the others will follow, of course if it’s deemed illegal by the supreme court, then all colleges will have to follow anyway.

“Congress is…and with all due respect, the side of Congress that would support doing away with race based admissions is predominantly made up of white men.”

At this point, only the supreme court can do away with race-based admission, and they’re a little more diverse, one black male, one Hispanic female, two white females. However it appears that the black male, Clarence Thomas, is the most conservative justice on the court.

“As for the lower SES Asians being discriminated against,”

It’s actually high SES Asians that are being discriminated against, as data10’s analysis shows, in post #4401, high income Asians have the worst chance of getting in to Harvard, low-income blacks have the best chance.

4407

@tpike12 I didn’t imply all white men think alike. You are choosing to read into that that way. But I’m not sure what your point was to begin with. That it’s perfectly fine for white men to have historically received preferences and to run our nation’s corporations and major institutions anyway because they don’t all think alike anyway? And just because a woman or minority gets picked for a man over something in corporate America, does that mean we have to scrutinize her credentials more than we would a white man that was promoted?

I guess the Supreme Court will decide whether race based admissions/Affirmative action is illegal or not. What about any diversity recruitment efforts? Are those all on the chopping block also @theloniusmonk ? Are you saying they should be?

And Congress controls our legislative process, so they definitely have an influence on all of this and social programs related to diversity.

I think URMs and women, as it relates to corporate America at least, might feel a bit intimidated in pursuing careers in industries that have historically ran by white men. Hence the “Fearless Girl” statue on Wall Street. Do you think that was a mistake, putting that statue there?

@theloniusmonk I fully realize that the high SES Asians are being discriminated against possibly the most of all…but I was responding to someone else’s post that pointed out that many lower SES Asians are also hurt in the process because they check the Asian box and are considered ORMs in many of the top schools (I’m paraphrasing).

As far as using Harvard as the benchmark…the point I was trying to make is that it seems so many plaintiffs are feeling discriminated against because they aren’t getting into the most elite schools, but yet there are so many great schools out there. With admission rates lower than 5%, no matter what your stats are, you’re probably not getting into Harvard and schools like Harvard. With or without AA.

4408

The Fearless Girl statue was a marketing effort by State Street to promote its new “Gender Diversity Index” product, whereby State Street would choose companies to invest in based mainly upon the number of women in company leadership roles, and nothing about whether the company has good fundamentals.

Since inception about 3 years ago, the Gender Diversity Index has lagged the S&P by about 2.2% per year. I mean, who could have predicted that company fundamentals matter?

Ironically, a few months later after Fearless Girl was put up, State Street paid a $5M settlement that it violated equal pay rights of its female and minority employees.

In other words, Fearless Girl was purely window dressing. In many ways, kind of like how affirmative action is practiced today.

4409

Coming from the mother of a college sophomore who just accepted an internship with a top Wall Street bank for this coming summer, I can assure you it means more to her and many others than what you describe.

Almost sounds like your rooting against future business leaders like my daughter, @hebegebe. I’m glad she didn’t read your post.

And in case you are questioning her credentials, I’d be happy to send you a (redacted) copy of her quite impressive resume. She’s holding her own in her male dominated business school. With zero help from AA.

4410

Oh, and she has a few white male friends that also have banking/financial internships this summer…every single one was the result of a connection, not the normal recruiting efforts. Hmmm…

4411

My D earned her private equity internship this summer purely on her merits as well, rather than through any connection or diversity program. So she is an aspiring business leader as well. And a feminist.

That’s the entire point. Qualified applicants can make it on their own merits. They don’t need a crutch.

4412

That’s great for your daughter. But there is a difference between the derogatory connotation of a crutch vs. creating awareness and outreach programs to recruit perfectly capable women and URM’s that might not have otherwise felt comfortable pursuing careers in white male dominated industries. And many big companies, the big banks and big 4 included are doing a lot to create work environments that will enable our daughters to climb the corporate ladder while maintaining their abilities to stay connected to their families, which is probably one of the biggest reasons why many women end up dropping out of the race to the very top positions (so I agree, it’s certainly not only about discrimination). I commend the companies for their efforts to attract and recruit women and URM’s.

4413

@collegemomjam wrote:

It sure seems like what you wrote below is saying that increasing diversity (women and minorities) results in diversity of thought, but if that is not what you believe then we agree that not all white men think alike.

The past is the past. It was a different time. You have a group of young Americans that are ready to be the most accepting generation in the history of America and you want to use punish them by using discrimination to balance historical discrimination?

Companies are eager and willing to hire talented women and minorities, why taint it with discrimination?

4414

I agree the companies are doing a lot, but there were comments being made that said white men were watching less deserving women and URM’s take their promotions away… how is that “…a group of young Americans that are ready to be the most accepting generation in the history of America”?

With that said, I would only hope that you are right and that this IS indeed the most accepting generation in the history of America, but I’m not so sure sometimes. It looked like we were headed that way, but there seems to be some backlash, unfounded in my opinion because I truly don’t see the victimization of the white people the way some others do.

4415

MODERATOR’S NOTE:
So the thread does not get shut down again, might I “suggest” that the personal examples of life’s not fair that don’t relate to college applications and affirmative action cease and that comments after mine deal with the thread topic.

4416

Hope this meet’s skieurope’s guidelines.

I agree with you. When it comes to diversity, I believe that outreach is completely appropriate. It is in a company’s best interest to find the largest pool of talented applicants, and if they are ignoring the 60% of the population that are women and URM, they will likely lose to a company that goes after the full 100%.

When it comes to college admissions, I am also strongly in favor of outreach programs to find talented students that might not have considered an elite college, or any college. Too few women considering STEM-focused schools? Encourage and financially support programs like Girls who Code. Too few men considering LACs? Create marketing materials that expound on the advantages of LACs, including great access to professors and lots of women to date :). Too few URMs applying? Create marketing programs aimed at them. MIT goes even further, having a range of programs that are intended to light a fire in women and URMs, including programs like MITES, PRIMES Circle, and WTP. I support all those efforts. In all these cases, I consider the outreach programs as increasing the pool of applicants that meet the academic bar. If that’s considered Affirmative Action, then it has my strong support.

What I don’t agree with is “lowering the academic bar” for reasons other than low income. And unfortunately we are seeing the lowering of the bar in terms of how Harvard does admissions (this focus on Harvard is only because that is where we have the best data right now). I consider this a crutch.

Note that I am also strongly in favor of a low-SES bump, which has the benefit of disproportionately helping URMs.

4417

I agree with a lot of what you say @hebegebe, I just think there might be factors in addition to that impact scores/grades in addition to low income that might justify a SLIGHTLY lower bar. It seems like for the tippy top schools, the lower-bar admits still have decent stats. I guess it just comes down to how we define “holistic” admissions, which I guess they tried to quantify with their personality assessments. I just feel like no matter how they select their classes, there will be “losers” because there just aren’t enough spots for the many, many qualified applicants.

4418

I think in the case of H, the bar is already set very high. It’s above 95% of applicants, in fact.

4419

@hebegebe Amazing post on outreach/marketing to increase achievement and interest my friend.

4420

“I also don’t get 32% for URM. The survey reports 10.7% Black and 6.7% Hispanic for a total of 17% between these two single-race categories.”

the website https://college.harvard.edu/admissions/admissions-statistics has AA at 15.2, Hispanic at 12.3, Native American and Native Hawaiian at 2.3, giving 29.8, plus an estimate of 3-4% for more than one is where I got the 32%. Honestly though, I think the website tends to overstate URM enrollment, to make it look like there are more URMs than there actually are, not just for Harvard but other colleges as well. The surveys and CDS are probably better, I agree.