"Race" in College Applications FAQ & Discussion 12

[collegemomjam]

@Happytimes2001 I don’t think you can say that about all kids coming in with lower stats…some might have to take easier majors (so what though?) and some might not get as great grades, but some will. Same for athletes, legacies, and anyone else not held to the same criteria as the highest caliber kids accepted. My point is that the SAT score in particular may NOT necessarily dictate their success as a student. (Pretty sure it’s been proven that GPA is a better gauge than scores anyway).

I bet there are lots of kids that got 700+s on the English writing that would NOT have gotten that if they didn’t have the tutor their parents hired. I’m sure of it. I hired tutors for my kids and there is no way they would have done as well without those tutors.

[Happytimes2001]

@collegemomjam Yes, GPA is definitely a better indicator. But I think that many kids get super high scores without a tutor. Some self study. The point is more, that public schools vary a lot across the US. Kids from some states and regions get a less than stellar education. Others get a great education. Some come into college with a lot of gaps. You just cannot compare a kid from an urban magnet score to one from a rural area that is sorely underfunded.

To think that everyone who got a high score had a tutor is not my type of thinking. I don’t think 700 are the scores were are talking about in any case. If you look at these scores the median is often much higher. You can check per school. I know SATs have changed a lot over the decades. That said, most of my Ivy league cohorts had scores in the top 2-5% and they had grades to match. Some had lower grads and higher scores and many had higher scores and lower grades ( I think they used to emphasis it more). I know a lot of Ivy league graduates and their kids. Few were weak students or had low scores. Most had high stats and a hook (alum, sports, some great EC or a combination). Alums were not weak academically . Most had parents who had graduated from Ivies in the 1960s and had a lot of wealth and attended some of the best schools so they were very prepared. Socio-Economically deprived kids were usually outstanding academically, else how would they get in? The sports kids often had tutors ( they still do) so that can help with the ramp on to college.
It’s funny to me that someone actually thinks people are walking into the Ivy league with scores in the 600-700 range, or some GPA that doesn’t meet the 50% percentile. I haven’t checked Harvard stats but I’d guess that they are pretty high. Also, most students accepted at Harvard are 1 or 2. It states in the case what those scores are. I read them the other day and they were as I would have expected.

[ChangeTheGame]

@ucbalumnus Your perception that there would still be those who will believe that racial preference exists even when it is removed is spot on.

@collegemomjam I do believe that the majority feel that minorities on elite college campuses have earned there way, but the data (especially standardized test scores) is part of the reason that the stigma exists. Personally, I get fuel from the slights of others, but it is much harder for me to deal with it when it possibly happens to my kids and their friends.

@Happytimes2001 One reason that I sometimes downplay standardized testing besides the context of an individual’s score also deals with my own experiences with my kids. My kids are just normal smart kids, not anywhere close to some of the genius level kids I see on CC. They have been unknowingly preparing for standardized testing since they were little kids and knowingly since the 7th grade. Not quite Tiger Dad levels, but little things over time to overcome some of theories on why African Americans struggle (like testing biases and being comfortable with the test format). They have probably “out-tested” their natural testing ability due to the resources available to them (including the most important resource, my wife who can teach almost any high school science and math) and the work they put in. They know how to take the SAT and ACT but can they critically think? I think so, but standardized testing has basically been “hacked” by families like my own trying to join the few who can score 1500+ SAT and 34+ ACT with one look, versus the years of preliminary work needed for my kids to get there.

[Data10]

What really needs to exist are more programs that prepare underserved kids. Programs that fill in the gaps for kids with big potential who are socio-economically deprived. There are some. But not enough. Filling the gap in the high school years is much better than placing kids in elite colleges who don’t have the support structure to do well and telling them they have aptitude. Aptitude is great. But the proper education is key.

Selective private colleges, such as Harvard, almost always have programs to support students who have relatively weaker HS backgrounds. They don’t just throw everyone in to the same intro freshman math, regardless of HS background, and see if they sink or swim. For example, Harvard has freshman take math and science placement tests that give a recommendation of what math and science courses to begin with. It might recommend any of the following – Math 1a; 1b; 19ab; 20; 21a,b; 23a,b; 25a,b; and 55a,b. They also have placement advisers that discuss the placement and results, as well as considering math course background and AP score from high school, as well as desired field of study and future goals, allowing the student to discuss what course is the best fit. A student who starts with the lowest level math 1a (single variable calculus at HS level) could be a physics major, as could a student who starts in math 55 (probably one of the most challenging freshman math courses in the United States). Also note that Harvard isn’t basing math sequence on math SAT score, nor are they primarily admitting students primarily based on SAT scores or claims of “aptitude”.

[Happytimes2001]

@Data10 Have you ever taken a low level Harvard math class as someone without a solid math background? If you did, you would realize that even if you are in the lowest level class you might not have the background to succeed. No one said they were going to lump everyone into the same level. It’s that you have to be at roughly the same level as others. Or let’s say you can take the basic math and you are a weak writer. The first week you would receive several writing assignments. Do you think you can catch up to a level of other students who are almost all very strong writers. You’d likely end up spending lots of time catching up. And I’d like to meet that physics student who started in the lowest level math class. I once was talking to a friend of mine who was a physics major at Harvard. She said she barely made it through due to having a weak math background. This coming from someone from a fine public school best state in education ratings wise. So…I know one data point isn’t a case. However, I don’t know anyone who majored in physics at Harvard or MIT who wasn’t stellar in math before they went to college.

Most kids in Asia are starting high school with Pre-Calc and some start at even higher levels. Many of these kids are also taking physics at Harvard. So you are basically stating that you can be several years behind others in math (let’s just say1-2 years behind the US kids with/without AP ) and still catch up and do well. Really?? How are you going to catch up while learning at a rapid pace? All while taking a difficult course load and surrounded by others who are strong.

No one claimed Harvard used SAT scores or other scores solely to determine their entering class. They know all of the schools by region. They know which ones have solid programs and which have mediocre classes. They also use SAT scores as a determinant for the applicant from schools which are not as well known. SAT’s tell only a small part of the story. But they are important.

The Freshman Class at Harvard is mainly made up of kids from excellent schools, some from mediocre schools who have some hook and very few from weak and underrated schools. There’s a reason that some public schools send 20% or more of their students to the Ivy league and others rarely have a single kid in any given year.

Personally, I wouldn’t send my kid to a school to be at the bottom of the class. I’d always want my kid to jump in and do well. Everyone knows what they are capable of. But to say that Harvard can support students with weak backgrounds is disingenuous. They might accept a few. But they are not going to “catch up” and soar as people hope. Even with a writing lab and help from great professors.

[UndeservingURM]

@ChangeTheGame

My daughter’s best friend (also an African American) is attending a top 3 ranked Ivy and it was frustrating to hear my daughter talk about what their peers thought about the “why” she got in despite being in the top 3%, 1500 SAT, and pretty good ECs.

I don’t know this woman but from what you’ve described, she doesn’t seem that amazing.

The problem is that most people with “good” stats (decent ECs, > 1500 SAT, ranked top 5%) but not amazing stats (won national competition or award, created start up, etc.) get rejected.

Non-URM with only “good” stats has like a 20% chance of getting accepted into HYPSM. An African American with only “good” stats has like a 90% chance of getting accepted into HYPSM.

If she was accepted and she’s African American with only “good” stats, there’s a good chance that she was accepted only because of her race because most people with only “good” stats get rejected.

[Data10]

@Data10 Have you ever taken a low level Harvard math class as someone without a solid math background? If you did, you would realize that even if you are in the lowest level class you might not have the background to succeed.

Have you ever looked at the course material and grade distribution for the lowest level math sequences at Harvard? If you did, you’d see that it is not rigorous compared to exams at many less selective colleges or even many HS AP calculus classes. Exams for such classes are often graded on scale where 90+% = A, like HS; rather than giving challenging questions with a curve, where you aren’t expect to get nearly all question correct. The classes are intentionally designed that way so students with a weak HS calculus background can repeat the material and get a solid single variable calculus background. You’d also see few get final course grades below B’s in such classes, and students failing is extremely rare, giving no suggestion of students not having the background to succeed.

Most kids in Asia are starting high school with Pre-Calc and some start at even higher levels. Many of these kids are also taking physics at Harvard. So you are basically stating that you can be several years behind others in math (let’s just say1-2 years behind the US kids with/without AP ) and still catch up and do well. Really?? How are you going to catch up while learning at a rapid pace? All while taking a difficult course load and surrounded by others who are strong.

Harvard does not require or expect students to skip calculus 1, even though some students do take as high as multivariable calculus, linear algebra, or beyond during HS. Most concentrations at Harvard don’t require a large number of classes. For example, physics was mentioned. It requires taking 12 courses, leaving 20 additional courses. Some of those additional course can be repeating HS calculus if needed. A possible math sequence could be Math 1a → 1b → 20 ->21. Math 1a/1b is the sequence to catch up for weaker backgrounds, and math 20/21 is multivariable calculus and linear algebra at a normal pace – not a fast pace like 23 and not with additional topics and/or rigorous proofs like math 25 and 55. Beyond Math 21 isn’t required for physics concentrations.

Harvard also offers a sequence more basic than 1a for students who don’t have a solid pre-calculus background. According to Harvard’s websites, they’ve had students starting in this lowest level sequence that repeats pre-calculus who have gone on to become math professors, after completing a math major/concentration at Harvard.

[UndeservingURM]

@collegemomjam

The sad reality is that most disadvantaged students that start at a lower place will never catch up with advantaged students that start out at a higher place in the professional and academic world. They will always be a step behind.

While Joe Disadvantaged is taking remedial classes, Jim Advantaged is studying or networking for IB/coding/quant interviews, publishing research papers, studying for GRE/LSAT/MCAT, etc.

That’s the way it is.

[ucbalumnus]

@Happytimes2001

Have you ever taken a low level Harvard math class as someone without a solid math background? If you did, you would realize that even if you are in the lowest level class you might not have the background to succeed.

Harvard’s lowest level math courses are:

Math Ma, Mb: two semester single variable calculus sequence equivalent to 1a with review of precalculus
Math 1a, 1b: ordinary two semester single variable calculus sequence

In other words, these are suitable for those with ordinary readiness for college level math (Ma and Mb are for those who are a bit behind).

It is the subsequent courses, for those who have had single variable calculus while in high school, where there are five levels:

Math 21a, 21b: ordinary multivariable calculus, linear algebra, differential equations – appears similar to ordinary sophomore level math elsewhere
Math 22a, 22b: 21a, 21b with more rigor
Math 23a, 23b: linear algebra and real analysis
Math 25a, 25b: theoretical linear algebra and real analysis
Math 55a, 55b: studies in algebra, group theory, real and complex analysis

The latter two sequences require familiarity with proofs (i.e. beyond just 5 on AP calculus BC or equivalent that the first three sequences list).

http://www.math.harvard.edu/courses/2018-2019.pdf

In other words, Harvard offers “ordinary” math courses (1a, 1b, 21a, 21b) as well as a semi-remedial course (Ma, Mb), in addition to four levels of honors math courses. It does not offer low-remedial math courses (e.g. elementary algebra, high school level geometry, intermediate algebra), but it is presumed that someone who can get admitted to Harvard (even with the greatest hooks) is beyond that. But the minimum math skill in the lower level math courses at Harvard is ordinary readiness for college level math, not top-end math skill like at Caltech and Harvey Mudd.

[OHMomof2]

@milee30 Because legacy and athlete preference - as ridiculous as many of us find them to be - are not illegal. Race preference is.

Even if those preferences disproportionately benefit white students?

Maybe colleges will need to find a way to do AA in a similar way as legacy/athlete/donor, not explicitly but with the same desired result.

@Happytimes2001 It’s funny to me that someone actually thinks people are walking into the Ivy league with scores in the 600-700 range,

Quite a lot of students are walking into the Ivy League with those scores. More than 12% at Harvard have under 700, more than 25% at Cornell. 20% at Brown. 18% at Penn. A not-insignificant percentage under 600 too.

[OHMomof2]

@milee30 Under the system that it appears Harvard uses, in rough terms they’re requiring the white and Asian nonathletes to have 1500+ SATs while the SAT bar for black students is closer to 1350. Which may make logical sense because those 1350 scores were very meaningful within that context. But the end result is that you end up with classrooms where all the white and Asian kids are at one level and it appears that the black kids are at another, less rigorous level.

It’s harder to see the white kids with the preferences that got them in with lower stats - legacy, donor. Athletes are generally known in smaller schools, perhaps not as obvious at Harvard sitting in a classroom. They’d all be on that “lower level” as well though.

@Happytimes2001 As Harvard cited in the case, the legacies are, for the most part, kids who are from highly educated parents and have had all of the opportunities one could expect. So they rarely need a leg up. They are in the pool of likely applicants.

The DOE investigation of Harvard in the 1990s found that legacies absolutely DO need that tip. Their SAT scores are on average lower than non-legacies - and this despite the privilege we presume they get being kids of H grad(s).

…the data collected by the Education Department contained some explosive information. It showed the athletes and so-called legacies who were actually accepted had lower SAT scores than the rest of the class and were also deemed less attractive candidates by the admissions officers conducting Harvard’s process.

https://www.politico.com/magazine/story/2018/10/21/harvard-admissions-affirmative-action-221669

In the Crimson back then - https://www.politico.com/f/?id=00000166-94a5-d2ba-a176-b5ad8ba30000 -

Athlete SAT scores averaged 1273, legacies 1369, non-athlete, non-legacies (everyone presumably including all the URM not in those categories) was 1405. Academic rating for the athlete group was 3 vs 2.4 for legacies and 2.19 for everyone else (lower number = better here).

[UndeservingURM]

More than 12% at Harvard have under 700,

That’s because more than 12% of the student body at Harvard is URM

I don’t have much of a problem with legacy admits. Legacy admits actually have a higher SAT scores than the average admit from the Harvard survey. A good percentage of them would have gotten in with or without legacy, because yes they lived a privileged life and in a meritocratic system, they would have the traits and accomplishments that would put them at the top. That’s the way it is. They’re not like the standard URM Ivy League admit whose only accomplishment is scoring in the top 15,000 of their year. Morally, it’s not as bad as racial discrimination.

Athletes for the main sports like soccer, football and basketball provide a ton of value to the campus. They deserve the boost. Other athletes, especially obscure rich sports, don’t.

Donors can subsidize the tuition for the poor and fund the campus and deserve some boost (assuming the money doesn’t just sit there in the college’s investment funds). Also, I support boosts for the rich and connected because their connections add value for other students. There just needs to be some limit on the number of z-listers given a boost.

[OHMomof2]

@UndeservingURM Legacy admits actually have a higher SAT scores than the average admit from the Harvard survey.

See my post right above yours. They do not.

[UndeservingURM]

@OHMomof2

Your source is from 1990.

Surveyed members of the Class of 2019 who identified as legacies reported higher best overall SAT scores—2269 on average—than their non-legacy peers, who reported SAT scores of 2221 on average.

https://features.thecrimson.com/2015/freshman-survey/makeup-narrative/

[OHMomof2]

Those are self reported on a survey filled out by 2/3 of the class @UndeservingURM - I’ll take the official information from 1990, I doubt much has changed since then but hopefully we’ll find out as the current case proceeds.

Plenty of data from other elites that legacies DO need their tip to get in, their scores are lower.

https://www.businessinsider.com/legacy-students-can-get-into-college-with-lower-sat-score-2016-10 is just one.

[Data10]

I don’t have much of a problem with legacy admits. Legacy admits actually have a higher SAT scores than the average admit from the Harvard survey. A good percentage of them would have gotten in with or without legacy, because yes they lived a privileged life and in a meritocratic system, they would have the traits and accomplishments that would put them at the top. That’s the way it is.

Legacies get a huge boost in admission. Harvard’s OIR analysis at http://samv91khoyt2i553a2t1s05i-wpengine.netdna-ssl.com/wp-content/uploads/2018/06/Doc-421-112-May-1-2013-Memorandum.pdff found the following regression coefficients for different hooks – the bigger the number, the more influential the hook. They found that legacy was more influential in admissions decisions than being Black.

Athlete: +6.33
Legacy: +2.40
Black: +2.37
Hispanic: +1.27
Low Income: +0.98
Asian: -0.37

The Plantiff’s analysis was more complete, including additional variables that separated the effects of legacy from things like Dean’s and Director’s special interest list and early action applicants, which both have a notable correlation with legacy. This reduced the measured effects of legacy to below Black, but it was still extremely influential, as listed below:

Athlete: +7.849 (0.153)
Black: +2.659 (0.104)
Legacy: +1.840 (0.082)
Hispanic: +1.419 (0.091)
Low Income: +1.083 (0.093)
Asian: -0.271 (0.071)

Looking at actual admission rates, academic rating is largely determined by stats. Applicants with similar stats tend to get similar academic ratings, such that there is an extremely high correlation between academic index and academic rat9ing. Among applicants with similar stats as indicated by a 2 academic rating, the link above mentions that legacies had ~55% admit rate – a higher admit rate than Black applicants with comparable stats (as indicated by AI) had in the plantiff’s analysis. In contrast, non-legacies had a ~14% admit rate. Non-legacies with a 2 academic rating had a similar overall admit rate to Asian applicants with a 2 academic rating.

Harvard’s expect simulated an alternative admission system without hooks and instead with a larger boost to SES disadvantaged applicants. He found the class composition would change as follows. Legacy had a larger decrease than all groups but athletes.

Athletes – Decreases by 93%
Legacy Students – Decreases by 70%
Black Students – Decreases by 51%
Hispanic Students – Decrease by 12%
Asian Students – Increases by 43%

All of this suggests that most legacies would not have been admitted without the hook, even though legacies as a whole tend to do well on the SAT, including among admitted students. The most recent freshman survey did not list legacy test scores, but it did mention that 46% of legacies reported an income of more than $500k per year. Many legacies have a variety of advantages over typical HS students that can influence score. Obviously there is also a genetic component. While SAT score may be a strength among legacies, that doesn’t mean that other more influential areas of the application are not weak compared to unhooked admitted students.

[Data10]

Non-URM with only “good” stats has like a 20% chance of getting accepted into HYPSM. An African American with only “good” stats has like a 90% chance of getting accepted into HYPSM.

Under the system that it appears Harvard uses, in rough terms they’re requiring the white and Asian nonathletes to have 1500+ SATs while the SAT bar for black students is closer to 1350. Which may make logical sense because those 1350 scores were very meaningful within that context. But the end result is that you end up with classrooms where all the white and Asian kids are at one level and it appears that the black kids are at another, less rigorous level.

I’m not sure where these numbers are coming from.

The lawsuit found that among otherwise unhooked Black applicants with the highest possible AI stat decile, the majority were rejected. Just having high stats alone is not enough The admit rate was much higher than White and Asian applicants, just no where near 90%.

In the most recent class listed in the lawsuit, White admitted students had an average SAT of ~745 per section, Hispanic admitted students had an average of ~730 per section, and Black admitted students had an average SAT of ~720 per section. That is a statistically significant difference, but nothing to suggest requiring 1500+ vs requiring 1350. There also isn’t anything approaching a requirement for students to have the listed SAT score averages. The lawsuit does not suggest Harvard is really hung up on score differences on the level of student A received a 745x2 = 1490, while student B received a 730x2 = 1460. Instead it’s one of several factors that contribute to the academic rating, which is one of many ratings that contribute to overall rating, which is considered in a holistic nature in the context of the full application, including many factors beyond the ratings to arrive at the decision. Many otherwise unhooked applicants are admitted with lower scores than listed, as is expected to occur with the admission process described above or generally in an average.

[milee30]

“Even if those preferences disproportionately benefit white students?”

@OHMomof2 - I think you are not understanding the context of my response. I personally disagree with and would change the system for all the preference examples listed - legacy, athlete, URM. My response wasn’t that only AA was a problem and the legacy + athlete preference is fine, it was in context of the discussion of the lawsuit. A lawsuit is only useful for things that are illegal and right now, the only strong mechanism outsiders might have to force change is suing over illegal practices. Unfortunately, the legacy and athlete preferences are legal choices so the discussion there would have to revolve around either working to enact change in the law or application of social pressure.

[milee30]

“It’s harder to see the white kids with the preferences that got them in with lower stats - legacy, donor. Athletes are generally known in smaller schools, perhaps not as obvious at Harvard sitting in a classroom. They’d all be on that “lower level” as well though.”

Yes, its unfair that the unqualified athletes or legacies find it easier to hide and if it were up to me those preferences would go away as well. Unfortunately doesn’t change the optics or the point, though. If a college is creating a system where there is one group of students that has an easily identifiable physical difference (could be race, could be gender, could be physical disability, just as examples) where the general group of that identifiable population has lower academic qualifications than the bulk of the other group, it sets that small different group up for failure and perpetuates existing negative stereotypes.

Using gender as an example, if a college requires male applicants to have math ability indicated by a 750+ on math yet only requires female applicants to have math ability to indicated by a 700+ on math, it sets the women up for failure. The women are easily identifiable. And although there will be some women who have 750+ ability and perform at the same level as the men, the optics will be that in most classes, the people at the bottom of the math/tech class will be women. Creating the impression that “women” are not as strong at math and are only there because of the AA/two track system. I think the harm from two track systems is as great or possibly greater than the benefit. Yes, it gives u

[ChangeTheGame]

@milee30 i think there is a problem with your premise of using gender. Part of the issue deals with the fact that under every metric I have ever seen, women do better in the classroom (GPAs) than men. Kids in the same classrooms all across America have young men with higher Math standardized test scores, but young women with better grades in those same classes. But standardized test scores will always dominate the discussion about who is worthy. In my own household my son is easily the better standardized test taker but he can not hold a candle in his sister in the classroom including their math classes. We obsess with one factor (standardized tests) when it tells nothing about one’s work ethic or time management skills but we discount GPA (I understand that high schools have different levels of rigor). If all students were taking a standardized test 1 time, I could see it being more effective in showing ones ability. But when a student starts at a math score of 600 and jumps up to 750 over 5+ testing cycles, does that mean that they have more innate math ability than a student who took the test once and got a math score of 700?