@calmom You could be right. The few test optional schools that I have ever looked at for ethnicity data seemed to have numbers slightly equal to or lower than Harvard’s current percentage, but their is nothing scientific about that. What would be most helpful is schools before test optional and the 1st year or 2 after going test option but it is still not a perfect comparison. U of Chicago’s data is going to be very interesting to me, but they are also still using racial preferences along with being test optional. In a landscape with no racial preferences, it is still seems like it is unknown which way the numbers will go with test optional policies.
.
The lawsuit is more complex than just test score differences. The 4 key claims the Plantiff makes are below. Harvard disputes all 4 of these claims. Plenty of documents from both sides are available for those who wish to review and consider the merits of their claims.
A – Harvard intentionally discriminates against Asians
B – Harvard engages in racial balancing
C – Harvard is not using race to achieve critical mass
D – Harvard did not seriously consider or use workable race neutral alternatives
When discussing claim A, the differences in admit rate by academic index were mentioned; but the primary focus more related to things like regression analysis with a large number of controls including scores, differences in personal qualities ratings, and Harvard’s past history.
Going test optional alone would not resolve these claims. The race neutral alternatives the Plantiff proposes more focus on giving a very strong preference to low SES applicants and not considering non-athletic hooks. The Plantiff’s expert claims the admissions percentages would changes as follows under their proposed changes (very strong low SES preference, no non-athletic hooks, no SCEA preference, choose equal number of admits from each College Board neighborhood cluster region). Harvard’s representative claims this approach wouldn’t be practical for a variety of reasons.
White: 40% → 32%
Asian: 24% → 31%
Hispanic: 14% → 20%
Black; 14% → 10%
Income Disadvantaged: 18% → 52%
Legacy: 15% → 4%
Avg SAT Score: 2244 → 2173
“I believe I concluded from the litigation data that the SAT score gap between black and white students in the admitted pool was less than 60 points. Less than 80 points separated the black and Asian students.”
The admitted pool is going to be a lot less than the applied pool, as you suggested the admits are the best applicants from each race. If Asians didn’t think they’d be discriminated against, a lot more with lower test scores would apply. Right now, generally speaking, Asians don’t apply unless they have a 34/1500, that’s the general guidance given to them. .
“Going test optional alone would not resolve these claims.”
Agree, if they’re found guilty and the Asian percentage isn’t up, Blum and whoever is bankrolling him will be back in court.
This dovetails with Greg Duncan’s findings at Northwestern:
“Mastery of early math skills predicts not only future math achievement, it also predicts future reading achievement,” Duncan said. “And it does so just as reliably as early literacy mastery of vocabulary, letters and phonetics predicts later reading success.” The opposite – reading skills predicting math success – does not hold up.
This is also consistent with my experience here in Ontario. Since we do not do SAT/ACT, I used to wonder how they fill highly desired programs…then a friend told me her prof let slip that senior high school calculus was the “invisible sieve”.
I think math simply taps into a higher order thinking skill.
There aren’t enough professional jobs for every person. 20% of jobs being for the educated/upper middle class seems about right actually.
No, I’m surprised that you have this misunderstanding as a regular poster in this thread. It’s that given the same performance/score on objective criteria (academics/extracurricular), Asians get worse scores on subjective criteria (personal/overall) and have lower rates of success.
Unfortunately the no punishment trend has harmed black and Hispanic students the most because the rate of misbehavior is higher within their communities (due to a variety of factors) and their communities are segregated so misbehaving black and hispanic students are usually disrupting the education of other black and Hispanic students.
I believe that this is the main driver for integration. Integration is a band-aid for broken discipline policies. Hispanic and black students benefit from being around non-Hispanic/black black students because non-Hispanic/black students have lower rates of misbehavior and disruption (while the non-Hispanic/black students experience negative effects from integration due to higher rates of misbehavior).
This can easily be fixed through tracking. Standardized test scores are good, cheap and easy-to-administer proxies for measuring rate of misbehavior. It can be a fairly equal opportunity system: make the threshold just high enough to filter out the delinquents (but is still a relatively low score) and then randomly choose from the students that meet this threshold.
The drawback is that there will be children left behind. But right now, there are full classes of students left behind because of the 1 or 2 disruptive students in the class that divert the teacher’s and students’ attention away from teaching and learning. Better a few students (the disruptive students and a few unfortunate false positives) than full classes of students.
Speaking of NYC K-12 schools, it looks like the mask is coming off, and the elites are simply waiving the white flag and admitting they will never be able to up the achievement levels. They want to move fully to diversity targets and deemphasize any actual results: virtue signaling at its worst.
Here is the even more troubling aspect:
There is plenty of money being thrown around everywhere in NYC schools - all schools, every school. The “resources” that are now going to be “redistributed” are children. If implemented, this should wound most of the few schools that are still excellent, and kill off the merely decent. (Presumably, because of the Hecht-Calandra statute, the superb specialized high schools should be fine, for now).
The whole system is only 15% white; to think of these kids as “resources” - which, make no mistake, is how they are being looked at - recalls Hoffer’s famous observation: “A ruling intelligentsia, whether in Europe, Asia or Africa, treats the masses as raw material to be experimented on, processed, and wasted at will.”
NYT coverage (quoted) here: https://www.nytimes.com/2019/02/12/nyregion/ny-public-schools-diversity.html
City Journal’s take here: https://www.city-journal.org/de-blasio-school-integration-task-force
Strictly speaking, jobs in computing do not necessarily need a bachelor’s degree if the job seeker has a proven record that shows that s/he has successfully self-educated the needed CS. However, most people can learn CS more effectively in the structured environment of a college bachelor’s degree program in CS. The self-educated people are uncommon, but not so rare that it is unheard of, and there are not occupational licensing barriers in most cases for computing like there are in some other jobs.
But also, CS is a relatively new and rapidly developing field. The volume of knowledge is greater, and frontier of knowledge in CS is further out now than it was in the Gates/Jobs era, so that many would find the structure of a bachelor’s degree program in CS to be desirable to learn the currently expected material.
Yes, there is credential creep, most egregiously in some jobs where the volume of job seekers is high enough that employers may require bachelor’s degrees (in anything) even though the job itself does not require either general thinking skills or major-related skills and knowledge that a bachelor’s degree is supposed indicate that the applicant has.
However, this does not mean that all instances of increasing credentials are useless credential creep, since some other fields now expect a greater volume of knowledge and skill than before. For example, health care has gotten more complex over the years, so that increasing percentages of those wanting to go into nursing seek bachelor’s degrees in nursing versus the associate’s degrees in nursing that used to be predominant.
My go to spot to see admissions data when it comes to African American students at leading schools is The Journal of Blacks in Higher Education (JBHE). Some schools are more mysterious with their data than others but you can get a pulse on how AA is used when compared to overall admissions rates (if the schools shared all of the requested data). One thing that I have noticed over the last 5 years is a steady increase in black students attending top schools so I am also sharing the data from 5 years ago (Incoming class of 2018 vs. 2013). Among the Liberal Arts colleges most of the largest gaps are seen out west with Harvey Mudd almost doubling the acceptance rate for African Americans (27.9%) versus the total class of 2018 data (14.5%). Carnegie Mellon and Rice University have the largest gaps (over 6%) for schools who reported all of the data requested between African Americans and total class of 2018 data.
On the middle class white family comment: de Blasio has been able to be so provocative on the SHSAT schools because the white people that attend the SHSAT schools are usually from lower or middle class immigrant families like the Asians at the schools. Upper class white students attend private schools and elite screened schools. If you look at the student list at the SHSAT schools, many of the white students have Polish, Russian or other clearly non-anglo European last names.
Here they are spending thousands of work hours (and potentially millions of dollars) trying to move students between schools instead of actually improving the schools.
Integration doesn’t even improve schools. Competitive schools with high achieving students aren’t better for the low achieving Hispanic/African American than regular schools. They are actually worse in some regards: the competitive schools expect their students to have self-motivation and to be at a certain academic level. There is a mismatch between the academic level that the school teaches at and the academic level of the low achieving diversity students. Students need to be taught at the level they are currently at.
Note that the acceptance rate gaps are much larger (in favor of African Americans) than they appear. The acceptance rates haven’t been normalized by the applicant quality. The average African American applicant is usually lower quality than the average overall applicant. In a race-neutral, equal opportunity scenario, African Americans applicants would be expected to have lower acceptance rates than overall. I posted a while back about how 50% (IIRC) of African American Harvard applicants are automatically disqualified based solely on low academic ratings.
@UndeservingURM Yep, I definitely get that those higher percentages of accepted African American students are despite having lower statistical numbers. Seeing UNC Chapel Hill on the list with a higher percentage of African Americans accepted versus the entire class when we also saw how much lower the GPA/standardized test scores were for African American students from the current lawsuit was an eye opener for me. UNC-CH has a larger statistical spread than Harvard does so it will be interesting to see how that case plays out.
There are over 5,000 African American students attending those 27 elite research universities listed for this year’s class. If the data on African American ACT scores stayed static (I have 2013 data for African American ACT scores), only 3750 black students scored a 28 or higher that year. ACT scores at the top end of the scale have really grown the last 5 years, so those numbers could have changed some, but the overall average ACT composite for Black students has basically stayed static at 16.9. Even if one said that there are 3750 equivalent SAT scores (unlikely since some of the students would have taken both tests and can not be counted twice), that would not cover the top 50 schools (It may get close with lower scoring athletes at the big D1 schools). HBCUs are also taking a percentage of those students (Howard University had a ACT/SAT 75th percentile that correlates to a 28 so that could be upwards of 300+ Black students taken from that pool of high scorers alone from 1 HBCU). Some Black top standardized test takers will also choose lower ranked PWIs and top ranked LACs as well. There is no question that AA is playing a large role in elite admissions for African American students in comparison to the statistics alone (essays and ECs are harder to compare) and the courts will have to decide what is too much of an admissions boost and if students of other races are being harmed from such practices.
The main point is that “[racial] preferences are not just discriminatory, they undermine their stated goals”. The example is given of URM “students that enroll in preference practicing schools who intend to major in STEM switch into softer majors at a high rate once they realize their fellow students are much better prepared to do the work. Had those students enrolled in schools that matched their level of preparation, they would be more likely to graduate with a STEM degree.”
I think that this matches what ChangeTheGame has stated about HBCs and why they do such a great job at producing scientists, doctors, lawyers, etc. Maybe money would be better spent not at recruiting URMs to HYPSM, etc, but giving the money to HBCs since they have a great track record, and maybe some of this money could be used to entice more whites/asians with or without limited means to consider going to these institutions.
The idea of matching the level of preparation is a very important one, not just for URMs but for everyone, and at all levels of schools. In fact one of the negative aspects of an MIT, caltech, or harvard is that it can take a very bright student who is very interested in math or physics and turn them away from majoring in the subject. If many of your peers have spent thousands of hours in the subject before entering college (e.g. average number of hours for math olympiad contestants) but you haven’t, you won’t catch up for several years (if ever). I’m not really a Malcolm Gladwell fan, but his idea in David and Goliath, that you are much more likely to succeed in life if you graduate in the top percentage of your class, from a second or third tier school, rather than in the bottom half of your class, from a tippy top, matches what I’ve seen.
I guess my post before #3952 wasn’t ok. It was removed since I copied too much of the article. But here it is with just the URL reference only:
I think a great point is made by Heather MacDonald in a WSJ opinion article entitled Diversity delusions at North Carolina: https://www.wsj.com/articles/diversity-delusions-at-north-carolina-11549829141?mod=article_inline
I read that article, and my thought was it would be great if they instituted the similar diversity policy in NBA/MLB/NFL.
@havesomeheart Welcome to the thread and thank you for the WSJ post. The author definitely mentioned some of my own concerns with using AA in admission policies. One of the things that I have seen a lot with AA is the “soft majors” conversation with some of those majors (like African American Studies) getting “killed” in any media format that it is mentioned in. I wish I had an understanding (data) of how going to elite institutions affected the majors of URM students. It would be telling if a much larger percentage of students at a HBCU majored in a STEM discipline versus African Americans that attend a top 25 institution. Current AA admissions policies at top schools will always have these questions and quantifying any benefits (besides higher graduation rates from students who would have high graduation rates anyway) is much harder than quantifying the negatives.
One of the older posts by @ChangeTheGame got me thinking. He mentioned that his friend suspected that Harvard would go “full holistic” if she were to lose the case. How would/should employers “counter” this response?
The employers can ask for SAT/ACT scores of course. My back-of-the-envelope approach is to focus not so much on the college but on the major. An elite can backdoor a student into the school, but unless the person is really academically strong, he will not come out with a degree in, let’s say, theoretical physics.
I find you can even differentiate students within the same major by their sub-specialty. Economic history is significantly easier than econometrics, for one. My children did their degrees in business/commerce. At the time, the program accepted 6.5 % of applicants. It still actively discourages weaker applicants from applying and the only hook is an attempt to maintain a 50/50 male to female split. Since the school only allow A grades be given to 25% of the class in the first 2 years, the students very quickly separate themselves, with the strongest majoring in finance, then accounting, then marketing, then the rest. Mine were interested in banking, but they had to reconsider when the grades came out.
I always find the obsessive focus on the schools here on CC puzzling. If admission is holistic, then it makes more sense to focus on the major, no?
The thread on the therapist was closed, because it veered into questions of race: http://talk.qa.collegeconfidential.com/discussion/comment/22008404/#Comment_22008404
These quotes in that thread caught my eye:
Fair points both. Fortunately, much useful data can be extracted from the reports in the Harvard litigation, representing six years of admissions cycles. Especially with respect to admit rates by academic stats.
Harvard breaks its entire applicant pool into deciles by a formula that includes GPA and test scores. These deciles are formed based on the entire applicant pool, so an applicant’s placing in a particular decile is an objective read of her stats against the entire pool, not simply a relative ranking based on her particular demographic. Note that data for recruited athletes are specifically excluded.
Here are the admit rates by race. Unhooked is the first figure. The corresponding rate for the preference applicants (legacy, development and faculty kids) is in parentheses:
Top 30% of academic stats:
White – 10.9% (53.4%)
Black – 49.8% (81.8%)
Hisp. — 25.8% (60.5%)
Asian – 8.8% (57.5%)
Top 20% of academic stats:
White – 12.9% (57.0%)
Black – 55.2% (85.7%)
Hisp. — 28.3% (67.4%)
Asian – 10.2% (59.5%)
Top 10% of academic stats:
White – 15.3% (57.1%)
Black – 56.1% (83.3%) (small n - 5/6 accepted over 6 years)
Hisp. — 31.3% (95.2%)
Asian – 12.7% (62.5%)
Just for fun, here are the admit rates for the bottom of the applicant pool, again with preference in parentheses:
Bottom 30% of academic stats:
White – 0.4% (13.5%)
Black – 1.3% (9.5%)
Hisp. — 0.7% (8.3%)
Asian – 0.3% (9.0%)
The legacy and development admit rates are eye-opening.
@SatchelSF - interesting data. I had estimated that about 50% of Ivy students are white, while 10% of all Ivy students are white recruited athletes. Your totals for white students at the Top 30% (10.9), Top 20% (12.9). and Top 10% (15.3) deciles comes to 39.1%, so it seems like the 40% of non-recruited whites is a good estimate.
Approximately 15% of all Ivy students are Jewish, so 37.5% of the non-recruited athlete white population is Jewish. If we distribute the Jewish students evenly across the deciles and remove 37.5% from your white student numbers, this is what they look like for non-recruited athletes.
Top 30% of academic stats:
Non-Jewish White – 6.82%
Black – 49.8%
Hisp. — 25.8%
Asian – 8.8%
Top 20% of academic stats:
Non-Jewish White – 8.07%
Black – 55.2%
Hisp. — 28.3%
Asian – 10.2%
Top 10% of academic stats:
Non-Jewish White – 9.57%
Black – 56.1%
Hisp. — 31.3%
Asian – 12.7%
As concerns the mother from the therapist article, if her son is a Non-Jewish white student and non-recruited athlete, he certainly will be discriminated against in the Ivies.