A new (and larger) Chetty study on elite college admissions is released today

The article was from the school newspaper, not the university itself, and much of the information in the article comes from the SFFA filings. In other words, the plaintiffs. That Harvard’s student newspaper mischaracterizes the rating as a “personality” rating is a mistake on their part.

And IMO the difference between a personal rating and personality rating is not mere semantics, especially when @fiftyfifty is always careful to put the word “personality” in quotes. While I don’t doubt @fiftyfifty was legitimately confused about the actual standard, IMO the insistence on mischaracterizing it on the part of others is in itself very telling.

From the Harvard Crimson:

If an alumni’s ratings don’t match an AO’s ratings, isn’t it possible that the the essay and the reports of principal/GC and teachers are partially responsible for the mismatch? I have no idea if that is true, I am just asking the question based on what @mtmind said and my own experience reading letters of rec and seeing how different they can be even for students who look the same on paper. In any case, I know that this thread is not the place to talk about that Supreme Court case, but as I mentioned many posts ago, I think it is really problematic when a teacher writes a negative letter of recommendation. If a teacher can’t write an enthusiastically positive rec, I think the fair thing to do is to tell the student so that the kid can ask a different teacher. Writing a lackluster rec for a student to whom you gave a strong grade is unfair and obnoxious in my mind.

1 Like

They say the quote is from Harvard’s filing, not the plaintiff. Yes, I realize the Crimson is the student newspaper. Are you implying that makes it less credible?

1 Like

FWIW, your quote of me aggregated a statement I made and the second half of a quote from the article on a related but different point.

Not any less credible than any of the other press/editorial sources cited, and perhaps more credible than some. But not a statement from the University.

I am so sorry! I was trying to edit out the reference to race since I know it is not allowed except in one particular thread in the politics forum. I was definitely not trying to change the meaning of your post by selective quoting. It sounds like I may have misunderstood the issue that you were raising, and if so, I’m happy to delete my post. LMK.

1 Like

Our kids’ HS is not offering the PSAT this year and as far as I know, it is not possible to take it at another local HS. Instead, our kids’ HS advised that if students want to try for National Merit they should plan for the alternate entry path by taking the SAT. My guess is that many students at our HS won’t end up taking any of these tests, since it takes more effort to sign up for a SAT sitting, and we are in CA and the UCs and CSUs don’t use the tests anyway.

“Personality” is the word preferred by critics because of its connotation of lightweightedness, superficialness, smacking of mere presentableness and good form, the opposite and enemy of real measurable accomplishment. The reason it’s so often put in scare quotes is to emphasize its dubiousness and even illegitimacy.

Humpty Dumpty told Alice that a word means what I say it means. He could also have said that the choice of a word with a loaded connotation can shape a meaning.

1 Like

Wouldn’t a legal filing (to the Supreme Court no less) by the university be at least as credible as a statement by the university? A statement may or may not be spin. A legal filing is subject to perjury. The paper says they are citing the university’s filing.

I agree, a teacher should not write an intentionally lackluster letter, it is better to decline to write a rec at all.

But a different problem is when teachers and GCs write an unintentionally lackluster letter. There are lots of studies that show that letter writers often unintentionally reinforce ethnic and gender stereotypes, for example praising some groups for hard work or reliability rather than accomplishment or creativity. Or praising members of other groups for being articulate or well mannered rather than intelligent. It’s a genuine problem. Add to this the fact that certain positions (e.g. captain of a male team sport) are seen as reliable signs of leadership while other positions (e.g. captain of math team) are often not. You can find recent comments on CC saying as much.

So this is just one example of why I worry about these type of subjective ratings. People think they understand a student’s true personal characteristics (or personality if you will), but maybe their thoughts are more an expression of biased ideas.

4 Likes

In fact that is one of the things the trial court observed, that the data indicated there were small but potentially significant differences in those areas.

If anything in the college application process should be made optional, it should be those LoRs. If a teacher doesn’t feel positively strongly about a student, s/he shouldn’t write the LoR or be required to write the LoR. Not only LoRs can be very subjective (especially if the teacher had little or infrequent interaction with the student), but they can also unconsciously convey personal biases.

I’m not sure there was any evidence of this happening in Harvard’s process. Certainly their instructions to reviewers suggested both athletic and academic team leadership should be considered.

1 Like

I almost did a spit-take at the notion that legal filings (at any U.S. court) lack spin, but that’s a different conversation.

As to the topic at hand, Harvard assigns a “personal rating” not a “personality” rating, and the statement you quote from Harvard’s filing doesn’t contradict this. Here is a more complete version of Harvard’s description from the lawsuit, with my bolds:

  1. The personal rating summarizes the applicant’s personal qualities based on all aspects of the application, including essays, letters of recommendation, the alumni interview report, personal and family hardship, and any other relevant information in the application. Ex. 1 at 164:11-165:2 (McGrath 2015 Dep.); Ex. 26 at 245:18-246:20 (Fitzsimmons Dep.).

  2. Many characteristics are valued in the assignment of the personal rating, and admissions officers may assign ratings based on their assessment of the applicant’s humor, sensitivity, grit, leadership, integrity, helpfulness, courage, kindness and many other qualities.

Why is it so important to you and others to insist that, contrary to Harvard’s statement, that Harvard assigns a “personality rating” rather than a “personal rating?”

Yes, I absolutely agree with all of the above, and I was going to write almost exactly what you said about reinforcing ethnic and gender stereotypes (using similar examples). I just wasn’t sure if that would go against the forum rules, which I admit that I don’t fully understand. I get the general rule, but I never know where the line is drawn in these discussions so I try to avoid discussions that allude to race and ethnicity --And believe me, it is very hard to avoid since I think about the impact of race on people’s perceptions a lot and it is a big area of concern when I worry about my children’s experience of their educational institutions in K-12, college, and beyond.

1 Like

I agree whenever individual human impressions are involved, there is going to be a possibility of those impressions reflecting bias.

But it is probably worth remembering that these entities are running schools, and their schools’ stakeholders include the faculty, and the faculty has a stake in wanting the sorts of students who make classes more enjoyable and rewarding to teach, more collaborative, and so on.

Recommendations from former teachers, sort of like recommendations from former employers, are a rather obvious source of information like that. And similarly former employer recommendations are also potentially subject to bias, but prospective employers often want them anyway, including because they think there is information uniquely available in that way. And I am not sure what the substitute would be in this case that could sufficiently satisfy stakeholders like the faculty.

While I agree that there are subjective elements to LoRs. there are subjective elements to all aspects of admissions. A larger issue in the admissions context (but one that gets much less attention) is that the LoRs are also relative to the rest of a particular school’s student population. So if a student attends a school with a high percentage of highly qualified, excellent students who are personally impressive, it will be very difficult for that student to shine relative to their peers. A GC/principal/teacher cannot honestly write that 60 students are the most impressive student, much less that they are the most impressive to attend the school in 10 years. This necessarily impacts some demographics more than others, but is consistent with the schools’ desire to view students in their context.

2 Likes

This is really sad. It denies students an opportunity to earn merit scholarships, and frankly is unethical, IMO of course. Because the PSAT is digital this year, I agree that not many other HSs are likely to let outsiders in to take the test.

3 Likes

Yes, I wasn’t arguing that colleges should eliminate letters of recommendation. I was saying that a teacher’s bias can come out in those letters and there is a possibility that a teacher can unknowingly convey stereotypes. Thus, I am curious if that is why there is sometimes a mismatch between alumni impressions and letters of rec.

But having worked in both K-12 schools and universities, I am very aware that some academically strong students can contribute to a successful classroom and some academically strong students can make the teaching experience miserable as well as disrupt the experience of other students. So I do think a student’s approach to learning and their contributions to a class dynamic matter a lot. To be honest, I wonder if grade inflation in high schools is sometimes driven by the desire of the teacher to not penalize a disruptive or problematic student. It is easier to just give the kid the higher grade than actually talk to the student (or her parents) about the problematic behavior so that the kid can address it. I suspect that in those cases, there are teachers who use their letters of rec to rank students applying to the same set of schools. But that approach is a disservice to the kid because it is not helping them learn classroom citizenship skills and it can even be misleading them into thinking the teacher will write a glowing letter when the letter is more likely to be lackluster.

Actually, I never suggested that Harvard or the Crimson called it a “personality” rating. You may be confusing me with another poster. And as far as I can tell, the Crimson (and I in turn of the Crimson) quoted Harvard correctly. Here’s exactly what I wrote:

The last link @fiftyfifty1 provided was from the Harvard paper. And it cited Harvard’s own legal filing which stated: “Harvard’s filing states that admissions officers review candidates’ “humor, sensitivity, grit, leadership, integrity, helpfulness, courage, kindness and many other qualities” when determining the personal rating.”

As you can see, my post correctly quoted it as a “personal rating” not personality.

I merely said, and still contend, that a reasonable person (not closely connected to this debate) might call rating a person on subjective things like “humor” “integrity” “grit” “kindness” etc. as “personality” and that defending that as something completely different than “personal” felt like semantics.

I don’t care what you call it. I am agnostic on either word. It seems like the salient point is that they provide a subject rating of the things in that list, whatever you want to call it. It appears that some people are triggered by the word personality and assign it all kinds of negative connotations, and thus need to make the distinction. Why is the discussion about the word choice rather than the substance of the rating?

2 Likes