Stanford, Harvard, Dartmouth, Yale, Penn, Brown, CalTech, JHU, and UT-Austin to Require Standardized Testing for Admissions

The issue is how one defines performing ‘poorly’. What verbal score equates to ‘poor’ performance? 2025 national mean SAT verbal score was 521 (out of 800).

Does scoring at the mean = performing ‘poorly’? I hope not, but I expect some here think that. Note that I am not saying a student with a 521 verbal score should be admitted to a highly rejective school. CB says college ready verbal SAT is 480, not sure what higher ed administrators say to that/if any use CB’s numbers or if they have their own.

Similarly the mean math score in 2025 was 508 (out of 800.) CB says college ready math score is 530 (ditto above comment.)

English language arts (particularly writing) at (native speaker) high school level may be less suited for evaluation by an hour or few exam than other subjects like math. English standardized testing may also be more game-able than math, as evidenced by vocabulary memorization used against older SAT formats, or the reverse engineered SAT writing essay formula that some test prep companies figured out (probably why the writing section is not used much by colleges).

Essay format exams used by colleges for college English placement may be the least bad of various methods, since those specific to a college are unlikely to be targeted for test prep or gaming the system. Actual college English composition courses may choose not to have final exams at all.

1 Like

Re: https://opportunityinsights.org/wp-content/uploads/2024/01/SAT_ACT_on_Grades.pdf

That seems to be different from what The College Board found, which is that HS GPA had a stronger correlation to college GPA than SAT did (but both together had greater correlation than either). They did note that HS GPA was weaker at more selective colleges, which is likely due to compression of HS GPA at the top for students at those colleges. Since the page that you linked refers to highly selective colleges, top compression of HS GPA applies there as well.

That may be why both the vocabulary section and the essay have been discontinued. Scoring high on SAT verbal does not mean that one will be a great writer or a lawyer. It will however provide some evidence that the student can handle the written material associated with college-level classes at a selective/rigorous school.

1 Like

Probably the vocabulary heavy versions of the SAT were that way because vocabulary was seen as a proxy of how well read one was. But test prep in the form of memorizing lists of SAT words made this proxy measure less useful in measuring what it was intended to correlate to.

For the record, I absolutely agree with you and since I have actually taught Humanities classes that suppose close reading, lots of writing, and discussion to classes with scores in the 450-low 500s and … other,high scoring groups I wholeheartedly agree, but I also think there are other ways to evaluate English readiness. That score is a very fallible proxy.
I wasn’t talking about my opinion though. I was specifically referring to what PSU found. They actually had a 4year study where they tracked test submitters and non test submitters and the only majors where that made a difference was majors that require calculus. The study was disseminated, discussed in Faculty senate, a white paper was issued, recommendations were provided, etc. So I wasn’t giving my vague opinion about tests, I was specifically referring to the fact PSU (a very large flagship) found that external correlation based on tests only matters for math. And they refined that to “unless the students has taken Calculus in HS” (data seem to show that even if Algebra2/precalculus foundations may not be perfect, students who took calculus in HS have enough of a math background to manage a major that require calculus even if they have to take a refresher course.)

NB: for the record, despite being “PSU champion”, I am not affiliated with PSU and I don’t even especially think they are the greatest possible college for future freshmen or anything like that. The former forum champion had to quit and I happened to know stuff about them so I stepped up because LOTS of people are interested and want/need answers, many of which it turns out I can provide.

^ this I think is why there’s a difference between English assessment and Math assessment. Retakes and resubmissions are GOOD in English. You want to read and reread, write and rewrite your paper, because that’s how you hone the skill and improve your critical thinking.

In fact, interestingly, that’s how PSU handles placement in its many versions of Freshman English: did you write frequently, complete multiple 4-5pp papers that you had to edit&rewrite? .. all the way down to “mostly in-class, short answer tests”. The adviser discusses this with the student to determine placement. And weirdly kids don’t pretend they feel comfortable writing and editing 5 page papers when they can’t whereas there’s a problem estimating Math preparation, so they had to take the ALEKS assessment test for correct placement…
The English SAT score also doesn’t matter that much because there are many versions of Freshman English and the process doesn’t just encompass that class.
There are the Schreyer English seminars (those kids had to write 3 long essays and 5 short ones and they were selected on this -300 out of 9,000 freshmen); 30H, for students who took IB English, AP English Lang and/or Lit, or wrote multiple research papers in HS. 5 versions of the English 15 class. ESL15 for students whose native or home language isn’t English (multiple versions). “Studio” version which is 15 with one on one tutorials. 4&5 which are for students with specific needs. Then all students have a communication class and all juniors have a “Writing in (field)” class, too.

Try 5 paragraphs
(Not kidding here.)
Reading/writing is becoming a more uncommon skill - not sure how to explain it. SAT English can’t detect it and I have no idea how you can test it.

3 Likes

Why would this be good for English, but bad for Math? One could make the same exact arguments - i.e. how one can improve critical thinking and problem solving skills related to math through repetition. It would be helpful to see the conclusions of the PSU study that says that SAT verbal scores are “useless” if there is a link somewhere.

That part of the SAT has not been called “verbal” since 2004 or so (when it was more vocabulary heavy).

1 Like

Do you mean 5 sentences?

But SAT RW questions seem to be based on reading or completing passages that are mostly 5 or fewer sentences.

They don’t know how to teach high school kids, and many are actually pretty poor at teaching undergraduates. They are also don’t understand standardized testing, nor do they actually understand WHY students are getting worse.

They are are observing a drop in the ability of students to answer the test questions that they are providing on their tests. That is the absolutely only thing that they have. They are not qualified to figure out the reasons for this, because they lack the knowledge.

The hubris that they are demonstrating is actually breathtaking. Imagine historians who have seen an increase in temperatures that are reported in newspapers since the early 20th century. Would we actually take them seriously if they would claim that they were the people who were best qualified to understand the reasons behind this change in temperatures, as well as being the people who are best qualified to develop ways to mitigate this climate change?

Yet, because the students are doing badly in their classes, STEM professors believe that makes them the people best qualified to figure out the processes behind the drop in math understanding, as well as being the people who are the best qualified to find a solution.

No, they are ONLY best qualified to figure out how well the students are prepared for these classes.

I will repeat this once again:

Most PhD programs in STEM do not put much effort (or any effort at all) in training their faculty to teach. In Engineering and CS it s worse than the rest. Historically, students in many PhD programs in engineering, biomedical sciences, etc, were actually discouraged from working as teaching assistants and the faculty advisors were encouraged to provide funding so that their students would only have to work as research assistants.

Things have become better, but this is how the vast majority of the present-day faculty in STEM at the UCs were trained. Faculty were hired primarily based on their research potential, and they were tenured primarily based on the research productivity and grand money. These are faculty at a set of R1 universities, not at colleges which engage primarily in teaching undergraduates. Most teach one undergrad class a year, or maybe a semester.

These are not a group of people with deep knowledge and understanding of pedagogy, nor are most of them really interesting in developing their teaching abilities.

So no, I do not trust that they know the reasons behind the drop in math abilities, nor do I believe that they have the best solution for the drop in math abilities.

To repeat: the only thing for which they are qualified is to identify and report the drop in math abilities.

So their claims that the reason behind this is the lack of SAT scores and that the solution for this is reinstating the SAT scores has little validity, unless it is supported by studies by people who understand pedagogy and sociology.

As an additional comment: if the drop in math abilities was the result of the dropping of SAT scores, there should have been a sudden drop in abilities as soon as the requirement was dropped, but very little movement after that. Yet there seems to have been a constant increase in the number of students who are not prepared.

I would also love to see their data. All I see is a claim that is being made by these faculty, without actual data to back them up. They’re STEM faculty, so they should understand the importance of providing fact-based evidence to support their claims.

In any case, if they believe that students aren’t prepared, the UC should ask people with the appropriate expertise to figure out why and to figure out the best way to address this. As it is, they are retreating to the old thought process of the wealthy and privileged, who believe that college education is only their right, and that the poor should only be allowed on sufferance.

A very good solution is summer courses for incoming students who are not prepared. Yes, you can actually bring students up to required expertise in pre-calc and geometry in a three month summer course.

However, the faculty who signed this petition seem to be more interested in gatekeeping than in anything else. That is why, I think, they demand the simplest gatekeeping solution. The status quo that existed before benefitted the wealthy, because the SAT benefit the wealthy. But, rather than find a solution to their problem that maybe doesn’t benefit the wealthy, they engage in mental gymnastics trying to “prove” that “No, SAT scores actually benefit the poor”.

I’m sorry, but that is the laziness of people who benefit from a situation in which college gatekeeping benefits wealthier students for admissions.

1 Like

Your argument is that the academic chasm between a historian & a climatologist is equal to that of a college-level teacher of a mathematics-focused subject to a high school-level teacher of a mathematics focused subject.

That seems so wildly ridiculous that I feel compelled to ask - were you joking and I just didn’t get the joke?

7 Likes

The concern addressed by the UC professors is not so much the lack of preparation of students—it is the lack of screening to exclude unprepared students. The concern is based on their classroom experience. The SAT is a screen that they want brought back

It is a simple point which your personal attack on them as a group of people favoring the wealthy doesn’t rebut

9 Likes

Exclude those unprepared for what? To go to college? To be a math major? Please be specific.

Are you talking about a minimum math score? What should the minimum score for math be?The national average? State avg? Something higher? Same min for a stem major vs English major?

Should the public colleges in California, including the UCs, bear no responsibility to provide a college education even for students who haven’t taken many math classes for whatever reason and/or had weak HS math teachers and/or didn’t have access to math beyond algebra 2?

In California in 2025, 26% of the HS class of 2025 took an SAT test, and 29% of the class of 2026 took an SAT test. About 4% of the class of 2025 took an ACT test.

Even looking across a two year time frame, test takers are likely a minority of both 2025 and 2026 classes. Do none of the majority without scores get to go to a public four year college in their home state? Or are just the UCs off the table? Should the UCs drop the ELC top 9% rule?

The UC leaders can’t make the test decision in a vacuum without addressing the issues I’ve highlighted…from middle school-HS math curriculum, teacher staffing, SAT/ACT test access, etc.

K-12 education in California is not where anyone wants it to be. First step is to decide if the UCs should be part of the solution or if they will turn their backs on students who are receiving inadequate K—12 educations.

The “chasm” between historical and climatological research exists solely in your mind.

Climatological research actually includes history, because it is the study of weather patterns over long periods of time, including the past. So not unly is there no “chasm” between these, any attempt to understand climate changes over the past few hundred years REQUIRES historical research.

I worked as an ecologist for 30 years, and almost of of my research skills are applicable to genealogical research, and traditional genealogical research is, in fact, historical research. As an ecologist I have used climatological research, and as a genealogist, I have engaged in historical research, so yes, I have a decent understanding of both of these, and I can repeat that not only is there no “chasm” between the two, there isn’t even a groove between the two.

What’s wildly ridiculous is thinking that doing research in physics or in mathematics magically bestows upon a person the ability to teach, as well as the magical ability to understand complex processes without doing any research.

I will repeat for the umpteenth time:

  • A PhD is a research degree, not a teaching degree.

  • A PhD in one topic does not bestow expertise in another topic, much less field.

  • Simply doing a PhD will not teach you how to teach college students, you need to learn how to teach independently of your research.

  • A PhD in Physics, Biology, Math, CS, etc will not teach you how to understand why your students are coming to your class unprepared. It won’t even teach you HOW to do the research that will provide you understanding of why they’re coming to the class unprepared.

Do these statements sound ridiculous or like a joke?

As I wrote - gatekeeping. That is EXACTLY what “screening to exclude” means.

They probably don’t think that they favor “the wealthy”. They favor kids who attended high schools which are very well resourced, kids who engaged in costly ECs that prepared them to talk about the topic taught in class, they prefer kids who don’t have to work after class and can do their work immediately, etc.

The overlap between these kid and wealthy kids is pretty high. By gatekeeping for these kids, they are essentially gatekeeping for the wealthy. They would have to be pretty obtuse to not understand this. On the other hand, yes, faculty in fields like math and physics (and other fields) can actually be that obtuse.

PS. “Screening to require remedial summer courses” is not gatekeeping, but it doesn’t seem that the faculty are looking for that.

Just let the UC schools have the data provided by the ACT/SAT! And then they can decide whether and how to use it. You obviously don’t trust them with the data.

1 Like

Probably the most widely taken test of this type in CA is the CAASPP test which is taken in 11th grade by about 95% of CA public school students. The UCs could require applicants to demonstrate first-year college math readiness by achieving a target score on any of a variety of tests (SAT, ACT, AP, IB, CLEP, CAASPP, etc.) so for example, score 3+ on CAASPP (meet or exceed the CAASPP standard for college math readiness), or 570+ on SAT math, or 23+ on ACT math, or 3+ on a math AP exam, etc.

As a start, I’m looking at the scores from the CSU multiple measures assessment for enrolling in a regular first year math class (rather than a “supported” first year math class); here is what the CSUs are looking at for non-STEM majors: https://csustudentsuccess.org/multiple-measures/math/non-stem

3 Likes

Right. An Academic Senate study group looked at using CAASPP (after the decision was made to drop SAT/ACT) and recommended to not go that route for a number of reasons, many similar to why SAT/ACT scores were dropped. Replacing one set of tests with another was not an improvement. Report link below. Not sure anything is different now? If(?) there has been a decline in math prep across the state since 2020/21 (assuming we aren’t talking about OOS matriculants), it seems like addressing the root cause of that makes more sense vs. closing tertiary education doors for students who are receiving subpar k-12 instruction.

I would personally have less of an issue using the score minimums that CSUs use for supportive math pathways, assuming that means supportive pathway options would exist.

IMO it’s the role of public colleges to be part of the solution to get students up to speed in math or whatever subject we are talking about. If the UC leaders don’t want that to be part of the UC’s mission that’s their prerogative. I just don’t see the issue with having a model like UT Austin (with supportive pathways) as I linked above, and the data show that effort has been objectively successful.

1 Like

Are you talking about letting admissions officers use SAT/ACT test scores in admissions, or something else? I have no pony in this race so not sure why you have inserted personal trust issues into the mix?

If the UCs want to use test scores again, either as optional or required, I would ‘trust’ AOs to use that data as part of a holistic review, but it’s not up to me. If the UCs go that route, IMO they do have to address the fact that most HS students aren’t currently taking these tests, and that’s a complicated issue.

The improvement here is that most CA students actually take the CAASPP, which addresses your previous point about CA students not taking the SAT/ACT.

I read this paper when we talked about it some time ago. The paper is looking at using CAASPP scores as a replacement for the usual use of the SAT/ACT to identify high performing students in admissions decisions (“the identification of the top 10% of students in the UC applicant pool”). Of course the CAASPP is not an improvement in this case, because it has the usual disadvantages of all standardized tests.

However, we are talking here about using these tests simply to assess college readiness with a minimum score that indicates meeting standards, rather than comparing students’ scores and looking for the highest ones. Do the UCs really have an obligation to admit students who do not score “Proficient” or higher on the CAASPP? Supportive pathway options exist in the CSUs and there is also nothing wrong with using our very strong community college pathway.

1 Like