No, that is NOT what happened. To begin with, individual faculty members do not see the applications of their students, and so have no idea whether a student was accepted TO or not. Second, they went test blind, so they have no idea what the SAT score of a student happened to be. Third, they had some anecdotes, not statistical evidence of a trend.
Finally, the petition was that a committee be formed to see whether the test blind policy was affecting the preparedness, and whether they should go to TO instead.
There is a VERY LARGE difference betwee:
âLetâs see whether going test blind has an effect on whether students are prepared or notâ
Versus
âBecause we have gone test-blind our students are now unprepared, we demand that the SAT requirement be reinstated immediately!!!â
I think we largely agree. Money and savvy will overcome almost any testing or evaluation schema. This forum in particular is about the SAT.
We agree that performance on the SAT does in part reflect the preparation by schools. But, in its new, abbreviated format, the test may be in some ways more âcrackableâ or ârandomâ in ways that are more amenable to test prep. K-12 educational unfairness is a tough nut, but learning how to apply algorithmically a few grammatical rules of punctuation seems like low hanging fruit, especially when access to free test prep is either too cumbersome or too limited.
South Korea is in fact quite close to a testing meritocracy. Sure, money still buys the best tutors and hagwon, but the tested subject matter is so comprehensive and difficult that the national exam is recognized as quite fair. Also, among South Koreans, education is so primary and the school you attend is so transformative that a large segment of the population does have the opportunity to prepare fairly as even poorer families will sacrifice all to improve childrensâ education.
In the US, the College Board (the money machine that it is) has it hard. On the one hand, they are under pressure from teachers not to ask difficult material due to the declining performance of most students. On the other hand, they must construct a test that will yield a normal distribution of scores. Looking at these two pressures, you come up with a system where things like grammar - not taught well in any school, but not considered exceptionally difficult - becomes a handy screening tool, even though syntactical rules are archaic and practically obsolete.
On the math section, this balance is struck using adaptive technology, so I also wonder if a detailed examination of the way the testâs adaptive technology and scoring work together can be used to get disadvantaged students over the âtough questionâ hump into the questions where answering every question perfectly isnât required to attain a good score.
If the meritocracy crowd and colleges are moving toward looking closer at the SAT, Iâm wondering whether someone smart can come up with a crash course for underprivileged kids that hits the obvious high value targets on the new SAT verbal and gets them over the hump in the math section. And make it free.
Sounds like you are suggesting to reverse-engineer the SAT adaptive technology to look for âquick fixesâ that can âbeat the test,â raising the scores of some disadvantaged students. Even if it can be done, this is just masking their math and verbal deficiencies and kicking the can down the road. What happen when they get to college and have to take exams without such fixes?
In addition, if these fixes are, as you said, free and widely available, the âbetterâ students will still do better, so SAT-percentile-wise these students will still be in the same spots, even if the score gap might appear to have closed (I believe College Board converts raw scores to the 200â800 scale in such a way that a certain percentile receives a certain score, so the gap might not close, at least not noticeably).
In a few years, College Board will mostly likely adapt their adaptive technology to stem these loopholes, so all the quick fixes will be for naught.
Well every little bit helps for those to whom few offer help and even fewer offer mercy. Whatâs wrong with trying at least something to help those who already start well behind the starting line in a savvy, intelligent way. Kind of the way many rich kids already benefit.
Of course, if both groups acted according to these beliefs, they would not spend money to live in better public school zones or pay private school tuition.
There are around 300 faculty members at Caltech, and 140 signed the petition. There are 20 faculty members on the undergraduate admissions committee, and they are the ones that have reviewed the application, not the other 280 faculty members, who first get to meet students when they walk into that faculty members class.
Even the faculty on the committee are highly unlikely to remember the SAT scores of the 200+ students who were accepted. Faculty can barely remember the names of students who they see three times a week in a class of 20 students. But you expect them to remember what SAT score each of the 200+ students had, 7 months after they reviewed the studentâs application for maybe half an hour?
Moreover, itâs a moot point, since even the faculty members on the admissions committee havenât seen the SAT scores of the student. Caltech was test blind years, so the student didnât even submit their test scores.
So no students submitted test results, and therefore the faculty who taught students arenât able to know whether a studentâs SAT score was, unless they asked the student.
Of course, they may well have done that, and that very well may have been on the petition. I donât know. However, if they had a good set of data of students who did worse and were also less prepared, they wouldnât be talking about âanecdotesâ, they would be talking about âdataâ
I didnât say they saw test scores, Iâm well aware CalTech is test blind.
The CalTech team is working on gathering data, I expect they will do whatâs right for them when it comes time to decide what to do wrt testing. Test blind was always a short term experiment for them, at which point they would assess how that went and then go from there.
I would be surprised if Top 100 schools didnât track candidates by town over time. From a technology standpoint, it is not that difficult. They already have the data, so it is a small step to run some analytics on it.
And that was not entirely the point I was making. Grade Inflation is completely out of control, particularly at preps and affluent suburban schools. Statistically this is hard to track, but anecdotally, almost every parent I know that sends their kids to prep or has kids in an affluent suburban school recognizes that the grade inflation is pervasive, especially for the upper half students.
This is exactly what many people predicted was going to happen when so many universities went TO and Test Blind. The incentives for every prep and affluent high school principal would be to inflate grades. Why wouldnât they? And as Charlie Munger famously said, âShow me the incentives and I will show you the outcomeâ. It is disappointing that top universities which have many academic departments studying human behavior couldnât figure out that this was going to happen. Or maybe they could, and they wanted this outcome, which is even more disappointing.
Grade inflation has been increasing for decades. Regardless, when looking at transcripts there are obvious differences across students (and schools) when one looks at course curricula and rigor. HS profiles help in making these determinations. IME there is not as much grade inflation at many prep schools or affluent publics as there is in a typical public HS, but I donât have data to support that, just based on reading apps all day.
The SAT may be imperfect, but this is why it is necessary.
I remember someone on this site asking about their chances for getting into Texas A&M. They werenât in the top 10% of their class, so they werenât an auto-admit. When discussing their stats, they did have a 4.0 UW GPA. If more than 10% of the graduating class of a high school has perfect grades, something is very wrong.
Iâm not arguing that this isnât true. However, you can control for this (somewhat) by using the GPA in the context of the high school, rather than by considering that they are standardized across all schools. You can do that for SATs as well, but the entire argument for the SATs is that they are unbiased and donât need to be controlled for family income and parental education.
I have never seen any admissions officer or anyone who is paying attention argue that. Ofc family income and parental education affect a studentâs SAT score. Itâs just that SAT scores are less affected than all the other factors considered in a wholistic review.
Each of these (GPA, rank, SATs) have strengths and weaknesses. Thatâs why colleges should ask for everything and then make a determination as to what to use and how to use it.
Colleges have had no problem letting potential applicants know that they are test optional. Big bold letters on their admission tabs etc.
They can apply the same effort to let students know that SATs are viewed both on an absolute basis and in context. They can put up stats about it like they do for every other metric they want to highlight, talk about it in their virtual and in-person tours, whatever. Make people other than CC posters very aware of how they use standardized tests.
Why canât schools have testing policies they want? Policies based on their schoolâs data, or those that serve their schoolâs enrollment goals best?
What do you feel you know that test optional or test blind enrollment VPs are all missing?