@Data10
It appears we have come full circle without agreement on the role that standardized testing plays as an equalizer for less affluent students, specifically non-URM lower middle class and middle class students.
For others who might be interested, here is a link to this topic who also believe that the SAT and ACT “can help level the playing field”.
https://www.peoplespolicyproject.org/2018/03/20/the-sat-can-level-the-playing-field-in-education/
[quote=“NearlyDone2024, post:75, topic:2097564”]
Being a poor non-URM is most certainly not an advantage.
It is. If you are a first-generation college student, you get a leg up. And that’s a good thing.
Also, racist much?
Regarding the correlation between income and tests scores, I am surprised that nobody has brought up the much higher rate of accommodations that wealthy kids get. There are wealth school districts in which 1/4 or more of the students have one type of accommodation or another for test taking. Wealthy kids also take the tests more often.
So, for example, one of the wealthiest schools districts in Illinois (perhaps the wealthiest?), New Trier, has about 1/4 of its students with accommodations for their tests, including extra time, etc, and on average, every student takes the ACT three times (one kid took it 13 times!).
As for prep, most of the people here are talking about outliers. Kids who get 1500 on their test without prep will not benefit much from prep. My kid did not do any real prep and did very well on her SATs.
However, kids who are getting 800-1200 and so forth can boost their scores by 200-300 points if they get private prep lessons. A kid who get 1200 on their first try may not be able to get 1520 with lessons alone, but give them lesson, extra time, and three tries, they can get around 1500, especially with super-scoring.
The SAT does not test innate abilities. It is nothing more than a test of whether students have a certain set of knowledge. It is a repeat of the tests which the students have taken over their previous high school years, but with a different methodology. As such, it does not provide much information on a students academics beyond that which is provided by the GPA.
Success at college courses requires the same set of skill and competencies which were required to succeed at high school classes. There is no requirement that a students engage in a series of standardized multiple choice tests.
Also, there are a large number of very smart, very talented people, who are able to do school work, produce amazing essays, research papers, and can deal with short, relatively low stake tests. However, test anxiety is a real thing, and is pretty common.
As for SAT leveling the playing field - that makes absolutely no sense.
Research has shown that Wealthier kids have extra help preparing for the tests, have better conditions under which to study for the tests, have more opportunities to take the test, are much more likely to have accommodations, needed or not, and are taking the tests in much better conditions. With all those advantages that the children of the wealthy have, how exactly can SAT tests be "leveling the playing field?
I do not find the argument that Freddie deBoer brings, which is that all of the factors used in admissions are unfair, but he thinks that the SAT scores are the least unfair. “I think that it’s not as bad as the rest” is not a good recommendation for the inclusion of anything in any decision making.
In somewhat different proportions. College typically needs more self motivation and better time management (over days or weeks, rather than within the time of a class or test period) than the much more supervised environment of high school.
Of course, SAT/ACT measure that even less than high school course work does.
Your earlier post talked about lower income students. However, a similar principle applies to lower-middle or middle income students as well, including those who are not URMs.
The previously linked college board study listed the following portions of students who SAT score was >1 SD below what would be predicted by their HS GPA. In the simple HS GPA + SAT model admission model that the College Board likes to use, this group would be the ones most likely to have their have their admission decision change from rejected to admitted, upon applying test optional. If $50k-$70k counts as lower-middle or middle, then the study finds $50k-$70k are twice as likely to be in the group that would benefit from test optional as >$100k kids - still a noteworthy difference.
Portion of Kids who Have SAT Scores >1 SD Lower Than Predicted by HS GPA
Less than $30k Income – 33% have much lower SAT than GPA
$30k to $50k Income – 26% have much lower SAT than GPA
$50k to $70k Income – 21% have much lower SAT than GPA
$70k to $100k Income – 18% have much lower SAT than GPA
More than $100k Income – 11% have much lower SAT than GPA
Other groups that are more likely to benefit from this test optional in this simple model include URMs, first gen, and women; as summarized below.
URMs
Hispanic – 30% have much lower SAT than GPA
Black – 28% have much lower SAT than GPA
Asian – 17% have much lower SAT than GPA
White – 15% have much lower SAT than GPA
Women
Women – 21% have much lower SAT than GPA
Men – 13% have much lower SAT than GPA
Of course the actual admissions system by colleges is far more complex than the College Board model. Colleges consider many factors in admission decisions besides just SAT score and HS GPA in isolation, particularly at selective, holistic colleges. The College Board study also has various other issues.
It’s also possible that colleges would not admit the most qualified test optional applicants based on their application and instead would primarily admit only kids from various desired groups such as desired hook-type groups. If we instead look at what colleges actually do, rather than making guesses about what they might be doing, the test optional and submitter students seem to fall in to a similar groupings as suggested by the CollegeBoard study and do not show evidence of test optional being limited to desired groups. For example, the NACAC review of >20 colleges found that:
–The lower income kids were significantly overrepresented among test optional admits, and higher income kids were overrpersented as submitters. The NACAC review defines income groups in terms of EFC instead of income. The median EFC (federal definition) appears to be slightly under $20k, suggesting a median family income on the order of ~$125k. Roughly 1/3 of the students fell into the “high need” group suggesting “high need” includes upper 5 figure incomes – including both lower income, as well as a good portion of middle income. 36% of non-submitters fell in to this lower-middle income group compared to 28% of submitters. While this is a significant difference, the differences became larger as income level decreases further from middle income to lower income, with Pell grants. At all 21 colleges, test optional enrollees had a lower average family income than test submitter enrollees.
–URMs were significantly overrepresnted among test optional admits. 23% of test optional admits were URMs compared to 10% of test submitters. While overrepresented, the majority of test optional admits still did not fall into any of the listed “diversity groups” – most were not URMs, not first gen, and not Pell. Non-URM middle income students appear to be getting admitted test optional, rather than just “diversity groups”.
–Women were slightly overrepresented as test optional applicants. I did not see a gender balance for test optional admits or enrolls in the NACAC review, only applicants. Other studies that do list gender balance by test submitter status show that women are overrepresented as test optional enrolls, as suggested by the College Board study. For example, the previously linked 25-years of test optional at Bates study found that 43% of women were admitted test optional, compared to 32% of men.
Test Optional policy seems to dubious. If a college thinks standardized tests are unimportant or have little value, it should just adopt a test blind policy to be fair to all students. It’s like those less-than-genuine EA programs practiced by some colleges that also offer ED: we’d like you to submit your applications (because we know you wouldn’t have otherwise), but…
I wish more grad schools would go test optional! The GRE is quite expensive and is not offered nearby for many students. D had to travel 1.5 hours to a testing center, for example. Had to wear a mask in a warm room for 4 hours too. The math section was onerous, as D had algebra in 7th grade, geometry in 8th, and no math at all in college. The GRE test prep offered at the flagship university 2 hours away costs $900, which is cheaper than many other courses, but still pricey for us as a middle class family trying to rustle up the funds to pay for year 12 of college tuition (3 kids). D self-studied but could have benefitted from math tutoring for sure.
Grad school admissions committees can see college GPA, how well students performed in classes in their soecific major, and can also consider research, internships and other career building experiences. In addition, they request 3 letters of recommendation from professors, a writing sample such as an academic paper, and a statement of purpose. Shouldn’t all that be sufficient?
“Success at college courses requires the same set of skill and competencies which were required to succeed at high school classes.”
That depends a lot on major, I was in engineering and a whole of kids that graduated in the top 10% of their class did not do well, a few (about 20%) had to transfer. I’m not saying high test scorers did better and maybe the skills are similar but the grading is tougher and the if there’s a curve, you’re competing against better students.
“SAT leveling the playing field - that makes absolutely no sense.”
Test scores can level the playing field among high schools especially if they are similar in SES. In fact the leveling happens before the applications are submitted, if you’re unhooked in the bay area, you’re not applying to Berkeley or Stanford unless you have over a 1500 (2250 in the old scoring system).
“As such, it does not provide much information on a students academics beyond that which is provided by the GPA.”
The GPA has many flaws which you conveniently overlook:
“The SAT does not test innate abilities. It is nothing more than a test of whether students have a certain set of knowledge.”
I agree on this, I’d rather the test be more on innate ability and not being able to study for or game.
Good points.
Based on the description at https://www.ets.org/gre/revised_general/prepare/quantitative_reasoning/ , the math topics seem like just high school math (algebra and geometry) similar to that use in the SAT, plus some introductory statistics. Not sure why it should be that onerous to a student who presumably knows enough math to have taken the SAT and/or ACT.
On the other hand, it is not obvious what or how much value the GRE general math section adds to making graduate admission decisions, compared to college record in the relevant subjects. Either the math involved is not relevant to the subject, or it is laughably too low level for the subject.
Any test that tries to measure whatever is defined as “innate ability” still depends on some learned knowledge as a baseline. Of course, once there are actual stakes tied to the test, those taking the test will try to figure out how to score better on it (test prep), and it is likely that those with more money can deploy more of it for this purpose for themselves or their kids.
@inthegarden: I saw you posted a comment about TO on another thread.
“Going TO when your kid goes to a famously wealthy and competitive high school with no grade inflation and superior teachers and students, whose dedicated college counselors have longstanding relationships with T-20 schools, whose teachers know just how to teach to the rubric of APs, whose students all have broadband for their 8-3 zoom classes… is a very different prospect than it is for those of us with our average schools in struggling communities, with no broadband, NO Zoom classes, a school for which a normal year sees zero or maybe one student at a T-20 school.”
Also regarding retaking the SAT: “D could have a chance to corroborate her top grades and rank (within the context of her mediocre environment) with a decent test score so that she’d have a CHANCE at a nice school”
Those are interesting takes. Do you have any additional thoughts on the effect of TO that you care to share?
No one thinks the SAT or ACT is better than grades, not even the testing companies themselves. Their strongest claim is that standardized tests plus grades are a marginally better predictor of academic success in the first year of college than grades alone. Notice how limited this claim is. Grades are and always will be the backbone of the college admissions process. Schools that go test optional are concluding that any marginal value added from standardized tests is simply not worth the additional paperwork and stress on students.
Standardized test scores aren’t an equalizer. It’s well documented that they systematically advantage the affluent as well as white and Asian students, and disadvantage everyone else. It’s been said that the SAT is a better predictor of household income than of anything else, including academic success.
I like the model used by Bryn Mawr and some other colleges. They consider any standardized test score the student chooses to submit: SAT, ACT, AP, IB, etc. That gives each applicant the opportunity to put their best foot forward. No test is advantaged over the others. In each case it’s just an additional data point in the applicant’s overall credentials. If an applicant chooses to sub,it no standardized test scores, that’s fine, too. Some very capable students just don’t do well on standardized tests; for them, a standardized test score would be a false indicator of their likelihood of academic success in college. Taking standardized tests is a highly specialized skill that has few real-world applications and is utterly irrelevant to how college courses are taught and tested.
However, some of those few real-world applications include gatekeeping for professional schools and professions. For example, someone aspiring to practice law will find the LSAT and bar exam as additional standardized test gates to pass through.
I would think that kids that score well on standardized tests will also perform better on any type of exam that tests mastery of the material. In addition, these kids can perform under pressure and time constraints and these skills definitely have real world applications.
“[Standardized] test scores are a key way admissions officers compare students from different high schools; they might take a chance on a kid from a little-known school — possibly one with a middling curriculum — if she has stellar scores. Without those scores, admissions officers will be less likely to take that gamble.”
(Per Jeffrey Selingo, the author of “Who Gets In and Why: A Year Inside College Admission.” He is former editor of the Chronicle of Higher Education and a special adviser and professor of practice at Arizona State University.)
https://www.washingtonpost.com/outlook/2020/10/05/college-admissions-sat-grades-covid/
Except in careers like medicine or law, standardized tests are not used to test mastery. The vast majority of work places test your mastery by giving you an example of the work you are supposed to do. They don’t care if you can solve a standardized mathematical problem that is written on a problem sheet, they want to know whether you can utilize this knowledge.
One of the most brilliant CS professor I know would have a difficult time passing many standardized exams, since he thinks relatively slowly. He has won multiple national awards for his work in his field, but that is because, like the vast majority of actual work, you don’t have to answer 100 questions on 10 different topics in 2 hours, but have days to weeks to months to years to figure out how to develop an algorithm to perform a specific function, find a solution to a specific problem, etc.
That is why I am skeptical that standardized tests actually test actual mastery of any topic.
A person can ace a standardized test and forget everything on it in a year. A person who has been required to utilize knowledge in a real life situation, and write about it, will actually have long-term mastery of the topic.
Sure it does. Many kids in low income schools don’t see an AP class or IB class until senior year – way too late for admissions. For such students, SAT/ACT is the only option.
There is no rush to take the GRE. Many grad student prospies take a gap year or two to work on research, and/or work to make some of their own money to pay for testing/prep.
Plenty of excellent online programs, such as Manhattan.
At my D’s grad program, her advisor told her that GRE was more of a separator for fellowships, i.e., some additional $$ over the fully-funded PhD program, than for admissions per se.