Many colleges, including the University of Chicago, now omit standardized admissions tests like the SAT as requirements for application. Is this justifiable?

Last year the University of Chicago made standardized admissions test, like the ACT and SAT, optional for students applying to get in. But this is only one school among over a thousand that is going test-optional. And, as this New York Times article reports (click on screenshot), a coalition from California, described as “advocacy groups and a largely black and Hispanic school district”, has filed suit against the entire University of California system, demanding that it stop using these standardized tests in its admissions. While Chicago makes it optional whether to include scores in your application, a victory in this lawsuit would make it illegal to ask for scores, which would eliminate them as a factor in college admissions and, if other schools follow suit, end the standardized-test industry. (The lawsuit asks for the UC system to “bar students from submitting test scores even if they want to.”)

The main reason for making tests optional, say the schools who do this, is to increase diversity; this is because the tests are said to be biased in two ways.

a.) (From the NYT): “The plaintiffs say the college entrance tests, the SAT and ACT, are biased against poor and mainly black and Hispanic students. By basing admissions decisions on those tests, they say, the system illegally discriminates against applicants on the basis of their race, wealth and disability, and denies them equal protection under the California Constitution.

They argue that the use of standardized tests has led to the creation of a vast test-prep industry that privileges affluent families who can afford to send their children to tutoring. Other measures — like grades and teacher recommendations — would provide a fairer way of judging students, the plaintiffs say.”

In other words, the tests give an unfair advantage to those who have the money to take test-preparation courses.

b.) Another complaint is that the very nature of the questions on these tests is biased against minorities because of the way they’re worded or what areas of knowledge they test.

The tests are also said to be useless because they don’t measure anything of interest to colleges, for the scores are claimed to show no correlation with any measure of success in college or after college.

Finally, revealing the big gap in scores between groups, with Asians at the top followed by whites, Hispanics, and blacks, exposes the embarrassing inequalities in American society. Without the tests, we simply don’t have to face this fact. But we should, for, as Marten Roorda says below, these differences diagnose a serious problem in American society, and do so in a quantitative way. I know of no other instrument that can do that.

I am not sure if I agree with the new test-optional policy for several reasons.  First, are they really “biased” in terms of wealth contributing to the size of scores? The results seem to be mixed. I’ve done a bit of reading over the past couple of days, and the effect of these for-fee tutoring courses seems small: an average gain of about 15-20 points on the math test and 4-10 points on the verbal. That’s out of a total of 800 points for each test (at least that was the total when I took the test.) Also, the SAT, at least, does provide practice tests for free, though those may have a lesser effect on scores than would intensive paid tutoring.

Still, the “wealth” effect seems to be minimal, and, at any rate, wealth already plays into college admissions in several other ways, favoring the children of donors and those children whose parents could afford private schools where they get a better education than in many public schools. It’s not just race that’s correlated with wealth.

Second, the SAT has ways of determining whether questions are biased in the way mentioned above, and they eliminate questions they find to be “ethnically biased.” Further, it would seem hard to have biased questions in the math section of the SAT (there are two parts; math and verbal). Finally, if questions really are biased because of minority status itself, one would have to explain the higher performance of Asians than of Hispanic and black students. The former do better than whites, the latter worse.

Finally, my (admittedly non-intensive) reading on whether the tests measure anything of value indicate that both high school grades and tests, when combined, are the best predictor of outcomes in college, and there may be a synergistic effects of the two criteria (the reports I’ve seen haven’t given statistics). [This article in Slate, noted by a commenter below, shows that SAT scores, especially but not only when combined with grades, are important in predicting college success.] Nothing is said about letters of recommendation, which of course are another ubiquitous requirement for college admissions. I suspect that those letters play a small role in determining who’s admitted. I haven’t been involved in college admissions, but others who have tell me that the letters are next to useless.

And then there are required essays, and I don’t know how important they are in admissions. Given that you can actually pay someone to write an essay for you, or help you with it, those also become indices of wealth, and should be omitted along with test scores.

But in fact the argument that scores bias decisions against minorities bears little weight for one important reason: colleges and universities already know that test scores of Hispanic and black applicants are lower than those of whites, and already compensate for it by not counting those scores as much for minorities applying for admissions. In other words, the scores are already being discounted to a large degree among different ethnic groups. (They would still seem to be of value, by the way, within groups, for example in deciding which Hispanic student you want to admit.) That the diversity argument rather than the predictability argument is the real reason for eliminating test scores, or making them optional, can be seen in this paragraph from the NYT report:

The University of Chicago, a top-ranked school, announced last year that it was going test-optional. This fall it reported that as a result of the new policy, along with more financial aid, outreach and mentoring, the entering class of 2019-20 had 24 percent more first-generation and low-income students and 56 percent more rural students than the previous year. About 10 percent of applicants did not submit test scores, a spokesman said.

Note that several things changed in one year: not just test-optional status, but also financial aid and mentoring. How can they ascribe these differences to making test scores optional without some kind of complicated multifactorial analysis?( The school didn’t report the percentage of minority students before and after the test-optional status.) While I applaud efforts to increase diversity here on many fronts, including wealth, race, rural vs. urban applicants, and even politics, why did the University’s previous efforts to increase this kind of diversity fail to suffice? I’m sure that the University of Chicago, like Harvard and other schools, already took the lower average test scores of some minorities into account when deciding whether to admit them. Were they still using them in a way that discriminated against these minorities? Knowing my University, I doubt it. We strive mightily to get minority students, but often fail, perhaps because the ones that look good will go to Harvard or Yale rather than Chicago.

My own theory is that the lower scores of Hispanics and blacks reflect an environment that is not conducive to getting a good education, which itself is a byproduct of present and past discrimination. In other words, equal opportunity is denied from the outset. The way to rectify this is to weight test scores by an applicant’s background. But, as I said, schools have already been doing this for some time! As Marten Roorda, chief executive of ACT says in the article:

“It is inappropriate to blame admissions testing for inequities in society. We don’t fire the doctor or throw away the thermometer when an illness has been diagnosed. Differences in test scores expose issues that need to be fixed in our educational system.”

Indeed—although of course Roorda has a personal investment in continuing the testing process.

Given the problems with grade-point averages, which of course vary substantially with the nature of the school, and with letters of recommendation (everybody’s is good), standardized tests are the only instruments colleges can use that provide a metric that applies to all applicants. Now if that metric proves to be of no value in predicting the quality of the student (and it won’t be perfect, of course, because some low-scorers are diligent and some high-scorers lazy), then yes, one must question its value, except for diagnosing societal problems. But I’m not yet convinced that the correlation between standardized test scores and all measures of college success is a big zero.

What looms here is the increasing erosion of all meritocratic standards for college admissions, for the meritocracy is said to be inimical to the attainment of diversity. But that’s not necessarily true if colleges weight test scores in such a way as to increase diversity, whether it be to rectify past histories of bias and oppression or to attain diversity (as the Bakke decision mandates) as an inherent good. And, as an advocate of affirmative action, one must accept that there might be tradeoffs between diversity and what are considered “meritocratic” criteria.

There is one big advantage of requiring test scores—two actually. One is to use them to evaluate students within a group, whether that group be rural students, veterans, first-generation students (those whose parents and grandparents didn’t go to colleges), blacks, Hispanics, and so on. I don’t think anyone has determined whether, within such groups, test scores have no correlation with college success. (Of course I may be wrong.) And, importantly, standardized test scores can be used, and have been used, to identify outstanding people whose other records might not have gained them admission.

Given that colleges already weight scores given the nature of the applicant, the only rationale I see for eliminating the tests is if they have a negligible correlation with success in college, and therefore are of no value in the application. That has not yet been demonstrated, at least to my satisfaction; in fact, the data suggest otherwise (see the Slate article noted above).

I’m not going to dwell on the other “inequities” in the system, including preferential admission of athletes and “legacies” (those who parents went to the same school, and are therefore more likely to donate money), for these admissions are less important than the kinds of diversity I emphasized above.  I just wish that colleges, when they eliminate tests or make them optional, would give the real reason why they do so. But it would be a cold day in July in Death Valley when that happens.

The use of test scores would seem to be a difficult issue, as the NYT notes, but I thought it had already been solved.

68 Comments

  1. Posted December 12, 2019 at 10:57 am | Permalink

    Good parenting will make up for most shortcomings of resources for test-preparation. Also I’ve seen disadvantages of parents pushing too much of their children’s education onto ‘tutors’. The result is usually disaffected wealthy kids who turn to depression and/or substance abuse.

    Being dyslexic, the test is incredibly unfair, but it’s still worth having as a data point for colleges. Good admissions committees weigh far more than just standardized tests.

    Standardized tests should remain. I think they hurt far less than people know with regard to getting into college. Colleges can know a lot more about a person (kid) than people think.

    • eric
      Posted December 12, 2019 at 3:03 pm | Permalink

      “Good parenting” slips in another prosperity bias though. I.e. it basically means that kids whose parents have a good single or double income – i.. prosperous enough to spend their time helping their kids learn will have a leg up over kids whose parent or parents have to work multiple low-income jobs just to get by.

      Having said that, I also don’t have much objection to the SATs. Back in my young fogie days, I raised my score by about 250 points by going to the library, checking out a book of past tests, and then doing a bunch of them. It took my time, but it didn’t take parental resources or money. A motivated student can get basically the same “prep” value as a rich person…it just takes a bit more self-motivation. Honestly, I think most of these test prep courses provide more motivation than learning. It’s like having a friend who goes running with you – it keeps you doing what you know you should be doing. But it doesn’t actually improve you, you’re doing that yourself.

      As for whether they correlate with class success…I’d love to see the statistics on that. I’m somewhat skeptical they don’t correlate at all (consider my case…what I did to get a better score is basically similar to I did to get better grades). However I’m not that skeptical and am very open to being convinced they’re useless in terms of measuring future academic performance.

  2. enl
    Posted December 12, 2019 at 11:02 am | Permalink

    I had missed this in the news (it is getting very hard for me to follow these days).

    Interesting and, unfortunately, unsurprising. After the Harvard suit (admissions practice upheld as acceptably non-discriminitory), this seems a fitting follow-up.

  3. Posted December 12, 2019 at 11:03 am | Permalink

    “My own theory is that the lower scores of Hispanics and blacks reflect an environment that is not conducive to getting a good education, which itself is a byproduct of present and past discrimination.”

    Is it really the case that there is currently widespread discrimination against those groups, in a way that does not apply to, say, Asians in the US?

    “The way to rectify this is to weight test scores by an applicant’s background.”

    Though if one does that and admits students who are not academically prepared for the course, then they struggle or drop out.

    • aljones909
      Posted December 12, 2019 at 12:41 pm | Permalink

      Why not have all applicants sit an IQ test?

      • Posted December 12, 2019 at 12:53 pm | Permalink

        No way! The scores would have markedly different means for different races. So people would freak out.

        • aljones909
          Posted December 12, 2019 at 5:01 pm | Permalink

          But.. IQ has crystallised by age 18. A very good predictor of how individuals will cope with a given course.

  4. KD
    Posted December 12, 2019 at 11:05 am | Permalink

    The problem is that these schools are worried that the Supreme Court is going to revisit Bakke, and evidence that you are admitting X with SAT’s 300 points lower than Y because Y is a member of a group with consistently “bad personalities” is potentially damaging. [As we saw in the Harvard litigation, Asian-Americans all have bad personalities.]

    Getting rid of the SAT helps hide evidence of purely race-conscious admissions, which are supposedly illegal (according to Bakke).

    Colleges can’t go public stating i.) yes, we illegally manipulate admissions to maintain certain demographic quotas in incoming classes, ii.) we are worried we might get sued for doing this, and iii.) we want to make it harder for the DOJ to prove discrimination if we get sued.

    • Chewy
      Posted December 12, 2019 at 8:19 pm | Permalink

      Your use of lower case roman numerals is culturally oppessive and ….

      8^)

  5. Jon Gallant
    Posted December 12, 2019 at 11:11 am | Permalink

    Some thought-provoking reading in this area by Lionel Shriver: The Asian American success story catastrophically undermines the assumptions behind affirmative action
    Lionel Shriver
    https://spectator.us/asians-doing-well-must-stopped/

    • KD
      Posted December 12, 2019 at 11:31 am | Permalink

      I could almost swallow the progressive story on race if Asian-Americans didn’t exist.

      However, once you are aware of Asian-Americans, it is very hard to chock testing up to “white supremacy” when Asian-American and Jewish students consistently blow out “decent, Christian, white people” on standardized tests.

      Moreover, if you watch most of the MSM coverage of race, they generally fail to even mention Asians when they are talking about Blacks and Hispanics and schools. This suggests to me that the editors are also aware that their narrative is contrived and are deliberately distorting the facts to create a false impression in the public. Very troubling, considering the negative long-term effects of scraping meritocracy in employment and college admissions

      • dd
        Posted December 13, 2019 at 12:04 am | Permalink

        I am hispanic and it’s fascinating to see how editors, especially at journals such as the NYTimes, often refuse to include Asians, and even Hispanics, in comparisons.

        It’s black and white all the way to get the most fitting “data” for the narrative.

        Happens again and again. One of the many reasons I tired of even the NYTimes and no longer subscribe.

  6. dd
    Posted December 12, 2019 at 11:17 am | Permalink

    Questions for readers:

    1. Can anyone offer a specific example of a biased question on a test or what would constitute one?

    2. Doesn’t the fact that Asians, many from poor background, do better than white people on SAT,etc show that the discrimination argument is weak?

    (This is the same phenomenon at Stuyvesant High School in NYC.)

    • Adam M.
      Posted December 12, 2019 at 1:03 pm | Permalink

      I can give some examples. When I was taking a test for admission to a private high school, one of the questions asked me to analogize between some sports – one of those “A is to B as C is to ?” questions. Well, one of the sports I had never heard of and for one of the others I had no idea what the rules were. (I wasn’t much of an athlete.) I’m pretty sure I got that question wrong.

      Another poster here gave an example about math questions the probability of drawing cards from a deck. Some people might not know what’s in a deck of cards.

      That said, I still got 148 out of 150 on that test and very few questions had any culturally specific content. I think they’ve been very careful to weed out such questions over the years and I doubt you could find any today. Not that it matters. I remember reading plenty of stories like the following: a fire department is sued for discriminating against blacks and Hispanics in its entrance exam. The judge admits he can’t find any bias in the test, but rules that the poorer results of those groups on the test prove that it must be biased and orders the department to hire the people who failed and give them years’ worth of backpay.

      • davelenny
        Posted December 12, 2019 at 1:27 pm | Permalink

        My extremely dim recollection of reading similar stories is that fire departments lost because they could not show the relevance of the tests to their employees’ duties. However, as I understand it, SATs have considerable predictive power.

      • Chewy
        Posted December 12, 2019 at 8:22 pm | Permalink

        I once missed a question (and I was Valedictorian in my rural/oppressed/class-shamed South Dakota small town school) because I didn’t know which goal was “defended” by a football team and so my yardage count was the reverse of the “true” answer. I complained. I lost. I’ve lived in shame and humiliation ever since.

        • merilee
          Posted December 12, 2019 at 11:28 pm | Permalink

          Thankfully I never got a football question as I would have missed it, too.😬

    • Posted December 13, 2019 at 11:43 am | Permalink

      There’s a good (albeit dated) episode of Diff’rent Strokes (!!!!) about this.

      (Rich, white) Drummond wants to send his adopted (African-American, grew up for years in the Harlem ghetto) sons to a prep school, but they flunk the admissions test because of culturally focused items of “general knowledge”. He’s incredulous that his old school could discriminate, when he realizes that “general” is not general at all – and has the boys set a test on “general knowledge from the hood” for the admissions officer.

  7. JezGrove
    Posted December 12, 2019 at 11:18 am | Permalink

    It is many years ago, but when I studied A-level Statistics we were told that some types of questions had been dropped because they made implicit cultural assumptions. If I recall correctly, and we’re talking about the early ’80s here, the main such assumption involved probability questions about the likelihood of drawing combinations of playing cards from a deck, and so on.

  8. Randall Schenck
    Posted December 12, 2019 at 11:19 am | Permalink

    I am not involved in the academic world so my thought may not mean much. I recall long time ago taking the test, I think it was during the summer after the Junior year of high school. It also might be a good idea to know the exact reasons why they started requiring the tests in the first place. So if those reasons no longer apply, that would explain the change.

    It looks like the whole business started in with the military way back in the 20s. But that does not explain why colleges all over the country begin requiring the test. With so much of our education in K-12 being in the hands of the states, the different educational levels are pretty wide. Also, the subjects taken during those years can vary quite a bit. So maybe the SAT allows for a different measurement that is beneficial to know. I am just guessing on this but before I stopped doing something I thought it might be a good idea to know why it started.

    I also think about another type of review that applies in some companies but not others. That is an annual PER or personnel evaluation report. Strangely enough, these were also something that was done in the military long before companies started doing them. The company I worked for spent many years changing the PER system until they finally made it much better and actually applied things like pay increases to them.

  9. Posted December 12, 2019 at 11:43 am | Permalink

    I know that the math portions of the ACT and SAT are useful for placement purposes (i. e., place into calculus or pre-calculus); they have some predictive power.

    • Posted December 12, 2019 at 11:44 am | Permalink

      I should point out that the situation might be different for elite universities than it is for more average universities.

  10. Roo
    Posted December 12, 2019 at 11:44 am | Permalink

    I think one variable to figure out before answering such a question is how valuable a degree from a prestigious university really is. I notice my conservative family members are increasingly skeptical about higher education these days, especially in its more gilded manifestations, and to some degree I wonder if they have a point. It’s ridiculously expensive, and even with a scholarship, students may incur a significant amount of debt. In some cases training from vocational programs results in high paying jobs (electricians, plumbers, hvac repairmen, etc., often do very well) and in some cases, unless you are looking for an extremely elite position, no one cares whether you went to Harvard or a state school (nurses, for example – if you are a competent nurse you can pretty much always find a job.)

    In other words, I don’t think the focus should be exclusively on how well a student does, but also on what the benefits to the students are. If there is a simple across the board advantage to going to a prestigious school, then I am more in favor of eschewing test scores and focusing on distributing those benefits more equally. If, on the other hand, these degrees are going to be a waste of time and money unless you have a specific reason for attending them (a special talent in mathematics and a school with a fantastic program to that effect, for example,) I think that should be taken into account, either via test scores or some other measure.

  11. Janos Simon
    Posted December 12, 2019 at 11:54 am | Permalink

    Two comments:

    1. Standardized tests have a role in calibrating transcripts. They can also guide — for better or worse — high school curricula. Knowing that interpreting simple graphs “will be on the SAT” will make (some) schools teach that skill.

    2. It is likely that small differences in scores are not relevant. Most viable applicants to elite schools will have very high scores — and I very much doubt that scoring in the 96th percentile vs the 95th has any statistical significance.

  12. Curtis
    Posted December 12, 2019 at 12:02 pm | Permalink

    The SAT has lots of problem but it is closer to being unbiased than anything else. A kid with a 1400 score is almost certainly smarter than one with a 1000. A kid with a 3.8 GPA is fairly likely to be either smarter or harder working than a kid with a 3.2 but there is a large degree of uncertainty depending on school and class selection.

    For historical context, China used the Imperial Exam for over a 1000 years as way to allow bright peasants to get government jobs. It was seen as way to overcome patronage as the deciding factor.

    SATs should be PART of the admission process.

    They also tell the applicant about their future. Certain degrees require different skill. If you get a 500 on the math section, you do not want to go into math or physics. If you get a 500 on the English, you should not be an English major.

    • EdwardM
      Posted December 12, 2019 at 12:19 pm | Permalink

      “A kid with a 1400 score is almost certainly smarter than one with a 1000.”

      Better educated yes, smarter maybe not.

      • Curtis
        Posted December 12, 2019 at 12:55 pm | Permalink

        SATs have a strong correlation with IQ tests (between 0.5 and 0.8 depending on who is measuring.) Have you ever met someone with a 1400 SAT score that wouldn’t consider quite bright? Have you ever met someone you consider quite bright with a SAT score of 1000?

        400 points is almost 2 standard deviations different. It means a lot.

        Let me ask you a question. Suppose you were a guidance counselor to two kids who want to go into some demanding STEM field. One had a 3.8 GPA and 1000 SAT score. The other had had 3.2 GPA and 1400 SAT score. Who do think has a better chance of succeeding in their career?

        • EdwardM
          Posted December 12, 2019 at 3:28 pm | Permalink

          You missed the point. SATs don’t measure intelligence – they’re not designed to. They are a measure of academic achievement apart from grades. Both are, of course, correlated with intelligence, but the mapping is not perfect. Someone with high intelligence who has been badly educated may do worse on an SAT than someone of less capable intellect who is well educated. There are many reasons for a poor education and historically membership in a racial or economic class is one of them.

        • NM
          Posted December 12, 2019 at 11:00 pm | Permalink

          Curtis, I would for sure say that the kid with a 1400 test score and 3.2 GPA was lazy and the other kid had a strong work ethic, thus, making him to likely candidate for success. It’s so much about work ethic, so much.

          • Posted December 13, 2019 at 4:35 am | Permalink

            And how exactly do you know this “for sure”?

            “So much about the work ethic”: are you speculating here without any data?

      • Curtis
        Posted December 12, 2019 at 1:16 pm | Permalink

        I know GRE are different but this shows the link between graduate majors and their GRE scores.

        Which majors have the best math scores? Solid state physics, physics, math, operation research as you would expect. High verbal scores are classical language, classics, history of science and philosophy as you would expect. The lowest in both are social work, physical education and criminology.

    • Justin Seabury
      Posted December 12, 2019 at 1:40 pm | Permalink

      GPA is an unfair way to measure anything. In high school I worked 20-30 hrs a week on top of going to school, but foolishly took harder classes to actually learn things. My GPA suffered as I did not have time to do the hrs of homework assigned by the harder classes and my level of being awake to learn was impacted by working beyond school time. I ended up with a C average as I passed most test with flying colors I was unable to keep up with assigned work.

      • Curtis
        Posted December 12, 2019 at 2:02 pm | Permalink

        Homework and grades is whole other topic. Last year my son got a C and D in his AP classes but got 4 and 5 on the AP tests.

  13. Roger
    Posted December 12, 2019 at 12:10 pm | Permalink

    On a side note, many libraries are eliminating late fees. It’s a brave new world out there folks.

    • Adam M.
      Posted December 12, 2019 at 12:52 pm | Permalink

      And transit systems are eliminating penalties for fare evasion, cities are refusing to prosecute shoplifting under $750 or $1000, etc. because certain racial groups commit those crimes much more often so the law (or its enforcement) must be racist…

      • Roger
        Posted December 12, 2019 at 1:07 pm | Permalink

        I should clarify my poor sentence. They’re eliminating the concept of late fees altogether. No more practice of charging late fees for overdue books.

        • Adam M.
          Posted December 12, 2019 at 1:09 pm | Permalink

          I understood ya. Sorry if my response didn’t seem relevant. 🙂

          • Roger
            Posted December 12, 2019 at 1:17 pm | Permalink

            My bad. I looked at my sentence and I was like, dang someone send me back to English class or something.

  14. Mark R.
    Posted December 12, 2019 at 12:23 pm | Permalink

    Of course, this issue has been exacerbated by the college admissions scandal centered around “counselor” William Singer. I’m sure there are still many out there who game the system.

    My thought is that we could improve Hispanic and black student outcomes if we as a nation stopped the idiotic practice of paying for K-12 education via property taxes. IMO, this inequality by far outweighs the inequality of ACT, SAT and other standardized testing in regards to college admission.

  15. Posted December 12, 2019 at 12:37 pm | Permalink

    My school has gone test-optional also. My general impression is that we’ve recognized that requiring a test sends a message that it’s important to score high or you won’t get in. (It’s also expensive, and may dissuade some potential students from applying.) I think a good admissions process should value more facets of the person than that – and I think we try to do that here.

    At Wake, every potential admit is interviewed, and with few exceptions interviewed in person. That is in part to weed out students who have others write essays or take standardized tests for them. But I think it’s also to get a sense of whether a student is a good fit. “Fit” is subjective – and a little unsatisfying as a decision making mechanism – but at some point you do have to make a decision about whom to admit.

    I think what most people want is for the admissions process to be “fair,” though the definition of what fair is varies from person to person. It’s unlikely we’ll ever achieve a system that everyone agrees is perfectly fair. I think the best we can hope for is to continue to improve.

    Jerry’s point about using exams to compare students within a group is interesting to consider here. Athletes, for example, are not necessarily competing with non-athletes for spots in the student body. But they *are* competing with each other. Whether you use an exam, an interview, or something else to select who gets to attend, you hope that your process selects students that have the potential to succeed in your academic environment. I’m sure there are kids who would be great college athletes but aren’t prepared for college academics. You then have to make the decision about whether you will watch out for those students and give them lots of support, or if you just move on to the next person. It would be nice if there were a cut and dried answer to this, but I don’t think there is.

  16. Peter
    Posted December 12, 2019 at 12:42 pm | Permalink

    Two psychology professors weigh in on the topic:

    David Z. Hambrick And Christopher Chabris: Yes, IQ Really Matters
    Critics of the SAT and other standardized testing are disregarding the data.
    https://slate.com/technology/2014/04/what-do-sat-and-iq-tests-measure-general-intelligence-predicts-school-and-life-success.html

    • Posted December 13, 2019 at 11:48 am | Permalink

      The problem with *some* critics of *IQ* testing in specific is that an IQ test has to be administered in a specific context, by a psychometrician or psychologist. School guidance counsellors are *not* in general such people, so the “IQ tests” that many people claim to have done in school are nothing of the kind. Similarly these “other aptitude tests” can have been written by who knows from schoolboards, etc. and (potentially) with all the problems that *that* would entail.

  17. davelenny
    Posted December 12, 2019 at 1:06 pm | Permalink

    Ii will be interesting to see if the increased 2019-20 admissions of 24% and 56% for various groups are matched several years from now by similar increases in graduations.

    If there is a substantial difference downwards, then the principal effects of changing admission procedures are likely to be increased frustration and indebtedness of people least able to afford debt.

  18. Adam M.
    Posted December 12, 2019 at 1:08 pm | Permalink

    Finally, revealing the big gap in scores between groups, with Asians at the top followed by whites, Hispanics, and blacks, exposes the embarrassing inequalities in American society.

    I don’t think we need to be embarrassed. These gaps exist all over the world and nobody has been able to close them, so I don’t believe they are best explained by anything particular to American society.

    • Posted December 12, 2019 at 1:22 pm | Permalink

      Well, for blacks they are the residuum of slavery, and I think we should be embarrassed about that, or at least, since we’re not practicing it any longer, do something about the legacy of slavery.

      • KD
        Posted December 12, 2019 at 4:10 pm | Permalink

        By giving racial preferences to the children of Nigerian doctors?

        • KD
          Posted December 12, 2019 at 4:14 pm | Permalink

          I don’t see anything wrong with affirmative action for actual Native Americans or the descendants of American slaves. That is, of course, not “diversity”, nor is it affirmative action as practiced by colleges and employers.

  19. rickflick
    Posted December 12, 2019 at 1:11 pm | Permalink

    It occurs to me that this issue is made much less important by the fact that there are many college opportunities across the country with varying levels of student expectation. If you can’t get into Harvard, so what? You can go to your local state university. If you can’t get in there, try the community college route. Once you’ve established you ability to navigate college level learning, you can still shift around to some extent to more prestigious institutions, if that’s your goal.
    So, yes someone will have a harder time getting hired by a top Manhattan law firm than some others based on the school they attended, but I don’t see that as a huge penalty. To some extent, life is what you make of it.

  20. Gareth Price
    Posted December 12, 2019 at 1:13 pm | Permalink

    On a slight tangent, I have had a lot of trouble trying to apply to graduate schools in the US because UK universities didn’t produce transcripts when I graduated. And, while I am not a lawyer, I have managed to convince myself that requiring transcripts might not be legal, given that almost the only people who can’t produce them are foreign nationals, which I believe is a protected class under the Civil Rights Act(?).

    • Posted December 13, 2019 at 11:50 am | Permalink

      Can’t you tell your UK school(s) to send a copy along to the US school(s)? That’s I did when I applied from Canada to both Canada and the US schools.

  21. merilee
    Posted December 12, 2019 at 1:45 pm | Permalink

    🐾🐾

  22. Posted December 12, 2019 at 1:56 pm | Permalink

    Let’s face it—the only thing SAT tests test is how well candidates do on SAT tests, which may have little to do with how well they do in college. Some qualified candidates across the board of race, economic background, etc. will score low on time-pressured tests. Still, this is useful information to know about a candidate, assuming that no admissions team worth its salt is going to make decisions based on SAT scores alone.

    There’s some evidence that candidates with test-phobia tend to do better on the more subjective essay section. But I’m reminded of the story about Niels Bohr, who, brilliant as he was, couldn’t grasp the concept of beginning, middle, and end: the last sentence in one of his student essays was “And I would also like to mention aluminum.”

    • Posted December 12, 2019 at 1:58 pm | Permalink

      Did you read the Slate article whose link I added after a commenter mentioned it? Why don’t you do that before you make statements like “the only thing SAT tests test is how well candidates do on SAT tests, which may have little to do with how well they doin college. Seriously, you seem to be saying what you’d like to believe without any indication that you know the literature.

      https://slate.com/technology/2014/04/what-do-sat-and-iq-tests-measure-general-intelligence-predicts-school-and-life-success.html

      • Posted December 12, 2019 at 3:13 pm | Permalink

        When I said that low SAT scores “may have little to do with how well they do in college” I meant that lacking the skill set required to perform well on time-pressured tests doesn’t necessarily preclude their having other qualities (imagination, aspiration, work ethic, etc.) that might compensate for this. The article you refer me to talks mainly about how high scores on SATs correlate with a high GPA in college, which isn’t surprising given that GPA is itself dependent in part on one’s ability to perform well on tests.

        You’re right, of course, that I was relying more on my own experience and intuition than on empirical evidence (big surprise!), but that I wasn’t dissing the SATs should have been clear from my comment that a poor performance on an SAT, if taken in concert with other indicators, “is useful information to know about a candidate”—e.g., in terms of making resources available to them if they’re admitted. On the whole, in fact, I thought I was agreeing with you.

  23. max blancke
    Posted December 12, 2019 at 2:57 pm | Permalink

    Admitting kids to conventional university programs when they are unprepared for the course of study helps nobody.
    I get the sense that an argument is being made that universities primarily use admission standards in order to exclude minorities, rather than excluding those who are unprepared to do the work.

    Perhaps we should have different admission standards for kids who are not going to pursue a traditional degree program. The grievance studies kids are not going to need the math background or the study habits that a student studying microbiology is going to need.

    This whole agenda seems to work like a virus. They might have very benign intentions, but the end result seems to always involve lowering standards to achieve equality of outcome.

    • ladyatheist
      Posted December 13, 2019 at 8:50 am | Permalink

      The GPA and recommendations from teachers are better predictors of success. If a student has a mediocre GPA and is very smart, they still have the option to demonstrate their ability this through test scores.

      For kids in low-performing schools, I think the SAT would still be a good choice, since they would have that to back up a GPA that might be looked down on because of the reputation of the school.

  24. Pim Wiersinga
    Posted December 12, 2019 at 4:22 pm | Permalink

    Generally speaking (as I’m at a distance from both the US and the age group), I would say that preparing aspiring students better is to be prefered over any lowering of standards. Th state should invest in this. ‘Socialism light’ is the way to dispel the halo of wokefulness surrounding such issues at one stroke. There’s a diabolical dynamic going on between meagre secondary education and unequal opportunities for kids to obtain private tutoring — which ought to be unnecessary.

  25. Posted December 12, 2019 at 7:51 pm | Permalink

    Students with family income > $200K/y have a 1:5 chance of scoring 1400 or better.

    For < $20K/y it's 1:50.

    That's one hell of a correlation.

    • rickflick
      Posted December 12, 2019 at 8:17 pm | Permalink

      Nature or nurture?

    • max blancke
      Posted December 12, 2019 at 10:05 pm | Permalink

      What if it works the other way? What if people with better reasoning skills and a strong work ethic are both likely to pass those characteristic to their kids and meet with success in life?

      Of course some people end up wealthy despite themselves, and plenty of smart, hard working people have adverse experiences. But in an egalitarian society with a strong economy, it would seem likely that such people would be successful. At least on average.

      • ladyatheist
        Posted December 13, 2019 at 8:45 am | Permalink

        So you think George W. Bush & Donald Trump are actually very smart because their parents are rich?

  26. Charles Sawicki
    Posted December 12, 2019 at 8:14 pm | Permalink

    “The University of Chicago, a top-ranked school, announced last year that it was going test-optional.”
    This is a very bad idea that will further degrade university education. The SAT math test covers pretty basic material. If you don’t do well, it suggests you will almost certainly have problems in any hard science.
    All the effort to help minorities and poor people by avoiding these standard measures is misdirected and will set people up to fail after admission.
    The only way to help the underprivileged is to stop avoiding the root problem and spend the money and invest the effort over a long period of time to improve their circumstances and educational opportunities. A big cut in defense would provide some of the required money. As a bonus, it would make stupid useless wars (Vietnam, Iraq and Afghanistan) harder to start.
    This isn’t a simple problem than can be fixed with patches applied after it’s largely too late for people to succeed.

  27. Posted December 12, 2019 at 8:25 pm | Permalink

    I went to graduate school in Economics at the University of Chicago back in the seventies. They used a winnowing model. They took in lots of students, like me, who could not probably get into Harvard. But the first year was brutal. Typically, only half the students survived to go on to second year. Some claimed it was cruel, but I thought it gave many a chance to prove themselves (then again, I made it through.)

    I wonder whether this model can be extended to undergraduates. Double the size of the entry class so students who don’t have stellar GPAs or 99 percentile SATS can get in, and then fail any students out if they do not perform.

    Contrast this to the current system where the tough part is getting in, but once you’re in everyone gets As or Bs.

    • Charles Sawicki
      Posted December 12, 2019 at 8:37 pm | Permalink

      The better state schools used similar systems. Unfortunately, this approach is no longer politically viable.

      • Posted December 12, 2019 at 8:43 pm | Permalink

        Why? Just asking. Because voters are angrier if their kids get failed out than if they are not let in?

  28. ladyatheist
    Posted December 13, 2019 at 8:43 am | Permalink

    I tend to agree with this direction. I found it a bit shocking at first, but there is evidence that GPA and transcripts are a much better predictor of success in college.

  29. Posted December 13, 2019 at 11:54 am | Permalink

    I have three problems with the SAT and GRE and such.

    (1) It does make for an economic inequality – both for the prep courses (that is an arms race against other students – after all, relative standing usually is what matters) and for the actual taking and sending of the test scores.
    (2) Worse, the ETS is a private corporation! This should be a service of the state, at least for public schools.
    (3) I am not convince they test the right thing – the GRE seemed to be a “race against time”, which I do not think should really be a thing. Surely we want people who think and have acquired skills that work? (Up to a point, of course.)


Post a Comment

Required fields are marked *
*
*

%d bloggers like this: