4 ms·
> the SAT is increasingly considered "racist" because it reveals racial disparities in learning Could you provide an example of that claim? Usually the SAT is
by wolverine876 5y ago
> the SAT is increasingly considered "racist" because it reveals racial disparities in learning
Could you provide an example of that claim? Usually the SAT is considered discriminatory because the way the testing is done favors wealthy, well-educated people, and gives negative results for equally talented people who lack those advantages. Colleges have found that the SAT is excluding a large part of the talent pool.
Also, SAT scores do not correlate well at all with college outcomes.
Why then use the SAT?
- ryan93 5y agoSAT math scores combined with GPA are 5 times more predictive than gpa alone at Berkeley. They also found the test was more predictive for low income students
- sjg007 5y agoFor the SAT the difference between winning and losing in college admission is if you can prepare for it. It's not a test you take cold. That is the biggest issue with it so it advantages those with more resources.
- walkhour 5y agoSAT correlates well with college performance and moreover must be used to obtain the best college performance predictors we have (or something highly correlated with SAT).
- stonogo 5y agoThe same task force you're quoting also recommended the SAT be revamped to reduce the inherent bias among the questions.
- ryan93 5y ago“black/white comparison on the SAT verbal test, some evidence of bias exists, but the bias is against white students on some questions against black students in other cases. Furthermore, our analysis of the results suggest that for this one, the effects are far too small to explain much of the SAT gap in test scores between black and white students.“ nonetheless the SAT math lets them find students who’s GPAs aren’t super high for personal reasons and also distinguish between students with the same GPA
- rayiner 5y ago> Could you provide an example of that claim? Usually the SAT is considered discriminatory because the way the testing is done favors wealthy, well-educated people, and gives negative results for equally talented people who lack those advantages. The alternatives, such as essays, are even more highly correlated with wealth: https://cepa.stanford.edu/content/essay-content-strongly-related-household-income-and-sat-scores-evidence-60000-undergraduate-applications https://cepa.stanford.edu/content/essay-content-strongly-rel... As to predicting outcomes, that’s largely due to score compression (the students at any given school are in a narrow score range). Folks across the score range aren’t being compared with each other in class performance. But look at the LSAT, where all students from different schools take the same bar exam. LSAT is highly predictive of bar exam performance. Students who score less than 150 (on a 120-180 scale) are virtually guaranteed to fail the bar. > Colleges have found that the SAT is excluding a large part of the talent pool. No, colleges have found that the SAT produces racial demographics they don’t like. Poor white kids have been excluded by the SAT for nearly a century and colleges never took action in response to that.
- judge2020 5y ago> LSAT is highly predictive of bar exam performance. Students who score less than 150 (on a 120-180 scale) are virtually guaranteed to fail the bar. Source? Or do you mean "taking LSAT at the same time as taking the bar for fun"? Because LSAT is typically done before you enter law school, while you take the bar after you get out.
- DiggyJohnson 5y agoI don’t have the source, but when I checked this out it captures LSAT performance and Bar Exam outcome at their normal milestones.
- wolverine876 5y ago> No, colleges have found that the SAT produces racial demographics they don’t like. Where does that come from, other than repetition? > look at the LSAT I'm not talking about the LSAT (and bar exam performance isn't a meaningful indicator of much - the great majority of aspiring lawyers pass their bar exams). > Poor white kids have been excluded by the SAT for nearly a century and colleges never took action in response to that. If that's true (and I don't know that it is, especially on the scale of what has happened to blacks), what does that does that have to do with racism against black kids? It seems everyone's great effort is not to address racism, but to deny it, against incredible evidence. Look at this entire discussion. It's incredible the effort that goes into denying racism, rather than doing something about it. Reactionary politics has swept the US and HN.
- civilized 5y agoAs the other comment points out, the obsession with the flaws of the SAT completely ignores the fact that every single other method of evaluation is even worse, both at identifying promising students and at equity. If you take it away, colleges will happily admit loads of "well-rounded" students with tons of expensive extracurriculars. They will be racially as balanced as possible, but their family incomes will all be very high. The kid from a poor broken home who somehow managed to score 1400 on the SAT will lose their best shot at standing out.
- hnthrowaway0315 5y agoExactly. It is still the least evil, one of the reasons Chinese are still keeping their infamous GaoKao. Removing it will create a huge political hitback.
- hackernewds 5y agoThis is the same argument supporting capitalism (which I agree with). Capitalism is the worst system, but it's still better than all else. I think taking away the SAT as criteria will actually enhance racial bias through subjective measures - than the flip
- wolverine876 5y agoThere's plenty of research you can read, as well what colleges have published. What do you think of it?
- wolverine876 5y agoAFAIK, the SAT is biased against poorer students, including in its structure, in its content (maybe they've fixed that), and because wealthy test-takers can afford test prep courses. What makes you say that the SAT is the best option for poor applicants?
- civilized 5y ago> AFAIK, the SAT is biased against poorer students, including in its structure, in its content (maybe they've fixed that), and because wealthy test-takers can afford test prep courses. The idea that the SAT is inherently biased against poorer students is controversial. AFAIK, the evidence for this requires advanced statistical analysis and expert interpretation and isn't widely accepted as definitive. Even among those experts who accept this evidence, there doesn't seem to be any clear sense of what is causing this bias or what can be done about it. It simply isn't the case that the SAT has some glaring bias that can easily be rectified, like questions assuming knowledge of water polo and horseback riding. Student skills in things like reading and math do not exist in a vacuum, but depend on prior exposure and knowledge, which robustly correlates with family/neighborhood/school environment and thus socioeconomic status. Efforts to create tests that are completely independent of those things have not produced useful results. Quality studies show that the benefit of test prep is modest to moderate, on the order of 30-100 points, and almost all of the benefit is achieved within 8 hours of preparation. Even that benefit is not widely accepted as simply retaking the test can increase scores by 60 points. [1] There is no evidence that expensive private tutoring leads to better outcomes than self-study with free alternatives such as Khan Academy. [2] > What makes you say that the SAT is the best option for poor applicants? It's helpful to look at it more as the "least bad" than the "best" -- we need to stop focusing exclusively on the issues with the SAT and think about the actual alternatives. For example, essays, extracurriculars, and prestigious brand-name private school affiliations are all valued by prestigious colleges, and can be bought legally and very easily by rich families. It is much harder to buy a high SAT score. To buy a high SAT score, a family could pay someone to impersonate their child, but this is very risky as the recent Varsity Blues scandal showed. Suppose we put ourselves in the shoes of a bright-but-poor high school student. We need a way to make ourselves stand out. We can sign up with Khan Academy or download some SAT prep materials at the public library and prepare for a few hours... or we can try to get into an expensive private school, get a private college essay consultant, join a rich-kid sport, take a trip to Africa, and other things that rich kids do to burnish their applications. Which of these seems more practical to you? If you look at the criticism of the SAT from universities, I think you'll find that they are mostly interested in attacking the SAT in isolation, not comparing with the alternatives. Comparing with the alternatives would require shedding light on their opaque and subjective admissions processes and the way in which those processes favor the rich. On the contrary, research suggests that leveraging standardized tests, and increasing participation in them, tends to open doors for poor students relative to status quo [4][5]. It's interesting that they are so favorable to a change which will make their processes even more subjective and opaque, especially given their history of using subjective and opaque processes to exclude undesirable minorities (yesterday Jews [3], today Asians). Although they have proven themselves untrustworthy, they are not interested in increasing transparency and objectivity of admissions processes. I think we should be hesitant to accept their reports and recommendations as being in completely good faith, given their strong interest in (and consistently observed behavior of) recruiting as wealthy a class as possible each year. [1] https://slate.com/technology/2019/04/sat-prep-courses-do-they-work-bias.html https://slate.com/technology/2019/04/sat-prep-courses-do-the... [2] https://www.seattletimes.com/explore/special-sections/sat-prep-courses-does-free-mean-better/ https://www.seattletimes.com/explore/special-sections/sat-pr... [3] https://scholar.harvard.edu/files/dobbin/files/2007_asq_karabel.pdf https://scholar.harvard.edu/files/dobbin/files/2007_asq_kara... [4] https://www.nytimes.com/2016/04/10/upshot/why-talented-black-and-hispanic-students-can-go-undiscovered.html?_r=0 https://www.nytimes.com/2016/04/10/upshot/why-talented-black... [5] https://www.nytimes.com/2017/07/14/upshot/how-universal-college-admission-tests-help-low-income-students.html https://www.nytimes.com/2017/07/14/upshot/how-universal-coll...
- skybrian 5y agoRather than saying "the SAT is discriminatory," maybe ask which questions on the SAT's are discriminatory, and why? And how could you fix them? This will be more revealing of what the problem might be. I believe a lot of problems that the SAT's used to have were fixed over the years? It's somewhat of a moving target. And if you can't make any standardized test that's fair, why is that?
- crackercrews 5y agoA prominent researcher noted that the following question was tested by the ETS but never made it onto the SAT: The actor's bearing on stage seemed ______; her movements were natural and her technique _____ A. unremitting...blasé B. fluid...tentative C. unstudied...contrived D. eclectic...uniform E. grandiose...controlled Apparently 8% more black students answered this question correctly than white students, and it was never moved from a potential question into the real test. The allegation is that the racial outcome of the question is the cause of its never making it onto the SAT. I think it never made it because the question is bad. None of the answers makes much sense (the correct answer is apparently C). 1: http://www.jayrosner.com/publication-onwhitepreferences.html http://www.jayrosner.com/publication-onwhitepreferences.html
- wolverine876 5y agoTo be clear, the thesis of the article you quoted from is the following: Every single question carefully preselected to appear on the test favors whites over blacks.
- throwuxiytayq 5y agoAs a non-American, this thesis simply doesn't pass the smell test unless you guys are living in literal KKK land, which - from be outside - doesn't seem to be the case.
- wolverine876 5y agoWhy not read the research? A 'smell test', especially by someone with no experience or expertise, doesn't indicate much. Any claim of racism - even raised by an expert who has done detailed research, in a society with overwhelming evidence of it historically and now, and in higher education admissions in particular - is always dismissed.
- chinchilla2020 5y agoSAT is one of the strongest correlations of academic success and future income. Where did you hear that it isn't? I worked in edtech analytics for a while where SAT among other factors were correlated with student success. SAT scores correlate with graduation rates, college GPA, job placement. Most elite universities also tend to receive applicants with higher SAT scores (A no brainer)
- wolverine876 5y agoI read research that it wasn't, and I believe many colleges claimed that it wasn't. Since neither of us is citing anything, I will keep an eye out. > Most elite universities also tend to receive applicants with higher SAT scores Many don't use SAT scores anymore.
- Jtype 5y agoI think you read propaganda.
- lifeplusplus 5y agoIt seems SAT merely reflects current academic skills. If they developed to that point due to privilege, the test isn't responsible for highlighting it. If college goal is to give limited space to most skilled students then it's an effective method to do that.. isn't it?
- wolverine876 5y ago> It seems SAT merely reflects current academic skills. What is that based on, and how do you define academic skills? It may reflect how much time and money you can spend on test prep courses (time is also a problem for poor students who tend to have jobs, take care of siblings for parents with multiple jobs and no day care, etc.). It may reflect test-taking skills in general, which aren't skills of value.
- lifeplusplus 5y agoThat's a possibility, that there is a difference between test taking skills and true academic skills, in that case maybe have students take one class and admission would based on percentile... ie. 1000 student in top percentile will be accepted full time other passing students will get a credit but not admission.