New UC study: SAT doesn't predict college success

<p>By the way ... epiphany, I love your posts ... very refreshing.</p>

<p>
[quote]
I do not think the schools use it to predict intellect. They use to see if you have developed the skill sets that seem to predict success. Let's say smartness vs intellect ... two different things in my mind.

[/quote]
Define smartness & intellect please.</p>

<p>Short and sweet ...</p>

<p>I think of intellect as an inherent intellectual capacity regardless of schooling/upbringing/ etc. What you were born with. </p>

<p>Smartness to me refers to a person's ability to take advantage of their given intellect and make use of it. </p>

<p>You can be smarter than a person with a higher intellect because you can make better use of what you have and be more successful. </p>

<p>I am sure there is a better way to articulate this but I am not as smart or intellectual as I would like to be! ;)</p>

<p>hmmmm ... faint feel of nature vs nurture???</p>

<p>
[quote]
What the UC study actually showed was that SAT1+SATII scores, combined, were consistently a better predictor of college success than HS GPA alone. They obscured that by never reporting that comparison, although it was present in the data. HS GPA was better than one or the other set of SAT tests, but never better than both combined. (Of course, there's a large area of correlation - kids with high GPA do tend to have correspondingly high test scores.)

[/quote]
</p>

<p>and if I remember correctly ... the differences between these variations was not very big ... and all of them were not particularly good predictors. So colleges are in a pretty tough spot ... they have a lot of pieces of info about applicants ... none of which alone is very good predictor of college performance ... and which seem to correlate to each other a lot. From the actions of the (the lion share of) colleges they seem to believe each piece of data helps them get a better handle on the applicants.</p>

<p>I don't think it really matters if the SAT is a good predictor or not at this point. I do agree that it is a coached test that is unfair to a point because you can coach it and the vocabulary you use in there you never use in normal life. However, there has to be some way of eliminating people otherwise colleges would take forever to determine who gets in. The SAT/ACT is a good way for them to automatically screen out some people (combined with other factors of course). What makes it different then you submitting your resume and it being screened out by a computer? You might have a 2.5 GPA and no internships in college but could end up doing better than that 3.5 GPA kid with tons of internships. It sucks that these tests are worth so much, but hey, until we come up with a better way to separate people, I think this will do.</p>

<p>Is SAT a great test? of course not. It is four hours long, and it is largely inane.</p>

<p>But this is a moot question. So what happens when we do take out the SAT? Do we rely on the GPA, which as we know, have pretty much no objective comparative value?</p>

<p>And effort does not equal desserts. I know of a lot of folks who put a lot more effort into English or History or whatever, for example, and it does not mean they deserve to get a better hand in becoming writers. Pure ability and aptitude are just as important, if not more so. Our opportunity distribution does have to make economic sense and not waste them in hopeless directions.</p>

<p>3togo - spot on. Not a strong predictive value from any of them; most predictive value comes from combining as much data as possible, and the standard of "success" against which the predictive value is gauged is less than compelling. College admissions is going to be a messy and imprecise art no matter what "standards" you try to use. I've given up complaining about it. (Although my daughter is a little peeved that kids she tutored got into schools she didn't. And the athletes she knows who are headed to Ivy League schools? Duck 'n cover! :) )</p>

<p>kluge:</p>

<p>perhaps my brain is a little foggy, but it was my impression that the original UC study showed that the former Subject Tests (includes M+W+elective) alone were the highest predictor of Frosh gpa...and that adding in gpa and/or SAT-1 only marginally increased that predictive value.</p>

<p>If so, there is a huge cost being perpetrated upon the state residents, many of which are poor, to take so many tests......</p>

<p>If the SAT didn't test intelligence, then everybody, with sufficient study time, should get a 2400, which would be ludicrous.</p>

<p>I think everyone ... within a certain level of basic intelligence ... probably could get 2400 with sufficient time ... more time for some. It is a useful tool though ... but not an IQ test.</p>

<p>It actually would be very interesting to have all of the 750 to 800 SAT test takers take an IQ test and see the spread of actual IQ.</p>

<p>MQD: That's very PC of you but, it's utter rubbish.</p>

<p>mia, no need to be so harsh.</p>

<p>What's utter rubbish? That SAT is not an IQ test? (It isn't; part of it once was.) Or that it would be interesting to see correlations side by side? Or that some test takers do as well with slightly more time? (Knowing that virtually no college-level test is like the SAT format -- for those of us who have actually attended selective colleges, done very well there, and even have graduate & professional degrees.) </p>

<p>She did say it was a useful tool. She probably meant in context, combined with other criteria.</p>

<p>disagree with mqd, about that many 2400s. A poor reader could take hours on the CR, but if they don't know vocab, the answer will never arrive. Ditto some math problems the classic average rate-distance problem (a train leaves Clevleland at the same time a train leaves Chicago), or planar geometry...if you know the xiggi formula, you can respond in 30 seconds. If you don't know the formula, it could take literally hours to create and solve.</p>

<p>Now, I would suggest a higher ACT score is achievable with more time if only bcos the ACT has less 'reasoning' imbedded in it.</p>

<p>Yes, well I do agree with that, blue. I was mainly referring to the other comments. SAT and ACTs are two different subspecies of a species. However, either one is in itself a poor predictor of college success. That's why students & parents would do well to remember the other 14 categories also important in college selection, as stated on the "geography" thread. </p>

<p>The bottom line, folks, is that if the colleges, including Elites, really found them as pertinent, decisive, predictive as some people believe them to be, the accepted scores would be in a very narrow band (narrower than currently), and the other factors listed on the CDS would mostly disappear or recede.</p>

<p>
[quote]
The bottom line, folks, is that if the colleges, including Elites, really found them as pertinent, decisive, predictive as some people believe them to be, the accepted scores would be in a very narrow band (narrower than currently), and the other factors listed on the CDS would mostly disappear or recede.

[/quote]
Predictive of what? And what makes you think that is actually the college's goal?</p>

<p>Thank you epiphany :)... my skin is thick ... I actually find this entertaining.</p>

<p>MIA ... read carefully</p>

<p>"I think everyone ... within a certain level of basic intelligence ... probably could get 2400 with sufficient time ... more time for some. It is a useful tool though ... but not an IQ test."</p>

<p>... certain level of basic intelligence ... which could be the top XX% crowd.</p>

<p>also ... "with sufficient time" ... It might take someone a few years but ...?</p>

<p>How can someone improve by 200 points after taking one prep class if it isn't a skills test. You can not improve on an IQ test. </p>

<p>I'll bet the disorganized do worse than the organized on the SAT ... is that a sign of superior intelligence if you are organized??? ... think of the absent minded professor. Being organized will certainly help you be a success in school and life ... but it has nothing to do with IQ.</p>

<p>I thought I was being the opposite of PC ... at least on CC. </p>

<p>As I have said before there are many types of "intelligences" the SAT tests a segment but is limited in scope and leaves out many other segments. </p>

<p>Getting a 2400 requires a certain level of intelligence and it is possible some of this higher IQ crowd does not test well ... so with practice they can get a 2400. Some folks will never get there. A person with 2050 could be as intelligent as the 2400 person ... but might be disorganized and over thinks the test or has ADHD and can not sit for so long. </p>

<p>SAT is a useful test but it is not an IQ test.</p>

<p>Post 96:<br>
Admissions is most definitely in part about predictability. An admissions ticket is not just an award for achievement, and high school is not college. You're not going to a second high school. You're entering a different, more independent environment, where a different configuration of factors will affect success.
Historically, predictability has been especially important to Elites. Read the history of admissions at the Elites. If anything, what was discovered was that long-term achievement in those extracurriculars which required discipline was determinant of success in and after college. Since the Elites care about how their students do after college, as much as during, you can bank on them looking at predictability factors.</p>

<p>How</a> to estimate your IQ based on your GRE or SAT scores</p>

<p>This chart correlates GRE or SAT scores to IQ.
You can click on the link to SAT 1 comparison for recent SAT scores, or there are other links for SAT's taken longer ago. </p>

<p>It is dead on accurate for my son (who has bad grades in H.S.)</p>

<p>SAT 1 (CR + M) = 1430 taken Jan 2008
IQ = 140 (tested at age 9)</p>

<p>Is it accurate for anyone else here?</p>