New UC study: SAT doesn't predict college success

Test Preparation
1

<p>Cash cow in trouble? Here we go again.</p>

<p>"For nearly 100 years, the SAT has been used by universities in their admissions process, but a new report out of the Center for Studies in Higher Education at UC Berkeley argues that the SAT is an inferior predictor of student success in college.</p>

<p>"The report, "Back to Basics: In Defense of Achievement in College Admissions," contends that admissions criteria which assess student mastery of curriculum, such as grades and achievement tests, are better indicators of a student's success in college than criteria that assess a student's ability to learn, like the SAT. The report was the conclusion of a series of studies conducted by the University of California over the course of 10 years."</p>

<p>California</a> Aggie // Jul 21, 08 // UC study shows SATs do not predict college success</p>

2

<p>What I don't understand about these discussions is why the ACT is never mentioned. Do the admissions committees believe the same about the ACT, or is it considered a better test? I know the ACT isn't popular in California, but the UCs never seem to place much value on its existence.</p>

3

<p>Maybe the ACT secretly funds these studies ;)</p>

<p>Kidding aside, this article mentions the SATs weaknesses in comparison to more traditional achievement tests, which I believe the ACT mirrors more closely. That may end up supporting that the ACT is a better examination, though I agree that they should have at least mentioned it somewhere.</p>

<p>I thought it was largely accepted that GPA alone was a stronger predictor of college success than any other single variable. This doesn't seem to mention whether or not some combination of GPA and SAT scores (with lower weighting, of course) predict college success more effectively to aid in finding that "diamond in the rough."</p>

4

<p>I don't think this is new. I've heard the same thing: "The SAT does not predict college success, the most accurate predictor of college success is success in high school" for YEARS - going back to when I did alumni admissions work in the 1980's. My only question is why is the SAT still around? I think "cash cow" is exactly right. Between the College Board, Princeton Review, Kaplan, and hundreds of small companies charging thousands of dollars for SAT review classes, and the fees the CB charges for every single little "service" they provide, this is a HUGE industry. Parents are desperate to get their kid into college so they're afraid to buck the system. But I'm convinced the only people benefitting from this system are the people making money from it (and maybe large colleges that are using SAT scores as a quick-and-easy way to do an initial screening/weed-out of applicants).</p>

<p>My sister-in-law teaches SAT prep. She says the test is coachable, and the kids she teaches do better on the test but they're not any smarter or better students because she's tutored them - they're just good and figuring out what the SAT wants and how to give it to them.</p>

<p>And BTW, I had a 1350 on the old SAT, and my son had a 2150 on the new one. With NO money spent on SAT prep. So this isn't sour grapes.</p>

5

<p>I taught the SAT many years ago. There is no question in my mind that the SAT has a very slight predictor of success at MOST, and in many cases, it actually has a negative correlation.</p>

<p>I have seen far too many kids who I knew were top notch, yet not do well on the test. The main reason is the timed nature of the test. There are a LOT of thoughtful kids who simply like to think things through. Einstein was one of these types of kids and would have been adversely affected.</p>

<p>Likewise, I see kids who score well and I know won't do well in school for a variety of reasons. Frankly, most people who have taught the SAT, that I have known, would have agreed with my sentiments. It is the colleges that are fooling themselves. Luckily, the colleges are waking up and realizing that the SAT isn't that good a predictor of college success.</p>

<p>As for the ACT and subject matter tests,I do believe that they are better predictors from what I have seen. However, they have the same problem regarding the timed nature of the test.</p>

<p>However, with all that said, I still doubt the veracity of UC's study. Let's face it: UC has a vested interest in affirmative action and in bringing in more minorities. Anything that will accomplish that seems to be acceptable to them.</p>

6

<p>hahahaha wasnt it the UC system that caused the SAT to be reformed with a writing section, eliminated analogies, no math comparisons etc to form what we call the "new SAT" </p>

<p>so the UC system changed the whole test, and is now complaining about it?? i find that ironic</p>

7

<p>I think that it's important to understand what the SAT measures. Think about college football prospects who are evaluated by the NFL. Pro teams measure their 40-yard dash times and the number of 225 lb. bench press repetitions they can do. Those who do particularly well in these tests see their stock rise when the NFL drafts that year's new talent. But does speed and strength accurately predict who becomes a great professional football player? They're important tools to be sure, but work ethic, desire, and self-discipline ultimately determine who can turn those abilities into actual performance.</p>

<p>The SAT is a measure of learning facility, as opposed to the ACT which tends more to measure content mastery. A lot of the right answers on the SAT can be deduced even without full mastery of the content, and a number of questions can be answered incorrectly, even by those who know the content, because the question adds an unusual twist or pitfall. These skills are the 40-yard dash times and bench press reps of the academic genre. Will those who perform well on the SAT have the highest freshman year grades? Well, their work ethic, desire, and self-discipline will have a lot to do with that.</p>

<p>But I have no doubt that those who score well on the SAT have a greater capacity to contribute to the quality of a classroom discussion or challenge their peers in residence hall discussions. That's a real value to the college environment, probably moreso than whether their talents translate into "A"s or "C"s. To push the football analogy further, a 170-pounder with a great work ethic can master all the technical skills of the game and have his performance graded highly by his coaches, but will ultimately find a limit to the impact that he can have on the team.</p>

<p>I think that like foot speed to an athlete, SAT skills are valuable resources to a college student, and I'm not surprised that colleges would covet those resources just as football coaches covet fast players. But ultimately, they too are raw materials to be developed.</p>

8

<p>I would say that the ACT is pretty likely to predict college success since it tests what students know/have learned in high school, whereas the SAT tests how intelligent they are.</p>

9

<p>I completely agree with what gadad said, good analogy. SAT score alone won't prove who does well in college and neither will GPA. Both items show the capacity of what a potential student may be able to do but really it depends on how motivated the student is.</p>

10

<p>The SAT tests how intelligent 'they' are? You're kidding.</p>

<p>Intelligence comes is many forms. Many people, like a girl I tutor in math, are highly capable in one area, but lack the ability to perform well in another. She's a published author, but struggles with geometry proofs. Should she not be accepted to a great university because her math SAT scores are sub-par? </p>

<p>College classes are akin to high school classes. Indeed, students who work hardest receive the highest grades. Do you truly believe a lazy student with As and a 2400 will be more successful in the long-run that a hard-working student with As and a ~2100? Think again.</p>

11

<p>
[quote]
I thought it was largely accepted that GPA alone was a stronger predictor of college success than any other single variable.

[/quote]
</p>

<p>The huge disparities between schools, teachers, courses, facilities, peers, grading policies, etc make GPA by itself a near useless mechanism - surely no one's going to argue that a 3.5 GPA at Thomas Jefferson or Stuyvesant High School in classes that are taught at the college level is the same as a 3.5 in high school where where half the students drop out and most of the graduating seniors read 3 or more grades below level. </p>

<p>Note that what they said is: "mastery of curriculum, such as grades and achievement tests", which is far more holistic than merely GPA. </p>

<p>The SAT is obviously an imperfect device - but it is one of the few objective measures available to adcoms, and is just another data point they can use. Every college wants, in addition to SAT scores, GPA, class rank, and course selection / difficulty, and uses all of these (and other factors) holistically. I doubt there are many colleges that make admissions decisions based solely on SAT scores, in either direction, except in extreme cases. </p>

<p>And likely, if/when they actually release the raw data, you'll see that SAT scores <em>strongly</em> correlate with GPA. </p>

<p>What the data will probably show is that a combination of GPA + Class Rank + Standardized Test Scores (including SAT) + course difficulty + (arbitrary S.W.A.G.factor) is better than SAT scores alone.</p>

<p>Which is kinda like saying blood pressure + blood cholesterol + smoking history + weight + age + family history are better predictors of heart attack than blood pressure alone. It doesn't make BP a useless measure , does it? </p>

<p>But there will be plenty of people with various vested interests or who just like being contrarian, who'll run around claiming this study shows that SAT scores are "useless" in predicting college success. </p>

<p>
[quote]
And BTW, I had a 1350 on the old SAT, and my son had a 2150 on the new one. With NO money spent on SAT prep. So this isn't sour grapes

[/quote]
</p>

<p>I'm sorry -- but what does this mean? That you did well on the SAT, but fared poorly in college? That your son did well on the SAT, but had a 2.0 GPA and did poorly in college? Or that he had a low GPA and did great (or vice versa)? That the test is coachable, despite the fact that you had no coaching? One fact, hanging in the air, really doesn't provide much illumination. Kinda like any other fact, like GPA or SAT score. It has to have context.</p>

12

<p>Difficulty of high school courses varies from high school to high school though. Some AP and honors classes are heavily curved and some are not. Schools have different grading scales and GPA conversions. Standardized tests have to be a part of college admissions in order to better compare students on a "fair playing field". </p>

<p>However, this is where I agree with the study. Key word: success. The SAT is not the right standardized test for college admissions. If anything, subject tests, the ACT, and IMO AP tests are better predictors of college success. The AP tests are considered to be college level exams so why not use them in admissions? Ok there are hefty fees as a result of taking numerous APs and many schools offer limited AP courses, but I'm sure the AP system could be overhauled. Also, if AP scores would be used for admission the grading scale would have to be changed to better distinguish students. </p>

<p>Overall though, standardized tests are necessary for college admissions just as they are necessary for grad school. However, maybe the SAT isn't the right test. </p>

<p>Also, I've been thinking about this as I go through the college admissions process myself. How important is the college you attend as an undergraduate? What's the big difference between that top 20 school and your state school? It's really what you do with your undergraduate education and experience. Are you willing to find research/internship opportunities, get involved, seek help from professors, and most importantly study? If so, and if desirable, you can attend a nice graduate school with the hope of securing a successful career in the future. </p>

<p>Maybe we're overemphasizing undergraduate admissions. Will getting that 2200+ really ensure or predict well... anything? Will getting into Harvard complete your life or be a sure shot for success? Probably not. It's really what you do with what you have. If you are that determined and motivated, a standardized test should not hold you back. </p>

<p>Anyways, the SAT isn't a predictor of college success or any success for that matter, but will somebody please do a study on the ACT for once :) lol.</p>

13

<p>Didn't the UCs recently agree with the college board's study saying the writing portion of the SAT was the best predictor of success in college? While, when you look at the details of the study, HS gpa remains the best predictor and the writing score is only slightly better than the CR score when they stand alone. There is a thread on here about those studies. Of course I find any study by the organiztion that runs the test suspect.</p>

14

<p>What I find interesting is that only a few years ago, UC found that the SAT IIs were a better predictor of success than the SAT Is. Are they doing away with both?? I am a little confused.</p>

<p>The</a> Harvard Crimson :: News :: Fitzsimmons Defends SAT II</p>

15

<p>Again, the SAT IIs are about content mastery, as (primarily) is the ACT. But as both a college administrator and a college parent, I find that I have a special appreciation for the SAT I. Let me see if I can explain why.</p>

<p>When I was in college, I was keenly aware that some of my fellow students had special intellectual abilities. They could often recognize a strategy from among a maze of options. They were the ones you went to with questions or areas of confusion. They were often more interesting to talk to about issues because they'd bring into the discussion angles that hadn't occurred to you, but which broadened and enhanced the conversation. I always figured that most of them were good students, though some were actually party animals who were more interested in kegs than term papers. But I couldn't really identify what it was that I admired; as with Justice Stewart's celebrated observation of pornography, I couldn't define what it was I was seeing, but I knew it when I saw it.</p>

<p>Then, in the Eighties, I became a fan of the TV show "MacGyver" and it brought this intellectual ability more clearly into focus. MacGyver was like a superhero whose powers were creative rather than physical. On each show, he'd find himself in a desperate situation and had to invent an elegant solution with whatever he had at hand, just like the Mission Control scientists who inventoried everything aboard the Apollo 13 command module in order to rig up air filters to save the crew. In the terminology of the day, that kind of skill was labeled "out-of-the-box thinking."</p>

<p>When my own kids began taking the PSAT, I started doing SAT questions with them and realized anew that those kinds of skills were what the SAT particularly assesses and rewards. Instead of giving you a recipe and saying "make this," it often attempts to supply you with raw ingredients and ask "can you find something worthwhile to make from these?" For instance, a regular test on Math content would simply ask you how to find the areas of various shapes. An SAT question might show you a drawing of a rectangle into which an adjacent circle and triangle neatly fit. The space in the rectangle outside of the circle and triangle would be shaded and you'd be asked to find the area of the shaded portion of the drawing. The drawing would include the barest information necessary in order to deduce the radius of the circle. Knowing the radius, you'd then be able to find the base and height of the triangle, and ultimately the dimensions of the larger rectangle. Then, like MacGyver, you'd have to figure all three areas, add the areas of the circle and triangle together and subtract them from the area of the rectangle to get your answer. Other another question, you might be given all the data about the per minute cost of long-distance phone calls in cents and then be asked to identify which formula would give the cost of a phone call of X minutes - in dollars. Knowing the content without attention to detail produces a wrong answer. </p>

<p>SAT Critical Reading and Writing questions typically don't ask matters of fact, they ask you to find a thought or element of writing in a paragraph and compare it to another paragraph with with a slightly different take on the same topic. This type of skill is a very valuable performance skill for freshman English courses, but only if you care enough to go to the trouble of employing it and actually turning in the assigned paper.</p>

<p>I believe that on its face, the SAT measures something of real value which may or may not predict something definable later on. But I'd be interested in seeing whether it correlates better with junior year grades than with freshman grades. During the freshman and sophomore years, most college students are taking core courses, continuing the same topics they studied in HS but at an elevated level. The opportunities for creative intellectual thought may be relatively constrained at that point. When they're finally choosing and pursuing their majors, as juniors and seniors, we might see more of the logical and deductive abilities that the SAT measures come into fruition.</p>

16

<p>Quote:
"I have seen far too many kids who I knew were top notch, yet not do well on the test. The main reason is the timed nature of the test. There are a LOT of thoughtful kids who simply like to think things through. Einstein was one of these types of kids and would have been adversely affected."</p>

<p>People who do well on the test think things through FASTER than the thoughtful kids you've known. Also, BS on the Einstein tidbit. You have no idea how Einstein would have done on the SAT I.</p>

17

<p>Here's a direct link to the UC Berkeley research paper:</p>

<p><a href="http://cshe.berkeley.edu/publications/docs/ROPSGeiserBasics-07-10-08.pdf%5B/url%5D"&gt;http://cshe.berkeley.edu/publications/docs/ROPSGeiserBasics-07-10-08.pdf&lt;/a&gt;&lt;/p>

<p>Here's a link to the 2001 results from the same research group (basically the same results, obviously before the "new and improved" SAT):</p>

<p><a href="http://www.ucop.edu/sas/research/researchandplanning/pdf/sat_study.pdf%5B/url%5D"&gt;http://www.ucop.edu/sas/research/researchandplanning/pdf/sat_study.pdf&lt;/a&gt;&lt;/p>

18

<p>"People who do well on the test think things through FASTER than the thoughtful kids you've known."</p>

<p>Just because they are "FASTER" does not necessarily mean they are smarter or more intelligent. They are just faster on the test. Test taking can be a learned skill ... as all of the prep courses mention. </p>

<p>Einstein also needed someone by his side all of the time to do his calculations because he was so error prone. He got the design but could not do the mechanics of the calculations accurately enough.</p>

<p>There are MANY types of intelligences ... test taking is one of many.</p>

19

<p>SAT test is not an I test.</p>

20

<p>Woops ... SAT test is not an IQ test ...</p>