Should the average SAT scores of a school help you in your college search?

College Search & Selection
1

<p>How important are average SATs in a school? What do SAT scores say about individuals and how should they be used when trying to find a school?</p>

<p><a href="http://www.pbs.org/wgbh/pages/frontline/shows/sats/etc/script.html%5B/url%5D"&gt;http://www.pbs.org/wgbh/pages/frontline/shows/sats/etc/script.html&lt;/a&gt;&lt;/p>

<p>"NICHOLAS LEMANN, Author, "The Big Test": The level of obsession over these tests is way out of proportion to what they actually measure. People don't realize this. It's not built to measure your innate worth or anything like that. It's built to predict 15 percent of the variants in freshman year grades in college. I mean, it's a fairly small thing for a test to do, is predict 15 percent of the variants in freshman year grades in college. But it does that.</p>

<p>"BOB SCHAEFFER, National Center for Fair and Open Testing: The sole scientific claim made by the SAT - when you get down to the bottom line and strip away all the rhetoric and nonsense - is its capacity to predict first-year grades. Well, young women get higher grades than young males across the country in colleges despite the fact that they earn lower SAT scores by about 40 points, on the average.</p>

<p>There's only two ways to square that circle. Either all the colleges in the country are wrong, they're biased towards girls and give them higher grades than they deserve, or there's something fundamentally flawed about the test."</p>

<p>"JOHN KATZMAN, Princeton Review: This is a test where everybody's saying, "Look, we're just being an incredibly fair society here. Everybody takes this test. And the better kids go to the better schools." And it's just bull****. You know, the better kids hire me."</p>

<p>"BOB SCHAEFFER, National Center for Fair and Open Testing: How do you know that the scores that you're seeing, whether they're the result of some kid walking in and taking the test cold on a Saturday morning, and the results of some other kid who's been tutored for $700 at the Princeton Review or Kaplan, or $1,500 for some tutor who comes to your house and drills you on the test? Those scores don't mean the same thing."
[<a href="http://www.pbs.org:%5B/url%5D"&gt;www.pbs.org:&lt;/a> Read reports on the score"</p>

<p><a href="http://www.fairtest.org/facts/satfact.htm%5B/url%5D"&gt;http://www.fairtest.org/facts/satfact.htm&lt;/a&gt;&lt;/p>

<h2>SAT Myths</h2>

<p>The Test Is a Common Yardstick</p>

<p>After years of describing the SAT as a "common yardstick," the test-makers have now flip-flopped, claiming "it is a myth that a test will provide a unitary, unequivocal yardstick for ranking on merit." The SAT has always favored students who can afford coaching over those who cannot, students from wealthy suburban schools over those from poor urban school systems, and males over females.</p>

<p><a href="http://cshe.berkeley.edu/publications/publications.php?id=265%5B/url%5D"&gt;http://cshe.berkeley.edu/publications/publications.php?id=265&lt;/a&gt;&lt;/p>

<p>(1) HSGPA is consistently the strongest predictor of four-year college outcomes for all academic disciplines, campuses and freshman cohorts in the UC sample; (2) surprisingly, the predictive weight associated with HSGPA increases after the freshman year, accounting for a greater proportion of variance in cumulative fourth-year than first-year college grades; and (3) as an admissions criterion, HSGPA has less adverse impact than standardized tests on disadvantaged and underrepresented minority students. The paper concludes with a discussion of the implications of these findings for admissions policy and argues for greater emphasis on the high-school record, and a corresponding de-emphasis on standardized tests, in college admissions.</p>

2

<p>Bates has looked at its students, those who took the SAT and those that didn't. </p>

<p>Bates didn't see much difference in the students.</p>

<p><a href="http://www.bates.edu/x150430.xml%5B/url%5D"&gt;http://www.bates.edu/x150430.xml&lt;/a&gt;&lt;/p>

<p>Does this mean SAT optional schools like Bates are made up of students with lower than average SAT scores (compared to their peers) and should be avoided?</p>

3

<p>Would going to school with the following students hurt your education?
<a href="http://www.berkeley.edu/news/media/releases/2003/10/23_admissions.shtml%5B/url%5D"&gt;http://www.berkeley.edu/news/media/releases/2003/10/23_admissions.shtml&lt;/a&gt;&lt;/p>

<p>Profiles of 41 students with SAT I scores of 1000 or below</p>

<ol>
<li><p>Weighted GPA 3.86, honor roll student each year of high school at API 1 school. Major scholarship winner. Active on student council (president, junior year), student representative to school district superintendent’s school advisory council, ROTC cadet. Low income ($12,000), family of five, parents high school graduates, single-parent family, took care of disabled grandparent on dialysis, lived independently away from parents during part of high school, worked 20 hours per week to pay rent and utilities.</p></li>
<li><p>Weighted GPA 4.20. Student was in top 4 percent of high school graduating class despite student being an immigrant and no English spoken at home. Participated in a rigorous internship program as part of academy curriculum. Two-year varsity athlete and numerous volunteer activities. Low income family ($15,440), with neither parent having education higher than high school graduation. </p></li>
<li><p>Weighted GPA 3.75, received scholarship to attend prestigious out-of-state private school summer program, active in four outreach programs, perfect attendance during high school, received various academic school awards, non-native speaker of English. Very low income ($1,365), family of 10, parents no high school, father on disability (formerly a farm worker), student worked 20-25 hours per week while achieving high grades.</p></li>
</ol>

4

<p>In my personal experience, I've been able to predict very accurately who would and wouldn't get high SAT scores, using my assessment of their intellectual ability.</p>

<p>I want to go to school with people whose average intelligence level is either the same or a little lower than mine, but not by much. I use average SAT scores to determine what schools fit that bill.</p>

5

<p>It must be noted that while a tutor helps alot, a simple online review/book review acomplishes the same thing.</p>

<p>I too can sense which kids will score high/low on the SAT. </p>

<p>While it's true that some kids are bad testers, the majority can take a test just fine. What IMO, the sat does, is it separates overachievers/kids who just learned info for the test from those who truly learned the concepts they were taught.</p>

<p>While a 4.0 gpa is a great achievement, I would rather attend a college with 3.8 students who earned that grade through a balance of intelligence and hard work, and not just an incredible amount of hard work.</p>

6

<p>A few weeks ago I heard a talk by the Dean of the School of Engineering at the private university my son attends. He announced that the new freshman class was the "best" and the "strongest" the school had ever seen. The evidence: the highest average SAT I score ever. No other evidence was offered. </p>

<p>Even if Deans of Admissions at some schools are ready to toss the SAT I, academic deans seem not to be. </p>

<p>I don't disagree that all the prepping, all the practice, the retakes and the superscoring reduce the usefulness of the information, but there are many within the ivory walls who still think it is meaningful as a predictor of who can handle a given curriculum.</p>

7

<p>dstark, I am familiar with the UC study, and have no problem with UC's decision to use HS GPA as the primary consideration in admission. I feel that, for a variety of reasons, that is a defensible policy decision.</p>

<p>But the results of the study are, in my opinion, misrepresented. Absent an "adjustment" factored into the results the combined SATI and SATII scores are a better predictor of college "success" (as defined in the UC study) than are HS grades. It's only if you separate out the different tests and compare them individually that the HS GPA excels. And the obvious (and unadjusted-for) external influences on the outcome (SAT math scores were inversely predictive for collegiate success - think about that for a while) were essentially ignored.</p>

<p>I accept that some bright kids are just "bad test takers". But I don't consider that to be a significant factor on a mass scale. I'll go out on a limb here: I think higher test scorers are, generally speaking, brighter than lower test scorers. A school with higher average SAT scored in their enrolled class will have brighter students, generally speaking. So, yes, I think a college's average SAT (or 25-75 percentile scores) really should be helpful in both college searches and admissions decisions.</p>

8

<p>In some cases, a family/student should consider SAT scores carefully. For example, a lot of applicants on cc are premed. Med schools care first and foremost about gpa. Attending Caltech, where the average SAT math score is ~799 (dunno the actual number, just making a point), and your math score is 600 will put you at a significant disdvantage gradewise. Goodbye med school dream. OTOH, that 600 student might do very well at a midtier UC.</p>

9

<p>Should you use average SAT scores as one way to decide where to apply? You don't have to be a literal slave to it, but you better have some idea where you fit unless you want to send out numerous "hail mary" applications and start praying. </p>

<p>I don't think it's enough to look at the averages as listed in the USNWR, though. I looked at a few potential colleges more closely by perusing the CDS data, available in a link under the "168 Notable Colleges" thread started by tokenadult and at collegenavigator.ed.gov. </p>

<p>Certain patterns do jump out. Quite a few of the liberal arts colleges really prefer high CR scores and are more lenient irt to math scores. For example, for Reed College, an applicant would get a big clue about their chances by looking at the SAT CR scores. Of the freshman scores, 67 percent in CR were over 700. For 600-699, it drops to less than half, 31 percent. So 98 percent of freshmen had CR scores over 600, mostly significantly so.</p>

<p>No matter what adcoms say about holistic admission, these numbers can tell a story that's hard to ignore. How many applications do you want to send that are essentially "going for it."</p>

10

<p><a href="http://www.time.com/time/2001/education/sat_test2.html%5B/url%5D"&gt;http://www.time.com/time/2001/education/sat_test2.html&lt;/a&gt;&lt;/p>

<p>Who was not worthy to have as a classmate?</p>

<p>I think Bill Bradley survived Princeton. He was an athlete. Maybe, Princeton wasn't a good school then. :)</p>

11

<p>Aren't most students who score low on the SAT I students who don't complete each section of the test in time? What percentage of students who take the SAT I complete all sections (as can be ascertained by what items have answers marked)?</p>

12

<p>Kluge, have you seen this study?</p>

<p><a href="http://www.equaljusticesociety.org/press_chron_highered_2004_05_21.html%5B/url%5D"&gt;http://www.equaljusticesociety.org/press_chron_highered_2004_05_21.html&lt;/a&gt;&lt;/p>

<p>Moreover, research also shows that SAT scores are weak predictors of college grades and graduation rates. The Equal Justice Society, where two of us work, along with other civil-rights groups and faculty members at Berkeley, released a report last fall documenting that Berkeley students with 900s on the SAT graduated 79 percent of the time, students with SAT's in the 1400s graduated 86 percent of the time, and even those small differences were again attributable to socioeconomic barriers rather than SAT scores. In the UC system, high-school grades explain a modest 15 percent of the differences in freshman grades, and adding the SAT boosts this to only 21 percent. Moores is at least partly correct when he says that "the path into UC is pretty straightforward: Work hard, take demanding courses, and demonstrate academic success." Regrettably, Moores profoundly misunderstands how his attack on comprehensive review sends the opposite message to California's youth as well as to its policy makers.</p>

<p>Similarly</p>

<p><a href="http://www.equaljusticesociety.org/press_latimes_2003_10_25.html%5B/url%5D"&gt;http://www.equaljusticesociety.org/press_latimes_2003_10_25.html&lt;/a&gt;&lt;/p>

<p>except...
"The report by the professors and civil rights groups upheld Berkeley's comprehensive review method. Such systems have long been used in some form by elite private colleges. The report cited data from a study by the former presidents of Princeton and Harvard universities, which showed nearly identical graduation rates among students with SATs below 1000 and those with scores above 1300 at 28 selective colleges."</p>

<p>I would like to see the study by the former presidents of Harvard and Princeton.</p>

13

<p><a href="http://education.berkeley.edu/accreditation/pr_essays_1b.html%5B/url%5D"&gt;http://education.berkeley.edu/accreditation/pr_essays_1b.html&lt;/a&gt;&lt;/p>

<p>"Despite their relative economic disadvantage, Pell grant recipients have high graduation rates that nearly match overall campus figures. According to a recent analysis, 82% of Pell Grant recipients entering as freshmen and 85% entering as transfers in 1992 have graduated. The overall figures are 86% and 87%, respectively. " </p>

<p>I couldn't help myself. I just had to throw that link in a post.</p>

14

<p>token:</p>

<p>perhaps you have the SAT confused with the ACT which is more of a speed test.</p>

15

<p>I'm asking an informational question. The College Board somewhere--ah, maybe it's here</p>

<p><a href="http://www.collegeboard.com/research/pdf/rn14_11427.pdf%5B/url%5D"&gt;http://www.collegeboard.com/research/pdf/rn14_11427.pdf&lt;/a> </p>

<p>yeah, that's it--looks at the issue of how many students complete the test form. I would submit that just as there is a real difference in college readiness between the student who scores 800s across all sections and a student who scores 200s across all sections, there is a less noticeable, but still important, difference between students who consistently finish the sections without feeling time pressure and students who never finish SAT sections in the allotted time. Reading speed matters for college study. </p>

<p>Most of the studies dstark mention suffer from the same flaw: they have restriction of range, artificially underestimating the usefulness of SAT scores in predicting criterion variables, because the studies are based on data from one college or one university system. SAT scores are meaningful enough that if you switched the students at North Hennepin Community College with the students at the University of Minnesota Twin Cities (to give an example familiar from my part of the country), you'd find some information value in SAT scores. </p>

<p>Recall that the thread title asks about using SAT scores to narrow down a college search. Would I disregard the issue if College A had an SAT score interquartile range that didn't overlap AT ALL with College B? No chance. I'd take that into account in deciding which college is the "reach" and which is the "safety" (acknowledging that there are other issues involved in admission probability) and in the extreme case a college might drop off the list, or be added on the list, mostly because of glaring differences with other colleges in SAT score interquartile ranges. It would make a difference to my son, so it would make a difference to me as a parent.</p>

16

<p>The SATs are the best and fairest way to see which schools are reaches, matches, safeties. The correlation between average SAT score and graduation rate is extremely high for a wide range of institutions. So, SAT does a great job of predicting success. I think the misperception that SATs don't predict success comes from efforts to predict individual student success. Predicting individual student success from SATs is complicated because smarter students take harder majors, for example. A student with 1350 SATs flunks out of engineering. Another student with 1350 SATs gradutes from communications with honors. Does this mean the SAT is invalid? No.</p>

17

<p>Collegehelp, you're good with numbers. </p>

<p>"The correlation between average SAT score and graduation rate is extremely high for a wide range of institutions."</p>

<p>Do you have these numbers for students at the same institution?</p>

<p>The correlation between wealth and income and SAT scores...do you have those numbers?</p>

18

<p>dstark - first, please discontinue the argument by anecdote. We all know that every person is at one point of numerous bell curves of aptitude, ability, opportunity, etc. "Joe Blow who got 1200 on his SAT's did better in life than John Doe who got 1350" -- so what?</p>

<p>As to the UC studies, I am familiar with the debate. Here's the fundamental sleight-of-hand which is easy to miss because the desired conclusion was touted and is not exactly inaccurate: Both HS GPA and test scores have a positive correlation with college success (as defined in the study.) There is a high degree of overlap in individuals between the two criteria: that is, students with high GPA's also tend to have high test scores (all together now: Duh.) So if you assess the predictive level of one (grades or test scores) you will see a certain level of correlation; adding in the other will increase the correlation but not by a lot; most of the added correlation is simply duplicative of what you've already got. So the statement that "adding test scores" to an assessment already based on HS GPA doesn't add much to the predictive level of your analysis is true.</p>

<p>What is also true, but un-touted, is that the converse also applies: If you base your assessment on tests alone you will get a certain level of correlation and adding HS GPA to that only adds a little to the predictive quality of tests alone, for the same reason that adding tests only adds a little to the predictive quality of grades alone: many of the students are essentially "symmetrical."</p>

<p>Moreover, if you look at the study's charts, the combined predictive quality of SAT + SATIIs is greater than the predictive quality of HS GPA - as a straight-up comparison. (The study was performed back in the 1600 days, so the writing part was included in the SATII component.) So if you had to choose just one standard: SATI + SATII test scores or HSGPA, in fact test scores would be a slightly better predictor of college success (as defined in the study) than HS GPA. That they didn't tout. </p>

<p>I support the UC policy as a matter of public policy: favoring grades over test scores is a way to assist students from lower socio-economic groups, which has a positive value. UC is a state school, after all. But as a matter of objective assessment of a student's academic "fit" for a particular university, the UC study fails to present a viable argument for ignoring test scores.</p>

19

<p>Let's first look at this at this as what the SATI is and what it isn't. The SATI has, unfortunately, been marketed as a predictor of college success, which it is, but as dstark has pointed out, that predictor is weak in physical science terms. However. an r of 22.5 is implied, and that is not so weak in social science areas where relevant factors are difficult, if not impossible, to isolate. As others have pointed out, the studies I've seen on this relationship have not adjusted for majors or the institutions at which one matriculates. A course of study as a pre-med major in curved distribution courses at Yale does not equal a course of study in circus arts at Florida State (not to pick on that institution or that major).</p>

<p>Regardless, it's not necessary to nail the SATI as a predictor (though it's useful) if one simply looks at the test items and infers what one can tell about an applicant from performance on those items.</p>

<p>For example, the current SATI critical reading test is, mainly, a test of vocabulary and reading comprehension. It is clearly culturally biased but, then, most of what one must read in college is culturally biased. If one does well on the CR section, it can be assumed that one has the skills necessary to read and comprehend college-level material. That's important, but it's not the be-all and end-all. Certainly, hard work and a dictionary can overcome lack of these skills coming in to college, and clearly, that happens often among those with good high-school GPAs.</p>

<p>(Note: The former SATI analogy section muddied up the test by measuring both vocabulary and the ability to tease out very subtle differences when comparing the relationship of word pairs. I miss that section, but if it had been kept, it probably deserved its own score.)</p>

<p>A college admissions department can tell a lot from SATI scores if its members bother to read and understand the questionnaire items and what good perfomance on those items can actually tell you. A high GPA, for instance, and a low CR score, might tell you that this person is likely to struggle for some time getting up to speed with college-level work, but is likely to catch up, in the end, even if it requires some remedial classes and/or tutoring. (Note: This is why I don't understand dstark's assertion that high school GPA does a better job of predicting grades after freshman year than during. I'd say that's exactly the result I would expect.)</p>

<p>The issue for admissions departments is how low is too low. If I were an ad. officer at Harvard, and I have a kid with straight As in a school that I know doesn't turn out many college bound students, and that kid's SAT CR score was 525, I would worry that he would be badly outclassed in his freshman year, and that it might shake his confidence. Besides, I have so many kids with similar grades, CRs of 750 and better, from schools that send a very high percentage on to four-year colleges. I would probably conclude that the CR 525 kid would be better off attending a different college.</p>

<p>Of course, this is only lead-in to the original question which was, "should one pay attention to SATI scores when choosning a college. My answer, based on what I've written, above, would be "yes."</p>

<p>Having taught at four institutions (plus been visiting faculty at two others), I know that the general abilities of the student body affect the level at which I can pitch the course. Material and tests that yielded a fairly normal distribution of grades at one institution yielded an unacceptably low distribution at an institution to which I moved. I had to rethink and refocus the entire course to match the abilities of my class (overall). Unfortunately, the really skilled kids in my classes then got (and still get) shortchanged.</p>

<p>It is not the least surprising that kids with good high school GPAs tend to perform better in college (using GPA as the measurement) than kids with low high school GPAs, assuming their SATI scores are comparable. Skills are not hard work, and hard work tends to triumph, in the end. I do think it's important, however, to make the point that college GPA is not a perfect measure of success. There are courses of study where one can memorize virtually everything one has to know, and these courses favor hard workers. This does not mean that brute force memorization is what colleges should be teaching, or how students should measure their suceess.</p>

<p>The bottom line here is that, for a really smart kid, attending an institution where classes have been pitched to those with lesser academic skills may not be a good idea. So, looking at SATI scores might very well be a useful thing to do when choosing a college.</p>

20

<p>Kluge, I'm not going to stop the anecdotes. :) </p>

<p>Kluge, you and I are focusing on two different things. I look at graduation rates of students and compare SAT scores at a school. At Bates, almost no difference at all. At Berkeley there is a small difference, but most of it (or all of it) can be explained by the income or wealth of a student's family.</p>

<p>SAT scores and SAT2 scores may be a better predictor of success than hsgpa. Hsgpa + SAT2 scores may be better than hsgpa alone too. </p>

<p>There was a study by Derek Bok ( I think it was Derek Bok :)) that said there was a difference in gpas when looking at SAT scores. For every 100 SAT points (1600 scale), gpa was .1 higher. Is that significant? Would I not want to sit in a class with someone who has a gpa .3 higher than mine? Or .3 lower?</p>

<p>There are people that look at SAT scores and think this kid is more capable or smarter than another. Maybe. Sometimes, many times, it is just circumstances that have led to a difference in SAT scores.</p>

<p>When a school, especially a school that can pick and choose who it wants, accepts kids with low SAT scores, maybe, they actually see something in the kids that SAT scores don't capture.
It doesn't mean the student body is weaker.</p>

<p>Now, if they are just picking students with low SAT scores and not much else going on either, that's a different story. :)</p>