<p>Out of curiosity I looked up the collegeboard "College Bound Seniors" reports for 1973 and 2008. I don't think the recentering really takes into account the explosion at the high end.</p>
<p>For the reasoning test:
In 1973, 2300 kids scored over 750 on the verbal. In 2008 it was around 24,000.
In 1973, 9800 scored over over 750 on the math. In 2008 over 40,500.</p>
<p>In the meantime, the amount of spaces at top schools has remained about the same.</p>
<p>No wonder these tests are somewhat useless in evaluating admissions chances at the very high end. I'm not a big fan of standardized tests, but don't you think they could sharpen this up a bit?</p>
<p>The subject test scores have crept up, but not quite as badly. People always scored fairly well on these.</p>
<p>Sharpening it up would only be necessary if the SAT was going to be used as the most important factor in college admissions – in the way that tests are used in some other countries.</p>
<p>But if students are going to be admitted on a holistic basis – with a wide variety of factors considered – I don’t see any problem with having the SAT be a blunt tool.</p>
<p>Yes, I am not a big fan of using these tests as the significant metric. But they sure take up a lot of discussion on this message board. I think they could be a little more discerning. I suppose a school that wanted to rely on test scores could just administer their own entrance exam.</p>
<p>Plus, what is the trend going to be? In 20 years will 100,000 kids taking these tests score over 2300? I’m too lazy to look up that type of thing.</p>
<p>These statistics are useless if not normalized against the total population. What are missing are the total numbers of children taking the SAT in 1973 and 2008. Divide the # above 750 by the total each year. Are the percentages different? How do the numbers vary from 1970-1975 and 2004-2009 (to get uncertainties)? Then, we will have a clear picture of whether or not there is a vast increase at the upper end of the range.</p>
It’s in the reports. You can click on the link. I should have mentioned there was not a big difference in test taking population.
1974 was a baby boom year. It looks like around 1.1 M kids took the test.
2008 was around 1.5 M. So it was more, but not that many more. Certainly not 5 to 10 times more.</p>
<p>In addition, if there are only, say 25K seats at elite universiteis, and that hasn’t changed at all, the raw number does make a difference if that’s where you want to go and if the school looks at your SAT score.</p>
<p>
Yes, but I’m pretty sure even that is creeping up. You could probably fill 2 or 3 freshman classes with perfect scorers.</p>
<p>I don’t have those statistics for 73-74, when I was a senior (hence my interest). But I’m pretty sure it was extremely rare. We would read about it in the national news.</p>
<p>^^^
My post above is incorrect as far as the perfect score. It is a lot fewer than I thought, only about 270.
So that does add some level of differentiation.</p>
<p>Yes the population of high scorers has significantly increased since 1973, but a corresponding increase in the score expectations of elite colleges has also occurred which marginalizes the effects of such a trend. </p>
<p>As Silverturtle has already pointed out, scores at the highest end of the spectrum are still quite rare in comparison to the total testing population.</p>
<p>For the math exam a couple stupid errors can drop your score from 800 to
750. Is that really a good way to write an exam? Why not just make it a little trickier overall?</p>
<p>But no wonder 90% of the kids I see posting here have scores over 750.</p>
<p>Is your info adjusted for the recentering which took place in the '80’s and 90’s? In the old days, when dinos roamed the earth, an 800 meant all were correct. Miss one = 770. Miss one, leave one blank = 760/750. Now a student can miss one-two-three? and still score an 800. Plus, recentering raised the ‘curve’.</p>
<p>
</p>
<p>I’m guessing your kid is a math whiz? :D</p>
<p>But the short answer is that I really don’t think “elite” colleges care once an applicant gets above the ~750 threshold. At that point, they KNOW that the kid has the smarts to successfully do the work, so all of the other stuff in the app determines whether they’ll be offered a spot.</p>
<p>the exam was recentered in the 90s, I think, and that added 70 points to verbal and 30 to math.</p>
<p>So, for example, I got an 800 on verbal and my mom got a 710. BUT if you recenter my mom’s score, she would have gotten a 780, so she should have, but would not show up in your population.</p>
<p>But yes, i concede that the advent of SAT prep classes, taking it multiple times in addition to the flynn effect is causing an increase in scores.</p>
<p>It’s also now the case that how kids approach the SATs is very different. Back in the day (70s), kids took them one, perhaps twice, often with very little prep. Among all the folks I know who took them in the 60s and 70s , none of them did any major prep work. A few memorized some words. And, many of these folks are very smart, got good scores and went (or were admitted to) elite schools.</p>
<p>I definitely also dislike how they test math on the SATs. When a significant portion have 800s on both the SAT and SAT II level 2, it really doesn’t differentiate their ability. </p>
<p>All the SAT Math measures is your ability to catch odd phrasings of questions, avoiding stupid mistakes, and reminding yourself this only deals with Pre-Cal concepts, so you shouldn’t be integrating or calculating derivatives.</p>
<p>I would definitely like to see more comprehensive exams that test problem solving abilities, with longer more complex problems (FRQs like AP tests, maybe).</p>
<p>I don’t know why there are two SAT II’s for math, when there could be just one that measures PreCal ability. Or there could be three, with one that covers Calculus ability, but that would crossover with the AP exams.</p>
<p>I definitely feel that they should try to change the test, so that from a 700-800, actually means something. If the problems were difficult enough, they could keep the similar scoring of today (where you miss 1-2 you still get a 780-800). I’d rather see them bump up the difficulty (not overly so), and change the scoring, so the average is a 600 (since 200 for signing your name), not a 700+. That way an 800 actually means something.</p>
<p>The Math section does not test any Pre-Calculus. </p>
<p>
</p>
<p>They want to have a test that is appropriate for students who have not finished Pre-Calculus before applying to colleges.</p>
<p>
</p>
<p>I have never heard of a test that does not measure someone’s ability to avoid stupid mistakes – that’s what retakes are for. Moreover, I have never encountered an oddly phrased question. For the vast majority of students, it’s not stupid mistakes that are keeping a student from scoring 800.</p>
<p>And I don’t know whether the inclusion of calculus on the test would necessarily make it more difficult; it would just isolate many students. For me, the AP Calculus exam was actually easier than the SAT Math section, as the latter required more problem solving even if it was more basic in the contents that it covered. (One is an achievement test; the other is an aptitude test.)</p>
No, but for the vast majority of students scoring a 770 or 750 it is stupid mistakes keeping them from scoring an 800. It certainly isn’t ability to solve the problems. THere aren’t any problems on the current exam that are so difficult a kid scoring 770 couldn’t solve it.</p>
<p>I tutored this test for a few years, and could routinely pull down an offical sample and score 800, but certain times I might miss one or two that I could have easily solved.</p>
<p>But I agree, that’s why people should retake them.</p>
<p>Also, you don’t necessarily need calculus to make an exam more difficult. I’m not sure, but I don’t believe the AMC tests all include calculus, and those are certainly tricky. I’m not saying the SAT should be at that level, but they certainly could put a little thought into it and develop some problems other than the cookie cutter junk that’s on it now.</p>
<p>This sounds reflective of the average student of very high math ability. He or she should be able to score 800 almost every time. If the one time he or she officially takes the test happens to be one of those rare exceptions (e.g., a “silly” mistake was made), then a retake is appropriate.</p>
Very rarely. Example please. And a real problem off an official test. Not something from some website unrelated to the CollegeBoard. Let me know which test it came off of.</p>