tag:blogger.com,1999:blog-7691251033406320222.post3698750161658107003..comments2024-03-26T04:19:38.862-07:00Comments on kitchen table math, the sequel: December SAT Scores (aka, My Buddha)Catherine Johnsonhttp://www.blogger.com/profile/03347093496361370174noreply@blogger.comBlogger12125tag:blogger.com,1999:blog-7691251033406320222.post-12735556203392223622011-12-24T13:16:38.702-08:002011-12-24T13:16:38.702-08:00The correlation between IQ and SAT depends on whic...The correlation between IQ and SAT depends on which IQ test you use, but there isn't that much debate about the numbers.<br /><br />One study (in 2004) is<br />Scholastic Assessment or g? The Relationship Between the Scholastic Assessment Test and General Cognitive Ability<br />by Meredith C. Frey and Douglas K. Detterman<br />http://pss.sagepub.com/content/15/6/373<br /><br />They got correlations of 0.82 and 0.483 (0.86 and 0.72 with appropriate statistical correction for artifacts of the sampling). These are very high correlations.<br /><br />Http://www.iapsych.com/iqmr/koening2008.pdf<br />shows similar correlations of the ACT with IQ measures though the SAT test had a somewhat higher correlation with the ASVAB test than the ACT did.<br /><br />I don't believe that there is much dispute about the high correlation between results on IQ tests and results on SAT and ACT tests. I believe the debate is about how to interpret that high correlation, and what it means about what the SAT, IQ, and ACT tests are really measuring.<br /><br />If someone can point me to studies which show a low correlation between SAT scores and widely-used intelligence tests, I'd be interested in seeing them.<br /><br />Note: the correlations are high, but not so high that there isn't still a huge amount of scatter. Any tests gives you just a snapshot of how you did on one day on a particular test. It is quite common for an intelligent person to do poorly on a test.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-89082564818674558822011-12-24T12:30:32.769-08:002011-12-24T12:30:32.769-08:00What is most entertaining is that there is serious...What is most entertaining is that there is serious disagreement (even among experts, it seems) about how well two sets of numbers (IQ test and SAT) correlate with each other.<br /><br />This really shouldn't be a matter of opinion :-)<br /><br />Mark RouloAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-40234702593550600652011-12-24T12:25:39.516-08:002011-12-24T12:25:39.516-08:00There are psychologists who think the SAT is a A G...There are psychologists who think the SAT is a <a href="http://www.nytimes.com/roomfordebate/2011/12/04/why-should-sats-matter/the-sat-is-a-good-intelligence-test" rel="nofollow">A Good Intelligence Test</a> , and there are others in that NYT debate who disagree.Bostoniannoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-24044397322577521422011-12-24T10:21:50.777-08:002011-12-24T10:21:50.777-08:00The SAT did once correlate closely with IQ measur...The SAT <i>did</i> once correlate closely with IQ measures (notably the Weschler and the Stanford-Binet) but this has not been true for some time. Mensa and other "high IQ" societies do not accept SAT scores after 1994 to meet eligibility requirements, precisely because they are no longer "intelligence" tests but also measure knowledge and taught skills to a much larger degree than the previous incarnations of the SAT.palisadeskhttps://www.blogger.com/profile/13700503881038569921noreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-8607721765486803432011-12-24T09:40:07.812-08:002011-12-24T09:40:07.812-08:00@Glen I think you're right. I underestimated ...@Glen I think you're right. I underestimated the amount of hours of "hard work" by a few years! <br /><br />And yes, I read voraciously (often a few books per week). I'm an obsessive (though not necessarily careful) reader. I'm more into quantity (unless I'm reading a beautiful novel). This made the reading a bit harder for me than I would have thought (i.e. I had to find the right pace -- fast enough to finish, slow enough to be careful).<br /><br />@SteveH With more time (add 10-15 minutes) I can score 18 or 19 out of 20 right consistently.Debbie Stierhttp://www.perfectscoreproject.comnoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-28221326413472174242011-12-24T06:20:59.262-08:002011-12-24T06:20:59.262-08:00Debbie, Did you ever take the test with more time?...Debbie, Did you ever take the test with more time? If so, does your score go up consistently? For many, it's easy to look at every question and know immediately that you can solve it. The performance timing of the test can be the killer. Math departments don't offer performance degrees. SAT speed is never expected in college or in real life.SteveHhttps://www.blogger.com/profile/03956560674752399562noreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-20184693542204306722011-12-24T06:06:06.673-08:002011-12-24T06:06:06.673-08:00The SAT is not an intelligence or aptitude test. I...The SAT is not an intelligence or aptitude test. It might be if it wasn't so important. Since it is, enormous effort is put into beating the test with preparation. This preparation arms race pushes the questions away from math because SAT math has such a limited domain. The test cannot tell the difference between aptitude and preparation.<br /><br />At the very low end, it's about math preparation, but this is not difficult math - the kind that would tell you anything about aptitude. The SAT makes the test more difficult by limiting time and making the questions tricky. This might reflect aptitude, but it also reflects preparation, and everyone prepares. <br /><br />At the top end, scores change dramatically with just a few right or wrong answers. This has clearly been described in other threads. Even if one has a high aptitude for math, he/she will easily be beaten without a proper preparation for the timing and tricks specific to the test.<br /><br />If you assume that everyone prepares, you might say that it once again reflects aptitude or a willingness to prepare (both useful attributes), but then you have the error range of the test. At the upper range, one or two errors can have a huge effect on your score; an effect that isn't entirely discounted when your college application is evaluated.<br /><br />Overall, it's a test of SAT preparation and a test of performance test taking. It does not have a lot to do with math aptitude. If they wanted that, they would allow you to substitute the SAT II Math for the regular SAT math.<br /><br />This is Bostonian's continued attempt to push this view in spite of all evidence to the contrary. Educational opportunity should be offered to those who prepare. Whether or not there is a correlation to IQ is immaterial. In that sense, one could argue that the SAT (based on preparation) is a proper test. However, the regular SAT is testing preparation in something other than math.SteveHhttps://www.blogger.com/profile/03956560674752399562noreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-80525082500847083532011-12-23T23:35:55.369-08:002011-12-23T23:35:55.369-08:00Debbie, I think your answer is in your posting. Yo...Debbie, I think your answer is in your posting. You say you "didn't learn a lick of math since 9th grade (until this project)", then you put in "dozens and dozens of joyful hours" over the past year on SAT math. And you got a 55%ile.<br /><br />Contrast that to your reading. How much reading would you say you've done since 9th grade? My guess is that you've done dozens and dozens of hours of reading nearly every MONTH since 9th grade (and probably many years before that.) And you scored a 99%ile. Hmm.<br /><br />Now consider your writing. You got a 97%ile. I'm going to go way out on a limb and guess that, since 9th grade, you've done a whale of a lot more writing than math, but not as much writing as reading. How did I do?<br /><br />I spent dozens and dozens of hours on math each MONTH, every month, 12 months a year, throughout middle school and high school. Lots of kids do. Every month or two they do as much math study as you've done in total since, what, the Jimmy Carter Administration? And they've been doing it for years. And you thought you'd unseat them in the SAT?<br /><br />I think this bolsters your claim that hard work pays off. It just takes more hard work than you expected. <br /><br />You think no one will believe that you're mathematically inclined? Well, why not? What this evidence suggests to me is that Debbie gets a 99%ile when Debbie does 99%ile-worth of work and gets less when she does less work, and that 99%ile, while possible, is a huge amount of work for Debbie, as it is for the rest of us.<br /><br />And I think that your intention to study math, at this age, without being forced to, is a plausible definition of "mathematically inclined."Glennoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-47668370299107601962011-12-23T19:06:14.430-08:002011-12-23T19:06:14.430-08:00There's a big difference between good in math ...There's a big difference between good in math and good at the math section of the SAT. <br /><br />The latter adds in all sorts of things that the first doesn't require at all. Which says something bad about the test, not about the person's math skills.Jennoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-44000055532631300952011-12-23T17:06:47.497-08:002011-12-23T17:06:47.497-08:00@Bostonian -- If it were an intelligence test, I w...@Bostonian -- If it were an intelligence test, I would have done better! I actually questioned my intelligence at one point during this process took an IQ test.<br /><br />I think Catherine's assessment is absolutely right.<br /><br />Here is the really weird aspect of the the math story for me:<br /><br />I actually still believe I'm good at math and am mathematically inclined. I know that NO ONE will believe me, but I'm not kidding -- math comes pretty easily to me -- EXCEPT on the SAT. <br /><br />This math story reminds me of when I was growing up, and my mother had instilled this belief in the whole family that if you weighed more than 110 pounds as a woman, you were overweight. Period. Didn't matter how tall you were, or how big your bones were -- over 110 lbs was a problem. (She is very small boned and thin -- and I'm built more like my father's side of the family -- i.e. hearty.)<br /><br />So when I started getting near that 110 mark in high school, she started getting very nervous that I was going to be fat, and she would bribe me with new clothes if I lost weight (and I was very thin in hs). <br /><br />I always say, it's a miracle I didn't get anorexia. <br /><br />Somehow, I ALWAYS knew that she was dead wrong about this 110 thing...so I let her just have it, and I sort of played into it, but I knew deep inside myself that I wasn't fat.<br /><br />That's sort of how I feel about this SAT math score. <br /><br />I KNOW I'm good at math! And, I know that I learned a ton of math on this journey, and even though it didn't show up in my score, I'm not convinced that it says I'm not good at math, nor does it reflect my aptitude.<br /><br />It probably reflects my lack of a good math education.<br /><br />Maybe I'm out of my mind, but every bone in my body tells me I'm right....and that I didn't go about this the right way, but then, I probably couldn't have done it the "right way" in the year I set out to do it. That was a bad calculation.Debbie Stierhttp://www.perfectscoreproject.comnoreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-55523161430416056442011-12-23T10:30:15.265-08:002011-12-23T10:30:15.265-08:00nix on that
You, Bostonian, have no ability to ju...nix on that<br /><br />You, Bostonian, have no ability to judge Debbie's 'math aptitude,' as you call it, based in a set of SAT scores and essentially no education in math beyond freshman year in high school (if that).<br /><br />The SAT is not a test of crystallized intelligence.<br /><br />It is a test of knowledge.<br /><br />Funky knowledge, but knowledge nonetheless.<br /><br />For the record, I improved my own math score by 110 points.<br /><br />I improved Chris's math score by 60 to 70 points.<br /><br />Debbie's experience is a brilliant illustration of the limits of 'teaching to the test': Debbie had a year of SAT prep, not a year of math.Catherine Johnsonhttps://www.blogger.com/profile/03347093496361370174noreply@blogger.comtag:blogger.com,1999:blog-7691251033406320222.post-50157636025861349802011-12-23T10:19:00.975-08:002011-12-23T10:19:00.975-08:00The SAT used to stand for "Scholastic Aptitud...The SAT used to stand for "Scholastic Aptitude Test", and you appear to have unusual verbal aptitude but only average math aptitude. The SAT is largely an intelligence test, and beyond a certain point studying for it will not improve one's score.Bostoniannoreply@blogger.com