College Board to Make Changes to SAT Exam

<p>Why does it always turn into my kids did X, so it must be true.</p>

<p>

</p>

<p>In general, people enjoy what they are good at. Smart people tend to be readers. I don’t think exhorting kids to read has much effect.</p>

<p>I sometimes wonder what my scores would have been like if I’d grown up in today’s world. Decades ago, TV wasn’t that enticing (certainly not after school), and there was no internet, dvd players, ipods, smartphones, texting, etc etc etc. I read because there was often not much else to do. Readers digest condensed books, the encyclopedia, and some “classics” that were lying around got consumed because that was the only “food for thought”, or entertainment. Of course, my recollection is that after junior high I read somewhat less, and during high school I loathed the assigned reading for English, and most other subjects. </p>

<p>Today’s kids are so overwhelmed with inputs from every direction, (most of them deleterious, IMO) that it’s no wonder the SAT decided to drop some of the vocabulary content. Its actually amazing to me that so many kids do get 800s on the verbal sections. And the availability of synopses of nearly everything has to have some effect on inclinations to read.</p>

<p>Just a different world, I guess. We aren’t going back.</p>

<p>

</p>

<p>While it does not change much to the spirit of the post, here is a small correction. Many people assume that the recentering of 1995 impacted both the verbal and the math section in a similar fashion. That is not exactly the case as the table below indicate.</p>

<p>Note than a V score of 710 became a 780 but that a M score of 710 was … lowered to 700! </p>

<p>STEM people cannot win! :)</p>

<p>Verbal
From/To
800 800 500 580
790 800 490 570
780 800 480 560
770 800 470 550
760 800 460 540
750 800 450 530
740 800 440 520
730 800 430 510
720 790 420 500
710 780 410 490
700 760 400 480
690 750 390 470 </p>

<p>Math
800 800 500 520
790 800 490 520
780 800 480 510
770 790 470 500
760 770 460 490
750 760 450 480
740 740 440 480
730 730 430 470
720 720 420 460
710 700 410 450
700 690 400 440 </p>

<p>Source: [SAT</a> I Individual Score Equivalents | Research and Development](<a href=“http://research.collegeboard.org/programs/sat/data/equivalence/sat-individual]SAT”>http://research.collegeboard.org/programs/sat/data/equivalence/sat-individual)</p>

<p>The selective universities that have many applicants with 800 Math SATs also have many applicants submitting SAT Subject Test Math Level 2 scores. In a CC thread <a href=“http://talk.collegeconfidential.com/sat-act-tests-test-preparation/107908-sat-ii-math-level-2-curve.html[/url]”>http://talk.collegeconfidential.com/sat-act-tests-test-preparation/107908-sat-ii-math-level-2-curve.html&lt;/a&gt; someone wrote that one can miss 6 questions on this test and still get an 800. The College Board could expand the scoring range for the SAT Subject Test Math Level 2 and equate a perfect score to say 900.</p>

<p>Can someone explain (Dick & Jane it, please) to this right brained parent why they curve the SAT?</p>

<p>The simplest explanation is that the curve serves to “equate” all tests on a historical basis. When a test is developed, the maker will use a number of questions from the database, but since it is not possible to have the overall difficulty of the test exactly the same as the previous editions, a curve is applied. </p>

<p>In so many words, Test Difficulty + Curve = All tests are comparable.</p>

<p>

</p>

<p>Would not serve any purpose, with the notable exception of giving a few misguided souls some moronic bragging right. The scale of the 200-800 standardized test is based on “averaging” the scores to a mean or median of 500. </p>

<p>Fwiw, how many questions do you think people can miss to get a perfect score on one of those “mythical” AP tests?</p>

<p>Xiggi – This pertains to tests given in the same calendar year, rather than comparing to years gone by, yes? So to be clear, there is a different test each testing month?</p>

<p>A lot longer than in the current year. In theory, one should be able to compare a test from 2006 (or longer) to one of 2013. </p>

<p>There are a number of tests given for each administration. The SAT is administered seven times a year in the United States and six times per year internationally. The tests can have different section orders, different experimental sections, and can also include repetition of sections that were not previously released to the public via QAs. </p>

<p>In addition, you have a number of additional tests such as Sunday (religious purposes) or mid-week for students who received added-time. It would not be unusual for a student taking one of the latter to see the exact same test that was given the previous Saturday.</p>

<p>Thanks for the explanations, Xiggi. One more question: is the curve is in place to protect the comparability of the test beyond each year?</p>

<p>

This is the problem. Despite what re-centering did or did not do to the math exam,there are still too many students achieveing astronomical scores to make it a very discerning test for the upper tail of the distribution.</p>

<p>In 1973 less than 10,000 students score over 750 on the Math portion. In 2012 over 11,000 scored 800.</p>

<p>CR has the same issue. In 1973 only around 2,000 students scored over 750. In 2012 over 8,000 scored 800.</p>

<p>I’d say these sort of ratios hold true for most of the top 5% of scores, and for most of the SAT IIs (formerly achievement tests) as well.</p>

<p>Note that I’m not saying it should be more than a rough cut, I’m stating no opinion. I’m just throwing out some facts.</p>

<p>What’s the difference between the number of test takers during the last 40 years? Obviously there are many more SAT takers today than four decades ago.</p>

<p>^^^
The 1973 stats represent around 1 million test takers. The 2012 stats represent around 1.6 million test takers.</p>

<p>I didn’t add it up precisely, but from this</p>

<p><a href=“http://media.collegeboard.com/digitalServices/pdf/research/SAT-Mathemathics-Percentile-Ranks-2012.pdf[/url]”>http://media.collegeboard.com/digitalServices/pdf/research/SAT-Mathemathics-Percentile-Ranks-2012.pdf&lt;/a&gt;&lt;/p&gt;

<p>It looks like around 50,000 students (a little less) scored 750 or above on the Math portion in 2012 as opposed to the 10,000 750 and above in 1973.</p>

<p>I think we also need to bear in mind that kids these days take the exam multiple times, and I think the data tables represent their highest scores. I think in 1973 most kids took the test once (at least I did) or maybe twice.</p>

<p>Agentninetynine, yes, you could say that!</p>

<p>

</p>

<p>Well, if needed, you could always look for 'em perfect scorers. That is a much smaller number. Considering that there are 300,000 applicants to the Ivies Plus only, the odds are pretty good that they could identify the “strivers.”</p>

<p>In 2006, out of 1.38 million seniors taking the SAT, 238 (roughly 1 in 5,000 students) receive a perfect score of 2400. In comparison, 216 seniors in the class of 2006 out of 1.21 million taking the ACT (about 1 in 5,600 students) receive a perfect composite score of 36.</p>

<p>In 2012, there were 360 perfect SAT score of 2400 out of 1,664,479. Of the 1,666,017 students taking the ACT in 2012, 704 students earned the perfect composite score of 36</p>

<p>^^^
Yes, you could do that if you wished. You would probably have to limit that to single sitting, because I think that is what those scores represent.</p>

<p>Of course, since each section has little value on its own, you almost have to consdier the composite scores, even if you aren’t particularly interested in anything but one section.</p>

<p>Didn’t used to be like that in the olden days (or not so olden days) as you can sort of see from the scores listed here</p>

<p><a href=“http://tech.mit.edu/V109/PDF/N15.pdf[/url]”>http://tech.mit.edu/V109/PDF/N15.pdf&lt;/a&gt;&lt;/p&gt;