<p>my friend said that the oct's SAT is the hardest of them all
is this true?</p>
<p>Lol, hope not, I'm planning to take them this october</p>
<p>total bull...actually OCTOBER might have the best curve.</p>
<p>lol how do you know that it's a total bull?</p>
<p>it's always 50% vs 50%, don't say it with such a confidence unless you're in ETS :)</p>
<p>hopefully it's easy tho, im taking it too!</p>
<p>ahha nah I'm just saying "bull" because if you think about it....curves are made AFTER the test and a lot of seniors take it then for like last chance type and they prolly do their best on that and at the same time...juniors are beginnign to take them so it statistically drags the curve more towards the lenient side (however this is not always the case and considering its new SAT it may be different)</p>
<p>but if you take a look at old 10 reals you'll notice something.</p>
<p>Ahhh, i see, well you DO make sense, i never paid much attention to the curves tho lol :)</p>
<p>lets pray .... that we can all get the scores we desire</p>
<p>Curves are made BEFORE the test. That is what the equating section determines. Curves are NOT based on who took the test the same month as you did. </p>
<p>(At least, that is what i've heard from this message board.)</p>
<p>hmm, i heard that they make it by taking the Median score of the test and compare to the previous test then they create the curve, i dunno</p>
<p>actually, curves are preset and made before the test. that's one of the biggest myths about the sat. it's all scaled b4 the sat is given.</p>
<p>hmm well it shall remain a myth until an ETS Employer clarifies it, which will never happen.</p>
<p>no matter if it is made b4 or after the test is given out, we all still have to study harddddddd :)</p>
<p>people forget about the equating sections. maybe i'm wrong abt this, but the way I understood it is that some of the equating questions show up on each testing administration, therefore the strength of the testing pool can be compared to previous testings (hence the curve will remain consistent).</p>
<p>alright I guess I was wrong.</p>
<p>"Next, the raw score is converted to the College Board 200-to-800 scaled score by a statistical process called equating. Equating adjusts for slight differences in difficulty between test editions, and ensures that a student's score of, say, 450 on one edition of a test reflects the same ability as a score of 450 on another edition of the test and that a student's score does not depend on how well others did on the same edition of the test." - Collegeboard (<a href="http://www.collegeboard.com/prof/counselors/tests/sat/scores/faq_scoring.html#quest01%5B/url%5D">http://www.collegeboard.com/prof/counselors/tests/sat/scores/faq_scoring.html#quest01</a>)</p>
<p>Every question on the SAT has been previewed on the SAT before that date. That's what the experimental sections are for. Basically, CollegeBoard, after making and refining their questions, put them in the experimental section and see how hard it is and if it, simply, a good question that provides some information about the student's abilities and that it does not have a high bias in favor of a certain group of test takers. Once they have all the information about the questions, they assemble a test that should be similar in difficulty to all the other SATs that have been given. Also, recently, ETS has been recycling old test questions, which works in essentially the same way as experimental sections.</p>
<p>In short, the "curve" is PRE-determined and the quality of the test takers on the day you take it has absolutely no relevance to what the "curve" is. They already know how hard every question is, so determining it as such would be unnecessary.</p>
<p>I'm sure that they have a formula, but they have not released it. The formula, though, just ensures that a score of, say, 1500, reflects the same level of ability as all other 1500s do.
Hope that helped</p>
<hr>
<p>AP STATISTICS CLASS FORMULA
the formula is here:</p>
<p>x bar = {[(xi - yi)^p(b) - (xa + ya)^p(a)] * standard deviation * Z (score)}/(population variance) * (n - 1)/500 * 800t</p>
<p>--> since it is based on log, whose base is ten, the resulting SAT curve fitting will always round to the nearest ten.</p>
<p>in which n is your raw score and t (either 78 for verbal and 60 for math) is the maximum raw score you can have</p>
<p>other variables are determined from the population statistic from the experimental section; your "real" test will always have been "experienced" by the former testtakers in their experimental section</p>
<p>Dude. The Dec 2004 Test? Probably best curve ever.</p>
<p>Who took that? (Just think). It was the second to last 1600 test and the last one the class of 2005 could submit to colleges. (or can they submit jan too?)</p>
<p>"Also, recently, ETS has been recycling old test questions, which works in essentially the same way as experimental sections."</p>
<p>Is that fair? What if I retake and get the same questions?</p>
<p>One of the old SAt had a 2 error margin for math and 5 for english...it was GREAT. 1995 may i think?</p>
<p>The curve wasn't that great for Dec 2004. I got 1560, and a 760 in verbal due to four mistakes. This isn't that generous compared to other old SAT curves. The math curve was also not great (800,780,760).</p>
<p>Haha ramsfan, I got a 1560 as well last month, with a 760 in verbal due to four... questions wrong, not necessarily mistakes. I just didn't know.. haha.</p>