<p>I agree with joker. It appears there’s a notable difference in curves from date to date. collegeboard wouldn’t intentionally make tests different difficulties. I think the curve is based on the scores of the test takers for that date</p>
<p>no the reason why the curves vary is because it is simply impossible to create two tests with identical difficulties. The only way to do that would be to have the exact same questions appear on both tests (which obviously wont happen).</p>
<p>Some questions are going to be harder than others and each question is assigned a difficulty rating (not the 1-5 rating given out but one that is more complex) and the overall difficulty of the test is determined using statistical formulas</p>
<p>@Stupefy - no CB does not want to make the test difficulties different but it is physically impossible for two tests to have the same difficulty (unless all the questions are identical)</p>
<p>I used to think that the curve was determined by the performance of the test takers but I see why it’s not now. One of my parents is a statistician told me this (after a lot of questioning and arguing 
)</p>
<p>Joker/stupefy, mabs is (mostly) correct.</p>
<p>There are certainly test dates with “better” curves (in that a given raw score results in a higher scaled score), but that is because the test was harder than average on that date. See [thread=760312]this thread[/thread] for a long discussion and a link to the College Board whitepaper on the subject. The CB tries to make the test about the same in difficulty but it isn’t possible ahead of time to do so accurately (without security issues like leaked test contents, besides other problems such as cost).</p>
<p>The curve corrects for small variations in the difficulty of the test from one test date to another. A large bunch of very strong (or very weak) test takers taking the test on a given date has no effect on the curve: the CB uses statistical methods specifically designed to account for that. These methods are applied using the non-graded sections (<em>after</em> the test is administered).</p>
<p>But if, in general, the test takers are less intelligent on a certain date, such as June, shouldn’t it be easier to score higher on the curve (regardless of the difficulty of the test)?</p>
<p>guys, seriously? this is all just silly speculation. nobody can say anything for sure.</p>
<p>Fig and Mabs seem to have a pretty good understanding of the “curve,” which is more accurately referred to as “scaling.” A scaled score is calculated based on the answer to two questions:
- Is this group of test-takers stronger or weaker than previous groups?
- Is this test form harder or easier than other test forms?</p>
<p>Based on the statistical answers to these questions, the raw scores are scaled. The official name for this process is “equating,” and it ensures that your score will be the same now matter how hard or easy an individual form or how able a particular test group is or isn’t.</p>
<p>So, let’s say you earned a 730 on an administration where the average score was a 600. If you then re-took it with a less able group, your would score “higher” on a “weaker” scale, and still end up in the same place. So you would still score a 730 + or - 20 to 30 or so, but the average score may be a 550.</p>
<p>I’ll leave with the only two “secrets” I know about getting the best score:
- Learn the “stuff” for the test.
- Take it when you are ready.</p>
<p>After the October test date, everyone was certain that -1 on math would be 790 or 780, since October is -always- the one with the best curve, and math is -always- 780 or 790.</p>
<p>I missed one. I got a 770.</p>
<p>Predictions are worthless.</p>
<p>People always seem to revive new threads and not notice they are old. :/</p>
<p>haha yea I just noticed that!</p>
<p>@stupefy, what drove you to do it?</p>
<p>^sorry I came across it when I googled “best SAT date” so I felt compelled to post. sorry :/</p>
<p>oh whoops, that mighta sounded kinda harsh (darn internet not being able to detect tone)</p>
<p>i was just curious is all :)</p>
<p>
And they also get mad when people ask questions that already have threads. As long as an old thread doesn’t have a question specific to the OP long bumps should be permissible IMO.</p>
<p>I kind of agree. But the person who revived it did not ask a question, they were just commenting.</p>
<p>Now that this is bumped anyway, I’m just wondering how November curves have been in the past if anyone has taken the Nov SAT before. I know the curves are based on the test and not the date, I’m just curious how the curves for the November tests have been before</p>
<p>^It is hard to know since the November tests aren’t released as QAS tests. However, the curve at the top end can generally be determined after scores are released. The 11/08 math curve in particular was 800-760-740-730 which (AFAIK) is the harshest curve to date on the new SAT.</p>
<p>Sorry for bumping a thread that is more than 3 years old, but I’m wondering this myself. I think I did pretty well on the June SAT. Just trying to decide whether it’s worth taking it again in September, when a ton of seniors will be racing to take it for the last time. Guess I’ll see when my score comes.</p>
<p>you are not competing with others taking the test on the same day. You are competing with all students taking the test in different days. Curves are just for adjusting the scores to the difficulty of the test to make it comparable. It does not matter if there are more well prepared students taking the test with you on the same day.</p>
<p>Billcsho is correct, as anyone can verify for him or herself on the College Board’s own website:</p>
<p>“Equating also ensures that a student’s score does not depend on how well others did on the same edition of the test.”</p>
<p>(Source: [How</a> the SAT is Scored - Overview of SAT Scoring](<a href=“The SAT – SAT Suite | College Board”>Your SAT Score Report Explained – SAT Suite | College Board))</p>
<p>The SAT does not have a “curve,” if you are using that term properly; equating is not at all the same thing.</p>
<p>^ Curve is not the official term, but it is generally used in this forum. It is basically the raw score to score chart that vary from test to test.</p>