<p>I’m starting to worry about the Dec. SAT. I’ve registered for it weeks ago and now i’ve just been told not to take it because that’s when all the seniors take it. Doesn’t a test date really matter?</p>
<p>How do they determine the curve?</p>
<p>I’m starting to worry about the Dec. SAT. I’ve registered for it weeks ago and now i’ve just been told not to take it because that’s when all the seniors take it. Doesn’t a test date really matter?</p>
<p>How do they determine the curve?</p>
<p>No, the test date does not matter. Please don’t worry about this. Just relax and do the best you can on the test. </p>
<p>Here is College Board’s explanation of how they make sure test scores are comparable from sitting to sitting.<br>
<a href=“http://professionals.collegeboard.com/research/pdf/rn14_11427.pdf[/url]”>http://professionals.collegeboard.com/research/pdf/rn14_11427.pdf</a></p>
<p>So what I inferred from the reading is that the “curve” is not determined by how well the test-takers did as a group but by how well the prior test-takers did on the experimental sections?</p>
<p>There is no “curve,” though everyone believes there is. There are two steps.
<ol>
<li> Equating is used to determined the relative strength of the current group of test takers compared to previous groups. A portion of the items on the test are “equaters” and have been used on a previous test.<br></li>
<li> Once the two groups have been equated, the test is scaled. This adjusts for the differences in difficulty of the the tests given to the current group and the equated group.</li>
</ol>
<p>The “experimental” section are items being pretested. The data from this section is used to determine the difficulty of items so that each test can be assembled to be as similar as possible in difficulty to previous tests. Scaling is required because sometimes items perform differently when they are “real” compared to “experimental”</p>
<p>Given the size of each testing population, it is possible to make the tests fair and reliable for each group. It doesn’t matter when you take the test.</p>
<p>It amazes me how many people who try to play so many angles never get the point. :-)</p>
<p>knguyen142,
I think that’s a nice nutshell summary. </p>
<p>BigIs,
It is funny how often this is misunderstood. Maybe they should include the equating/scaling explanation on the CR section and see how people score.</p>