January SAT I curve?

Test Preparation
1

<p>Are SAT I curves based on the particular session taken? For example, is the January test next month based only on the scores of test takers that particular sitting, or is it based on a larger population or number of sittings?</p>

<p>A related question: if it is based only on those who test in January, is the curve easier in January than other months?</p>

<p>Thanks in advance</p>

2

<p>I believe the curve depends on how good the people that took the SAT in that particular month did. If the SAT that was made by CC was very difficult, then there’s a better curve than when people did well in general. Hope that makes sense. I think that’s right anyways so month’s don’t matter. You can’t predict the curve ahead of time. Only after everyone has taken it for that month.</p>

3

<p>The curves are made before the test is administered.</p>

4

<p>Ah, I didn’t know that. So the test takers won’t know about the curve until after they receive their scores?</p>

5

<p>

</p>

<p>I wish everyone who makes this claim would carefully read what they are writing and think about it.</p>

<p>The scores cannot be scaled until two pieces of information are gathered.

  1. How hard the test actually was when taken by the test group.
  2. How strong were the students who took the test at a particular time.</p>

<p>Only after the administration are these data available.</p>

<p>Tests are not curved. They are scaled so their scores will be comparable to other test forms and administrations. Even teachers who say they are “curving” tests are not really curving them, just adding points to raise grades. </p>

<p>Because of scaling, it does not matter which test date is chosen, despite all the people who “know” better.</p>

6

<p>I use the words curving and scaling interchangeably. Apparently, they have different meanings.</p>

<p>I read this article about equating.
<a href=“http://professionals.collegeboard.com/research/pdf/rn14_11427.pdf[/url]”>Higher Education Professionals | College Board;

7

<p>As the paper indicates, equating and scaling are two different steps and have very specific meanings.</p>

<p>Equating can’t occur until there is data from the exam. Scaling can’t occur until the test is equated.</p>

<p>The only thing close to what people think of as “curving” is that the test is assembled to the same difficulty as previous tests. The items are pretested so that this can be done.</p>

<p>Given the large sample size of test takers for the SAT, the grades will distribute in a standard curve pretty much by themselves. No “curving” is necessary.</p>

8

<p>Hmm… I was right for a change. :D</p>

9

<p>How 'bout that. See, a little research pays off. ;)</p>