<p>That's not the actual curve.</p>
<p>its the curve on the collegeboard.com practice test.</p>
<p>Good work, pentasa.</p>
<p>Hey so if that's the practice test curve...let's figure something out. i didnt take that one (i took the 8 out of the book) but was the online harder/easier than the real thing? anyone know? if we know that maybe we can equate. lol</p>
<p>Too much uncertainty:
AS AUTHOR OF THIS THREAD,
I DECLARE THE CURVE POSTED IN THE BEGINNING OF THIS THREAD AS UNRELIABLE
It would hurt if CR is harsher than what was put in the curve at the beginning of this thread</p>
<p>CB obviously feels the need to make the equating method secret, which is completely stupid in my opinion. I mean isn't equating, by definition, to make equal or reduce to an average....so that would mean they would want to make the same percentage of people get the same scores every year, but what they said doesn't agree with this. I read somewhere (can't remember where, i think a "getting into college" book) that they find the average score for that year, then all scores are based on the average, forming a "perfect" 200-800 curve.</p>
<p>Ticklemepink, please refer to p.882 of The Official SAT Study Guide for an explanation of the curve. I've posted it on another thread ("Can Somebody Explain The Curve For Me") here; don't feel like reposting.</p>
<p>Thanks for figuring that out, pentasa. Fireflyscout, I just read the explanation and I think you've interpreted it incorrectly. The "minor differences in difficulty" don't refer to the difficult, medium, and hard ratings of the problems. </p>
<p>Plus, if your interpretation is true, why would mere "score ranges" be provided for tests that have not been previously administered? Underneath the explanation, the book clearly states that "since this test has not been previously administered, score ranges are provided for each possible raw score."</p>
<p>That's corny. PR just took curve from PRACT. TEST?
This is wayy to confusing, I'll wait until apr. 11 for a curve
I declare this curve unreliable so far.</p>
<p>nice find pentasa, so i think its fairly obvious to everyone that xindian's curve is total princeton review BS?</p>
<p>54-800, 53-790, 52-760... i refuse to believe this, lol</p>
<p>i want a 780!</p>
<p>OK, let's review this thread and set everyone straight.</p>
<ol>
<li><p>The curve that is posted by the OP is simply speculative. It may end up being close to the real one that will be made public in ONE MONTH, but it remains to be entirely irrelevant to the Match 12. </p></li>
<li><p>The day that PR is able to write a predictive test which is not pure BS is the day that I MIGHT start believing what they write. In the meantime, I leave them in their world of complete speculation and deceit. </p></li>
<li><p>Tickle, it is your comment, "Sorry, hate to be blunt, but... that's wrong" that is in fact entirely wrong, and so is your questioning of FFS' interpretation. </p></li>
<li><p>Fireflyscout's statement has been absolutely correct to this day, "The curve is not based on the raw scores of the testtakers, but rather on the relative difficulty of the questions on the test."</p></li>
<li><p>The above statement has been correct for the test before March 12, 2005. While we do not know much about the grading and equating of March 12, chances are that TCB will maintain its modus operandi. </p></li>
<li><p>Discussions about curves and relative difficulty of the tests are an absolute exercise in futility. Most anyone can create a curve that has a semblance of reality. Believing that the "posted" curve is real is not too smart.</p></li>
</ol>
<p>All this argument over the curve is sort of meaningless, since the curve can be anything at this point.</p>
<p>I agree that the curve for this test is determined quite a while ago. But the key is how it's determined. As CB said, the curve is not to fit the student's scores into a nice bell curve. They even admitted that the average for their tests is never 500. The curve is only to equate the test scores of different administrations. Therefore, there is what's equivalent of a reference test administration, with a reference scale. Based on that, the curves of all other administrations would be adjusted according to whether this particular administration is more or less difficult than the reference test.</p>
<p>Using that logic, let's apply it to the very FIRST administration of a new test. You can either say that there is no reference test, or think of the first test as the reference test. Therefore, the curve of this test can be entirely arbitrary, since it does not have to be adjusted to another reference test. All future tests must be adjusted to have scores equivalent to the ones of this test, but the First test does not have to be. The curve can be anything.</p>
<p>hey guys, lol i feel special now haha, um anyways let me give you a briefing of this curve...ok so this curve was given to all princeton review students by the princeton review "headquarters," the letter in the mail said this is the curve that princeton review will use from now on to scale there tests, after reviewing the paper itself, it said 'Source: College Board Field Test for the New SAT"...Now as for everyone's guess on whether curves are pre-determined or whether they are done after people have taken the test, they are pre-determined! Next, i do not randomly post curves for everyone's astonishment or to kill peoples hopes or make people happy, the reason i picked this curve is because it seems VERY likely and it has showed the trend most sat curves in the past year have shown, including psat. Lastly, the reason i mentioned the CB collegeboard book was because i was curious on how this curve corressponded with the collegeboard "ranges," ironically enough, each raw scaled score was half of the range, so a '700-800' was a 750 on my sheet...As you can see, for the aforementioned reason, i decided to use this curve and let everyone get a "most likely" (in my opinion 90%) glimpse of what they are to expect on the march 12th sat "curve"</p>
<p>oh man... gone are my dreams of getting a 2300 GODDAMNIT</p>
<p>wow, okay i can't believe people are investigating this, i'm going to scan this sheet and put it up on the web, and give everyone a link so people DO NOT THINK this is princeton review bull sh!t, i could care less about princeton review, i posted this curve in order to give everyone SOME insight on what to MOST LIKELY expect for the new sat...damn everyone...get a life lol</p>
<p>LET ME MAKE THIS CLEAR TO EVERYONE, NOT ONCE DID I SAY ON ANY ONE OF MY POSTS THAT THE CURVE I POSTED IT THE OFFICIAL CURVE, NOT ONCE, so pentsa, if princeton review got this curve from the college board FIELD TEST, LIKE I SAID 80 TIMES, then thats great for them, the curve is what you can, once again, let me say is conspciously, MOST LIKELY, expect on the test..sheesh.</p>
<p>I think everybody's really confused here:</p>
<p>The equating works like this. When ETS (NOT CB) makes the test, they attempt to make it at the same difficulty level as the other tests. When they score the tests, they score the equating sections too. Since the equating section has a known score distribution, they simply match the score distribution of the test to the score distribution of the equating section. To put it simplistically:</p>
<p>If people do well on the test and poorly on the equating section, then they have to give a harsher grade.</p>
<p>if people do poorly on equating and well on the test, then they have to give a better grade.</p>
<p>Obviously it's a LOT more complicated than this, and many other factors are taken into account. The reason it's not public is that you probably couldn't follow the statistical methods used for the equating. It might actually be available somewhere if you were really to search. Clearly, the grades are not predetermined, or the equating section would be unnecessary.</p>
<p>In the end, the grades match not percentiles but rather the grade that you would have gotten on any other administration. This is true even between the Jan SAT and the March SAT, which is why colleges shouldn't be saying "apples to oranges".</p>
<p>By the way, the equating section is NOT to try out new questions. They use small focus groups for that.</p>
<p>Hope this clears things up.</p>
<p>No one should criticize Xindianx for this curve
It's actually a pretty awesome curve</p>
<p>I think this curve will reflect March SAT by 90%</p>
<p>Yertle, I appreciate your explaining the process! Definitely much clearer than what I'd been trying to explain. Hopefully now people will understand it.</p>
<p>ok so is the equating section the not scored experimental section?</p>