June SAT Curve Predictions

<p>I'm really in the dark when curve predictions come in (as are most people seeing as its nearly impossible to get it exact).</p>

<p>what im wondering is if there is any way to tell if the curve is supposed to be ridiculous (as in the january and march tests) or more average (like may's)</p>

<p>June is my fourth and final attempt at the SAT and after getting screwed by the March test i wanna go into this one having an idea of a possible curve</p>

<p>P.S. ive been using Xiggi's method of prep for June</p>

<p>you cannot know what the curve will be like. it is a standardized test</p>

<p>I would say more average.</p>

<p>How can you define getting "screwed" by a test curve?</p>

<p>My guess is the reason the May test was so forgiving is due to the fact that many of the 'smartest' kids did not take the exam because of AP tests.</p>

<p>^that is not how the curve works. The curve is PRE-determined based on overall difficulty (they rank each individual question by difficulty) and is not affected by how test-takers perform that day. this discussion is completely useless given that fact</p>

<p>this is the main curve that standardized tests employ. i do not know if they slightly adjust the curve based on performance but it is not significant.</p>

<p>@Nick567 "screwed by march curve" definition- my same raw scores i got on this test yield much higher scaled scores on SAT's from other months and according to all of the BB Book sample test scaled ranges (almost 150 points total)</p>

<p>@crazybandit yes i know thats why i said its nearly impossible thats why the question is asking for predictions lol </p>

<p>@apyyyy yeah i understand that. what im thinking and hoping for is that many smart kids dont take June SAT or do poorly cause its in summer and too lazy (((: one can dream</p>

<p>@crazybandit you're wrong. the curve is made AFTER the test is administered based off of student performance because the average for each portion (M, CR, and W) need to be around 500</p>

<p>
[quote]
@apyyyy yeah i understand that. what im thinking and hoping for is that many smart kids dont take June SAT or do poorly cause its in summer and too lazy (((: one can dream

[/quote]
</p>

<p>like i said, how well students do does not affect the curve. there is no point for predictions. if anything it is just superstition</p>

<p>
[quote]
you're wrong. the curve is made AFTER the test is administered based off of student performance

[/quote]
</p>

<p>And your source is?</p>

<p>The</a> SAT Scoring Scale</p>

<p>thats the quickest i could find. ive seen examples and evidence all over though</p>

<p>at least thats what ive always gone by and ive seen averages for SAT scaled scores from CR, W, and M (from links people have posted on CC) and they are 99% of the time between 490 and 510</p>

<p>The SAT curving process is not a straight-forward "let-make-500-the-average" adjustment. Here is what the College Board says:</p>

<p>
[quote]
We do a statistical analysis to make sure the test is an accurate representation of your skills. The unscored section of the test also helps us ensure the test is fair. Questions in the unscored section are not factored into your SAT score.</p>

<p>In our statistical analysis, 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. 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>Every SAT includes a 25-minute section, which doesn't count toward your final score. It may be a critical reading, mathematics, or multiple-choice writing section.</p>

<p>We do this because it helps us assess questions for next year's test, and it ensures that the SAT accurately reflects your skills. Also, the unscored section helps us account for minor differences in difficulty across all the different forms of the test.

[/quote]
</p>

<p>"450 on one edition of a test reflects the same ability as a score of 450 on another edition of the test"</p>

<p>so thats saying that on a "harder" test the curve has to be more generous to allow a 450 to equal a 450 on an "easier" test with a harsher curve. im not saying the curve is based off of making the average 500 (thats due to the bell curve). but the average WILL always be 500 BECAUSE of the curve.</p>

<p>Like on my March SAT math i skipped none and i got 3 wrong (all medium level silly mistakes -___- and obviously out of 54) so raw 50. that yielded a 720 on March's Test. and -3 on the reading was a 740! thats the lowest ive ever heard of. why would that same raw be higher on another test? just because of TCB's evaluation of difficulty? what if there was a month that they considered the test the be really easy and there were no 800s which is where the bell curve comes into play making the highest one an 800.</p>

<p>idk maybe im wrong i just find it hard to believe the curve could be predetermined. one thing im a strong believer in is that TCB is gradually making all of their tests (SAT, SATII, AP) easier so that the curves can be more harsh.</p>

<p>"Equating ensures that the different forms of the test or the level of ability of the students with whom you are tested do not affect your score." thats from collegeboard also. i do understand it but something is stopping me from believing it. like why has march provided a harsher curve 99% of the time (my reasoning: more "smart" juniors take it making the curve more harsh)</p>

<p>well, if the College Board ranks the individual questions based on difficulty, what makes it hard to believe that it also ranks the entire test based on overall difficulty? how students perform is accounted for in "experimental" sections from previous exams. that is their purpose. it is not true that the march SAT always has a harsher curve.</p>

<p>when people predict curves and make statements on the basis of difficulty, they are referring to how hard the test actually was, not how other students performed (although obviously there is a correlation that is unaccounted for in the curving process). meaning predicting the curve before you take the test is futile</p>

<p>i agree with you crazybandit. (mainly because of the realization that the difficulty of the test is a prediction of how students will perform)</p>

<p>having said that. what do you guys think will happen with the June SAT?? more towards the easy side (harsh curve) or difficult (easier curve)</p>

<p>^ Predictions of difficulty are meaningless.</p>

<p>Here's my usual curve info: </p>

<p>1) first read this</a> College Board white paper for the technical information about the curve / equating / standardization of the SAT. This document is the primary resource.</p>

<p>2) next read [thread=760312]this thread[/thread] for a long but example-filled discussion of the curve</p>

<p>3) assuming you skipped (1) and (2), here is the executive summary:</p>

<p>The curve adjusts solely for the difficulty of the test: a harder test will have a nicer curve and an easier test will have a harsher curve. Someone who gets a 45/54 raw math score on a harder test will receive a better scaled score than a 45/54 on an easier test.</p>

<p>The curve is not adjusted so as to make the average score 500 or any other particular number. Likewise, the number of people receiving 800s is not fixed or predetermined.</p>

<p>The curve is not determined before the test is administered (although the difficulty of each question independently is known beforehand).</p>

<p>The curve does not depend on or account for the quality (or lack thereof) of the people taking the test in a given month.</p>

<p>In the end, the curve makes the test as fair as possible by making a 640 (say) mean the same no matter when you took the test.</p>

<p>So, predicting an upcoming curve based on previous history (see the sat curves chart linked to in countless threads) or choosing to take the SAT in a particular month due to a perceived pattern in the curves (I don't know of any pattern anyway) doesn't make much sense.</p>

<p>^ Your first link is not working.</p>

<p>^Thanks for the heads up , forgot the CB redesigned the web site, sigh. New link:</p>

<p>College</a> Board Equating / Scaling / Curving White Paper</p>

<p>Also of interest is this slightly older but useful [url=<a href="http://www.collegeboard.org/html/pdf/howsatmade.pdf%5Ddocument%5B/url"&gt;www.collegeboard.org/html/pdf/howsatmade.pdf]document[/url&lt;/a&gt;] on the process each question goes through to determine difficulty etc.</p>