<p>Hi, I am international student and I heard the word SAT for the first time this september :) I have seen loads of people talking about scoring curves. I would like to know what it means, how its calculated, stuff like that.</p>
<p>I don`t know what is a steep curve,but in the SAT,you always lose 1/4 of the point,when you answer incorrectly to a multiple choice question.
If you skip a question,you lose nothing.</p>
<p>I am not sure what a "curve" means with regard to the SAT.</p>
<p>For regular classroom tests in academic subjects like Physics, European History, Calculus,
Philosophy, etc. a curve is a numerial correction applied to every grade so that most or all of the class do not fail or get a poor grade. It almost but not always gives everyone in the class some points. </p>
<p>For a test that has 100 possible points, the simplest curve is to take the highest score,
add enough points for that to be a 100, and then give that many points to each student on that test.</p>
<p>Eample: The highest score is 94. That student gets 6 points for a 100 grade. A student with an 80 gets an 86, the failing 64 becomes a passing 70. The student who made a 42 gets a 48 (and drops the course LOL).</p>
<p>In some engineering classes, the high grade might be 34 out of 100. That student would get 66 points and an A. Everyone would get 66 points for a simple curve. The professor might choose to employ a sliding curve so that the score of 10 would not be passing.</p>
<p>There are a variety of algorithms for curving grades. The professor chooses, not the student.</p>
<p>A "curve wrecker" is someone who makes a high grade when everyone else has a low grade.</p>
<p>Grading on a curve, especially a complex one, allows the professor to measure the classes knowledge and progression without spending a lot of time crafting a test many students can pass without the curve. I recall an engineering class where every test was essentially the same questions with minor changes in order, dimension, and format to thwart rote memorization. The first testing yeilded a very low average. The final required no curve for everyone to pass with at least a C.</p>
<p>For the sat, I want to make a correction to what Ivan said. Each section has a certain amount of questions. Each question is worth one point. That said, if you leave a question blank you lose one point. If you get it wrong you lose 1.25 points. So therefore, if you leave 12 blank you lose 12 points. If you get 12 wrong, you lose 15 points. Each point is roughly worth 10 scaled points (out of 800) from 670-720 and is worth more from 720-800 especially for math. Below 670 I believe you lose 10 points for every two raw points. So someone who leaves 14 blank and someone who leaves 15 blank might get the same score.</p>
<p>Last year, for my AP bio class we had a general curve and a spread curve. First he would add 100- highest score to everyones grade. And then depending on the scores on the graph he would add a percentage. Usually from 20-30 percent. Sometimes up to 40 percent</p>
<p>The SAT curve is not like the one many people are familiar with from school: it is <em>not</em> designed to ensure that the average score is 500, say, or that only a certain number of people get 800s. It is meant to correct for small variations in test difficulty. In principle, if in a particular month only very unprepared students took the test, the average score could be 400 and the highest score 700.</p>
<p>@SuperMarioJ
I don understand. How can I lose points for leaving a question blank? Suppose I don answer any question and leave everything blank, my score would be in negative??</p>
<p>I don't necessarily agree with reading the paragraphs above, just make sure to look down at the chart that talks about raw scores and how it converts to the SAT score. You calculate your raw score by this:</p>
<p>Number Right (does not include wrong or blank answers) - (1/4 x number wrong)</p>
<p>So let's say for critical reading, your raw score is 65. Now, there's a range of scores like 740-800 for a raw score of 65, but Collegeboard recommends doing 740+800/2 to achieve a balanced score during practice tests. This helps with creating a more accurate prediction of your score before test day.</p>
<p>The curve is only used to make sure that this scoring plan is relatively equal for all different test administrations, so that it wouldn't matter what month you'd take it. That's the other part of the meaning of "standardized test." :D</p>
<p>Thanks for the link. But I am afraid I still don't understand. If the scoring pattern sis going to be the same for all tests, why do some people say some curves are tough while others are easy?</p>
<p>Good question, and it's because they have misleading terminology to describe it. They adjust the curve for the difficulty of the test also, hence the standardizing across all test dates. So let's say you had a particular difficult test day, and the curve will be slightly more lenient (not sure by how much). The converse will also happen, an 'easier" test day, and the curve will be more harsh. It has nothing to do with the people who take the test, it's all about how the SAT was created for that particular day.</p>
<p>@ neethu16
It's actually pretty simple.
There are two scores.
One is your raw score. The other is the scaled score(out of 800)
The raw score can not be higher than the number of questions, for math I believe that number is 54.
So 54-number blank- (number wrong X 1.25) = your raw score
Here is the chart.</p>
<p>Btw guys this scale is from January SAT 2008 --</p>
<p>If a test is harder for example, it might take more wrong questions to get 1 700. I remember one test that I left 4 blank I got 700. I remember another time when I got 5 wrong but got 710 because the curve was generous(the test was harder).</p>
<p>Basically, the SAT is a little like the MCAT. You are competing with those who are taking the test when you are. Actually, not many people might agree with me on this. I think the collegeboard has a percentage that has to get a question wrong before they consider it hard. That's why the order of difficulty is bs. Difficulty depends on the number of students that get it wrong, not the complexity of the question or the concept tested on the individual question. for example, the difficulty ranks for a test could go like this, 1,1,2,1,1,1,2,3,4,3,4,5,4,4,5,5
I don't think there is much more to explain</p>