<p>I see a lot of people on here saying that, if you have no idea what the answer is, you should omit answers instead of filling them in. This is not necessarily true. For every wrong answer, you lose 1/4th of a point. For every correct answer, you gain 1 point.</p>
<p>You have an 80% chance of losing 1/4th of a point and a 20% chance of gaining a point per guessed question. Let's say you guess randomly on 50 questions. On average, you will get 40 wrong and 10 correct. All 40 wrong answers will add up to -10 points. All 10 correct answers will add up to +10. -10+10 = 0. On average, you will break even. That's the same as if you answered none of the questions. Therefore, guessing on one question is mathematically just as good as not answering the question.</p>
<p>There are other factors that I won't get into much now, though, like the time it takes to bubble in an answer, or how bubbling in every answer makes taking the test more straightforward because you don't have to worry about accidentally filling it in in the future (at least, for me). There's also the fact that people take the test more than once, so you might as well guess to try to get a better score than before.</p>
<p>I'm sure most people know this, but a surprising amount don't.</p>
<p>This is not that simple… Such statistical analysis is valid only when you are unsure about a huge number of questions, which will rarely, if ever, happen to someone on CC… First of all, everyone can usually eliminate at least one answer, so guessing would be (statistically speaking) always advantageous. Second, for some people (read “those who will for sure have doubts in no more than two questions”), getting a question wrong or leaving it in blank means exactly the same (as 0.5 rounds up)
Now let’s say you are unsure about and will randonly guess on 3 questions. Guessing will probably earn you 1 right question (because you cannot get 0.75 question right) and 2 wrong, a net earn of 1 - 0.5 = 0,5 raw point. On the other hand, in CR, for example, the questions are designed to be tricky, so, if you do not understand the text (unless you guess completely randomly, that is, without even reading the alternatives), you are very likely to get it wrong, even with an educated guess. This discussion is a lot more complex, and certainly, as people won’t take SAT endless times to wait for statistical results to show their effectiveness, luck (random variations) plays a far more important role, so I would say it’s more a matter of personal preference. I particularly never leave anything in blank.</p>