How to outsmart multiple choice test

William Poundstone, author of "Rock Breaks Scissors: A Practical Guide to Outguessing and Outwitting Almost Everybody," claims to have found several common patterns in multiple-choice tests, including computer-randomized exams like the SATs.
After examining 100 tests - 2,456 questions in total - from varied sources, including middle school, high school, college, and professional school exams; drivers' tests; licensing exams for firefighters and radio operators; and even newspaper quizzes, Poundstone says he found statistical patterns across all sources.

First, ignore conventional wisdom.
You've probably been given test-taking advice along the lines of "always guess the middle answer if you don't know," or "avoid any answer that uses the words never, always, all, or none," at some point in your life. However, according to Poundstone, this conventional wisdom doesn't hold up against statistics. In fact, he found that the answers "none of the above" or "all of the above" were correct 52% of the time. Choosing one of these answers gives you a 90% improvement over random guessing.
Look at the surrounding answers.
Poundstone found correct answer choices hardly repeated consecutively, so looking at the answers of the questions you do know will help you figure out the ones you're stuck on. For example, if you're stuck on question #2, but know that the answer to #1 is A and the answer to #3 is D, those choices can probably be eliminated for #2. Of course, "knowledge trumps outguessing," Poundstone reminds us. Cross out answers you know are wrong based on facts first.
Choose the longest answer.
Poundstone also noticed that the longest answer on multiple-choice tests was usually correct. "Test makers have to make sure that right answers are indisputably right," he says. "Often this demands some qualifying language. They may not try so hard with wrong answers." If one choice is noticeably longer than its counterparts, it's likely the correct answer.
Eliminate the outliers.
Some exams, like the SATs, are randomized using computers, negating any patterns usually found in the order of the answers. However, no matter their order, answer choices that are incongruent with the rest are usually wrong, according to Poundstone. He gives the following sample answers from an SAT practice test, without including the question:
A. haphazard...radical
B. inherent...controversial
C. improvised...startling
D. methodical...revolutionary
E. derivative...gradual
Because the meaning of "gradual" stands out from the other words in the right column, choice E can be eliminated. Poundstone then points out that "haphazard" and "improvised," have almost identical meanings. Because these choices are so close in meaning, A and C can also be eliminated, allowing you to narrow down over half the answers without even reading the question. "It's hard to see how one could be unambiguously correct and the other unambiguously wrong," he says. For the record, the correct answer is D.