Who wants a handful of free LSAT points?
July 20th, 2007 by Mike Barrett
Last week we talked about guessing on the SAT, and saw how the standard guessing advice doesn’t actually work against real SAT test questions. This week, we’ll take a look at guessing on the LSAT. (This strategy applies to the ACT as well.)
The LSAT is a very different test from the SAT, and one of the biggest differences is that LSAT-takers don’t lose points for marking wrong answers the way SAT-takers do. As a result, most LSAT-takers just go ahead and mark the answer they feel the best about for every question. Unfortunately, this leaves a lot of points on the table. If you know how to take advantage of the LSAT’s design, you can score 5 or 6 extra LSAT points that you don’t really “deserve.”
The first step in exploiting this design flaw on the LSAT is developing a feeling for when you’re right or wrong. This might sound like a crazy idea, but bear with me.
The LSAT, like all standardized tests, is designed according to rules and patterns, because it has to find a way to cover the same materials without using exactly the same questions every time. If you work with the test for long enough and develop a familiarity with these rules and patterns, it won’t be long before you know for sure when you’re answering a question correctly. I call it “seeing the halo.” (You can learn how to do this in my LSAT Guide, of course, but you can also get there by studying and analyzing real LSAT questions on your own. It just might take a lot longer.)
However you get there, you’ll want to come to a point where you know for sure if you’re right when you mark an answer choice on the LSAT, as opposed to when you’re taking a flyer on something. This sensitivity to correct answer choices is an essential part of mastering any standardized test, and it comes from understanding the rules and patterns of the test.
Once you’ve developed the ability to answer LSAT questions with absolute certainty, you’re most of the way home. The next step is understanding a fundamental design element of the LSAT:
LSAT answer choices are usually distributed unevenly within a section.
When you know these two things, guessing your way to a score that’s several points higher than you deserve is a piece of cake, as long as you can train yourself to do something counter-intuitive but very, very easy.
What you’ll need to do is leave blank the questions that you’re not sure about. Then, when you’ve finished with all the questions that you’re absolutely certain about in a particular section, you go back and “poll” the answer choices. In other words, you look at the answer choices that you’re completely certain about, and count up the number of (A)s, (B)s, (C)s, (D)s, and (E)s. You’re looking for the two or three answer choices that appear most often.
Then you go back to the questions you left blank so you can fill them in. (Remember—you NEVER want to leave an LSAT question blank!) Wherever possible, mark the answer choices for the letters that appeared the most often in your poll.
So if your poll reveals that (B) appeared the most often in the questions you knew for certain, and you come to a question that completely mystifies you, you mark (B)—since the LSAT answer choices are distributed unevenly within a section, the answer choice that appears most often in your poll is also slightly more likely to be correct for a question where you have no understanding at all of what’s going on.
But what if you do have some idea? Say you’re down to two answer choics but you can’t tell which one is better. If one of the answer choices appears more often than the other one in your poll, that’s the one you pick.
I know this sounds counter-intuitive and simplistic. It is. But it works much better than you’d expect.
If you use this technique on all the questions on the LSAT that you don’t feel sure of, you’ll end up with 5 or 6 extra points on average, which is no small thing.
Of course this is only one of many LSAT guessing strategies you can use—there’s also hunch guessing, constant guessing, and complementary guessing. You’ll want to pick the one that works the best for you.
The absolute WORST thing you can do is not address the idea of guessing at all in your LSAT preparation. Most LSAT-takers just wander through the test without ever really being sure of anything. For them, there’s no clear distinction between guessing and not guessing: Every answer they put down is about 70% conviction and 30% guess.
These test-takers lose out in all kinds of exciting ways: They never learn to approach the LSAT with certainty, which is the most damaging thing, and they never learn to apply strategies like the one I just taught you to pick up a few extra points and max out their potential.
So if you want to do the best you can on the LSAT, you need to take care of two things. First, make sure you think of the LSAT as a precise test with clear, correct answer choices that you can learn to recognize with certainty by applying the same habits over and over again. Second, make sure you understand the test’s design, which will allow you to apply this advanced guessing strategy (and many other strategies as well, of course).
If you’d like my help with your LSAT prep, you know where to find it—click the LSAT Guide Order Page link to the left. Whatever you decide, I wish you the best of luck!
Tell Others About This Article
This entry was posted on Friday, July 20th, 2007 at 1:17 pm and is filed under LSAT. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
July 20th, 2007 at 1:29 pm
Wow, man, I was just looking at your web site when you added this. Good stuff. I think I’ve been guessing the wrong way this whole time lol.
July 22nd, 2007 at 6:28 am
um, actually guessing wouldn’t be a good idea on the LSAT, trust me!!!!!!!!!!!!!!1
Most importantly study hard and if you are to guess, guess the same letter every time! like DDDDDDD; you are bound to get at least one or three right rather than getting them all wrong.
July 22nd, 2007 at 10:46 am
@Morgan–
That’s an interesting point, and it raises some important issues relating to test design.
If you’re going to guess on a set of several consecutive questions (say, for example, on questions 1 - 10 of a given section), the likelihood that guessing the same answer choice every time will result in correct answers for some of those consecutive questions increases as the number of consecutive answer choices increases, and increases slightly faster than the likelihood that guessing randomly for each question will result in correct answer choices.
But if you’re going to save the questions you can’t answer until the end of the time for a particular section and then go back and guess on them all at once, you’re probably going to be guessing on mostly non-consecutive questions (say, for example, numbers 1, 3, 4, 8, 10, 16, 17, 20). In this situation, guessing the same answer choice every time is no more likely to get you correct answers than guessing randomly.
How can that be?
The LSAT, like all of the major standardized tests, doesn’t allow a particular answer choice to be used more than a certain number of times in a row. On the LSAT, the upper limit on consecutive answer choices appears to be 4 (it should be noted that there might, in fact, not be a limit at all–if the LSAT’s answer-choice-ordering algorithm is truly random, it might simply be the case that it just hasn’t generated 5 consecutive identical answer choices yet). In a nutshell, all this means is that when you’re going to guess on a set of LSAT questions, the number of questions in the set and whether or not they’re consecutive will both affect your odds of success.
If the answer choices aren’t consecutive, then it doesn’t matter whether you guess the same answer choice every time or not, as long as the guesses you make are random.
But there’s another problem–guessing on large numbers of consecutive questions may increase the likelihood that the “DDDDDDD” strategy will generate correct answer choices, but it makes it impossible to score high on a given section because it can’t generate enough of them (assuming, again, that no more than 4 consecutive answers can use the same answer choice, which seems to be the case on the LSAT).
So the “DDDDDDD” strategy is slightly better than random guessing in the unlikely event that you have many consecutive questions to guess on. Where you have non-consecutive questions to guess on, it’s no better than random guessing. Further, it gets you fewer extra points as you answer more and more questions correctly, because answering more questions correctly leaves fewer questions to guess on.
This is what makes the guessing strategy I described in the original article so great: It works better than random guessing in any situation, and the odds that you’ll get extra points using it actually increase as you answer questions correctly. This works because your poll of the known answer choices becomes more reliable as more answer choices are included in it; even when there aren’t a lot of answer choices to poll, though, the strategy in the article is still a little bit more effective than random guessing or “DDDDDDD” guessing.
I do agree with you, though, on the idea that “guessing” is the wrong way to approach the LSAT (or any standardized test) in general. This is why I insist that people develop the ability to know for certain when they’re right about a question–LSAT-takers have to be aware of the rules and patterns on the test, and have to know that every single question has one definite answer, even if they can’t identify it. Focusing on learning the test’s rules and patterns allows test-takers to approach the majority of questions without guessing, and without having to cope with the uncertainty that the word “guessing” seems to suggest. The guessing strategy I recommended in the article is only to be used to answer the occasional questions that a student has difficulty with. (Not that the rules and patterns don’t apply to every question–they do. But test-takers are human, and under the pressures of test day they can’t be expected to perform flawlessly. For this reason it’s smart to have a backup plan for the questions you can’t figure out, and that’s what the guessing strategies are for.)