A few days ago, my colleague Jon Denning wrote an excellent blog about how to best review your LSAT practice questions. His main point was that students often focus on whether they got the questions right, not on how confident they were in their choice of answers. This is problematic for several reasons:
- On the one hand, if you got the question wrong, you can easily dismiss your mistake ("I was stuck between two answer choices, so I guess I chose the wrong one.") Just because you were able to narrow it down to two possible answers doesn't mean you almost got it right. You didn't. They still tricked you, and you fell for it. Unfortunately, now that you know what the correct answer choice is, you can easily buy into it, without any further introspection on your part.
- On the other hand, If you got the question right, but were not at all certain in your answer, you can easily miss a similar question in the future. Paradoxically, this is an even less desirable outcome than the one described above, because you probably won't even review the questions you did well on: you will either forget that you struggled on some of them, or you will pat yourself on the back for getting lucky. Consequently, you will forfeit the opportunity to analyze a potentially confusing situation, making you liable to missing a similar question in the future.
Clearly, then, the confidence with which you answer your questions is just as important as whether or not you get the answers right, because low-confidence questions - even if answered correctly - can give you insight into underlying conceptual weaknesses, such as causal or conditional reasoning, numbers and percentages, etc. I like Jon's solution of quantifiably noting exactly how you certain feel about each question using a numbering system of 1-5, but I suspect this system would work best on homework problem sets or untimed practice sections. If you are being timed (or are just lazy), you can simplify your approach as follows:
On a timed practice test, simply circle any question where your estimated probability of success is less than 50%.
That way, even if you get the question right, you will still get the opportunity to analyze it more closely after scoring your test. Granted, this advice is probably better suited to RC and LR questions, because the confidence with which you approach a LG question is almost always a reflection of your mastery of the game itself, not of your understanding of each specific question in it (though there are exceptions to this). In general, given the correct setup and understanding of how the rules operate, the confidence with which you approach the questions should be relatively high. And inversely, if you misunderstand a rule or a key element in the scenario, the questions simply won't work.
One way to measure your confidence in Logic Games is to analyze how quickly you are answering the questions. A game that took you 12 minutes to complete is a game you should thoroughly review, even if you got all the questions right. Yes, you can plug-and-chug your way through every damn question, but that doesn't mean you did well on the game itself, even if you got the questions right. Struggling to complete the questions is almost always an indication that you're missing a key piece of the puzzle (an inference, a template, etc.), warranting a closer look at the puzzle itself.
Photo courtesy of Susana Fernandez.