To 170+ hopefuls: stop taking so many practice tests.
Posted: Fri Aug 24, 2012 9:34 am
Hi all, it's been a while since I was on these forums in your position- prepping for the LSAT, putting in crazy 6 hour days and striving for a 170+. The information on TLS was a huge help for me, so now I'm back to share some of my own wisdom. For full disclosure, I started at a diagnostic of about a 153 and ended up with a 174 on the real thing. Now I teach LSAT courses for a major test prep company. After my own experience making a huge improvement, and teaching other students to do the same, here's some advice that I believe all students can use, especially those aiming for the 170s.
My general thesis is that students who aim for 170+ (and especially TLSers, though this goes hand in hand) place way too much of an emphasis on practice exams. Instead, here's what you should do:
How to improve your score (the one that will count, not PT#XX)
-Review the concepts
you should review the ENTIRE method for a question type before starting a problem set. My preference is the memorize it- every single piece of the process, AND the common right and wrong answers (for example, before a starting a set of main point questions, I would recite: "Look for the conclusion, and don't worry if it's valid or not. Watch out for sub-conclusions- "Thus" and "Therefore" are often traps. Look for a shift in the author's attitude, and remember, the conclusion is usually not the last line). It's not good enough to just do a bunch of questions of one type- you need to be thinking about the correct strategy the whole time while you're doing questions, and consciously be trying to apply them. "Oh, a Sufficient Condition question: I should look for a new term in the conclusion that isn't mentioned in the premises, and find an answer that creates a link between that term and the premises."
-Isolated practice by type
Almost all test-prep courses and books, regardless of company, are structured a certain way: isolate each question type one by one, master the LSAT in piece by piece, and then put it together later. They're all like this for a reason- there is no super secret 170+ strategy, where I tell advanced students to take practice tests and tell the rest to go piece by piece. All students should practice concepts one at a time because it's more effective for students to master the LSAT in parts, instead of trying to do it all at once. It's hard enough to learn flaw questions doing 100 of them in a row, let alone switching between flaw, must be true, and parallel, and sufficient/necessary. Even if you're trying to get an elite score, the strategy is no different. You've taken 10 practice tests and you average a 171? That just means you're screwing up under time, and you need to take more tests right? You'll get it right the next time? Consistently averaging a 171 means you're consistently getting 9-10 questions wrong, and there's something fundamentally flawed in your logic skills that you need to find. It can be difficult to identify- students always expect their weaknesses to manifest themselves as a question type- ("8 of my 10 wrong answers were strengthen") but for 170+ hopefuls, that usually doesn't happen. It's more likely that its something smaller than spans multiple question types- identifying conclusions correctly, diagramming, etc.
-Untimed practice
If you can't get it PERFECT in slow motion, how can you possibly get close to perfect under a time limit? When you practice, you should aim for 100% accuracy, and take all the time that we need to get there. If that means you only do 20 questions one day instead of 100, so be it. Quality over quantity. While you're prepping, you don't get any real points for getting the questions right- your goal shouldn't simply be to select the right answer. What you're trying to accomplish while drilling questions is to learn from them, and take away lessons for the future. Timing yourself doesn't allow you to do that. For example in logic games, I tell students to time themselves, but starting at 0:00. I say don't look at the clock- take as long as you need to understand the game, finish it, and just use the timer to see how long it took you. Maybe a game takes you 13 minutes one day, and you got 5/7. Then you redo that game 3 days later, go 5/7 again, but did it in 11 minutes. Then 6/7 in 9 minutes, etc. A timed practice test wouldn't have allowed you to make that improvement. You probably would have given up at some point due to the pressure of time and need to work on the other games.
-Reviewing the content
It's not enough to just do the questions. Like I said, you need to review them so you can learn from them. When I was prepping, I liked to take notes on questions: I write down what I did wrong ("I didn't pay attention to the cannot be together rule") and then give advice for the future ("cannot be together rules create options!"). When you answer a question incorrectly (especially in LR and RC), you've done TWO things wrong: you've read a logically incorrect answer and said it was correct, and you've read the correct answer and determined it was wrong. It's up to you to prove to yourself, in concrete terms (not just "oh, i get it") what logical deficiency makes the wrong answer 100% wrong and what by what logic is the correct answer 100% right. You also need to review the questions you answered correctly, to make sure you were correct for the right reasons. If you had a 50/50 question where you're deciding between A and B, liked A more, and correctly chose A, you haven't learned all you can from that question. You need to find out what makes B 100% wrong. Because the next 50/50, you may not be so lucky.
-Reviewing yourself
You can understand the concepts fully, but not be applying them correctly. For example, plenty of students understand what makes a statement strong or weak but don't use it to eliminate answers. When reviewing questions, I regularly ask myself:
-What did I do wrong? Why did I like the answer that I chose?
-How could I have avoided it?
-What should I have done to get it right?
-What should I do moving forward?
How NOT to improve your score:
-Blowing through questions
I meet a lot of students who took a different prep course or self studied and have done almost every LSAT question, but are still doing terrible. That's because you need QUALITY practice, not just quantity. Avoid the mentality that says to yourself, "I will be better after I finish these 100 questions." There's more to it than that.
-Timing yourself
A common misconception is that in order to test faster, you need to practice fast. No. You go faster by being better at logic. You go faster by having a better mastery of the concepts. You go faster by having broken down so many questions untimed, that you can quickly recognize patterns from previous questions. Yes, timed practice does have a few benefits- of course, at some point you're going to need to establish the right pace, and practice going at that pace a few times. You'll also reduce some test day anxiety by getting used to the clock. But none of this doesn't MAKE you faster, and students will too often believe that "time management" is the reason they're missing questions. But the best way to get faster is gain a better mastery of the underlying concepts, and that requires going slow.
-Taking practice exams
Practice exams are the most overrated part of LSAT prep. I understand why people want to take them- eventually on game day you are going to take a test, so you need to practice taking a test. But here's the problem. You do not learn anything from taking a practice test. At the end of a 4 hour practice exam, you are not better at understanding and applying LSAT concepts than you were before you took it. All you get from a practice exam is a snapshot of how good you are at the LSAT that day. You've audited yourself and gotten a result- maybe you bombed and are more motivated, or maybe you did well and you have some peace of mind. But you're not any better. You could have taken those 4 hours and drilled concepts one by one. Untimed, isolated practice. That's where improvement comes from.
Disclaimer
Don't get me wrong, practice exams are indeed an important part of LSAT study. Eventually, you need to take all those concepts that you've mastered and practice doing them all at the same time- switching back and forth from question type to question type, flipping from one strategy to another. You need to practice working under stress, keeping your cool, and maintaining your desired pace. But this should not happen until you are extremely accurate in an untimed setting. This is ESPECIALLY true for 170+ hopefuls. Those few points that you have left to gain are the most difficult. I tell my students, a 170 requires about 90% accuracy. How do you get 90% accuracy in a timed setting? By shooting for 100% accuracy in an untimed setting.
My general thesis is that students who aim for 170+ (and especially TLSers, though this goes hand in hand) place way too much of an emphasis on practice exams. Instead, here's what you should do:
How to improve your score (the one that will count, not PT#XX)
-Review the concepts
you should review the ENTIRE method for a question type before starting a problem set. My preference is the memorize it- every single piece of the process, AND the common right and wrong answers (for example, before a starting a set of main point questions, I would recite: "Look for the conclusion, and don't worry if it's valid or not. Watch out for sub-conclusions- "Thus" and "Therefore" are often traps. Look for a shift in the author's attitude, and remember, the conclusion is usually not the last line). It's not good enough to just do a bunch of questions of one type- you need to be thinking about the correct strategy the whole time while you're doing questions, and consciously be trying to apply them. "Oh, a Sufficient Condition question: I should look for a new term in the conclusion that isn't mentioned in the premises, and find an answer that creates a link between that term and the premises."
-Isolated practice by type
Almost all test-prep courses and books, regardless of company, are structured a certain way: isolate each question type one by one, master the LSAT in piece by piece, and then put it together later. They're all like this for a reason- there is no super secret 170+ strategy, where I tell advanced students to take practice tests and tell the rest to go piece by piece. All students should practice concepts one at a time because it's more effective for students to master the LSAT in parts, instead of trying to do it all at once. It's hard enough to learn flaw questions doing 100 of them in a row, let alone switching between flaw, must be true, and parallel, and sufficient/necessary. Even if you're trying to get an elite score, the strategy is no different. You've taken 10 practice tests and you average a 171? That just means you're screwing up under time, and you need to take more tests right? You'll get it right the next time? Consistently averaging a 171 means you're consistently getting 9-10 questions wrong, and there's something fundamentally flawed in your logic skills that you need to find. It can be difficult to identify- students always expect their weaknesses to manifest themselves as a question type- ("8 of my 10 wrong answers were strengthen") but for 170+ hopefuls, that usually doesn't happen. It's more likely that its something smaller than spans multiple question types- identifying conclusions correctly, diagramming, etc.
-Untimed practice
If you can't get it PERFECT in slow motion, how can you possibly get close to perfect under a time limit? When you practice, you should aim for 100% accuracy, and take all the time that we need to get there. If that means you only do 20 questions one day instead of 100, so be it. Quality over quantity. While you're prepping, you don't get any real points for getting the questions right- your goal shouldn't simply be to select the right answer. What you're trying to accomplish while drilling questions is to learn from them, and take away lessons for the future. Timing yourself doesn't allow you to do that. For example in logic games, I tell students to time themselves, but starting at 0:00. I say don't look at the clock- take as long as you need to understand the game, finish it, and just use the timer to see how long it took you. Maybe a game takes you 13 minutes one day, and you got 5/7. Then you redo that game 3 days later, go 5/7 again, but did it in 11 minutes. Then 6/7 in 9 minutes, etc. A timed practice test wouldn't have allowed you to make that improvement. You probably would have given up at some point due to the pressure of time and need to work on the other games.
-Reviewing the content
It's not enough to just do the questions. Like I said, you need to review them so you can learn from them. When I was prepping, I liked to take notes on questions: I write down what I did wrong ("I didn't pay attention to the cannot be together rule") and then give advice for the future ("cannot be together rules create options!"). When you answer a question incorrectly (especially in LR and RC), you've done TWO things wrong: you've read a logically incorrect answer and said it was correct, and you've read the correct answer and determined it was wrong. It's up to you to prove to yourself, in concrete terms (not just "oh, i get it") what logical deficiency makes the wrong answer 100% wrong and what by what logic is the correct answer 100% right. You also need to review the questions you answered correctly, to make sure you were correct for the right reasons. If you had a 50/50 question where you're deciding between A and B, liked A more, and correctly chose A, you haven't learned all you can from that question. You need to find out what makes B 100% wrong. Because the next 50/50, you may not be so lucky.
-Reviewing yourself
You can understand the concepts fully, but not be applying them correctly. For example, plenty of students understand what makes a statement strong or weak but don't use it to eliminate answers. When reviewing questions, I regularly ask myself:
-What did I do wrong? Why did I like the answer that I chose?
-How could I have avoided it?
-What should I have done to get it right?
-What should I do moving forward?
How NOT to improve your score:
-Blowing through questions
I meet a lot of students who took a different prep course or self studied and have done almost every LSAT question, but are still doing terrible. That's because you need QUALITY practice, not just quantity. Avoid the mentality that says to yourself, "I will be better after I finish these 100 questions." There's more to it than that.
-Timing yourself
A common misconception is that in order to test faster, you need to practice fast. No. You go faster by being better at logic. You go faster by having a better mastery of the concepts. You go faster by having broken down so many questions untimed, that you can quickly recognize patterns from previous questions. Yes, timed practice does have a few benefits- of course, at some point you're going to need to establish the right pace, and practice going at that pace a few times. You'll also reduce some test day anxiety by getting used to the clock. But none of this doesn't MAKE you faster, and students will too often believe that "time management" is the reason they're missing questions. But the best way to get faster is gain a better mastery of the underlying concepts, and that requires going slow.
-Taking practice exams
Practice exams are the most overrated part of LSAT prep. I understand why people want to take them- eventually on game day you are going to take a test, so you need to practice taking a test. But here's the problem. You do not learn anything from taking a practice test. At the end of a 4 hour practice exam, you are not better at understanding and applying LSAT concepts than you were before you took it. All you get from a practice exam is a snapshot of how good you are at the LSAT that day. You've audited yourself and gotten a result- maybe you bombed and are more motivated, or maybe you did well and you have some peace of mind. But you're not any better. You could have taken those 4 hours and drilled concepts one by one. Untimed, isolated practice. That's where improvement comes from.
Disclaimer
Don't get me wrong, practice exams are indeed an important part of LSAT study. Eventually, you need to take all those concepts that you've mastered and practice doing them all at the same time- switching back and forth from question type to question type, flipping from one strategy to another. You need to practice working under stress, keeping your cool, and maintaining your desired pace. But this should not happen until you are extremely accurate in an untimed setting. This is ESPECIALLY true for 170+ hopefuls. Those few points that you have left to gain are the most difficult. I tell my students, a 170 requires about 90% accuracy. How do you get 90% accuracy in a timed setting? By shooting for 100% accuracy in an untimed setting.
