5 ms·

The LSAT is what's known as a "learnable" exam, which means that taking it without preparation is 100% not indicative of the final score one could get. This is

by dvt 9y ago

The LSAT is what's known as a "learnable" exam, which means that taking it without preparation is 100% not indicative of the final score one could get. This is well-reported all over law school forums where people took it cold and got a 140 and after months of study ended up with a 177. Or started with a 155 and ended with a 165. The LSAT (especially the logic games) is mostly pattern-recognition and understanding the "type" of problem you're being asked to solve. Getting a high score (e.g. over 170) the first time around is probably impossible because the questions are formulated so weirdly (in other words, you'd need to be familiar with the format to do well). Further, you'd probably not be used to the time constraint and end up guessing on whatever you didn't have time to answer. This article is just stupid bravado, but then again, it's Vice and they're not exactly known for hard-hitting journalism. NB: 158 going in cold is very high (which is why the article seems like humblebragging to me). A 20 point increase is doable with an intense studying regimen. And with a 178, you can write your ticket to basically any law school (given a 3.5+ undergrad GPA).

zippergz 9y ago
The GPA is an important caveat. I scored in the 99th percentile on the LSAT, and still couldn't get into any of the top tier schools I applied to because I slacked off too much in undergrad.
- dvt 9y ago
  Yeah, with a sub-3.0 GPA even with like a 180 on the LSAT, you're basically out of the t14 (except for a few splitter-friendly schools if you get lucky). The good news is that if you scored in the 99th percentile on the LSAT, you probably don't need to go to law school to be successful :)
  - zippergz 9y ago
    Yes, in fact I view this as one of the great "dodged bullets" of my life. Because I wasn't accepted by my top choice schools (I had several who I wasn't interested in beating my door down), I decided not to go to law school at all. In retrospect, that was the best possible outcome.
    - sgpl 9y ago
      Do you have an email I could reach you at? I'd love to ask you a few questions privately!
  - geebee 9y ago
    Could be. I'd be interesting in knowing if there's a big difference between sub 3.0 STEM GPAs and humanities. The problem with humanities is that there's just no need to consider you, they are already going to reject a large number of people with high grades and test scores. I suppose if you were a STEM major, the schools might need consider applications that aren't perfect if they want a decent representation from the sciences and engineering. I'd be interesting in seeing some data on this. At a place like Berkeley or other difficult STEM programs with low GPAs, I'm prepared to believe that a 2.9 GPA doesn't indicate someone would have any trouble with law school. It's very different from running this GPA in history at a top private. It's also worth noting that elite law schools have vanishingly low attrition rates - Columbia, for instance, has a 1L attrition rate of 0.3%. Yale often has no 1L attrition. Although I don't have data by engineering schools, my experience in a grad program at Berkeley is that such low attrition would be inconceivable in STEM, even at the MS level (and without question at the PhD level, where attrition can be as high as 50%, really, over 100 times higher than at an elite law school).
jacobolus 9y ago
I know a guy (with BA in physics) who almost aced the LSAT going in cold (i.e. had never seen a practice test), on a dare, without any intention of becoming a lawyer. At some point a decade ago I was vaguely curious to try the same, but not curious enough to overcome laziness and actually sign up for the test (and nobody dared me). My impression is that people who spend a lot of time thinking critically about text, have a solid understanding of grammar, and have done lots of logic puzzles (say, work through Smullyan’s books for fun) and other mathematics problems tend to do pretty well without any specific training. One reason many Americans have trouble with tests like the LSAT is that we spend very little time or effort up through high school on solving real math problems (as compared to trivial mechanical exercises), with the result that students’ technical reading comprehension and critical thinking also tends to be pretty poor. See e.g. http://www.de.ufpe.br/~toom/travel/sweden05/WP-SWEDEN-NEW.pdf http://www.de.ufpe.br/~toom/travel/sweden05/WP-SWEDEN-NEW.pd...
- watwut 9y ago
  I am non-American and pretty much everyone I knew who wanted to study in America prepared for those tests extensively for months. Yes, if you did the same style of problems a lot without lsat being target, then you are practically prepared. However, just being good in math in general and good in solving hard math exercises wont make it. You would spend a lot of time thinking and trying to figure things out, while if you prepare you will instantly recognize problem and "solve it fast". The only difference there is that Americans see test being learnable as a bad thing and talk about preparing on it as almost cheating. Here it was common sense, of course all tests are learnable and of course you gotta learn to do well.
  - jacobolus 9y ago
    Believe it or not, there exist people who can skim for key points, make fast mental diagrams, “instantly recognize” what task is being asked, and quickly solve a wide variety of standardized test questions of types they have not explicitly seen before. These people are not superhuman or anything, they just have spent a lot of time doing related kinds of tasks (close reading, competitive debate, code debugging, recreational math problems, etc.) Recreational logic puzzles get way harder than anything that shows up on the LSAT, and there are students out there who work through books of hard logic puzzles for fun, the way other people might read novels or play video games or whatever. As an individual, studying for standardized tests isn’t “cheating” or a “bad thing”. It’s often the most rational use of time for folks with specific professional ambitions. But having everyone in the society spend a significant amount of time and stress on a handful of particular standardized tests is IMO a huge waste of collective attention and focus [though perhaps there isn’t any alternative gatekeeping system for law schools which won’t be equally gamed in a time-wasting way]. One problem with the increasing numbers of standardized tests at every level of education is that time spent training for a specific test in e.g. 5th grade has very little long-term benefit for a student. The standard school curriculum (and frankly a lot of explicit test prep) doesn’t really do a very efficient or effective job of training the kinds of skills that would help someone do well on novel types of standardized test questions. Instead of training reading comprehension, working memory, generic problem solving heuristics, quick recall/matching, problem-solving time management, etc., they tend to focus on the particular features of specific types of problems, and teach memorized recipes or collections of factual information. Obviously this is a reasonable approach if the goals of the training are very narrow and specific. It doesn’t seem to me like the best generic training to do successful intellectual work. A lot (hundreds if not thousands of papers) has been written about this subject in the scholarly literature by professors who were frustrated that their high-achieving undergraduate {math, physics, engineering, ...} students seem to be able to memorize recipes for solving specific problems very well and ace their usual tests, but have great difficulty synthesizing their knowledge or solving non-standard problems, and then spent some time and effort trying to understand why that is.
brooklyn_ashey 9y ago
As a former LSAT tutor, I can tell you that 158 isn't humblebragging for a journalist. Most of my tast-taking students were in the 150s before tutoring. I was a test prep tutor for other tests for a few years before I added the LSAT to my repertoire. Most students had no idea how to set up charts or mental pictures for the logic problems. While I understood this aspect right away, I needed to work on speed before I helped anyone else. Later in life, when learned how to code, I realized that coders would do well at this section. If I'd known how to code back then, I'd have used algorithmic thinkg techniques to help students improve their LSAT score.
- matt_wulfeck 9y ago
  LSAT logic questions are in many ways similar to common interview questions: listen to the problem, identify key words, apply known solving pattern.
- scotty79 9y ago
  When I did Mensa exams I also felt that coders might have advantage there.
- wahern 9y ago
  > Most students had no idea how to set up charts or mental > pictures for the logic problems. I took Introduction to Symbolic Logic my first year in college. It was a tiny class, which was a real shame. Many years later, I remember my LSAT instructor spending the first two sessions explaining, e.g., Venn diagrams to the class. I was like, "is this guy serious?" "Who could possibly need a refresher on this stuff?" But I guess I was the odd one in terms of formal exposure. Symbolic logic should be required for every college student. I can't think of any better class for helping produce well-rounded students with solid critical thinking skills, regardless of their academic focus. The class is easy and it has almost no direct application whatsoever. But that's the point. Every college student should be able to get through it without sweating, even if it seems initially daunting. And it has no direct application precisely because it's a distillation of some of the very basic mechanisms of rational thinking--mechanisms that everybody and everything glosses over. It's the one class that puts the really simple stuff front-and-center, which after a semester of lessons makes it so much easier to recognize the logical patterns (or lack thereof) elsewhere. Practicing logic negations for example is incredibly useful, especially in the humanities, but even in the hard sciences. The only downside is producing more of those annoying people who apply formal logic in every context, oblivious to the limitations. But that's better than the current state of affairs, I think. And I guess those annoying people keep everybody on their toes.

by dvt 9y ago

zippergz 9y ago

The GPA is an important caveat. I scored in the 99th percentile on the LSAT, and still couldn't get into any of the top tier schools I applied to because I slacked off too much in undergrad.

dvt 9y ago
Yeah, with a sub-3.0 GPA even with like a 180 on the LSAT, you're basically out of the t14 (except for a few splitter-friendly schools if you get lucky). The good news is that if you scored in the 99th percentile on the LSAT, you probably don't need to go to law school to be successful :)
- zippergz 9y ago
  Yes, in fact I view this as one of the great "dodged bullets" of my life. Because I wasn't accepted by my top choice schools (I had several who I wasn't interested in beating my door down), I decided not to go to law school at all. In retrospect, that was the best possible outcome.
  - sgpl 9y ago
    Do you have an email I could reach you at? I'd love to ask you a few questions privately!
- geebee 9y ago
  Could be. I'd be interesting in knowing if there's a big difference between sub 3.0 STEM GPAs and humanities. The problem with humanities is that there's just no need to consider you, they are already going to reject a large number of people with high grades and test scores. I suppose if you were a STEM major, the schools might need consider applications that aren't perfect if they want a decent representation from the sciences and engineering. I'd be interesting in seeing some data on this. At a place like Berkeley or other difficult STEM programs with low GPAs, I'm prepared to believe that a 2.9 GPA doesn't indicate someone would have any trouble with law school. It's very different from running this GPA in history at a top private. It's also worth noting that elite law schools have vanishingly low attrition rates - Columbia, for instance, has a 1L attrition rate of 0.3%. Yale often has no 1L attrition. Although I don't have data by engineering schools, my experience in a grad program at Berkeley is that such low attrition would be inconceivable in STEM, even at the MS level (and without question at the PhD level, where attrition can be as high as 50%, really, over 100 times higher than at an elite law school).

jacobolus 9y ago

I know a guy (with BA in physics) who almost aced the LSAT going in cold (i.e. had never seen a practice test), on a dare, without any intention of becoming a lawyer. At some point a decade ago I was vaguely curious to try the same, but not curious enough to overcome laziness and actually sign up for the test (and nobody dared me). My impression is that people who spend a lot of time thinking critically about text, have a solid understanding of grammar, and have done lots of logic puzzles (say, work through Smullyan’s books for fun) and other mathematics problems tend to do pretty well without any specific training. One reason many Americans have trouble with tests like the LSAT is that we spend very little time or effort up through high school on solving real math problems (as compared to trivial mechanical exercises), with the result that students’ technical reading comprehension and critical thinking also tends to be pretty poor. See e.g. http://www.de.ufpe.br/~toom/travel/sweden05/WP-SWEDEN-NEW.pdf http://www.de.ufpe.br/~toom/travel/sweden05/WP-SWEDEN-NEW.pd...

watwut 9y ago
I am non-American and pretty much everyone I knew who wanted to study in America prepared for those tests extensively for months. Yes, if you did the same style of problems a lot without lsat being target, then you are practically prepared. However, just being good in math in general and good in solving hard math exercises wont make it. You would spend a lot of time thinking and trying to figure things out, while if you prepare you will instantly recognize problem and "solve it fast". The only difference there is that Americans see test being learnable as a bad thing and talk about preparing on it as almost cheating. Here it was common sense, of course all tests are learnable and of course you gotta learn to do well.
- jacobolus 9y ago
  Believe it or not, there exist people who can skim for key points, make fast mental diagrams, “instantly recognize” what task is being asked, and quickly solve a wide variety of standardized test questions of types they have not explicitly seen before. These people are not superhuman or anything, they just have spent a lot of time doing related kinds of tasks (close reading, competitive debate, code debugging, recreational math problems, etc.) Recreational logic puzzles get way harder than anything that shows up on the LSAT, and there are students out there who work through books of hard logic puzzles for fun, the way other people might read novels or play video games or whatever. As an individual, studying for standardized tests isn’t “cheating” or a “bad thing”. It’s often the most rational use of time for folks with specific professional ambitions. But having everyone in the society spend a significant amount of time and stress on a handful of particular standardized tests is IMO a huge waste of collective attention and focus [though perhaps there isn’t any alternative gatekeeping system for law schools which won’t be equally gamed in a time-wasting way]. One problem with the increasing numbers of standardized tests at every level of education is that time spent training for a specific test in e.g. 5th grade has very little long-term benefit for a student. The standard school curriculum (and frankly a lot of explicit test prep) doesn’t really do a very efficient or effective job of training the kinds of skills that would help someone do well on novel types of standardized test questions. Instead of training reading comprehension, working memory, generic problem solving heuristics, quick recall/matching, problem-solving time management, etc., they tend to focus on the particular features of specific types of problems, and teach memorized recipes or collections of factual information. Obviously this is a reasonable approach if the goals of the training are very narrow and specific. It doesn’t seem to me like the best generic training to do successful intellectual work. A lot (hundreds if not thousands of papers) has been written about this subject in the scholarly literature by professors who were frustrated that their high-achieving undergraduate {math, physics, engineering, ...} students seem to be able to memorize recipes for solving specific problems very well and ace their usual tests, but have great difficulty synthesizing their knowledge or solving non-standard problems, and then spent some time and effort trying to understand why that is.

brooklyn_ashey 9y ago

As a former LSAT tutor, I can tell you that 158 isn't humblebragging for a journalist. Most of my tast-taking students were in the 150s before tutoring. I was a test prep tutor for other tests for a few years before I added the LSAT to my repertoire. Most students had no idea how to set up charts or mental pictures for the logic problems. While I understood this aspect right away, I needed to work on speed before I helped anyone else. Later in life, when learned how to code, I realized that coders would do well at this section. If I'd known how to code back then, I'd have used algorithmic thinkg techniques to help students improve their LSAT score.

matt_wulfeck 9y ago
LSAT logic questions are in many ways similar to common interview questions: listen to the problem, identify key words, apply known solving pattern.
scotty79 9y ago
When I did Mensa exams I also felt that coders might have advantage there.
wahern 9y ago
> Most students had no idea how to set up charts or mental > pictures for the logic problems. I took Introduction to Symbolic Logic my first year in college. It was a tiny class, which was a real shame. Many years later, I remember my LSAT instructor spending the first two sessions explaining, e.g., Venn diagrams to the class. I was like, "is this guy serious?" "Who could possibly need a refresher on this stuff?" But I guess I was the odd one in terms of formal exposure. Symbolic logic should be required for every college student. I can't think of any better class for helping produce well-rounded students with solid critical thinking skills, regardless of their academic focus. The class is easy and it has almost no direct application whatsoever. But that's the point. Every college student should be able to get through it without sweating, even if it seems initially daunting. And it has no direct application precisely because it's a distillation of some of the very basic mechanisms of rational thinking--mechanisms that everybody and everything glosses over. It's the one class that puts the really simple stuff front-and-center, which after a semester of lessons makes it so much easier to recognize the logical patterns (or lack thereof) elsewhere. Practicing logic negations for example is incredibly useful, especially in the humanities, but even in the hard sciences. The only downside is producing more of those annoying people who apply formal logic in every context, oblivious to the limitations. But that's better than the current state of affairs, I think. And I guess those annoying people keep everybody on their toes.