http://www.wsj.com/article_email/lsats-grip-on-law-school-admissions-loosens-1455964203-lMyQjAxMTA2NTI2MjYyMjI5WjInteresting article. Part of me does wish the LSAT was more similar to the GRE like offering the LSAT on a rolling basis and making it computer based so results can get back faster. To do that though, I think you'd have to get rid of the games section. As much as the games are "fun" and relatively "easy" points once you fool proof them, it is a useless section that should be replaced with a math section like on the GRE, but that's a totally separate discussion.
Also, many law schools are trying to find ways to combat the declining application problem. Getting rid of the LSAT barrier and accepting the GRE (which far more people take) might cause someone on the margin to decide to apply to law school (and boost revenue numbers via application fees and stuff). It's not that uncommon for people to wake up one day, decide to take the LSAT, and apply to law school (as weird as that sounds); making it easier for people to apply in general by accepting a broader test is just going to encourage those people more I think.
Lastly, LOL at this part: "The entirety of the LSAT was meant to mimic the law school experience..."
Comments
@Accounts Playable, far as your point about the games, I disagree with your notion that they're useless. Yes, there's a lot of repetition involved, and LR/RC represent much more concrete skills in terms of reading ability and how to evaluate arguments. But in my opinion, games are all about understanding convoluted systems, and being able to efficiently cut to the heart of many moving parts to see how the pieces of a system interact. I believe this soft skill that can be applied in other areas of life
Secondly, I strongly disagree with the idea that LG should be replaced with a math section. LG exists to test students' analytical reasoning skills. Who's to say that a math section would be a better measure of those skills? If anything, a math section would move us further away from the actual skills required for success in law school.
Not to mention, utilizing a math section would likely discourage applicants rather than encourage them (not sure why we're trying to encourage applicants but that's a question for another time). Have you ever looked into studies that identify gender biases that exist among teachers of young students particularly as they relate to math? Utilizing the GRE as a barrier to law school admission *could* create the type of unequal distribution of the genders that we see in the STEM fields.
Even if the end goal is attracting more students to the legal profession, I'm not sure that the GRE would achieve that goal. Yes, many more students take the GRE but it's also accepted by a vast number of graduate programs.
I'm actually interested to know how they conducted this study especially because, to my knowledge, the GRE and the LSAT are fundamentally different exams. Wouldn't it be an illogical comparison?
"After analyzing test scores from 78 current and former Arizona Law students, the company determined it had at least as much predictive value as the LSAT."
Is it just me, or did anyone else think "unrepresentative sample" after reading this? lol.
I completely disagree. In my opinion, the LSAT would be greatly improved with a math section, and it doesn't necessarily need to be exactly like the one found on the GRE. Logic is applied mathematics, and the games section would be improved if it further owned up to that. Right now, the games section is a wasted section due to its odd level of abstraction in my opinion.
Lawyers are going to have to deal with math problems in law school (i.e. tax law) and outside of law school (i.e. billing clients). In this respect, replacing the games section with a math section or making the games section more mathematical would be the most practical change the test makers can do. The LSAT correlates very poorly already with law school grades and with any meaningful measure of "success" after law school. Enhancing the test in a practical aspect would do nothing but help the test.
As for the LSAT correlating poorly with grades, where's the support for this? Genuinely curious as the only study I've ever seen published was one that seemed to support a correlation (although admittedly it was put out by LSAC).
Two examples (tax law and billing) just aren't enough to justify a change to a math section, at least in my mind. Why test something that applies to such a narrow field (math) when you could test something that applies to literally every specialization (analytical reasoning presented in abstract terms)?
1.) Article about the predictive power of the LSAT: http://blogs.wsj.com/law/2015/07/13/new-study-tries-to-predict-law-school-grades/
The LSAT does correlate slightly with first year grades (which I didn't know), but the value put on the LSAT is overblown.
2.) I'm not saying that law doesn't deal with abstraction, it does. But, the LSAT should be practical before it's abstract. Adding a math section is inherently more practical.
3.) Tax law and billing are just two examples. The overwhelming amount of forensic evidence from a forensic accountant is going to be mathematical in nature. The overwhelming amount of scientific forensic evidence is going to statistical in nature (i.e. hypothesis testing). Lawyers use math to calculate probabilities/calculating the benefits of strategy. Here is a list of quite of few cases that depend on some sense of mathematical evidence that the lawyer needed to understand: http://www.llrx.com/features/calculatingjustice.htm
4.) I think there is confusion about what type of math I want to see on the LSAT. I don't believe calculus or trig or things like that would be especially useful to test either. But, a good understanding of statistics is absolutely necessary, and it is noticeably lacking in the legal profession. This lack of mathematical understanding is a problem: https://news.illinois.edu/blog/view/6367/204846. Even Richard Posner wrote something on his excellent blog about it a few years ago (unfortunately, I can't find the link).
From the GRE website: Data analysis topics include basic descriptive statistics, such as mean, median, mode, range, standard deviation, interquartile range, quartiles and percentiles; interpretation of data in tables and graphs, such as line graphs, bar graphs, circle graphs, boxplots, scatterplots and frequency distributions; elementary probability, such as probabilities of compound events and independent events; random variables and probability distributions, including normal distributions; and counting methods, such as combinations, permutations and Venn diagrams. These topics are typically taught in high school algebra courses or introductory statistics courses. Inferential statistics is not tested.
I don't think this is perfect, but it's a great start, and I think it is far better than the game section. In fact, every introductory logic book that I've read has at least one chapter about inductive arguments and probabilities. I think it's odd that the LSAT tends to shy away from these types of problems.
I think the GRE is a far easier exam. I think the main reason law schools like U of A are using the GRE is to entice more applicants to apply there, which would presumably boost application fee revenue. You might have a person take the GRE so they can get into the MBA program but if that school also allows you to apply to the JD program as well with the GRE, you might just apply and see what happens. I don't that that's a good idea to do, but law schools are really hurting right now revenue wise.
@Alejandro
No, I like the LR and RC for the most part. Ideally, I'd scrap the writing section and just add the math section there. My recommendation for eliminating the games and replacing it with a math section was under the framework of the LSAT being a computerized exam (which is what I would like to see happen along with offering it on a rolling basis, like the CPA exam). I really think it would be too hard to keep the game section if the LSAT were computerized, although it wouldn't be impossible. In my mind, though, the game section is so heavily dependent on scratch paper, it would be difficult to keep it unless the LSAC overcomes its aversion to scratch paper.
If I had to choose a section to ax in order to incorporate the math section, it would be the games section; I just think there is more value in a math section than a games section. In my opinion (and Professor Orin Kerr's), the games section's value is debatable (I do concede that the commenters on the article don't think it is, but unfortunately, not very many of them talk about the alternative of offering a math section, which would be worth it I think) http://volokh.com/2007/12/27/should-the-lsat-have-a-logic-games-section/
The usefulness of statistical reasoning in the legal world (and for law school) is undeniable though, and for that reason alone, I think it should trump the game section. It's just more practical.
Lastly, I found the Richard Posner article that I had read a few years ago; it wasn't on his excellent law blog (Becker-Posner blog), but it was in the University of Miami Law Review: http://lawreview.law.miami.edu/225-growing-distaste-math-science-legal-profession-consequences-flow-therefrom/
I will pay upwards of $250k to learn the law, learn how to think like a lawyer, learn from my professors and fellow students, challenge my own notions and opinions, improve as a person, fulfill my dream, etc. I wont pay to compare my LSAT score with other people and will certainly pity those who do, if any.
Still, the comparison is interesting because the GMAT's Critical Reasoning questions are very similar to the LSAT's LR questions, albeit easier.
As far as some of the harder LR and RC questions, I really don't see the point in them and how they really show a candidates aptitude to succeed. To me, a 165 applicant is just as qualified as a 180. Sure, being able to dissect super complex/dense text is a tough skill to acquire and is very useful in the legal profession. However, at a certain point, when the questions get ridiculously super hard, this will not help someone in their legal career or in law school. I have heard this same remark from several judges/lawyers.
I think they need a verbal section on LSAT where the student is actually speaking. Imagine a section where a student had to articulate, in their own words, a complicated idea or theory to a panel of reviewers. In my opinion, that type of test would be a way better predictor of law school success, given the nature of the Socratic method.
While I would obviously love that, having engaged in rhetorical/advocacy activities for essentially the duration of my teenage and adult years, I think it'd ultimately test something that doesn't have much of a purpose. Pretty much all legal work is done in written motions. Most lawyers never see the inside of a courtroom, and even fewer do so in a context that requires extended articulation of a legal concept (aka not just being present for a hearing to schedule another motion deadline). Trial lawyers are a very specific breed, with a set of very specialized skills on top of their baseline legal acumen. Those specialized skills shouldn't be the focus of a generalized test like the LSAT.
If they REALLY wanted to introduce a verbal component, I think the best way to do so would be to do something like they did on my Spanish AP test way back in the day, where I had to tape-record myself speaking the language in response to a prompt and was graded on my fluency. It would alleviate a lot of logistical problems and simultaneously avoid the "actually can't do it or just nervous facing a firing squad?" issue. Another idea along these lines would be to score the essay section. The problem with all of these ideas is that it brings a completely arbitrary element to scoring unless there is a VERY strictly defined rubric, in which case people 'in the know' would know exactly how to game the test and the exact same complaints about "L$AT $core" would persist, worse than ever.