Law School Discussion

Nine Years of Discussion
;

Author Topic: Is LSAT really good at predicting law school success?  (Read 8709 times)

burghblast

  • Sr. Citizen
  • ****
  • Posts: 2178
    • Yahoo Instant Messenger - tforsean
    • View Profile
Re: Is LSAT really good at predicting law school success?
« Reply #50 on: March 14, 2006, 10:27:54 PM »
No it's not.  My average LSAT score is 172 but I have all A+'s in every class.  So the margin of error on the LSAT must be -8 points. 

cyberrev

  • Guest
Re: Is LSAT really good at predicting law school success?
« Reply #51 on: March 14, 2006, 10:31:56 PM »
Your wife, father, brother and dog could all be dying of cancer


well i'm done talking with clueless children.  

it's like trying to teach a pig to sing.  you waste your time and annoy the pig.




Aerst2

  • Full Member
  • ***
  • Posts: 33
    • View Profile
Re: Is LSAT really good at predicting law school success?
« Reply #52 on: March 14, 2006, 11:49:21 PM »
Quote
An aced final exam is an aced final exam... as long as the student doesn't cheat, no one cares how the final exam was aced.  Likewise, a 165 is a 165... and will beat the bejesus out of a 155 whining about how "natural" he or she is.
Here again, how can this be interpreted other than you are saying "[the LSAT] is a qualified measurement regardless of prep?"

Like I said above, I'm confused you either saying the LSAT is predictive only for those who prep, or its predictive for everyone regardless of prep but no one has any 'natural ability', or LSAT is more predictive for those who prep. These seem to be mutually exclusive options

I say, like before, doesn't that mean the test is just arbitrary?

I know you are not responding to me, but it seems like he is saying that whether or not you prep for the LSAT is actually part of the test. The idea is this: the test is a rough, rough test of aptitude. Someone good at the test without studying has a somewhat better chance of being a good lawyer. The test is also easy to prep for - meaning that even someone without an inherent or learned ability has an opportunity to show how they can make up for their lack by studying. In that regard, it is fair to both groups. People who are already apt (for whatever reason) can get by without studying. People who are not, can study. People who don't bother to take practice tests, who don't bother to study, or who are unable to prep for a test even with tons of available material are weeded out. It seems to me that those are likely to be the worst law students.

Also - studies can show whether or not the LSAT is actually predictive for law school pretty easily. It's a simple statistical test (I don't know the answer.) You should keep in mind, though, that studies can't show the true importance of the LSAT: whether or not you will make a good lawyer, not law student. As far as a student being a good lawyer, it is my opinion that whether or not the student chose to actually study for the LSAT is has some validity as an indicator.
165 / 3.94

Deus Ex Machina

  • Sr. Citizen
  • ****
  • Posts: 615
  • Magnus frater spectat te
    • View Profile
    • Take the Roger Water's Cynicism test!
Re: Is LSAT really good at predicting law school success?
« Reply #53 on: March 15, 2006, 01:29:58 AM »
Aye, see my former post on this topic
Let's just say that if complete and utter chaos were lightning, then he'd be the sort to stand on a hilltop in a thunderstorm wearing wet copper armor and shouting 'All gods are bastards.'-T.Pratchett

redemption

  • Guest
Re: Is LSAT really good at predicting law school success?
« Reply #54 on: March 15, 2006, 07:45:14 AM »
Damn. I read this whole thread just to say this...and you beat me to it.

Your wife, father, brother and dog could all be dying of cancer just after your house burned down and Bush got re-elected, and it is still not a valid excuse.

If you can't do that, you deserve a low score.


Wow. This is a really lame post. You ought to be ashamed of yourself.

plaintext

  • Guest
Re: Is LSAT really good at predicting law school success?
« Reply #55 on: March 15, 2006, 08:55:32 AM »
I'm a low LSAT scorer here which explains my inability to understand this thread.  Someone please help me.

Premise:
1) LSAT is a good PREDICTOR of 1L grades, based on a .4 correlation
2) Typical LSAT range of a school is roughly 5-7 points
3) Standard error of the LSAT for an individual is 7 points w/ a 68% confidence, and 12 points at a 95% confidence

So, how does a measuring device predict grades to such a high correlation, where the inherent precision is much courser than the range in which it measures?  An exaggerated analogy is using a telescope to measure the distance across subatomic particles.  (if anyone cares, im referencing Nyquist)

Also, I wonder why the logistics of administering an objective test is never really discussed.  When devising an test that can be administered efficiently for thousands of people, certain sacrifices must be made.  It's my belief that the LSAT may in fact overtest, beyond what is needed, certain skills (strengthen, weaken), and not test enough the more relevant skills (parallel reasoning, principle questions).  There's also a certain amount of gaming the question types such that statistically the raw scores will correspond to certain scaled scores... this means adding some of the easier less relevant questions (strengthen, weaken), and reducing other question types.

Additionally, LSAC makes it clear the LSAT tests a SUBSET of the those skills.  It's a question of how much of the total set this subset comprises.  There are certain skills wholly ommitted, such as pattern matching and inductive reasoning, simply because it's difficult to test in an objective, timely manner (although they do appear in IQ tests), but more importantly to the level of discrimination the LSAT is used in the admissions process (ie: LSAC needs approximately 100 questions).

It's more easy to undersand from this   OneLSkills =  0.4LSAT + 0.6Other_NotTested

So it's certainly possible for someone to have deficient LSAT skills, superior not tested skills, and overall be more fit for law school studies than the counterpart with a higher LSAT.  Given the admissions process, this person would be culled from the process.. a false-negative so to speak.  There are also those high lsat scorers with (unreasonably) low GPAs such as a 2.7 who are given chances... false postives.  Again though, considering the logistics of test administration, there will always be a group like this (although it will vary, based on the skillset), so someone is always SOL.  In all likelihood, Spearman's G is the more probably explanation and this argument may not hold.

The correlation argument is the most amusing.  One assumes because a high correlation exists, then it must test the same skills addressed.  Geezus, if there was a higher correlation with hand and strokes one used to wipe their ass, and this was disguised in fancy math and numbers, I'd expect some people to accept and defend it vigilantly.  Someone else mentioned a higher correlation with socioeconomic status.. this is reality, but again it can't be tested objectively, nor defended ethically.

Anyway, we're all stuck with the LSAT for the conceivable future.  The legal profession prestige-whore driven, at times beyond the limit of reasonableness, so just accept it as a critical procedural hoop that you must do, and leave it at that :)


(Oh, and to satisfy the pro crowd, I'm only writing this because I'm a disgruntled, waitlisted, rejected, low lsat, TTT scorer =)

redemption

  • Guest
Re: Is LSAT really good at predicting law school success?
« Reply #56 on: March 15, 2006, 09:04:24 AM »
You put some effort into this post. I'm sure you'll get lengthy responses from other people.

For my part, I think that this entire conversation is a red herring. Whether or not the LSAT predicts grade performance in Law School is irrelevant to its utility as an admissions tool.

likewise

  • Sr. Citizen
  • ****
  • Posts: 1055
    • View Profile
Re: Is LSAT really good at predicting law school success?
« Reply #57 on: March 15, 2006, 09:15:55 AM »


the lsat doesnt measure work ethic, the ability to overcome adversity, or desire.  it doesnt measure integrity, experience, motivation, or a host of other non-tangible attributes that allow someone to surpass the one-size-fits-all expectations that have been decreed because of their results on a standardized test. 


This, I like a lot. 

Folks on this thread have suggested deferring application for a year.  That's just not an option for me.  This is my time.  I did the best I could on that given Saturday.  Is my performance on the test indicative of my ability to perform in law school?  Probably not.  I DID NOT prepare much at all. 

That's why I centered my application around overcoming adversity, experience, and work ethic.  And it's worked, to a degree.  I got into Temple, when almost all others with my LSAT / GPA / Index were rejected.  Others have similar admissions success stories.

This test IS a relatively good predictor of grades early in LS.  It IS, unfortunately, the best tool LSs have to compare candidates with diverse backgrounds. Consequently, if you blew the LSAT (relatively, of course), just accept what you got, write an addendum re: your score (if you choose), accentuate the positive aspects of your character on your apps, and roll the dice.

plaintext

  • Guest
Re: Is LSAT really good at predicting law school success?
« Reply #58 on: March 15, 2006, 09:36:45 AM »
You put some effort into this post. I'm sure you'll get lengthy responses from other people.

For my part, I think that this entire conversation is a red herring. Whether or not the LSAT predicts grade performance in Law School is irrelevant to its utility as an admissions tool.

i agree.. the question of administrative vs. individual fairness.  also to be honest, there's a subtle equivocation in my prior post.. 

"V"

  • Sr. Citizen
  • ****
  • Posts: 617
  • Vi Veri Vniversum Vivus Vici
    • View Profile
Re: Is LSAT really good at predicting law school success?
« Reply #59 on: March 19, 2006, 10:13:05 PM »
You put some effort into this post. I'm sure you'll get lengthy responses from other people.

For my part, I think that this entire conversation is a red herring. Whether or not the LSAT predicts grade performance in Law School is irrelevant to its utility as an admissions tool.

I knew you weren't REALLY leaving. You can't leave, you're too sucked in now - you're a PART of our spinning wheel machine now, HA HA HA HA!