Law School Discussion

Are some law schools "easier" than others?

->Soon

  • *****
  • 18818
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #30 on: September 16, 2007, 09:47:55 AM »
grrrrrr

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #31 on: September 16, 2007, 07:50:27 PM »
Tag for later when I have time to read all of this.

I doubt it'll be worth your time.  :D

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #32 on: September 16, 2007, 09:49:26 PM »
Oops, left this tab open for a while.

I know you were partly joking, but I did see an interesting angle in that assertion, and thought I'd note it.  The truth is, anyone who would brag about their grades in LS is clearly a feminine hygiene product, and there are also many people who do very well and yet keep it quiet.

I agree, except that there aren't "many" people -- at least not if "very well" is defined as straight As.  As I mentioned, after only one year, ten classes, there are no more than five people in my class, one percent, with all A- through A+ grades.  Everyone else has at least one B+ or below.  So while I agree that doing well, in general, is not random, I do think it's hard to figure out what, besides luck, separates these four or five from the other forty or fifty people who also have mostly As and A-.  

Well, I've never given you any reason to assume that I had both a high GPA and a high LSAT.  Moreover, I specifically qualified my statement by introducing educational obstacles, etc., into the predictive equation.  However, I do think the issues of aptitude and effort are highly relevant when others are claiming that law school grading is completely subjective or random (admittedly, two distinct concepts, but I hear a lot of people also claim the latter.) 

The truth is that academic success in law school, as elsewhere, does largely come down to aptitude and (carefully applied) effort.  Some of that aptitude can be measured by the LSAT, some cannot.  As noted, it probably comes down (mainly) to 1) how "law-school" bright you are, 2) how well you do on issue-spotters and 3) how hard/smart you study.  (Do you disagree with that interpretation?)

The joy of debate aside, I think this issue is important to future students, as it is in fact quite easy to grow discouraged and feel overwhelmed during first year.  Recognizing that exams aren't completely random, and that you can efficiently prepare for them, is an important insight.  I again advise all students to review model exams and answers where possible, read books on how to approach LS exams, etc.  There are never any guarantees, of course, but you'll also generally do better if you commit to law school studies than if you screw around. (I don't think anyone disputes that.)

As far as the objectivity of the LSAT vs. LS (or most other) exams, this also appears an interesting related point, especially given all the hate the LSAT gets on this board.  To the extent people resent any subjectivity that does exist in LS (and other essay exams), it is presumably worth nothing that this is one problem the LSAT really have.  I, for one, appreciate that fact, as I'm sure Matthies does as well.

Finally, I'm certainly not claiming that LSAT+GPA has 100% predictive power, but they do have significant predictive power, and it's difficult to find other concrete factors that predict LS success as well.  This alone makes clear that LS grading is not completely subjective or random, which is all that I was saying.  It is also, presumably, a relevant point to note when it arises, given how it relates to many other issues.  If we ignore such predictive factors in admissions, for example, we're apt to end up with even more people frustrated with their law school performance, something I've witnessed first-hand. 

However, you're certainly welcome to ignore such posts if you disagree or find them unnecessary.  :)

Either I have been misunderstanding you, or you have missed my (admittedly minor) point.  I agree that law school grades are neither completely random nor completely subjective (I find this concept difficult to pin down, but I'm using it in the sense that there would be significant variation in how someone with the same skill, knowledge, and effort would do on different exams, graded by different professors, for the same subject); indeed, though I failed once, I believe it is always possible to study for a B+, assuming a B curve -- which means that grades are definitely quite predictable up to a certain threshold where other factors (such as those I mention) play a bigger role.

My point about the LSAT/UGPA index was simply that it didn't predict whether someone would get a B+ average or an A- average, an A- or an A.  It's not that precise of a predictor.  The correlation between the index and first-year average in the last LSAC report is 0.47 +/- 0.09.  This is high enough to establish a connection between LSAT/UGPA and LGPA (especially first-year GPA), but it is not high enough to suggest that, say, within score bands, the person on the bottom end will be the A- to the person on the top end's A.  I'm sure Piggy or someone could do a better job of explaining this than I can with my one trimester of statistics back in 1990.  (I'd also like to point out that the only studies of these data are conducted by LSAC iself, hardly a neutral arbiter given its self-interest in the universal LSAT requirement in law school admissions.)

As an aside, one provocative recent study (Henderson 2004, I believe) suggests that the reason the LSAT is so predictive may not be that it is a reliable gauge of any of the skills and attributes you mention (which I do not dismiss) or, more discouragingly, a straight index of race, class, and parental education, but rather that the LSAT tests one's ability to take time-pressured exams and time is a significant variable in the in-class exams that are likely to determine first-year grades.

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #33 on: September 16, 2007, 10:45:59 PM »
As an aside, one provocative recent study (Henderson 2004, I believe) suggests that the reason the LSAT is so predictive may not be that it is a reliable gauge of any of the skills and attributes you mention (which I do not dismiss) or, more discouragingly, a straight index of race, class, and parental education, but rather that the LSAT tests one's ability to take time-pressured exams and time is a significant variable in the in-class exams that are likely to determine first-year grades.

I find it odd that anyone would find that surprising.  I always figured that was the main reason why the LSAT correlates at all with law school grades. 

Well, the surprising part is how little the LSAT correlated with take-home and paper grades.  UGPA still correlated with both, at about the same rate as it did with in-class exams.  If LSAT is really a test of reasoning ability at all, you would expect it to have at least as strong a correlation with the more organized and clear arguments students write out on longer exams, no?  I guess one response is that people with less reasoning ability are able to catch up to their more able classmates when they have more time.

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #34 on: September 16, 2007, 11:06:59 PM »
Plus you have to remember that when you're studying the correlation between GPA and LSAT, you can only compare the grades of people who go to the same school, and people who go to the same school will have similar LSAT scores.     

I don't understand the point of this.  I mean, they will have similar LSAT scores yet... the LSAT is correlated at almost 0.3 with first-year GPA.  So even within their range of LSAT scores, there is a significant correlation between LSAT and test results.  That result is negligible when you switch testing formats.

I think if they have reasonable time limits, take-home exams and papers are better assessments since the skills tested (argument, issue-spotting, research, and editing) correspond more with actual lawyering than in-class issue spotters, on which people generally spew rules [from] their outlines and spend more time on counterarguments for the purpose of exhaustion than they do on carefully reasoning out their choices and exclusions.

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #35 on: September 16, 2007, 11:27:54 PM »
On the second point, maybe.  I'm sure there are good and bad points for each format, even though I'd be curious why profs don't assign 8-hour take homes and give you a 3,000 word limit instead of assigning you a 3 hour issue spotter and giving you a 3,000 word limit.  You'd think there be some reason for their consistently doing the latter? 

You're right, and I've puzzled on this many times.  My best guesses are institutional resistance and fear of rampant collaboration/cheating.

Quote from: HR
On the first point, all I am trying to say (probably not well) is that the whole idea of measuring correlation between the LSAT and LSGPA is flawed.  A kid with a 170 may finish in the top 5% at Harvard, while the kid with the 176 finishes at the median.  A kid with a 158 may finish at the bottom of Texas Tech, while the kid with the 153 finishes in the top 5%.  In both cases the person with the lower LSAT score had a higher GPA than the person with the higher LSAT score, and I'm sure that happens all the time, but what does it tell us about correlation?  Not much since we will never know how the kid with the 170 would have fared against the kid with the 153 or the 158, or how the 176 kid would have done at Texas Tech. 

I still don't totally understand.  I agree both that it is not a very good predictor on the individual level and that schools' choice of LSAT as an admissions criterion on the basis that it predicts grades is rather dubious given that the curve effectively establishes the distribution of the grades, if not student consistency (and I see little reason that a school would want some students with all As and some students with all Cs any more than it would want the majority of students to have a mix).  Nonetheless, I think you've misinterpreted the data somewhat.  The LSAC studies consistently show that, all things being equal, the people with the 75%ile UGPA/LSAT indexes are reasonably likely to end up in the top quartile of the class, and the people with the 25%ile UGPA/LSAT indexes are reasonably likely to end up in the bottom quartile of the class.  (This is especially true when looking at white and Asian students.) Sure, a kid with a 170 at Harvard can end up at the top of the class (and this probably happens with some frequency) and a kid with a 177 can end up at the bottom.  The LSAC studies suggest that the reverse is more likely.  They also suggest that the kid with the 170 would end up near the bottom of the class at Harvard, at the median at Penn, and near the top at Fordham.     

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #36 on: September 17, 2007, 12:09:07 AM »
I don't see how they could predict this though, or how if they really believe this they could come up with the magic "0.3" or whatever.  Maybe I'm just cynical, but people can twist statistics to mean anything.  Maybe someone who gets a 176 but jacks around because he's at Harvard would actually do really well if he were made to work harder at a school with a harsher curve.  Or maybe the person with the 153 who is at the top of his class at TX Tech would wilt under the pressure at HLS?  I have no idea, and neither do statisticians.  I bet I could produce a study that proved there was no correlation just as easily as I could prove a correlation of 0.8 or something. 

Yes, I think a lot of it is how the data points are defined, frankly.  But I hardly understand statistics well enough to argue this point.

Another issue is that at some schools, the LSAT doesn't seem as predictive as it is at others.  So is that because in the upper and lower ranges LSAT is somehow more or less predictive than it is in the middle?  Or is it because some admissions committees do more supple evaluations of applications that bring out indicia of success beyond numerical credentials?  Or is it because the curricula, pedagogy, and testing are somehow different at these schools, and grades there depend less on the characteristics the LSAT assesses?  I have no idea.  I'm just babbling at this late hour.

cui bono?

  • ****
  • 3785
  • blah bliggetty blah
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #37 on: September 17, 2007, 07:03:58 AM »
The higher ranked the school the easier it is.
Why?
Law school grade curves.

If your going to base how difficult a school is I think you have to look at how hard it is to do well there.

Lower ranked schools have harsher mean/median curves. Higher ranked schools have easier, or no curve. I know Lewis and Clarke (T2) used to have a C- mean/median curve, while Georgetown has a B+ curve, and I think Harvard and Yale donít curve at all. Lower ranked schools lower grades by curving to artificially inflate those at the top of the class. I.E. its ďharderĒ to have a cumulative 3.7 GPA with a C- curve than with a B+ curve. A 70 person class on a C- curve may only allow 3-4 Aís for the entire section, while a B+ school may allow 15-2o Aís.

This where class rank comes in, because law school curves can vary immensely between schools you have class rank to show where you shake out compared to your peers. Except at many schools, like mine, ranks donít come out until the first week of the next school year, so for 1L jobs it alls based on your GPA. A 3.1 might actually be top 15-20% at your C- school, but at a school with a B+ curve thatís going to be bottom 50% or worse. People like to think that every firm knows every schools grading curve, they donít, your 3.2 at a C- school looks crappy compared to a 3.7 with no curve.

Why is there almost zero academic attrition at top schools? Because its almost impossible to fail out. Show up, write some gibbersish and you will get a C+ at a school with a B+ curve, do that at a school with a C- curve and your getting a D or worse, they have to give out Dís and Fís there to make the curve work out because the mean is so low. Our cruve is a 3.0 mean meadian, I have only ever seen 1 F in four years, and the person did not turn in the exam. law schools don't give F's, unless the curve makes you give F's.

The hardest part about Yale Law School is getting in, the easiest part is having a high GPA, the easiest part about Cooley law is getting in, the hardest part is having a high GPA.

I donít think brain power has much to do with it, some, but not enough to offset the curves. Law exams are completely, 100% subjective. Your grade in part has to do with your knowledge of the material, but also in part how well you write, when your exam got graded (first or last) and what the prof had for lunch that day.


TITCR.  :) A lot of ppl don't realize that.  The standard to move from a 1L to 2L and then from a 2L to 3L at a lower tiered school is actually much higher than at a higher ranking school. 

->Soon

  • *****
  • 18818
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #38 on: September 17, 2007, 09:35:05 AM »
uh, was the consensus that some law schools are easier than others?

Miss P

  • *****
  • 19300
    • View Profile
Re: Are some law schools "easier" than others?
« Reply #39 on: September 17, 2007, 10:12:20 AM »


I have had some discussions with professors on this subject. The general consensus of what I have been told is this:

1)Professors tend to give the type of exam they preferred to take in law school (be that open book, closed book, take home)

2)70-100 3,000 word exams would be a huge effort to grade if not for the relatively easy (to grade) issue spotting grade criteria

3)There is a general feeling among law school administers and professors that the role of law school, at least the first year, is to prepare student to take the bar, not to practice law. It is generally thought that time sensitive closed book exams better reflect the kind of preparation for the bar one will need. While the opposite type of exam better reflects the actual work you will do as a lawyer. Our school, after an absolutely miserable showing on the February bar, has now mandated all closed book in class exams. My understanding is this is for all classes, but Iím not sure as it just started. I think past the first year this will do more harm than good for most students.


I think you're missing a key point, which is that while the issue-spotters may be easier to grade, they are also more objective than grading 100 three-hour dissertations on whether or not a bunch of 1Ls think strict liability is superior to negligence.  Not only would a bunch of 1Ls probably not have much intelligent to say on the subject, they would probably just parrot the professors' views in order to get on his good side.   

Why not eight-hour take-home issue-spotters where students can organize their answers more thoughtfully, analogize and distinguish cases more carefully, refer to specific authorities, and spell check, though? I had three of these my first year and I thought they were excellent exams.  Of course, I am very slow.