Doing Better at a "Worse" School Forum

[jn7]

here's a puzzle for you: let's say you're choosing between 3 schools, with LSAT medians of 160, 160, and 170. you have a 165 lsat. but you retake and get a 175. you visit school #3 and the dean reveals a goat in her office. do you switch schools?

[OutCold]

All it takes is one prof that doesn't like your formatting or writing style to give you a B at whatever lower-tier school you chose to attend. No more top 5%, game over. Everyone who starts one of these threads misses the point of going to a T14 in the first place: so that when arbitrary grading leaves you at median, you aren't up shit creek without a paddle.

[jarofsoup]

I think if someone goes to a worse school thinking they will do better b/c they have a higher than average LSAT/GPA score would say a lot about social darwinism.

The LSAC runs metrics on all this stuff:

http://www.lsac.org/docs/default-source ... -11-02.pdf

I do not think it is very predictive. Hence the large amount of transfer students in the market that rock it as a 1L and then usually rock it as a transfer.

Honestly, I think that GPA is much more predictive. Law school exams do not resemble the LSAT at all. They are mostly essay exams and some multiple choice. So, the exam really tests the wrong skill set.

[Tiago Splitter]

Honestly, I think that GPA is much more predictive. Law school exams do not resemble the LSAT at all.

I'd argue that the LSAT replicates the law school exam experience much better than undergrad classes, but that's just my opinion. What isn't my opinion, but is actually a fact found by LSAC, is that the LSAT is more predictive of first year grades.

[Yazzzay]

I don't understand how a correlation can come into play in cases like where a person retakes the LSAT and increases a substantial amount. I took it twice, did horribly, retook my third time and increased 11 points (still not ridiculously high, but a lot better). Also people study under very different circumstances (school, work, or only LSAT). So which of my scores is more indicative of my ability to succeed in law school? I don't think I know.

I'm a 0L though so I can't really talk, but I think there has to be a number of outliers to that theory.

[arklaw13]

Honestly, I think that GPA is much more predictive. Law school exams do not resemble the LSAT at all. They are mostly essay exams and some multiple choice. So, the exam really tests the wrong skill set.

I agree that UG performance does have predictive value for how well you do in law school, but I don't think that raw ugpa is a good representation of that. A 4.0 in comparative literature at Brown is a lot easier to get than a 3.5 in electrical engineering at MIT or a 3.6 from West Point. If there were a decent way to weight upga based on institutional rigor then I think it might be a somewhat reliable indicator.

[daleearnhardt123]

I don't understand how a correlation can come into play in cases like where a person retakes the LSAT and increases a substantial amount. I took it twice, did horribly, retook my third time and increased 11 points (still not ridiculously high, but a lot better). Also people study under very different circumstances (school, work, or only LSAT). So which of my scores is more indicative of my ability to succeed in law school? I don't think I know.

I'm a 0L though so I can't really talk, but I think there has to be a number of outliers to that theory.

The correlation here is no doubt much weaker. But that doesn't matter - Law Schools are after #s to report to USNWR, they aren't nearly as concerned with correlation.

[daleearnhardt123]

Honestly, I think that GPA is much more predictive. Law school exams do not resemble the LSAT at all. They are mostly essay exams and some multiple choice. So, the exam really tests the wrong skill set.

I agree that UG performance does have predictive value for how well you do in law school, but I don't think that raw ugpa is a good representation of that. A 4.0 in comparative literature at Brown is a lot easier to get than a 3.5 in electrical engineering at MIT or a 3.6 from West Point. If there were a decent way to weight upga based on institutional rigor then I think it might be a somewhat reliable indicator.

UGPA reveals absolutely nothing other than a student's willingness to put in effort and study. And at some undergraduate institutions, it doesn't even reveal that much.

[Tiago Splitter]

I don't understand how a correlation can come into play in cases like where a person retakes the LSAT and increases a substantial amount. I took it twice, did horribly, retook my third time and increased 11 points (still not ridiculously high, but a lot better). Also people study under very different circumstances (school, work, or only LSAT). So which of my scores is more indicative of my ability to succeed in law school? I don't think I know.

I'm a 0L though so I can't really talk, but I think there has to be a number of outliers to that theory.

For any one student the numbers are almost meaningless. But over a large population a correlation will appear where people with higher LSAT's and GPA's will have better grades. Makes sense.

To your question, there are a lot of people who put in a ton of effort and score horribly, even with retakes. There are also people who do really well with minimal effort on one shot. The effects across a large group will cancel out for the most part, but your point is part of why this analysis is so useless for any one individual.

[ookoshi]

Barring massive tuition differences (like free ride versus sticker)...
If you're in at HYSCCN, you're an idiot for trading down to a T14.
If you're in at a T-14, you're an idiot for trading down to a T1.
If you're in at a T1, you're an idiot for trading down to a T2.

If you think trading one T-14 for a lower ranked one, or one T1 for a lower ranked one is going to result in a significant difference in the quality of your competition in the top 10%, you're also an idiot.

[woosah]

even if it was, the correlation vs rankings should balance out anyway. Top ten % at wustl vs median at duke vs bottom third at Harvard etc.

I know I said I was done, but I thought of something else that might be interesting. For the sake of argument, accept these two assumptions:


Assumption 1: Employers are able to perfectly calculate "talent" level, regardless of the school the candidate went to. So, in other words, if a candidate who is Top 20% at GULC is "equal to" a candidate who is Top 50% at Harvard, then employers would be able to discern that. Similarly, if a candidate who is Top 5% at WUSTL is "equal" to a candidate who is Top 20% at GULC, then employers would be able to discern that. That would mean that all of these candidates (Top 50% at H, Top 20% at GULC, Top 5% at WUSTL) would have an equal chance of employment.

Assumption 2: There is a 1.00 correlation between (your LSAT/GPA relative to a law school's median LSAT/GPA) and (your class rank in that law school). So, if the LSAT/GPA median at a school is "x" and your LSAT/GPA is greater than x, then you would be above the 50th percentile by an amount that is exactly proportional to the amount that you are above the law school's LSAT/GPA.


Implications: If these two assumptions were true, then (if your only consideration is employment percentage) it wouldn't matter AT ALL where you chose to go to law school. Your employment prospects would be the same regardless of where you chose to go to school. Your "talent" level would not change, and since employers have perfected their ability to evaluate "talent," you would end up with similar prospects. Ostensibly, that would mean that you should just choose from among the schools that have offered you a full scholarship. (That may put an end to the whole "Ruby at Chicago vs. Yale" debate)


Now, obviously these assumptions aren't true. I'm not claiming that they are. BUT, it seems to me that the closer you get to these assumptions being true, the closer you get to the implications being true.

[kartelite]

even if it was, the correlation vs rankings should balance out anyway. Top ten % at wustl vs median at duke vs bottom third at Harvard etc.

I know I said I was done, but I thought of something else that might be interesting. For the sake of argument, accept these two assumptions:


Assumption 1: Employers are able to perfectly calculate "talent" level, regardless of the school the candidate went to. So, in other words, if a candidate who is Top 20% at GULC is "equal to" a candidate who is Top 50% at Harvard, then employers would be able to discern that. Similarly, if a candidate who is Top 5% at WUSTL is "equal" to a candidate who is Top 20% at GULC, then employers would be able to discern that. That would mean that all of these candidates (Top 50% at H, Top 20% at GULC, Top 5% at WUSTL) would have an equal chance of employment.

Assumption 2: There is a 1.00 correlation between (your LSAT/GPA relative to a law school's median LSAT/GPA) and (your class rank in that law school). So, if the LSAT/GPA median at a school is "x" and your LSAT/GPA is greater than x, then you would be above the 50th percentile by an amount that is exactly proportional to the amount that you are above the law school's LSAT/GPA.


Implications: If these two assumptions were true, then (if your only consideration is employment percentage) it wouldn't matter AT ALL where you chose to go to law school. Your employment prospects would be the same regardless of where you chose to go to school. Your "talent" level would not change, and since employers have perfected their ability to evaluate "talent," you would end up with similar prospects. Ostensibly, that would mean that you should just choose from among the schools that have offered you a full scholarship. (That may put an end to the whole "Ruby at Chicago vs. Yale" debate)


Now, obviously these assumptions aren't true. I'm not claiming that they are. BUT, it seems to me that the closer you get to these assumptions being true, the closer you get to the implications being true.

Assumption 3: Employers make hiring decisions based purely on "talent" level, not prestige of the school/its network or other factors. Since clients do choose law firms based somewhat on the "pedigree" of each firm's attorneys, rather than information about their actual abilities, this could matter. Firm A could have 20 lawyers from HLS and Firm B could have 20 equally capable lawyers from GULC, and I think Firm A would generally have an easier time bringing in business.

[objecion]

Anecdotally. First thing I noticed at my T6 was that everyone was so damn smart (compared to UG). During my two SAs, with a mixed bag of SA schools (some outside the T6/14) with only minor exceptions everyone was diligent. As SAs we were aware of who was from where, but I do not believe the firm cared or treated anyone differently because of their school. At that level once you’re in, you’re in.

If what you're hoping for is a big drop off at a lower ranked school, meaning that you’re brilliant and everyone else is slow, good luck with that assumption. One of my best friends (across town at a tier1) is the smartest person I know (she had a 4.0 GPA but a sucky low 160s LSAT).

[ManoftheHour]

Go big or go home. Whether you go to Yale or Cooley, you should be the best. Ginormous fish, big pond, yo.

[cron1834]

LSAT explains something like 13% of the variance. That's not a very large number. In that context, jbagelboy is right - since job prospects diminish way faster than student quality, it's straight-up irrational to trade Yale for GULC given how small a portion of variance LSAT explains. It's mathematically proven idiocy.

[Pneumonia]

I love how people that are convinced by this argument always assume that they'll be like, the ONLY person at the local T2 that thought of it. Also

the argument that student quality diminishes at a much slower rate than employment prospects carries the day with this question 99/100 times.

[woosah]

Assumption 3: Employers make hiring decisions based purely on "talent" level, not prestige of the school/its network or other factors. Since clients do choose law firms based somewhat on the "pedigree" of each firm's attorneys, rather than information about their actual abilities, this could matter. Firm A could have 20 lawyers from HLS and Firm B could have 20 equally capable lawyers from GULC, and I think Firm A would generally have an easier time bringing in business.

Yes, you're right. I hadn't thought of this.

[star fox]

Any slight edge that you may gain by going to a school with lower medians is more than negated by the gap in employment outcomes.

[patogordo]

Any slight edge that you may gain by going to a school with lower medians is more than negated by the gap in employment outcomes.

hmm are you sure? i'm reserving judgment until i hear this from a few more people.

[yossarian]

I am under my school's 25th LSAT percentile, and I am in the top 4%. THERE IS LITTLE TO NO CORRELATION WITH LAW SCHOOLS EXAMS AND LSAT SCORES

One anecdotal story isn't statistically significant.

You should retake.

This.

[Tiago Splitter]

Any slight edge that you may gain by going to a school with lower medians is more than negated by the gap in employment outcomes.

hmm are you sure? i'm reserving judgment until i hear this from a few more people.

Maybe we should all agree to mix it up and start answering this question the other way from now on. Keep things interesting.

[rickgrimes69]

tl;dr: going to a worse school bc you think you will do better is statistically and logically retarded bc even if you rank slightly higher at a lower ranked school (big "if") employment prospects drop exponentially faster than applicant quality

/ upset that I missed this thread

[Volake]

What posters seem to be neglecting is that this "small pond" factor (you're likely to do slightly better at a school with lower student LSAT scores and UGPA), although never going to be enough to justify the lower school alone, could make a difference at the margins. In conjunction with a significantly lower net cost or other factors that make the lower-ranked school more desirable, the "small pond" factor could be decisive in close cases.

[bjsesq]

What posters seem to be neglecting is that this "small pond" factor (you're likely to do slightly better at a school with lower student LSAT scores and UGPA), although never going to be enough to justify the lower school alone, could make a difference at the margins. In conjunction with a significantly lower net cost or other factors that make the lower-ranked school more desirable, the "small pond" factor could be decisive in close cases.

It is highly unpredictable. I get a retrospective application, but before hand?

[Pneumonia]

What posters seem to be neglecting is that this "small pond" factor (you're likely to do slightly better at a school with lower student LSAT scores and UGPA), although never going to be enough to justify the lower school alone, could make a difference at the margins. In conjunction with a significantly lower net cost or other factors that make the lower-ranked school more desirable, the "small pond" factor could be decisive in close cases.

Please, describe such a case