UG MATTERS! Don't believe the hype Forum

[cavalier1138]

I only used that combination in response to A. Nony Mouse who used it in his tl;dr point a few posts above:

My tl;dr point is: UG won't keep you out where you're competitive and won't get you in where you're not. So what difference should it be making to your application? Someone with a 3.6 and a 167 still isn't getting into HYS even if they went to HYS.

I was following along with his hypothetical of a 3.6 and 167. I am well aware that other combinations exist

Isn't. Nony said that someone with a 3.6/167 isn't getting HYS. Your response was then to ask who is getting in with those numbers, which was a patently absurd way to address that point.

The answer to your question, as someone already said, is that anyone who got in with those numbers (which we don't actually know has happened) was a URM.

ETA: Also, to respond to your question about who is getting rejected from HYS with a 3.8/175, the answer is that very few people are. I would wager that pretty much everyone with those numbers gets Harvard. Yale and Stanford tend to be much pickier, and as mentioned above, those are schools where your softs can start making a difference (once you have the numbers).

[lavarman84]

Also, does anyone really believe that adcomms just ignore schools with notorious grade inflation?

Largely yes.

What TTT did you go to? But seriously, if anyone has evidence that grade inflation is ignored I want to see it.

[ivankasta]

Also, does anyone really believe that adcomms just ignore schools with notorious grade inflation?

Largely yes.

What TTT did you go to? But seriously, if anyone has evidence that grade inflation is ignored I want to see it.

Here's how I see it: grade inflation DOES count as a soft. Your class GPA rank (%) is displayed right in the middle of the CAS report. If I were an adcom and I saw two students, one with a 3.65 (rank 22%) and another one with a 3.60 (rank 16%) I think I would look at the later more positively. Of course, the one true god is USNWR, so if their median is 3.63, they will admit the 3.65. But I do think it may come into play for splitters. When both GPAs fall outside of the 25 or 75, then they are affecting the school's numbers the same way and admissions can take inflation into account.

I don't even know if I believe that, but I'd like to haha. I am biased: my GPA is 3.69 which is 9% at my school while most other schools have their 10% at ~3.8+; so take all that with a grain of salt

[miss_nomer]

Here's how I see it: grade inflation DOES count as a soft. Your class GPA rank (%) is displayed right in the middle of the CAS report. If I were an adcom and I saw two students, one with a 3.65 (rank 22%) and another one with a 3.60 (rank 16%) I think I would look at the later more positively. Of course, the one true god is USNWR, so if their median is 3.63, they will admit the 3.65. But I do think it may come into play for splitters. When both GPAs fall outside of the 25 or 75, then they are affecting the school's numbers the same way and admissions can take inflation into account.

I don't even know if I believe that, but I'd like to haha. I am biased: my GPA is 3.69 which is 9% at my school while most other schools have their 10% at ~3.8+; so take all that with a grain of salt

Counterpoint that proves nothing: barely top 1/3 gpa at inflated undergrad. Admitted @ HYS

[core_four]

how is class rank calculated for people w/ multiple institutions? can I boost my ug class rank by inflating my lsdas gpa w/ a bunch of easy CC courses?

[cavalier1138]

how is class rank calculated for people w/ multiple institutions? can I boost my ug class rank by inflating my lsdas gpa w/ a bunch of easy CC courses?

LSAC only reports your rank (it's really more of a percentile score) for your degree-granting institution. But no one actually cares about that. What they do care about is your overall LSAC GPA, and yes, taking easy CC courses will boost that, as long as they're taken before you get your degree. Once you get your bachelor's, your GPA is set in stone.

[Platopus]

.

[ivankasta]

LSAC only reports your rank (it's really more of a percentile score) for your degree-granting institution. But no one actually cares about that.

Class rank doesn't matter, schools care about your LSAC GPA, that's it. So yes, taking some easy CC course to pad the GPA will help.

Disclaimer* I'm a 0L and what you're about to read is speculation:

I think it does matter, on the margins. Based on my experience in data analysis and institutional research, I think that there's a chance that class rank is a more valid predictor of 1L grades than raw uGPA is, due simply to the fact that you remove a HUGE external variable (your undergrad's grading policy) from the measurement. Taking it a step further, I'd be confident that a single number combining both class rank and your undergrad's average admitted SAT score would be a more valid predictor than raw uGPA alone. Something like this could account for the variability in both grading scales and in competitiveness between different schools.

Reducing unrelated variability (aka standardizing) is what makes both the LSAT and the GRE significantly better predictors of 1L grades than uGPA.

Now USNWR does not use a very sophisticated ranking system for law schools, so as far as ranking is concerned, predictive construct validity doesn't come into the picture. What USNWR says matters is what matters. But on the margins, schools are often deciding between two applicants whose numbers will have the exact same effect on their medians. For these cases, the school has a vested interest in using the most valid prediction model possible.

I know most schools publish an admissions index through LSAC, which is a weighted sum of LSAT and uGPA, but I would hope that they have internal institutional researchers who have developed a more complex and more predictive model. But who knows, probably not.

But at the end of the day, I do think that at least two numbers besides uGPA and LSAT on the CAS report have significant predictive value: class rank % and average undergrad LSAT score. If that's the case, I hope they (along with non-quantifiable factors) are taken into account at the margins.

[cavalier1138]

LSAC only reports your rank (it's really more of a percentile score) for your degree-granting institution. But no one actually cares about that.

Class rank doesn't matter, schools care about your LSAC GPA, that's it. So yes, taking some easy CC course to pad the GPA will help.

Disclaimer* I'm a 0L and what you're about to read is speculation:

I think it does matter, on the margins. Based on my experience in data analysis and institutional research, I think that there's a chance that class rank is a more valid predictor of 1L grades than raw uGPA is, due simply to the fact that you remove a HUGE external variable (your undergrad's grading policy) from the measurement. Taking it a step further, I'd be confident that a single number combining both class rank and your undergrad's average admitted SAT score would be a more valid predictor than raw uGPA alone. Something like this could account for the variability in both grading scales and in competitiveness between different schools.

Reducing unrelated variability (aka standardizing) is what makes both the LSAT and the GRE significantly better predictors of 1L grades than uGPA.

Now USNWR does not use a very sophisticated ranking system for law schools, so as far as ranking is concerned, predictive construct validity doesn't come into the picture. What USNWR says matters is what matters. But on the margins, schools are often deciding between two applicants whose numbers will have the exact same effect on their medians. For these cases, the school has a vested interest in using the most valid prediction model possible.

I know most schools publish an admissions index through LSAC, which is a weighted sum of LSAT and uGPA, but I would hope that they have internal institutional researchers who have developed a more complex and more predictive model. But who knows, probably not.

But at the end of the day, I do think that at least two numbers besides uGPA and LSAT on the CAS report have significant predictive value: class rank % and average undergrad LSAT score. If that's the case, I hope they (along with non-quantifiable factors) are taken into account at the margins.

Yes, that is entirely speculative. So it's weird that you presented it as though it had some sort of factual backing.

There's no evidence to suggest that adcomms even look at your class rank, much less give it any weight in the process (even on the margins). There's also nothing to suggest that your undergraduate class rank is a good predictor for law school performance. It is simply not important. And the average undergraduate LSAT score is even less important; I'm not even sure how that would be factored in.

[ivankasta]

Yes, that is entirely speculative. So it's weird that you presented it as though it had some sort of factual backing.

There's no evidence to suggest that adcomms even look at your class rank, much less give it any weight in the process (even on the margins). There's also nothing to suggest that your undergraduate class rank is a good predictor for law school performance. It is simply not important. And the average undergraduate LSAT score is even less important; I'm not even sure how that would be factored in.

I don't see how you got that impression. I started half of my statements with "I think" just to reiterate the part about it being my own speculation.

As I said, my speculation was based on experience in developing predictive models. Any measurement's construct validity will increase as you remove the effects of unrelated variables.

This is a fact: LSAT and GRE both have Pearson correlation coefficients in regards to 1L grades of around .4 [accounting for 16% of variance] while uGPA has one of ~.26 [accounting for 6.8%] (According to LSAC/Harvard/University of Arizona law. I will cite if you'd like). Why is this the case when GRE and LSAT are such different tests? I think it's because they are both standardized (aka, many unrelated variables removed) while uGPA is still plagued with variables like a school's grading policy. Class rank MIGHT have an advantage over uGPA in that it removes the effect of a school's grading system.

Let's not kid ourselves, both of us are 0L's or 1L's and neither of us has been in the room when law school adcoms are making a decision. I'm not claiming, as you do, to know anything about what factors adcoms like to look at when making hard decisions. All I'm saying is that based on my experience and the evidence I've cited, I would think that facts about an undergraduate institution might outperform uGPA alone in a predictive model. Whether that implies adcoms actually use it, I can't say.

[Future Ex-Engineer]

TL;DR

UG doesn't matter. get all A's and 170+ if you want >50% chance at HYS

[UVA2B]

Yes, that is entirely speculative. So it's weird that you presented it as though it had some sort of factual backing.

There's no evidence to suggest that adcomms even look at your class rank, much less give it any weight in the process (even on the margins). There's also nothing to suggest that your undergraduate class rank is a good predictor for law school performance. It is simply not important. And the average undergraduate LSAT score is even less important; I'm not even sure how that would be factored in.

I don't see how you got that impression. I started half of my statements with "I think" just to reiterate the part about it being my own speculation.

As I said, my speculation was based on experience in developing predictive models. Any measurement's construct validity will increase as you remove the effects of unrelated variables.

This is a fact: LSAT and GRE both have Pearson correlation coefficients in regards to 1L grades of around .4 [accounting for 16% of variance] while uGPA has one of ~.26 [accounting for 6.8%] (According to LSAC/Harvard/University of Arizona law. I will cite if you'd like). Why is this the case when GRE and LSAT are such different tests? I think it's because they are both standardized (aka, many unrelated variables removed) while uGPA is still plagued with variables like a school's grading policy. Class rank MIGHT have an advantage over uGPA in that it removes the effect of a school's grading system.

Let's not kid ourselves, both of us are 0L's and neither of us has been in the room when law school adcoms are making a decision. I'm not claiming, as you do, to know anything about what factors adcoms like to look at when making hard decisions. All I'm saying is that based on my experience and the evidence I've cited, I would think that facts about an undergraduate institution might outperform uGPA alone in a predictive model. Whether that implies adcoms actually use it, I can't say.

I think I get what you're going for here, but what you're pointing to is an isolated factor that goes into the rest of the "other stuff a school considers outside uGPA/LSAT (or possibly GRE if trends continue)." Schools definitely factor in other things besides uGPA/LSAT in making decisions, but to weight one statistically over another is speculation. There are most likely instances where grade inflation/deflation/class rank help a candidate over the admissions hump, just like there are cases where resume, PS, URM status, ED application, donor parent, influential alumni making phone calls on your behalf, etc. help a candidate over the admissions hump.

Admissions is more holistic than we broadly give credit to around TLS, but that's because that other stuff can't be quantified so as to make it reliable to gauge how those things might factor into an admissions decision. What cav is likely trying to point out (to the extent I can put words in his non-0L mouth) is that in the majority of cases, class rank will not have an impact, even if there are a minority of cases where class rank might play some small part.

[cavalier1138]

Let's not kid ourselves, both of us are 0L's and neither of us has been in the room when law school adcoms are making a decision. I'm not claiming, as you do, to know anything about what factors adcoms like to look at when making hard decisions. All I'm saying is that based on my experience and the evidence I've cited, I would think that facts about an undergraduate institution might outperform uGPA alone in a predictive model. Whether that implies adcoms actually use it, I can't say.

1. Not a 0L.
2. Admissions advisors have directly contradicted the idea that anything from your undergraduate transcript besides your LSAC GPA has any kind of strong bearing on your application (see Spivey's linked post earlier in the thread).

It seems like you have a bit of a thing for trying to find ways to contradict established wisdom on a subject based on instinct. Just focus on getting your application in the best shape it can be and not on whether you're going to figure out the secret key to bucking the trend.

[A. Nony Mouse]

But the other issue with uGPA is comparison across schools. Yes, someone with a 3.4 may be top 1/3rd in one school and top 1/5 at another (making up numbers), but those are different schools with different student bodies and possibly different basic requirements, and the students may well be in different majors. It tells you something about how a given student performed compared to the other students in their school, but you're still stuck comparing across schools.

[Future Ex-Engineer]

The idea that class rank in regards to the entire college/university matters is hilariously laughable.

Just use a little bit of brain here - why in the world would a communication studies major with a 4.0 (aka top of the class) be looked at more favorably than a bio/chem/electrical engineer with a 3.8 (which could easily be the best engineering GPA for a given graduation class)? This is all aside from the fact that a 4.0 would look better for USNWR, there would be no 'soft' bump.

[ivankasta]

1. Not a 0L.
2. Admissions advisors have directly contradicted the idea that anything from your undergraduate transcript besides your LSAC GPA has any kind of strong bearing on your application (see Spivey's linked post earlier in the thread).

It seems like you have a bit of a thing for trying to find ways to contradict established wisdom on a subject based on instinct. Just focus on getting your application in the best shape it can be and not on whether you're going to figure out the secret key to bucking the trend.

1. Sorry about that. I don't mean to be adversarial, reading back I see it kind of came off that way.

2. Right. I agree with the conventional wisdom that the LSAT/uGPA are by far the strongest influences on admissions. What I was trying to do, long-windedly I admit, was to make a case for undergrad factors like class rank or school average lsat score as soft factors, with influence similar in scale to other widely recognized 'softs'.

I am guilty of playing devils advocate against parts of the admissions system I don't understand, like in that post you linked. I mainly do it to help myself understand these things better.

[etramak]

The idea that class rank in regards to the entire college/university matters is hilariously laughable.

Just use a little bit of brain here - why in the world would a communication studies major with a 4.0 (aka top of the class) be looked at more favorably than a bio/chem/electrical engineer with a 3.8 (which could easily be the best engineering GPA for a given graduation class)? This is all aside from the fact that a 4.0 would look better for USNWR, there would be no 'soft' bump.

The top 100 GPAs in my class got a $250 reward. It was all Comm and Psych majors and Business school kids.

[Future Ex-Engineer]

The idea that class rank in regards to the entire college/university matters is hilariously laughable.

Just use a little bit of brain here - why in the world would a communication studies major with a 4.0 (aka top of the class) be looked at more favorably than a bio/chem/electrical engineer with a 3.8 (which could easily be the best engineering GPA for a given graduation class)? This is all aside from the fact that a 4.0 would look better for USNWR, there would be no 'soft' bump.

The top 100 GPAs in my class got a $250 reward. It was all Comm and Psych majors and Business school kids.

I think we can all agree that's a shitty reward system...except for the fact that the hard science majors probably jobs for the most part

[etramak]

Bumping this thread since I looked at an academic summary report for the first time today. I didn't realize they included GPA distribution for your uni was included in the report. What percentage of students receiving 3.8+ would y'all consider inflated? My school is 15%, which seems rather high.

[cavalier1138]

Bumping this thread since I looked at an academic summary report for the first time today. I didn't realize they included GPA distribution for your uni was included in the report. What percentage of students receiving 3.8+ would y'all consider inflated? My school is 15%, which seems rather high.

I promise that no matter how many 0Ls insist otherwise, no one is really looking at this information.

[stego]

At one point in the cycle do ad comms start caring more about the new median they are constructing vs. last year's?

Like say an applicant is on the waitlist, they're above (or below) a school's LSAT median from the previous year's 509, but when the ad comm looks at who has accepted their offers so far it appears the applicant won't have any effect on the median, the 25th, or the 75th, just based on where the numbers fall. do they therefore not care as much?