U. of Illinois Law suspends Dean of Admissions Forum

[splittinghairs]

From the completely unsubstantiated rumor bin that I thought of today and will now share:

Illinois has a program called iLeap that is an experimental thing that the ABA is allowing on a five-year test basis. Essentially it allows Illinois undergrads to apply and be admitted to Illinois law school without taking the LSAT. They use ACT scores and various other hoops that other applicants wouldn't have to go through. This started with the class of 2013, I'm not sure on the number of students admitted.

What if the discrepancy in the medians is the result of Illinois admitting reverse splitters through the iLeap program and then just not reporting their LSAT scores? Wouldn't that be a plausible explanation? The program would essentially allow you to pad the stats with a bunch of 3.8's and 3.9's, while having no effect on the LSAT number at all. Those students would likely take the LSAT eventually anyway to see if they could get into a better school, but if they bomb it, then they probably fall back on their Illinois acceptance, right? Then does Illinois report those scores or not? In hypothetical defense of our admissions department, why should they? Those students were already admitted.

Just a thought I had. I doubt we ever find out the full story here, I don't get the sense that Bruce Smith is the sort of guy that runs an organization that leaks.

Exactly

Also, how does this explain the GPA being way off?

I can't. But there isn't nearly the GPA discrepancy that there is with the LSAT. 163 was a jaw-dropper.

going from 3.8 to 3.7 is a HUGE difference

[Indifferent]

and the bottom half will be saddled with huge debt and no decent job prospects whatsoever

Plenty of people at Illinois, like those at other similarly ranked schools, network their way into decent mid and small law firms that pay enough that the students can easily service their debts. They might not be making 160k off the bat, but saying they have "no decent job prospects whatsoever" is inaccurate, unless your definition of "decent job prospects" is limited to NLJ 250 employment.

[Helmholtz]

But there isn't nearly the GPA discrepancy that there is with the LSAT.

I agree, but .11 is not exactly "small."

[DerrickRose]

Engaging in this rabbit trail one last time, the "game" is not played with the expectation that all admitted, and deposited, students matriculate. To the very day of orientation there is movement on the waitlist, drop-outs, and the like. Admissions officers don't play with fire that way unless they're very bad at their job.

Again, purely engaging in an unsubstantiated hypothetical here: The iLeap kids are admitted very early in the process, and I believe that have to put down some sort of a deposit before the cycle really begins in earnest. And if those kids who got into Illinois based on their 3.9 GPAs go out and get 158's on their LSAT, I think you have a little more certainty on your yields with that student than one whose application randomly falls into the office in April.

[Indifferent]

184 to be exact.

In which 12 students is...less than 7%. Holy lord I'm bad at math.

Still not nearly as bad as the admissions office at UIUC.

too soon?

[DerrickRose]

But there isn't nearly the GPA discrepancy that there is with the LSAT.

I agree, but .11 is not exactly "small."

No, not at all.

Again, I'm not trying to troll on my school's behalf here, I'm as disgusted as the rest of us.

[DerrickRose]

184 to be exact.

In which 12 students is...less than 7%. Holy lord I'm bad at math.

Still not nearly as bad as the admissions office at UIUC.

too soon?

-> -> ->

[Helmholtz]

Engaging in this rabbit trail one last time, the "game" is not played with the expectation that all admitted, and deposited, students matriculate. To the very day of orientation there is movement on the waitlist, drop-outs, and the like. Admissions officers don't play with fire that way unless they're very bad at their job.

Again, purely engaging in an unsubstantiated hypothetical here: The iLeap kids are admitted very early in the process, and I believe that have to put down some sort of a deposit before the cycle really begins in earnest. And if those kids who got into Illinois based on their 3.9 GPAs go out and get 158's on their LSAT, I think you have a little more certainty on your yields with that student than one whose application randomly falls into the office in April.

But if they're worried about medians, it's the people near the middle who count, right?

E.g. if you have 200 students in your class, and 95 of them have a 168 or above, and 95 of them have a 163 or below, your big concern is to find at least ten more people who have a 168 or above so your median can be 168. And you don't really care how much above 168 they are (ignoring the schools' desire to boost 75th percentile numbers). It doesn't really matter if some GPA-heavy splitters got a 140 on their LSAT. You already have them locked in, are using them for your GPA, and you don't really care what their LSAT is (unless they're luckily at the 168 mark). I might not be thinking about this right. I have to admit that I don't really know how math works.

[romothesavior]

Illinois has a program called iLeap that is an experimental thing that the ABA is allowing on a five-year test basis. Essentially it allows Illinois undergrads to apply and be admitted to Illinois law school without taking the LSAT. They use ACT scores and various other hoops that other applicants wouldn't have to go through. This started with the class of 2013, I'm not sure on the number of students admitted.

What if the discrepancy in the medians is the result of Illinois admitting reverse splitters through the iLeap program and then just not reporting their LSAT scores? Wouldn't that be a plausible explanation? The program would essentially allow you to pad the stats with a bunch of 3.8's and 3.9's, while having no effect on the LSAT number at all. Those students would likely take the LSAT eventually anyway to see if they could get into a better school, but if they bomb it, then they probably fall back on their Illinois acceptance, right? Then does Illinois report those scores or not? In hypothetical defense of our admissions department, why should they? Those students were already admitted.

Huh...that might actually make a hell of a lot of sense.

Yes it does. Until you read this post:

See, here's what I don't get: a "mistake" in both fields is really two separate mistakes, and one of those mistakes (the five point LSAT jump) is too large to make and not to notice. That narrative only makes sense if we presume the Admissions Office is profoundly inept. But the fixing of both numbers is a similarly difficult narrative to swallow, because there's a 0.000000000001% chance that you can inflate your median LSAT by five points and no one ever notices or tells. Even if it is a year or two later, you will get caught, and you will suffer far more disgrace than any bad recruiting year. So that narrative only makes sense if we presume the Admissions Office is so mindbogglingly stupid as to defy description.

If it was just an LSAT mistake, then I could see DR's account being true. But not both numbers. There is no way this wasn't straight up lying.

Also, I could be wrong, but if you do the iLEAP, aren't you precluded from taking the LSAT? I thought I heard you were.

[Tiago Splitter]

E.g. if you have 200 students in your class, and 95 of them have a 168 or above, and 95 of them have a 163 or below, your big concern is to find at least ten more people who have a 168 or above so your median can be 168. And you don't really care how much above 168 they are (ignoring the schools' desire to boost 75th percentile numbers). It doesn't really matter if some GPA-heavy splitters got a 140 on their LSAT. You already have them locked in, are using them for your GPA, and you don't really care what their LSAT is (unless they're luckily at the 168 mark). I might not be thinking about this right. I have to admit that I don't really know how math works.

You are thinking about it correctly, but the key is that some people have to be above both medians. Ultimately some of the 168 people need to have 3.8+ GPAs to make the numbers work. If those 3.8, 168 folks don't take the full ride, UIUC is SOL.

[romothesavior]

It seems to me that since GPA and LSAT are the two most critical numbers adcoms work with and report, they are most likely carefully reviewed many times for accuracy.

Many, many, many times. I spoke with a former admissions officer today about this scandal who said (even before the new numbers came out) that there was almost no way this was a mistake. Adcomms more or less sleep with their LSAT spreadsheet under their pillow. It is something you monitor almost every day, especially towards the end of the semester when you are trying to decide who to admit off the WL.

[splittinghairs]

Illinois has a program called iLeap that is an experimental thing that the ABA is allowing on a five-year test basis. Essentially it allows Illinois undergrads to apply and be admitted to Illinois law school without taking the LSAT. They use ACT scores and various other hoops that other applicants wouldn't have to go through. This started with the class of 2013, I'm not sure on the number of students admitted.

What if the discrepancy in the medians is the result of Illinois admitting reverse splitters through the iLeap program and then just not reporting their LSAT scores? Wouldn't that be a plausible explanation? The program would essentially allow you to pad the stats with a bunch of 3.8's and 3.9's, while having no effect on the LSAT number at all. Those students would likely take the LSAT eventually anyway to see if they could get into a better school, but if they bomb it, then they probably fall back on their Illinois acceptance, right? Then does Illinois report those scores or not? In hypothetical defense of our admissions department, why should they? Those students were already admitted.

Huh...that might actually make a hell of a lot of sense.

Yes it does. Until you read this post:

See, here's what I don't get: a "mistake" in both fields is really two separate mistakes, and one of those mistakes (the five point LSAT jump) is too large to make and not to notice. That narrative only makes sense if we presume the Admissions Office is profoundly inept. But the fixing of both numbers is a similarly difficult narrative to swallow, because there's a 0.000000000001% chance that you can inflate your median LSAT by five points and no one ever notices or tells. Even if it is a year or two later, you will get caught, and you will suffer far more disgrace than any bad recruiting year. So that narrative only makes sense if we presume the Admissions Office is so mindbogglingly stupid as to defy description.

If it was just an LSAT mistake, then I could see DR's account being true. But not both numbers. There is no way this wasn't straight up lying.

Also, I could be wrong, but if you do the iLEAP, aren't you precluded from taking the LSAT? I thought I heard you were.

yes but if they wanted to see if they could get into better schools, theyd wanna at least try the lsat

[DerrickRose]

But if they're worried about medians, it's the people near the middle who count, right?

E.g. if you have 200 students in your class, and 95 of them have a 168 or above, and 95 of them have a 163 or below, your big concern is to find at least ten more people who have a 168 or above so your median can be 168. And you don't really care how much above 168 they are (ignoring the schools' desire to boost 75th percentile numbers). It doesn't really matter if some GPA-heavy splitters got a 140 on their LSAT. You already have them locked in, are using them for your GPA, and you don't really care what their LSAT is (unless they're luckily at the 168 mark). I might not be thinking about this right. I have to admit that I don't really know how math works.

I think it would go something like this:

Class of 200 students, for ease of calculation's sake. You have 10 iLeap kids who you admit beforehand, all of whom are over your 75% GPA, and whom you assume you won't have to report the LSAT scores of, if they even take the LSAT at all.

You then get 96 students with an LSAT of 168 or more, then the rest you don't care about their LSAT's, so you take all reverse splitters you are taking for their GPA. Obviously with uncertain yields it's not that clean and easy, but we know from LSN that they were rejecting 167's left and right.

Assuming you don't have to report the iLeap kids, your median of reported LSAT scores is 168. But then all of the sudden, someone tells you that you have to report all of the 158's those extra students got, and you slide the median down 10 students, which, because you heavily leveraged yourself to get to 168, all of the sudden crashes the median down to 163.

[romothesavior]

yes but if they wanted to see if they could get into better schools, theyd wanna at least try the lsat

True, but if they took it and bombed it, Illinois would have every right to rescind their acceptance. I like where DR's head is at, but at the end of the day I don't think it is what happened.

[DerrickRose]

You are thinking about it correctly, but the key is that some people have to be above both medians. Ultimately some of the 168 people need to have 3.8+ GPAs to make the numbers work. If those 3.8, 168 folks don't take the full ride, UIUC is SOL.

I wish I understood that better in undergrad, I would have studied harder.

*realizes that if he had studied harder, he probably would have understood that better in undergrad*

[splittinghairs]

But if they're worried about medians, it's the people near the middle who count, right?

E.g. if you have 200 students in your class, and 95 of them have a 168 or above, and 95 of them have a 163 or below, your big concern is to find at least ten more people who have a 168 or above so your median can be 168. And you don't really care how much above 168 they are (ignoring the schools' desire to boost 75th percentile numbers). It doesn't really matter if some GPA-heavy splitters got a 140 on their LSAT. You already have them locked in, are using them for your GPA, and you don't really care what their LSAT is (unless they're luckily at the 168 mark). I might not be thinking about this right. I have to admit that I don't really know how math works.

I think it would go something like this:

Class of 200 students, for ease of calculation's sake. You have 10 iLeap kids who you admit beforehand, all of whom are over your 75% GPA, and whom you assume you won't have to report the LSAT scores of, if they even take the LSAT at all.

You then get 96 students with an LSAT of 168 or more, then the rest you don't care about their LSAT's, so you take all reverse splitters you are taking for their GPA. Obviously with uncertain yields it's not that clean and easy, but we know from LSN that they were rejecting 167's left and right.

Assuming you don't have to report the iLeap kids, your median of reported LSAT scores is 168. But then all of the sudden, someone tells you that you have to report all of the 158's those extra students got, and you slide the median down 10 students, which, because you heavily leveraged yourself to get to 168, all of the sudden crashes the median down to 163.

So after what like 3 or 4 years of doing this ileap program they just suddenly find out this year that they have to report their lsat?

[vanwinkle]

Illinois has a program called iLeap that is an experimental thing that the ABA is allowing on a five-year test basis. Essentially it allows Illinois undergrads to apply and be admitted to Illinois law school without taking the LSAT. They use ACT scores and various other hoops that other applicants wouldn't have to go through. This started with the class of 2013, I'm not sure on the number of students admitted.

What if the discrepancy in the medians is the result of Illinois admitting reverse splitters through the iLeap program and then just not reporting their LSAT scores? Wouldn't that be a plausible explanation? The program would essentially allow you to pad the stats with a bunch of 3.8's and 3.9's, while having no effect on the LSAT number at all. Those students would likely take the LSAT eventually anyway to see if they could get into a better school, but if they bomb it, then they probably fall back on their Illinois acceptance, right? Then does Illinois report those scores or not? In hypothetical defense of our admissions department, why should they? Those students were already admitted.

Huh...that might actually make a hell of a lot of sense.

Yes it does. Until you read this post:

See, here's what I don't get: a "mistake" in both fields is really two separate mistakes, and one of those mistakes (the five point LSAT jump) is too large to make and not to notice. That narrative only makes sense if we presume the Admissions Office is profoundly inept. But the fixing of both numbers is a similarly difficult narrative to swallow, because there's a 0.000000000001% chance that you can inflate your median LSAT by five points and no one ever notices or tells. Even if it is a year or two later, you will get caught, and you will suffer far more disgrace than any bad recruiting year. So that narrative only makes sense if we presume the Admissions Office is so mindbogglingly stupid as to defy description.

If it was just an LSAT mistake, then I could see DR's account being true. But not both numbers. There is no way this wasn't straight up lying.

Also, I could be wrong, but if you do the iLEAP, aren't you precluded from taking the LSAT? I thought I heard you were.

It's possible they excluded both numbers from iLEAP people. That would be one single decision that resulted in different numbers in both categories.

[2014]

From the completely unsubstantiated rumor bin that I thought of today and will now share:

Illinois has a program called iLeap that is an experimental thing that the ABA is allowing on a five-year test basis. Essentially it allows Illinois undergrads to apply and be admitted to Illinois law school without taking the LSAT. They use ACT scores and various other hoops that other applicants wouldn't have to go through. This started with the class of 2013, I'm not sure on the number of students admitted.

What if the discrepancy in the medians is the result of Illinois admitting reverse splitters through the iLeap program and then just not reporting their LSAT scores? Wouldn't that be a plausible explanation? The program would essentially allow you to pad the stats with a bunch of 3.8's and 3.9's, while having no effect on the LSAT number at all. Those students would likely take the LSAT eventually anyway to see if they could get into a better school, but if they bomb it, then they probably fall back on their Illinois acceptance, right? Then does Illinois report those scores or not? In hypothetical defense of our admissions department, why should they? Those students were already admitted.

Just a thought I had. I doubt we ever find out the full story here, I don't get the sense that Bruce Smith is the sort of guy that runs an organization that leaks.

Also, how does this explain the GPA being way off?

Perhaps they were including the iLeapers in their data which was boosting their GPA numbers (because the 12 them were presumably 3.8+) and their lack of LSAT was simply not included in the calculation so the GPA median was based on say 200 GPAs and the LSAT based on 188 LSATs.

By removing the 12 high GPAs from the calculation, the median would fall.

DF's explanation suffices for the LSAT drop.

[Helmholtz]

But if they're worried about medians, it's the people near the middle who count, right?

E.g. if you have 200 students in your class, and 95 of them have a 168 or above, and 95 of them have a 163 or below, your big concern is to find at least ten more people who have a 168 or above so your median can be 168. And you don't really care how much above 168 they are (ignoring the schools' desire to boost 75th percentile numbers). It doesn't really matter if some GPA-heavy splitters got a 140 on their LSAT. You already have them locked in, are using them for your GPA, and you don't really care what their LSAT is (unless they're luckily at the 168 mark). I might not be thinking about this right. I have to admit that I don't really know how math works.

I think it would go something like this:

Class of 200 students, for ease of calculation's sake. You have 10 iLeap kids who you admit beforehand, all of whom are over your 75% GPA, and whom you assume you won't have to report the LSAT scores of, if they even take the LSAT at all.

You then get 96 students with an LSAT of 168 or more, then the rest you don't care about their LSAT's, so you take all reverse splitters you are taking for their GPA. Obviously with uncertain yields it's not that clean and easy, but we know from LSN that they were rejecting 167's left and right.

Assuming you don't have to report the iLeap kids, your median of reported LSAT scores is 168. But then all of the sudden, someone tells you that you have to report all of the 158's those extra students got, and you slide the median down 10 students, which, because you heavily leveraged yourself to get to 168, all of the sudden crashes the median down to 163.

I'm skeptical that a school would be like, "oh shit, you mean we have to report every student's LSAT; I HAD NO IDEA, THIS IS GOING TO DESTROY OUR MEDIAN, I PROBABLY SHOULD HAVE CHECKED BEFORE BETTING EVERYTHING ON THIS ONE FACT."

[splittinghairs]

It's possible they excluded both numbers from iLEAP people. That would be one single decision that resulted in different numbers in both categories.

no if that were the case the median GPA should have gone up, because those ileap kids have like GPA of 3.9+

[DerrickRose]

If it was just an LSAT mistake, then I could see DR's account being true. But not both numbers. There is no way this wasn't straight up lying.

Also, I could be wrong, but if you do the iLEAP, aren't you precluded from taking the LSAT? I thought I heard you were.

You can still take the LSAT if you're in iLeap. But I do think there is a deposit you have to pay before that year's admission process starts. But still, you'd be a fool not to give it a shot.

With the GPA thing though, aren't there about a million ways you could fudge that?

[CanadianWolf]

Two pages earlier another poster wrote that law school applications were almost universally down this cycle & that this might partially explain Illinois numbers drop. My understanding is that applications to Illinois law were up this past cycle due to no application fee & generous scholarship offers in addition to a top 25 ranking.

[Helmholtz]

Perhaps they were including the iLeapers in their data which was boosting their GPA numbers (because the 12 them were presumably 3.8+) and their lack of LSAT was simply not included in the calculation so the GPA median was based on say 200 GPAs and the LSAT based on 188 LSATs.

By removing the 12 high GPAs from the calculation, the median would fall.

So you're saying that the school possibly calculated in the iLeap students' GPAs, but not their LSATs? And when the third-party investigators unearthed the real numbers, they didn't calculate the iLeap students' GPAs into the calculation, but did so for their LSATs?

confused

[romothesavior]

What is really surprising is that they shrunk the class by so much and still got destroyed on their LSAT median. You'd think they'd be able to easily maintain it by throwing more money at the top candidates, which is what we did at WUSTL this year. It would have taken a miserable year of admissions to manage a 5 point drop.

I'm not the best at math, but I did understand DF's example of how this 5-point drop could happen. But not everyone at UIUC is either at a 163 or a 168. UIUC probably takes 10+ students in every class with a 164, 165, 166, or 167. If a few of your 167/168s matriculated elsewhere, that would only bump you down a point or two. But a 5 point drop? That's outrageous. I wouldn't be surprised if UIUC has been fudging for years.

[DerrickRose]

I'm skeptical that a school would be like, "oh shit, you mean we have to report every student's LSAT; I HAD NO IDEA, THIS IS GOING TO DESTROY OUR MEDIAN, I PROBABLY SHOULD HAVE CHECKED BEFORE BETTING EVERYTHING ON THIS ONE FACT."

Or more like, "Wow, this iLeap thing is a real gray area, I've got an idea..."

Because really, you admit a student, they pay their deposit, and then later on you find out they took the LSAT and got a 156? What's an adcomm to do? That has no real analog to a normal admissions process.