U. of Illinois Law suspends Dean of Admissions Forum

[Cornelius]

I'm sure I'm missing something, but I can't for the life of me figure out the hesitancy for auditing medians. If the ABA gave me a thousand bucks and a 24 pack of mountain dew I could probably do all 200+ schools for the last three years in a week.

A week? Import to excel -> use median formula. It could be done in a day.

[masochist]

I'm sure I'm missing something, but I can't for the life of me figure out the hesitancy for auditing medians. If the ABA gave me a thousand bucks and a 24 pack of mountain dew I could probably do all 200+ schools for the last three years in a week.

Not sure if srs.

it's not that hard...if the LSAC has the average of every undergrad institution, surely they can quickly figure out the medians for each law school. Hell, why isn't the LSAC in charge of putting out the LSAT/GPA medians already? Seems like that'd end all the controversies immediately.

Does LSAC ever find out what schools we attend? I know they have the info on application, but I don't think they get it any info after that point, right? I am not sure they would even be allowed access because of FERPA. ABA can get around FERPA because it is an accrediting body, but LSAC is just a data distribution service.


I think the ABA or the regional accrediting bodies for the Universities as a whole would be the only ones allowed access to individual-level data for an audit. Maybe HLC could be convinced to step in if the ABA remains asleep at the switch.

[Patriot1208]

I'm sure I'm missing something, but I can't for the life of me figure out the hesitancy for auditing medians. If the ABA gave me a thousand bucks and a 24 pack of mountain dew I could probably do all 200+ schools for the last three years in a week.

A week? Import to excel -> use median formula. It could be done in a day.

I don't think you know what "auditing" means.

BUT HE'S A STATISTICS GENIUS!!!

[Cornelius]

I'm sure I'm missing something, but I can't for the life of me figure out the hesitancy for auditing medians. If the ABA gave me a thousand bucks and a 24 pack of mountain dew I could probably do all 200+ schools for the last three years in a week.

A week? Import to excel -> use median formula. It could be done in a day.

I don't think you know what "auditing" means.

If you had the official scores from LSAC and cross referenced that with the student body, it would be pretty easy to identify a difference in medians.

I think you're overestimating the difficulty of confirming whether or not reported medians are the true medians. This isn't a financial entity with a multitude of accounts that need to be verified, with individual transactions confirmed.

BUT HE'S A STATISTICS GENIUS!!!

Yay taking a sarcastic remark out of context. Statistics and auditing are not the same thing. Then again, I did work in the public accounting field.

[chrisbru]

I'm sure I'm missing something, but I can't for the life of me figure out the hesitancy for auditing medians. If the ABA gave me a thousand bucks and a 24 pack of mountain dew I could probably do all 200+ schools for the last three years in a week.

A week? Import to excel -> use median formula. It could be done in a day.

I don't think you know what "auditing" means.

BUT HE'S A STATISTICS GENIUS!!!

[birdlaw117]

Statistics and auditing are not the same thing. Then again, I did work in the public accounting field.

How often did you complete an audit of hundreds of locations in on day? You must be an Auditing Genius!!

[Cornelius]

Statistics and auditing are not the same thing. Then again, I did work in the public accounting field.

How often did you complete an audit of hundreds of locations in on day? You must be an Auditing Genius!!

Considering what you're talking about involves matters more complicated by a number of orders of magnitude than about 200 lists with a couple hundred numbers on them, for which each actual median can be found in the amount of time it takes you to type =median(A:A). To use the word "audit" is a misnomer, verify would be more accurate.

[birdlaw117]

Statistics and auditing are not the same thing. Then again, I did work in the public accounting field.

How often did you complete an audit of hundreds of locations in on day? You must be an Auditing Genius!!

Considering what you're talking about involves matters more complicated by a number of orders of magnitude than about 200 lists with a couple hundred numbers on them, for which each actual median can be found in the amount of time it takes you to type =median(A:A). To use the word "audit" is a misnomer, verify would be more accurate.

Pretty sure you would be curious that the numbers already in the excel sheet are accurate. Just a thought. Otherwise I'd probably just fill out an excel sheet with 250 "180s" and "4.0s", hand it off to you, and you'll put it in your little formula and say I'm the God of Admissions.

[Cornelius]

Statistics and auditing are not the same thing. Then again, I did work in the public accounting field.

How often did you complete an audit of hundreds of locations in on day? You must be an Auditing Genius!!

Considering what you're talking about involves matters more complicated by a number of orders of magnitude than about 200 lists with a couple hundred numbers on them, for which each actual median can be found in the amount of time it takes you to type =median(A:A). To use the word "audit" is a misnomer, verify would be more accurate.

Pretty sure you would be curious that the numbers already in the excel sheet are accurate. Just a thought. Otherwise I'd probably just fill out an excel sheet with 250 "180s" and "4.0s", hand it off to you, and you'll put it in your little formula and say I'm the God of Admissions.

If you had the official scores from LSAC and cross referenced that with the student body, it would be pretty easy to identify a difference in medians.

Reading is fun.

[birdlaw117]

If you had the official scores from LSAC and cross referenced that with the student body, it would be pretty easy to identify a difference in medians.

Reading is fun.

But you don't. Reality sucks.

[Cornelius]

If you had the official scores from LSAC and cross referenced that with the student body, it would be pretty easy to identify a difference in medians.

Reading is fun.

But you don't. Reality sucks.

But could we? I wouldn't be surprised if, in the event of a wide-scale audit, LSAC released the data to the ABA.

Thanks for playing, read my posts before you offer lame criticism.

[birdlaw117]

If you had the official scores from LSAC and cross referenced that with the student body, it would be pretty easy to identify a difference in medians.

Reading is fun.

But you don't. Reality sucks.

But could we? I wouldn't be surprised if, in the event of a wide-scale audit, LSAC released the data to the ABA.

Thanks for playing, read my posts before you want to offer lame criticism.

In the even of a wide-scaled audit it would take longer than an afternoon. Look it isn't a terribly complicated audit, but it's not watching the powerball numbers pop out of a machine either. You have to make sure the students each school says is matriculating actually is, make sure people getting off WLs don't skew things, cross-reference it with LSAC data, etc.

[Cornelius]

In the even of a wide-scaled audit it would take longer than an afternoon. Look it isn't a terribly complicated audit, but it's not watching the powerball numbers pop out of a machine either. You have to make sure the students each school says is matriculating actually is, make sure people getting off WLs don't skew things, cross-reference it with LSAC data, etc.

Fair enough. My original "in an afternoon" post referenced what would be, in my opinion, a preliminary check. Obviously you would have to dig deeper to be thorough and do the things you mentioned - verifying they're actually enrolled, etc. Though why any school would have the forethought of falsifying in that way when they can, currently, just say "here's our medians" and nobody checks them, I'm not sure.

[vanwinkle]

Wouldn't an actual audit require going into individual student records? If someone's committing fraud, they could just as easily commit fraud by generating an entire faulty database, fudge numbers until the median actually comes out to 168 or whatever they want, and auditing their database would mean nothing unless you could get into the actual student records and verify their reported scores were real. But even then, accessing student records at the school is meaningless because the people committing the fraud are the people who control those records, and could doctor the physical records as well. After all, if you're already committing a fraud, wouldn't you want to cover it up?

To verify the actual student records weren't doctored, we've now got to gain access to these students' LSDAS reports, or at least to their LSAC scores. While FERPA creates a legal privacy exception for schools, it doesn't appear to apply to LSAC, which is a non-profit corporation.

LSAC's website says this about your LSAC score, under the "confidentiality" section: "Scores are released only to you and to the law schools to which you have applied. They will not be released to a parent, spouse, friend, or any other person." It doesn't sound like there's an exception for providing scores directly to the ABA. LSAC might be liable for some kind of privacy tort if they released LSAC scores to the ABA without first obtaining students' permission.

That means that, in order for the ABA to run a proper audit of a law school, they'd need to get permission to access the LSAC records of many of its students. That doesn't sound very practical, especially since those students will know an audit means nothing good for their school. Either their school passes the audit, which just maintains the status quo, or their school fails the audit, which gives their school a black eye and potentially hurts their employability.

[NancyBotwin]

Statistics and auditing are not the same thing. Then again, I did work in the public accounting field.

How often did you complete an audit of hundreds of locations in on day? You must be an Auditing Genius!!

[birdlaw117]

Wouldn't an actual audit require going into individual student records? If someone's committing fraud, they could just as easily commit fraud by generating an entire faulty database, fudge numbers until the median actually comes out to 168 or whatever they want, and auditing their database would mean nothing unless you could get into the actual student records and verify their reported scores were real. But even then, accessing student records at the school is meaningless because the people committing the fraud are the people who control those records, and could doctor the physical records as well. After all, if you're already committing a fraud, wouldn't you want to cover it up?

To verify the actual student records weren't doctored, we've now got to gain access to these students' LSDAS reports, or at least to their LSAC scores. While FERPA creates a legal privacy exception for schools, it doesn't appear to apply to LSAC, which is a non-profit corporation.

LSAC's website says this about your LSAC score, under the "confidentiality" section: "Scores are released only to you and to the law schools to which you have applied. They will not be released to a parent, spouse, friend, or any other person." It doesn't sound like there's an exception for providing scores directly to the ABA. LSAC might be liable for some kind of privacy tort if they released LSAC scores to the ABA without first obtaining students' permission.

That means that, in order for the ABA to run a proper audit of a law school, they'd need to get permission to access the LSAC records of many of its students. That doesn't sound very practical, especially since those students will know an audit means nothing good for their school. Either their school passes the audit, which just maintains the status quo, or their school fails the audit, which gives their school a black eye and potentially hurts their employability.

Spot fucking on.

[Cornelius]

Wouldn't an actual audit require going into individual student records? If someone's committing fraud, they could just as easily commit fraud by generating an entire faulty database, fudge numbers until the median actually comes out to 168 or whatever they want, and auditing their database would mean nothing unless you could get into the actual student records and verify their reported scores were real. But even then, accessing student records at the school is meaningless because the people committing the fraud are the people who control those records, and could doctor the physical records as well. After all, if you're already committing a fraud, wouldn't you want to cover it up?

To verify the actual student records weren't doctored, we've now got to gain access to these students' LSDAS reports, or at least to their LSAC scores. While FERPA creates a legal privacy exception for schools, it doesn't appear to apply to LSAC, which is a non-profit corporation.

LSAC's website says this about your LSAC score, under the "confidentiality" section: "Scores are released only to you and to the law schools to which you have applied. They will not be released to a parent, spouse, friend, or any other person." It doesn't sound like there's an exception for providing scores directly to the ABA. LSAC might be liable for some kind of privacy tort if they released LSAC scores to the ABA without first obtaining students' permission.

That means that, in order for the ABA to run a proper audit of a law school, they'd need to get permission to access the LSAC records of many of its students. That doesn't sound very practical, especially since those students will know an audit means nothing good for their school. Either their school passes the audit, which just maintains the status quo, or their school fails the audit, which gives their school a black eye and potentially hurts their employability.

It seems they just got the records for Illinois, so evidently that's not a problem. Unless they just took for granted that Illinois didn't create a fictional database of students and LSAT scores. I'd be surprised if that was the case given all the lying and whatnot. Can't really expect any honesty coming out of Illinois, I'm sure the LSAT scores and GPA numbers were externally verified, and how else could they do that without the LSAC data.

[Mickey Quicknumbers]

Wouldn't an actual audit require going into individual student records? If someone's committing fraud, they could just as easily commit fraud by generating an entire faulty database, fudge numbers until the median actually comes out to 168 or whatever they want, and auditing their database would mean nothing unless you could get into the actual student records and verify their reported scores were real. But even then, accessing student records at the school is meaningless because the people committing the fraud are the people who control those records, and could doctor the physical records as well. After all, if you're already committing a fraud, wouldn't you want to cover it up?

To verify the actual student records weren't doctored, we've now got to gain access to these students' LSDAS reports, or at least to their LSAC scores. While FERPA creates a legal privacy exception for schools, it doesn't appear to apply to LSAC, which is a non-profit corporation.

LSAC's website says this about your LSAC score, under the "confidentiality" section: "Scores are released only to you and to the law schools to which you have applied. They will not be released to a parent, spouse, friend, or any other person." It doesn't sound like there's an exception for providing scores directly to the ABA. LSAC might be liable for some kind of privacy tort if they released LSAC scores to the ABA without first obtaining students' permission.

That means that, in order for the ABA to run a proper audit of a law school, they'd need to get permission to access the LSAC records of many of its students. That doesn't sound very practical, especially since those students will know an audit means nothing good for their school. Either their school passes the audit, which just maintains the status quo, or their school fails the audit, which gives their school a black eye and potentially hurts their employability.

Whatever man, I could still do it in a week.

[Patriot1208]

Wouldn't an actual audit require going into individual student records? If someone's committing fraud, they could just as easily commit fraud by generating an entire faulty database, fudge numbers until the median actually comes out to 168 or whatever they want, and auditing their database would mean nothing unless you could get into the actual student records and verify their reported scores were real. But even then, accessing student records at the school is meaningless because the people committing the fraud are the people who control those records, and could doctor the physical records as well. After all, if you're already committing a fraud, wouldn't you want to cover it up?

To verify the actual student records weren't doctored, we've now got to gain access to these students' LSDAS reports, or at least to their LSAC scores. While FERPA creates a legal privacy exception for schools, it doesn't appear to apply to LSAC, which is a non-profit corporation.

LSAC's website says this about your LSAC score, under the "confidentiality" section: "Scores are released only to you and to the law schools to which you have applied. They will not be released to a parent, spouse, friend, or any other person." It doesn't sound like there's an exception for providing scores directly to the ABA. LSAC might be liable for some kind of privacy tort if they released LSAC scores to the ABA without first obtaining students' permission.

That means that, in order for the ABA to run a proper audit of a law school, they'd need to get permission to access the LSAC records of many of its students. That doesn't sound very practical, especially since those students will know an audit means nothing good for their school. Either their school passes the audit, which just maintains the status quo, or their school fails the audit, which gives their school a black eye and potentially hurts their employability.

Whatever man, I could still do it in a week.

Mickey Quicknumbers - the unshakeable human calculator

[minnbills]

What kind of sanctions do you A) think UIUC should get and B) what do you think they actually will get?

[Aberzombie1892]

What kind of sanctions do you A) think UIUC should get and B) what do you think they actually will get?

I think that the ABA SHOULD put UIUC's accreditation on probation/provisional accreditation (whatever new law schools are put on) for a period of 5 years. If UIUC messes up again for any reason during that time period, the school should lose ABA accreditation for a predetermined number of years (maybe another 5?). Also, US News should not rank the school for at least 1 year (this makes a huge difference; ask UC-Hastings).

In reality, UIUC may get censured and what not (re: nothing).

[birdlaw117]

I swear a much simpler version (than VW's post) would be to have LSAC report LSAT/GPA using a unique-yet-anonymous key (e.g., the LSAC#), and then compare those numbers to the unique key of actually matriculating students. Run each school as a report in a relational database program, and see which schools have differing medians. You could still have a simple baseline comparison to self-reporting without leaving your desk.

Now, of those schools with any variance, you'd have to go on site and investigate. But the bulk of schools could be double-checked by a small team with laptops, and it would only take a few days (and, likely, avoid the privacy tort liability).

But schools could give the LSAC numbers of student they accept but don't matriculate. Yes, they would be able to find out if a student is double-counted. You also have deferrals and people that simply don't attend any school. I do, however, think that a school would have to be pretty bold (read: stupid) to try it.

You could run a check for double-counted students and then investigate from there though.

[whuts4lunch]

It has been going on for years. http://www.law.illinois.edu/news/article/1627

[Aqualibrium]

It has been going on for years. http://www.law.illinois.edu/news/article/1627

wow so 165, 165, 167, 163

[Kilpatrick]

It has been going on for years. http://www.law.illinois.edu/news/article/1627

You really thought two days later you would be the first to post that?