LSAC settles LSAT Disability Lawsuit Forum

[papercut]

Regardless of the policy reasoning behind extra time, or the flagged versus non-flagged scores issue, one of the outcomes of this will be a surge in LSAT's taken with disability accommodation. Since US News will no longer be able to discount accommodated LSAT's due to no special markings being given, they will be worth just as much to a law school's ranking as regular LSAT's. It's quite likely that the vast majority of accommodated testers (At least initially) will have real disabilities that negatively affect their performance. Yet it's also worth noting that people respond to incentives, and in the realm of law school applications opening an avenue to game the LSAT is perhaps the most powerful. It's why people pay thousands of dollars for prep courses, and it's why this system is rife for abuse.

Watch as contrary to the current trend towards lower medians, scores instead trend upward rapidly in the next couple of years due to an increasing awareness of this lowest-common-denominator standard of proof. Which post-graduate test has the broadest scope and weakest verification of disabilities that the LSAT will be forced to match (The GRE, GMAT, MCAT)? I'm profoundly thankful I'll be attending this cycle, and not the next. Splitters are screwed from here on out, HYS to 175+ medians.

If they do get a surge in inflated LSATs (at least until the "curve" catches up) the schools are likely to suffer with lower bar passage rates, which will hurt their rankings. I hope.

[Onomatopoeia]

Regardless of the policy reasoning behind extra time, or the flagged versus non-flagged scores issue, one of the outcomes of this will be a surge in LSAT's taken with disability accommodation. Since US News will no longer be able to discount accommodated LSAT's due to no special markings being given, they will be worth just as much to a law school's ranking as regular LSAT's. It's quite likely that the vast majority of accommodated testers (At least initially) will have real disabilities that negatively affect their performance. Yet it's also worth noting that people respond to incentives, and in the realm of law school applications opening an avenue to game the LSAT is perhaps the most powerful. It's why people pay thousands of dollars for prep courses, and it's why this system is rife for abuse.

Watch as contrary to the current trend towards lower medians, scores instead trend upward rapidly in the next couple of years due to an increasing awareness of this lowest-common-denominator standard of proof. Which post-graduate test has the broadest scope and weakest verification of disabilities that the LSAT will be forced to match (The GRE, GMAT, MCAT)? I'm profoundly thankful I'll be attending this cycle, and not the next. Splitters are screwed from here on out, HYS to 175+ medians.

If they do get a surge in inflated LSATs (at least until the "curve" catches up) the schools are likely to suffer with lower bar passage rates, which will hurt their rankings. I hope.

How exactly will the curve catch up? Everyone with a subpar gpa seems to be fucked bro. The curve 'catching up' seems to entail a tougher curve as a greater percentage of lsat test takers will be taking the test with more time. If you exclude the accommodated testers from the test curve then the curve will not change. Actually how would that work, lol. Presumably more people will be scoring higher, but they will not be considered when calculating the curve. So will law schools just expect less of a bell curve with scores. Will the percentiles per score stay constant. How would percentiles be impacted if you don't include accommodated testers in the curve --> It seems like if you DONT flag scores, you need to include all test takers within the curve.

The only out is creating an alternate exam, examples include: more of a technical test or a subjective exam (essay based) - both of which are less easily duped with the granting of accommodations

[papercut]

Regardless of the policy reasoning behind extra time, or the flagged versus non-flagged scores issue, one of the outcomes of this will be a surge in LSAT's taken with disability accommodation. Since US News will no longer be able to discount accommodated LSAT's due to no special markings being given, they will be worth just as much to a law school's ranking as regular LSAT's. It's quite likely that the vast majority of accommodated testers (At least initially) will have real disabilities that negatively affect their performance. Yet it's also worth noting that people respond to incentives, and in the realm of law school applications opening an avenue to game the LSAT is perhaps the most powerful. It's why people pay thousands of dollars for prep courses, and it's why this system is rife for abuse.

Watch as contrary to the current trend towards lower medians, scores instead trend upward rapidly in the next couple of years due to an increasing awareness of this lowest-common-denominator standard of proof. Which post-graduate test has the broadest scope and weakest verification of disabilities that the LSAT will be forced to match (The GRE, GMAT, MCAT)? I'm profoundly thankful I'll be attending this cycle, and not the next. Splitters are screwed from here on out, HYS to 175+ medians.

If they do get a surge in inflated LSATs (at least until the "curve" catches up) the schools are likely to suffer with lower bar passage rates, which will hurt their rankings. I hope.

How exactly will the curve catch up? Everyone with a subpar gpa seems to be fucked bro. The curve 'catching up' seems to entail a tougher curve as a greater percentage of lsat test takers will be taking the test with more time. If you exclude the accommodated testers from the test curve then the curve will not change. Actually how would that work, lol. Presumably more people will be scoring higher, but they will not be considered when calculating the curve. So will law schools just expect less of a bell curve with scores. Will the percentiles per score stay constant. How would percentiles be impacted if you don't include accommodated testers in the curve --> It seems like if you DONT flag scores, you need to include all test takers within the curve.

The only out is creating an alternate exam, examples include: more of a technical test or a subjective exam (essay based) - both of which are less easily duped with the granting of accommodations

Yeah I think this sucks. I was just musing about how it might hurt the schools to admit a huge percentage of applicants with inflated scores.

The alternatives you mention sound like decent. But an alternate exam is probably really expensive to put together.

[Onomatopoeia]

Regardless of the policy reasoning behind extra time, or the flagged versus non-flagged scores issue, one of the outcomes of this will be a surge in LSAT's taken with disability accommodation. Since US News will no longer be able to discount accommodated LSAT's due to no special markings being given, they will be worth just as much to a law school's ranking as regular LSAT's. It's quite likely that the vast majority of accommodated testers (At least initially) will have real disabilities that negatively affect their performance. Yet it's also worth noting that people respond to incentives, and in the realm of law school applications opening an avenue to game the LSAT is perhaps the most powerful. It's why people pay thousands of dollars for prep courses, and it's why this system is rife for abuse.

Watch as contrary to the current trend towards lower medians, scores instead trend upward rapidly in the next couple of years due to an increasing awareness of this lowest-common-denominator standard of proof. Which post-graduate test has the broadest scope and weakest verification of disabilities that the LSAT will be forced to match (The GRE, GMAT, MCAT)? I'm profoundly thankful I'll be attending this cycle, and not the next. Splitters are screwed from here on out, HYS to 175+ medians.

If they do get a surge in inflated LSATs (at least until the "curve" catches up) the schools are likely to suffer with lower bar passage rates, which will hurt their rankings. I hope.

How exactly will the curve catch up? Everyone with a subpar gpa seems to be fucked bro. The curve 'catching up' seems to entail a tougher curve as a greater percentage of lsat test takers will be taking the test with more time. If you exclude the accommodated testers from the test curve then the curve will not change. Actually how would that work, lol. Presumably more people will be scoring higher, but they will not be considered when calculating the curve. So will law schools just expect less of a bell curve with scores. Will the percentiles per score stay constant. How would percentiles be impacted if you don't include accommodated testers in the curve --> It seems like if you DONT flag scores, you need to include all test takers within the curve.

The only out is creating an alternate exam, examples include: more of a technical test or a subjective exam (essay based) - both of which are less easily duped with the granting of accommodations

Yeah I think this sucks. I was just musing about how it might hurt the schools to admit a huge percentage of applicants with inflated scores.

The alternatives you mention sound like decent. But an alternate exam is probably really expensive to put together.

the applicant will lose before the law school. every law school will have a hit on their bar passage rate if they don't wise up to the shift in scores. and i think they will focus on gpa more, if anything. the lsat median will be no more of an issue than it ever was, and the bar passage worries will not be more of a worry than they ever were bc presumably inflated scores will hurt bar passage rates uniformly across all schools. this whole thing is sooo dumb

[papercut]

the applicant will lose before the law school. every law school will have a hit on their bar passage rate if they don't wise up to the shift in scores. and i think they will focus on gpa more, if anything. the lsat median will be no more of an issue than it ever was, and the bar passage worries will not be more of a worry than they ever were bc presumably inflated scores will hurt bar passage rates uniformly across all schools. this whole thing is sooo dumb

Yeah man this was very disappointing news to me too.

The way schools treat GPAs is absolutely atrocious. LSAC uses a 4.3 scale even for 4.0 GPAs and schools don't seem to care about the difference. Then there are the grade inflation shit shows where most people get an A, and again the schools don't seem to care about the difference.

It's hard to not be cynical about LS admissions.

[Onomatopoeia]

the applicant will lose before the law school. every law school will have a hit on their bar passage rate if they don't wise up to the shift in scores. and i think they will focus on gpa more, if anything. the lsat median will be no more of an issue than it ever was, and the bar passage worries will not be more of a worry than they ever were bc presumably inflated scores will hurt bar passage rates uniformly across all schools. this whole thing is sooo dumb

Yeah man this was very disappointing news to me too.

The way schools treat GPAs is absolutely atrocious. LSAC uses a 4.3 scale even for 4.0 GPAs and schools don't seem to care about the difference. Then there are the grade inflation shit shows where most people get an A, and again the schools don't seem to care about the difference.

It's hard to not be cynical about LS admissions.

we could always turn this cynicism into profit by opening up shop. We will call it Bornagain & Papercut School of Law

[papercut]

the applicant will lose before the law school. every law school will have a hit on their bar passage rate if they don't wise up to the shift in scores. and i think they will focus on gpa more, if anything. the lsat median will be no more of an issue than it ever was, and the bar passage worries will not be more of a worry than they ever were bc presumably inflated scores will hurt bar passage rates uniformly across all schools. this whole thing is sooo dumb

Yeah man this was very disappointing news to me too.

The way schools treat GPAs is absolutely atrocious. LSAC uses a 4.3 scale even for 4.0 GPAs and schools don't seem to care about the difference. Then there are the grade inflation shit shows where most people get an A, and again the schools don't seem to care about the difference.

It's hard to not be cynical about LS admissions.

we could always turn this cynicism into profit by opening up shop. We will call it Bornagain & Papercut School of Law

If those guys can do it, we can too. I'm sure there's a Vietnamese hairdresser that's looking to rent out their space for the mornings.

[Onomatopoeia]

id just like to add that getting accommodations for the bar exam in your state should not be too difficult per my research, so again, only applicants heavily relying on the lsat (splitters, international applicants) to stand out for law school admissions will be losers in general - more so than they are now, thats for sure

[Jeffort]

Regardless of the policy reasoning behind extra time, or the flagged versus non-flagged scores issue, one of the outcomes of this will be a surge in LSAT's taken with disability accommodation. Since US News will no longer be able to discount accommodated LSAT's due to no special markings being given, they will be worth just as much to a law school's ranking as regular LSAT's. It's quite likely that the vast majority of accommodated testers (At least initially) will have real disabilities that negatively affect their performance. Yet it's also worth noting that people respond to incentives, and in the realm of law school applications opening an avenue to game the LSAT is perhaps the most powerful. It's why people pay thousands of dollars for prep courses, and it's why this system is rife for abuse.

Watch as contrary to the current trend towards lower medians, scores instead trend upward rapidly in the next couple of years due to an increasing awareness of this lowest-common-denominator standard of proof. Which post-graduate test has the broadest scope and weakest verification of disabilities that the LSAT will be forced to match (The GRE, GMAT, MCAT)? I'm profoundly thankful I'll be attending this cycle, and not the next. Splitters are screwed from here on out, HYS to 175+ medians.

If they do get a surge in inflated LSATs (at least until the "curve" catches up) the schools are likely to suffer with lower bar passage rates, which will hurt their rankings. I hope.

How exactly will the curve catch up? Everyone with a subpar gpa seems to be fucked bro. The curve 'catching up' seems to entail a tougher curve as a greater percentage of lsat test takers will be taking the test with more time. If you exclude the accommodated testers from the test curve then the curve will not change. Actually how would that work, lol. Presumably more people will be scoring higher, but they will not be considered when calculating the curve. So will law schools just expect less of a bell curve with scores. Will the percentiles per score stay constant. How would percentiles be impacted if you don't include accommodated testers in the curve --> It seems like if you DONT flag scores, you need to include all test takers within the curve.

The only out is creating an alternate exam, examples include: more of a technical test or a subjective exam (essay based) - both of which are less easily duped with the granting of accommodations

Yeah I think this sucks. I was just musing about how it might hurt the schools to admit a huge percentage of applicants with inflated scores.

The alternatives you mention sound like decent. But an alternate exam is probably really expensive to put together.

Gotta clear up one important fact about 'the curve'/scoring scale for each test form and how the sections are equated.

Accommodated LSAT test takers are not administered an experimental section, so their performance has ZERO influence on the 'curve'/scoring scales/difficulty level determinations/etc. used for equating to assemble test forms and determine score scale conversion chart for each test form, so you guys can toss out that worry if a bunch of people end up gaming their way into high LSAT scores through extra time they don't legitimately deserve. Also, accommodated scores are not included in calculating the % ranking chart for scaled scores. Just because LSAC will no longer put an * on score reports doesn't mean they're going to or have to include accommodated scores in the % rank calculations, so this won't affect that either.

I think the terms of this settlement agreement suck due to the potentials for abuse it opens up that people will certainly attempt to exploit given the high stakes involved in LS admissions with LSAT scores. The language about having to honor previously given extra time accommodations is a little vague. I'm not sure if it means that if the SAT gave double time that LSAC has to also give double time. The language seems open to interpretation that LSAC has to give appropriate extra time, but still has discretion to determine how much in light of the differences in the exam types and influence of time on skills being tested.

The article Graeme linked on page 1 about the strict rules for getting special accommodations on the SAT that got put into place a while ago gave me comfort that this won't expose the LSAT to being easily exploited illegitimately for extra time, but I'm not clear about how many other tests this agreement covers besides the obvious ones like the SAT and don't yet know if any of them are super lax so that someone could literally do everything they needed in like 2-3 months with some stupid obscure test. There are tons of specialty exams for all sorts of professions, from plumbing up to nuclear engineering and all sorts of obscure stuff. I'm worried about there being an easy to get accoms quick test this agreement forces LSAC to copy unquestioned.

In the ADA and also in the 2002 previous DOJ LSAC settlement agreement there is a bunch of important definitions and discussion about accommodating only for things unrelated to the skills the test is meant to measure to insure that disabilities don't impair their ability to demonstrate their skill level due to an aspect of how the test is administered that isn't directly part of the skills/abilities the test is designed to measure. As mentioned already, there is a big difference in how much extra time can help a score on one type of test vs another. Extra time isn't going to help for a knowledge based test if you don't know the stuff, but the LSAT is specifically a skills/performance test intended to measure them under heavily timed/time limited conditions, partly because all the information you need to know to get the question correct other than common sense and common general knowledge is given to you and it's a multiple choice test, making timing in important part of mental skills the test is meant to measure. If LSAC has to give the same proportion of extra time some SAT people got, it will surely give an unfair advantage to many people.

People with legit disabilities that deserve special accommodations should for sure get them, ones that are proportionate to compensate for their specific limitations as they apply to the specific test, how it's administered, which skills the test is meant to measure in fair ways to level the playing field. I fear this auto-approve rule is going to end up giving many people an unfair advantage they don't deserve unless LSAC has some flexibility in determining how much extra time to give.

Funny thing is, the people that do end up gaming it successfully to get into a top law school with an artificially inflated LSAT score from scamming extra time will mostly likely actually be totally screwing themselves in the arse long term. Good chance people that gamed their way into a high LSAT score and a top LS that wouldn't have without the extra time will end up getting shitty grades/end up at/close to the bottom of the class curve at graduation, making getting a big law or any other high paying law job right out of LS pretty hard. People from top schools that graduated with low class rank have trouble finding jobs even though prestigious LS pedigree. Being in the bottom third or whatever of the LS class GPA is looked down at pretty badly by employers no matter what school your degree is from, at least for big law/high starting salary right out of school jobs and especially for clerkships and lots of other important stuff. Bar pass rate might be lower for them too.

LSAC cares a lot about the integrity of the test and has stood strong all these years against tons of lawsuits to keep it fair and ungameble, so I expect they're going to do whatever they can to make sure this doesn't hurt the integrity/reliability of the test and admissions system. It'll be interesting to watch how it plays out over the next few years. It's obvious a lot more people are going to apply for extra time and many will explore ways to game extra time even though they don't deserve it, I just hope LSAC and the law schools team up in good ways to make sure the scammers don't make it through.

Given all the detailed personal questions LS apps already include these days compared to years ago, it wouldn't surprise me if LSs simply just add another few questions to their apps about having been given spec accoms before or not, or whether the student might ask for it, or whatever types of questions they could legally ask to figure it out and get around the new no * rule. Hopefully LSAC will track the LS performance of the accommodated scores students under the new policy and take action to modify the policy once they find statistical proof that it's being abused like the SAT people figured out in the article graeme linked.

Testing the same skills in a similar format without time pressure wouldn't actually be that hard to do with some modifications to the LSAT. Same stimulus and such, just not multiple choice. written answers instead or something, just not multiple choice where you can use POE. It won't happen though because grading would be a bitch with labor and consistency so it won't happen. I think add/adhd is a nonsense excuse for extra time for most people except the true extreme cases, which are only a small proportion of all people currently diagnosed as add. It's one of the more commonly over diagnosed conditions these days. From what I've seen with many students, add kids tend to do better and improve faster than others, whether medicated or not, due to their quick thinking minds and being used to jumping from thinking about one thing to another rapidly like you do in LR and LGs, but that's just anecdotal from my experience.

[papercut]

180 post Jeffort.

[Onomatopoeia]

Jeffort what you say resonates with me. I knew that accommodated testers weren't included in the curve before but with flagging scores no longer being there, not including them in the curve would be a travesty if law schools had no way of seeing accommodated scores and evaluating them as such. If people want to assume that accommodation only levels the playing field then I don't see why they would be against including accommodated testers within the curve. Law schools should ask if applicants received accommodated testing like you said. But it gets to the whole issue of whether law schools can ask these things. The ADA has been and will continue to be on law schools' asses. And the whole idea that accommodated testers will flunk out... I'm not so sure. The Lsat isn't a perfect predictor, and accommodations for law school and the bar exam are not that difficult to come by. Screw the idea that there is a stigma stopping people from asking for these accommodations when hundreds of thousands of dollars are on the line.

[ScottRiqui]

180 post Jeffort.

[Jeffort]

Jeffort what you say resonates with me. I knew that accommodated testers weren't included in the curve before but with flagging scores no longer being there, not including them in the curve would be a travesty if law schools had no way of seeing accommodated scores and evaluating them as such. If people want to assume that accommodation only levels the playing field then I don't see why they would be against including accommodated testers within the curve. Law schools should ask if applicants received accommodated testing like you said. But it gets to the whole issue of whether law schools can ask these things. The ADA has been and will continue to be on law schools' asses. And the whole idea that accommodated testers will flunk out... I'm not so sure. The Lsat isn't a perfect predictor, and accommodations for law school and the bar exam are not that difficult to come by. Screw the idea that there is a stigma stopping people from asking for these accommodations when hundreds of thousands of dollars are on the line.

Test equating using pre-tested experimental sections wouldn't be psychometrically valid if they administered experimental sections to accommodated test takers. That's why LSAC never has. If you get spec accom, you only take four sections and the writing sample. This settlement doesn't change that. If LSAC wants to include the scores in the percentile rank chart that is published with each score release they can, but they currently don't since they aren't normal conditions scores and not representative of the general test taker pool the chart is meant to represent.

I need to read the agreement more when I have time. I would be surprised if there isn't something in there to give LSAC some flexibility/discretion with applicants that look fishy, such as people asking for extra time that just took the GZAP or whatever a couple months earlier with special accoms that were easy to get and that's all the student has to back up extra time request to LSAC and the other test is unrelated to any academic or professional activities/goals/life of the applicant. I wouldn't think the LSAC lawyers and directors would agree to something that ties their hands when flooded with obviously trying to game the system demands for extra time that are blatantly gamed with some other test with lax rules.

[NYSprague]

I want to see the result in practice. The basis for this suit was that legitimate needs weren't being addressed. I feel that was a disservice to people who had disabilities.
You guys are worried about the system being gamed, by why so much?
Shouldn't you worry about getting your own high score? If you can do that without extra time, this shouldn't affect you.

I guess I don't agree this will open the floodgates to many high scores until I actually see it happening.

I also don't think that high scores due to accommodations will mean someone will get low grades in law school. But I've never agreed with the idea that the LSAT predicts grades or performance.

I don't think that schools can ask about disabilities. People have to self-identify.

I guess it bothers me that people don't see this, in a large part, as an positive correction for people who have been wrongly denied extra time. It also bothers me that people want to judge for themselves who deserves extra time without any expertise and assume people don't really need it. That is exactly what LSAC was doing.

Maybe because I've been out of school for so long I have a different perspective.

[papercut]

Shouldn't you worry about getting your own high score? If you can do that without extra time, this shouldn't affect you.

In law school admissions, if you boost one person you necessarily nerf another.

But I've never agreed with the idea that the LSAT predicts grades or performance.

You can disagree all you want but LSAT + GPA gives you a monster correlation with 1L grades, especially when you consider that it is social science we're dealing with.

[haus]

I guess it bothers me that people don't see this, in a large part, as an positive correction for people who have been wrongly denied extra time. It also bothers me that people want to judge for themselves who deserves extra time without any expertise and assume people don't really need it. That is exactly what LSAC was doing.

Perhaps people are OK with the idea that the playing field has been uneven as long as they are not the one's who are short changed by it.

[ScottRiqui]

You can disagree all you want but LSAT + GPA gives you a monster correlation with 1L grades, especially when you consider that it is social science we're dealing with.

I wouldn't go this far. The median correlation coefficient between LSAT + uGPA and 1L grades is only 0.48, which is actually pretty loose. As an example, here's a scatter plot with a similar correlation (0.5):



There's an obvious overall positive correlation, but a lot of the data points are significantly under- or over-performing the prediction.

Just because the combination of LSAT + GPA is the single best predictor we have, that doesn't mean it's necessarily a very good predictor - it just means everything else is worse.

[papercut]

You can disagree all you want but LSAT + GPA gives you a monster correlation with 1L grades, especially when you consider that it is social science we're dealing with.

I wouldn't go this far. The median correlation coefficient between LSAT + uGPA and 1L grades is only 0.48, which is actually pretty loose. As an example, here's a scatter plot with a similar correlation (0.5):



There's an obvious overall positive correlation, but a lot of the data points are significantly under- or over-performing the prediction.

Just because the combination of LSAT + GPA is the single best predictor we have, that doesn't mean it's necessarily a very good predictor - it just means everything else is worse.

What are you comparing this to? This is social science we're talking about here. Social scientists get wet over a .3 correlation coefficient.

The range of correlations for LSAT + GPA and 1L grades also depends on the law school. Not sure what makes the difference, but at some law schools it's quite large (again for social science), I think it's .6? Too lazy to look it up.

[ScottRiqui]

What are you comparing this to? This is social science we're talking about here. Social scientists get wet over a .3 correlation coefficient.

The range of correlations for LSAT + GPA and 1L grades also depends on the law school. Not sure what makes the difference, but at some law schools it's quite large (again for social science), I think it's .6? Too lazy to look it up.

Just because social scientists are happy to get a coefficient of 0.3, that doesn't make it any more predictive than if it were a chemist or a physicist getting the same coefficient on their experiments - the correlation is still fairly loose. We can't let ourselves mistake "the best we have" for "good" - a weak correlation is still a weak correlation, no matter what field you're in.

Yes, the coefficients for various schools ranged from 0.3 to 0.62, but I haven't seen anything that suggests that the coefficient at any particular school stays constant over time, so the median coefficient is more useful than the extremes.

Also, this particular settlement isn't going to fuck with uGPAs, so we really should be talking about the correlation between 1L grades and LSAT alone, which has a median coefficient of only 0.36. At that point, all you can really say about LSAT relative to 1L grades is "more is better, in general". The predictive power for any single applicant is still shaky.

[cotiger]

Funny thing is, the people that do end up gaming it successfully to get into a top law school with an artificially inflated LSAT score from scamming extra time will mostly likely actually be totally screwing themselves in the arse long term. Good chance people that gamed their way into a high LSAT score and a top LS that wouldn't have without the extra time will end up getting shitty grades/end up at/close to the bottom of the class curve at graduation, making getting a big law or any other high paying law job right out of LS pretty hard. People from top schools that graduated with low class rank have trouble finding jobs even though prestigious LS pedigree. Being in the bottom third or whatever of the LS class GPA is looked down at pretty badly by employers no matter what school your degree is from, at least for big law/high starting salary right out of school jobs and especially for clerkships and lots of other important stuff. Bar pass rate might be lower for them too.

I think this is a way overstated impact. One of the major TLS mantras is that placement drops off a whole lot faster than student body quality, so it doesn't make sense to go to a lower school with the idea that you'll do better. For instance, you'd much rather be around "top" 70% at Northwestern than "top" 55% at UIUC. Getting a higher LSAT puts you in a better position regardless of your ability level.

LSAC cares a lot about the integrity of the test and has stood strong all these years against tons of lawsuits to keep it fair and ungameble, so I expect they're going to do whatever they can to make sure this doesn't hurt the integrity/reliability of the test and admissions system. It'll be interesting to watch how it plays out over the next few years. It's obvious a lot more people are going to apply for extra time and many will explore ways to game extra time even though they don't deserve it, I just hope LSAC and the law schools team up in good ways to make sure the scammers don't make it through.

If the problem was scammers, this would be a lot easier to work through. The problem is the hordes of well-intentioned well-off parents who just want the best for their children. Little Johnny has been rambunctious and isn't paying much attention in school->Take Johnny to the doctor to see if you can help him->doctor prescribes some medication to help Johnny pay attention->boom, you've got your documented disability->accommodation for life.

Look, if someone thinks they have such a severe disability that they can't take the same test as everyone else, that's fine. But cognitive ability is absolutely of legitimate interest to grad schools looking for students, and it should be indicated that they took the easier test. It would then be up to the schools to place those test results in context with the rest of the application to determine if they think that applicant would succeed at their institution.

[LSAT Hacks (Graeme)]

Great post Jeffort. Neither the LSAC nor the DOJ are dummies, so hopefully there's something in the agreement to prevent the scenario I described.

[cotiger]

Great post Jeffort. Neither the LSAC nor the DOJ are dummies, so hopefully there's something in the agreement to prevent the scenario I described.

The fact that they aren't dummies doesn't change the fact that the way that they interpret ADA compliance (low bar+no flagging) will cause effects we've all described. Ultimately the DOJ isn't in the business of making sure that the LSAT is a valid test, they're in the business of enforcing the law.

[cotiger]

Some research!

"Separate predictive validity analyses were conducted for test takers classified as having Attention Deficit/Hyperactivity Disorder, Learning Disabilities, Neurological Impairment, and Visual Impairment... Results suggest that LSAT scores earned under accommodated testing conditions that included extra testing time are not comparable to LSAT scorers earned under standard timing conditions as evidenced by a tendency of the former to overpredict FYAs. Results for individual groups were consistent with the overall group result."

http://eric.ed.gov/?id=ED469183

[NYSprague]

In New York students have to be tested and the results accepted to get accommodations. It isn't just a prescription.

[LSAT Hacks (Graeme)]

Some research!

"Separate predictive validity analyses were conducted for test takers classified as having Attention Deficit/Hyperactivity Disorder, Learning Disabilities, Neurological Impairment, and Visual Impairment... Results suggest that LSAT scores earned under accommodated testing conditions that included extra testing time are not comparable to LSAT scorers earned under standard timing conditions as evidenced by a tendency of the former to overpredict FYAs. Results for individual groups were consistent with the overall group result."

http://eric.ed.gov/?id=ED469183

Good catch. That's an LSAC report. Here's an updated version, full text. They give particular emphasis to extra time. The report says non-time accommodations preserve predictive validity, while time accommodations remove it.

http://www.lsac.org/docs/default-source ... -09-01.pdf