LSAT Flex score inflation? Forum

[DMVdude]

Hi everyone,

I was wondering what everyone's thoughts are on how the LSAT flex will affect the distribution of scores. I feel like having 3 sections instead of 4/5 will make the test less mentally strenuous and might result in higher average scores? If so, are average scores at top schools going to rise significantly? I would appreciate hearing others' thoughts on this as I haven't been able to find anything on this issue.

Thanks.

[IntellectualMode]

Hi everyone,

I was wondering what everyone's thoughts are on how the LSAT flex will affect the distribution of scores. I feel like having 3 sections instead of 4/5 will make the test less mentally strenuous and might result in higher average scores? If so, are average scores at top schools going to rise significantly? I would appreciate hearing others' thoughts on this as I haven't been able to find anything on this issue.

Thanks.

This is a really good question!

Generally speaking, you are right in that the LSAT-Flex will be less mentally strenuous.
Doing 3 fully scored sections as opposed to 5 sections, 4 of which are scored and 1 that isn't, takes a lot less longer than a full LSAT paper - that's a no brainer.

A classic LSAT is a bit like a marathon - most test takers spend months if not years training themselves to perform well on a test that lasts for a good few hours and is quite mentally draining.
So to cut out 2 full sections of the test may mean that test takers have more energy and a higher level of concentration, which could lead to an improved performance on the test.

For the distribution of scores, LSAC aren't working with the usual 99-101 total questions, they'd be working with a much smaller amount of total questions between the mid to high 70s.
After readjusting the percentiles and considering the possibility for increased performance amongst test takers, it's likely that the average LSAT score will go up for test takers who are taking the LSAT-Flex exam.

As for its impact on average LSAT scores at top schools, it's possible it might have an effect but it depends on a lot of different considerations:

- LSAT-Flex test takers are going to represent a certain proportion out of all students applying to law schools for a given year.

- Members of Law schools admissions teams may realise that it's a possibility that test takers from these
administrations potentially have higher average LSAT scores than test takers who did not. Whether or not
an individual law school would actively take such a suggestion into account in their admissions process is
unknown.

We really can't say anything for definite re impact at this stage. We'll just have to wait and see.

[nixy]

- Members of Law schools admissions teams may realise that it's a possibility that test takers from these administrations potentially have higher average LSAT scores than test takers who did not. Whether or not an individual law school would actively take such a suggestion into account in their admissions process is unknown.

There's no reason to think they'd do anything other than happily snatch up as many high scores as they can. Until/unless they have to omit those scores from consideration for the USNWR rankings, they don't have any incentive to do otherwise.

[IntellectualMode]

- Members of Law schools admissions teams may realise that it's a possibility that test takers from these administrations potentially have higher average LSAT scores than test takers who did not. Whether or not an individual law school would actively take such a suggestion into account in their admissions process is unknown.

There's no reason to think they'd do anything other than happily snatch up as many high scores as they can. Until/unless they have to omit those scores from consideration for the USNWR rankings, they don't have any incentive to do otherwise.

Perhaps, but just because there's no incentive doesn't mean they wouldn't take it into account.

[nixy]

- Members of Law schools admissions teams may realise that it's a possibility that test takers from these administrations potentially have higher average LSAT scores than test takers who did not. Whether or not an individual law school would actively take such a suggestion into account in their admissions process is unknown.

There's no reason to think they'd do anything other than happily snatch up as many high scores as they can. Until/unless they have to omit those scores from consideration for the USNWR rankings, they don't have any incentive to do otherwise.

Perhaps, but just because there's no incentive doesn't mean they wouldn't take it into account.

Then let me put it this way - why would they "take it into account" when they have every incentive just to take the (presumably) higher scores to pad their stats? (I'm assuming the concern here is that the flex-LSAT scores will somehow be "unfairly" high compared to the regular ones.)

[IntellectualMode]

- Members of Law schools admissions teams may realise that it's a possibility that test takers from these administrations potentially have higher average LSAT scores than test takers who did not. Whether or not an individual law school would actively take such a suggestion into account in their admissions process is unknown.

There's no reason to think they'd do anything other than happily snatch up as many high scores as they can. Until/unless they have to omit those scores from consideration for the USNWR rankings, they don't have any incentive to do otherwise.

Perhaps, but just because there's no incentive doesn't mean they wouldn't take it into account.

Then let me put it this way - why would they "take it into account" when they have every incentive just to take the (presumably) higher scores to pad their stats? (I'm assuming the concern here is that the flex-LSAT scores will somehow be "unfairly" high compared to the regular ones.)

Well, within an admissions cycle, it could be a concern of providing an unfair playing field.
For example, If we assumed that admissions were based on nothing but the LSAT score, and we were to say that an LSAT-Flex student who got a 180 and a LSAT student who got a 180 are in completely the same positions, is that fair?

The LSAT student had the full 5-section test and ascertained the top percentile score, and the LSAT-Flex student did exactly the same but they only had a 3-section test.
If the Flex student had completed a 5-section test, it is completely unknown whether that student would've actually ascertained their 180.

This obviously isn't a real scenario because admissions takes into account other application component, but do you see what I'm trying to get at?

[nixy]

Yes, I completely understand what you're getting at. I just don't think it's going to make any difference. It's like how your GPA is what it is, regardless of whether your major was underwater basketweaving at the easiest community college around or rocket science at MIT.

(Also I think you mean "obtain" not "ascertain.")

[LBJ's Hair]

A little confused here. Let's assume the Flex is "easier." Why would that result in "higher" LSAT scores?

An LSAT score is a percentile, not a raw score. You need, iirc, to be in the 97.5-98th percentile to break a 170, regardless of how "easy" or "hard" the test is for a given sitting. That's why the raw score required to break 170 varies from test to test. On "easy" tests, you might need a -12 to get a 170. On "hard" tests, a -15. If the Flex is really easy, the raw score required to break 170 will be a lot lower, but that doesn't mean proportionally more people will be getting 170+s. Those are awarded to only the top 2.5% or so of test-takers, regardless of raw score.

There are, I guess, some edge case scenarios where a wildly non-normal distribution of raw scores could result in strange LSAT score distributions. Say the Flex was so easy (or so many people cheated or w/e) that 10 percent of all Flex test-takers got a perfect score. All of those students would get a 180, the next highest raw score would earn a 165 or so (which corresponds to the 90th percentile), and no one would get anything in between.

But barring that, I don't see why the Flex is any different than any other LSAT administration. What am I missing?

As an aside: LSAT score averages have increased at top law schools over the past few cycles, but that's because more people have taken the test (resulting in, necessarily, more 170+ scores) and law schools haven't increased classes sizes. It's not because the test is "easier."

[dvlthndr]

I see two possible concerns, but I don't think either of them make the test "easier."

First, the test is "shorter" and you don't need as much endurance to complete it. This rewards/punishes a certain kind of student, but I don't think it makes it any easier in absolute terms.

Second, as people have noted, having fewer questions can cause "clumping" in the highest/lowest part of the distribution. You will see more people with a perfect 180, but you will also be kicked out of the high 170 range with fewer wrong answers (e.g., getting 3 questions wrong might result in a 176 versus a 177). That said, the percent of people with 170+ range scores won't change, and I don't think you can really call it "score inflation."

Crazy things can happen when a standardized test is "too easy" (e.g., the distribution for the SAT "Math II" subject test is bonkers since a full fifth of test-takers get a perfect score), but you aren't going to see something like that happen with the LSAT just because they shrunk the size of it.

[AfroKant]

To some extent, I understand the worry about the possibility of inflation, but I think more simply, nobody who couldn't get/would have not gotten 170+ on the regular LSAT will magically allow them to suddenly reach that score. Whether the test is comprised of 3 or 5 sections, you still need a certain degree of aptitude/cognitive ability/studying and effort/etc. to really break into the 95th percentile and higher, and that aptitude/ability/however you want to characterize it does not really seem to be contingent upon time constraints and/or the amount of sections required, IMO.

[DMVdude]

- Members of Law schools admissions teams may realise that it's a possibility that test takers from these administrations potentially have higher average LSAT scores than test takers who did not. Whether or not an individual law school would actively take such a suggestion into account in their admissions process is unknown.

There's no reason to think they'd do anything other than happily snatch up as many high scores as they can. Until/unless they have to omit those scores from consideration for the USNWR rankings, they don't have any incentive to do otherwise.

Has there been any news on USNWR's consideration of LSAT Flex scores?

[DMVdude]

I'm currently scheduled for the July test. My main two worries are, if I take the Flex, that my scores will be seen as somewhat less legitimate, and if I don't take the Flex version, I will be at somewhat of a disadvantage because of higher averages.

I'm right in the 170-175 range on 4 section PTs right now, so a bunch of 180s on the flex would definitely hurt my position. I'm primarily concerned because my GPA is slightly lower than the average of my ideal schools, so having a higher than average LSAT score for those schools is especially important to my application. I suppose my score will balance out with others in the same score range regardless.

Ultimately, does anyone have a strong opinion for or against taking the Flex? Assuming I have a choice...

[DMVdude]

A little confused here. Let's assume the Flex is "easier." Why would that result in "higher" LSAT scores?

An LSAT score is a percentile, not a raw score. You need, iirc, to be in the 97.5-98th percentile to break a 170, regardless of how "easy" or "hard" the test is for a given sitting. That's why the raw score required to break 170 varies from test to test. On "easy" tests, you might need a -12 to get a 170. On "hard" tests, a -15. If the Flex is really easy, the raw score required to break 170 will be a lot lower, but that doesn't mean proportionally more people will be getting 170+s. Those are awarded to only the top 2.5% or so of test-takers, regardless of raw score.

There are, I guess, some edge case scenarios where a wildly non-normal distribution of raw scores could result in strange LSAT score distributions. Say the Flex was so easy (or so many people cheated or w/e) that 10 percent of all Flex test-takers got a perfect score. All of those students would get a 180, the next highest raw score would earn a 165 or so (which corresponds to the 90th percentile), and no one would get anything in between.

But barring that, I don't see why the Flex is any different than any other LSAT administration. What am I missing?

As an aside: LSAT score averages have increased at top law schools over the past few cycles, but that's because more people have taken the test (resulting in, necessarily, more 170+ scores) and law schools haven't increased classes sizes. It's not because the test is "easier."

Thanks for the input. Very helpful

[mysonx3]

If the Flex student had completed a 5-section test, it is completely unknown whether that student would've actually ascertained their 180.

The other side of this coin is that, if the 5-section student had completed the Flex, it is completely unknown whether that student would have still gotten a 180.

You seem to be operating under two assumptions:

- That a 180 on the Flex doesn't necessarily equate to a 180 on the 5-section test (which is a justified assumption, since we just don't know)
- That a 180 on the 5-section DOES equate to a 180 on the Flex (which is not justified because, again, we just don't know)

[abcanada]

A little confused here. Let's assume the Flex is "easier." Why would that result in "higher" LSAT scores?

An LSAT score is a percentile, not a raw score. You need, iirc, to be in the 97.5-98th percentile to break a 170, regardless of how "easy" or "hard" the test is for a given sitting. That's why the raw score required to break 170 varies from test to test. On "easy" tests, you might need a -12 to get a 170. On "hard" tests, a -15. If the Flex is really easy, the raw score required to break 170 will be a lot lower, but that doesn't mean proportionally more people will be getting 170+s. Those are awarded to only the top 2.5% or so of test-takers, regardless of raw score.

There are, I guess, some edge case scenarios where a wildly non-normal distribution of raw scores could result in strange LSAT score distributions. Say the Flex was so easy (or so many people cheated or w/e) that 10 percent of all Flex test-takers got a perfect score. All of those students would get a 180, the next highest raw score would earn a 165 or so (which corresponds to the 90th percentile), and no one would get anything in between.

But barring that, I don't see why the Flex is any different than any other LSAT administration. What am I missing?

As an aside: LSAT score averages have increased at top law schools over the past few cycles, but that's because more people have taken the test (resulting in, necessarily, more 170+ scores) and law schools haven't increased classes sizes. It's not because the test is "easier."

I had the same questions as OP. Is the above info factual? If so, the test possibly being "easier" is meaningless.

If I am understanding this correctly, take this hypothetical - everyone BOMBS the test and the 97th percentile only getting 70/101 questions correct. Would that taker still get a 170? Curious as this would clam the nerves a bit of school score medians shooting up 1 point each this year.

[dvlthndr]

A little confused here. Let's assume the Flex is "easier." Why would that result in "higher" LSAT scores?

An LSAT score is a percentile, not a raw score. You need, iirc, to be in the 97.5-98th percentile to break a 170, regardless of how "easy" or "hard" the test is for a given sitting. That's why the raw score required to break 170 varies from test to test. On "easy" tests, you might need a -12 to get a 170. On "hard" tests, a -15. If the Flex is really easy, the raw score required to break 170 will be a lot lower, but that doesn't mean proportionally more people will be getting 170+s. Those are awarded to only the top 2.5% or so of test-takers, regardless of raw score.

There are, I guess, some edge case scenarios where a wildly non-normal distribution of raw scores could result in strange LSAT score distributions. Say the Flex was so easy (or so many people cheated or w/e) that 10 percent of all Flex test-takers got a perfect score. All of those students would get a 180, the next highest raw score would earn a 165 or so (which corresponds to the 90th percentile), and no one would get anything in between.

But barring that, I don't see why the Flex is any different than any other LSAT administration. What am I missing?

As an aside: LSAT score averages have increased at top law schools over the past few cycles, but that's because more people have taken the test (resulting in, necessarily, more 170+ scores) and law schools haven't increased classes sizes. It's not because the test is "easier."

I had the same questions as OP. Is the above info factual? If so, the test possibly being "easier" is meaningless.

If I am understanding this correctly, take this hypothetical - everyone BOMBS the test and the 97th percentile only getting 70/101 questions correct. Would that taker still get a 170? Curious as this would clam the nerves a bit of school score medians shooting up 1 point each this year.

Yes. That information was accurate. If you want to get a 170, that means you need to be in the top ~2.5 percent of test takers. For LSAT flex, that will probably mean getting about 68/75 questions correct. But if everybody bombs, a lower raw score could be sufficient to get you into the 170's. This notion that the test is "easier" is meaningless.

If you are worried about medians, LSAC keeps careful records on the number of test takers and applicants in every cycle. See https://www.lsac.org/data-research. So far in the 2019-2020 cycle, we have ~3350 people with a 170+, compared to ~3100 in the 2018-2019 cycle. I don't think that will be enough to move the medians. For perspective, the lull in the 2014-2016 range (when LSAT medians were a point or so lower) had ~2500 or so applicants with a 170+ score. That doesn't mean it can't happen. But it usually takes an awful lot to move the needle one way or the other.