Poll: What will the 2021-2022 application cycle be like?

d.h.isais

Share your thoughts!

d.h.isais

Thank you in advance to everyone who contributes to this poll!

VerdantZephyr

I will point out that it is not possible for a strictly curved test to have higher scores than another with an identical curve regardless of whether the test is longer or shorter with more variability or less. It doesn't even matter if it is easier, it is a curve. The only logical explanation is that there are more people taking more tests. That or they decided to completely destroy the validity of the LSAT for no reason.

VerdantZephyr

Sorry, that may have come off as snarky rather than joking.

d.h.isais

@VerdantZephyr said:
I will point out that it is not possible for a strictly curved test to have higher scores than another with an identical curve regardless of whether the test is longer or shorter with more variability or less. It doesn't even matter if it is easier, it is a curve. The only logical explanation is that there are more people taking more tests. That or they decided to completely destroy the validity of the LSAT for no reason.

Hey Verdant, I actually think you might be mistaken. From what I understand, the test is NOT "curved." Rather, it is "equated" in advance of anyone actually taking a specific test, which is to say that LSAC creates a scale for anticipated scores in order to approximately achieve a bell curve. The key word here is "approximately.* Since the test isn't curved, but equated, it is actually possible to have a test administration with a distribution of scores that does not fit a bell curve.

In other words. If the test were curved, then there would be a distribution of scores resembling a "bell curve" for every test administration. But because it is equated, the distribution of scores will not necessarily resemble an ideal bell curve every administration. In fact, since it is equated, you could have test administrations in which there are a disproportionate number of 170+ scorers. From what I understand, that is what is happening this year.

If I'm wrong about any of this, then please correct me.

VerdantZephyr

The curve is changed for each test. If you listen to the power score podcasts they'll talk about whether the curve is -8 or 9 or 10. However, even if what you say is true, LSAC, as much as it sometimes seems otherwise, is run by very smart people who are very good at math. If the test does have an amount of wiggle room because it is "equated" rather than curved, (this sounds strange to me because it is far easier to curve than equate as you describe it) that amount of difference would be very small and they would adjust it before releasing scores if the results were drammatically off of what a curve would be. They certainly would adjust their system after a given test is there were starkly invalid results rather than continuing to use a bad system for nearly a year. So that said, even if that was untrue, such a flaw still couldn't explain why scores meant to equate to a percentile have as much as doubled on applications. Small errors are conceivable, multiplicative errors are not. What does make far more sense is the number of applicants with lots of time on their hands to study and retake. Millions are out of work and there is always a strong correlation between unemployment/economic downturn and applications to graduate and professional schools. There probably also exist a subset who already had scores from the past that are now deciding to go after all despite postponing. I think there are actually a lot of reasons for increased applications and scores, not all of which are understood, but I certainly don't think that the organization which has created a reliable and scaled test for generations has massively screwed the pooch and accidentally doubled high scores on every test this year. I'll also point out that abnormally high numbers have been occurring since the beginning of the cycle rather than increasing dramatically with the number of Flex administrations. Undoubtedly some of those high numbers occurred before the FLEX started.