lsat curve? Forum

[warandpeace]

is there ever one lsat test that has more of a generous curve than others (october, february, etc)?

[LSAT Blog]

December typically has the most generous, June typically has the least generous.

[john1990]

All lsat curves are different and they depend solely on how the takers of the test performed. The LSAC will assign the same score to the same percentile everytime ex(the 95th%=168) so if test takers score less accross the board, you will need less correct answers to get the same score, as in my example where 5% must score a 168+. However, the curve is not based on when you take the test, but how well everyone taking that test performed

[Jeffort]

All lsat curves are different and they depend solely on how the takers of the test performed. The LSAC will assign the same score to the same percentile everytime ex(the 95th%=168) so if test takers score less accross the board, you will need less correct answers to get the same score, as in my example where 5% must score a 168+. However, the curve is not based on when you take the test, but how well everyone taking that test performed

John1990 is wrong, very very wrong.

[bball1997]

All lsat curves are different and they depend solely on how the takers of the test performed. The LSAC will assign the same score to the same percentile everytime ex(the 95th%=168) so if test takers score less accross the board, you will need less correct answers to get the same score, as in my example where 5% must score a 168+. However, the curve is not based on when you take the test, but how well everyone taking that test performed

Dude I'm pretty sure the whole thing is pre-set.

[mille295]

No, from what my Kaplan teachers told me... It really does depend on how everyone does on the test. You can get the same number of questions correct but a different scaled score depending on a lot of other factors, including how others did on the test. It depends on how "logically difficult" a test is. There's no way for them to pre-set it.

Look on the LSAC website. Somewhere on there it explains it.

[vissidarte27]

It is pre-set based on the responses of the people who had each of the sections as experimental sections a couple of years ago. It has nothing to do with how well anyone else in the room scored today.

[incompetentia]

"Curve" is set before the test. Process is similar to a curve, but based on the results from the individual sections when they were experimentals.

[d34d9823]

"Curve" is set before the test. Process is similar to a curve, but based on the results from the individual sections when they were experimentals.

In fact, this is why LSAC refers to the process as "equating", and not as a "curve".

[youarereadingthis]

No, from what my Kaplan teachers told me... It really does depend on how everyone does on the test. You can get the same number of questions correct but a different scaled score depending on a lot of other factors, including how others did on the test. It depends on how "logically difficult" a test is. There's no way for them to pre-set it.

Look on the LSAC website. Somewhere on there it explains it.

Your Kaplan teacher is wrong. The LSAT is not curved. The scale is equated prior to the day of the exam.

[kwais]

curve is pre-set fo sho

[soj]

from what my Kaplan teachers told me

[emptyflare]

I'm curious:

Hypothetically speaking, let's suppose todays test takers did much worse or much better than those who have taken the same sections as experimental sessions before. Does the LSAC revise their initial, given curve in response to this?

[CJJ96812]

EDITED BY TLS MODS

[incompetentia]

I'm curious:

Hypothetically speaking, let's suppose todays test takers did much worse or much better than those who have taken the same sections as experimental sessions before. Does the LSAC revise their initial, given curve in response to this?

Pretty sure the answer is no. This is why the percentiles vary from test to test.


They're pretty damn good at what they do, though. Predictions are probably pretty accurate as a rule

[d34d9823]

I'm curious:

Hypothetically speaking, let's suppose todays test takers did much worse or much better than those who have taken the same sections as experimental sessions before. Does the LSAC revise their initial, given curve in response to this?

Pretty sure the answer is no. This is why the percentiles vary from test to test.


They're pretty damn good at what they do, though. Predictions are probably pretty accurate as a rule

Also, I think they would take this as a sign that the pool of testers was stronger or weaker than when they gave the experimental. There definitely are fluctuations in the strength of the candidates, which is why they use the equating process to ensure that a 175 now is the same as a 175 in October.

[suspicious android]

No, from what my Kaplan teachers told me...

This argument has been done about 500 times on this forum, and amusingly (to me anyway), whenever someone says "My teacher told me that the scale is based on test-day performance" it's a Kaplan teacher who says it. I wonder if they actually tell the teachers that during training, or is that just what happens when you pay less than half what competitors pay and you don't even strictly require instructors to take a real LSAT.

[CJJ96812]

EDITED BY TLS MODS

[mille295]

No, from what my Kaplan teachers told me...

This argument has been done about 500 times on this forum, and amusingly (to me anyway), whenever someone says "My teacher told me that the scale is based on test-day performance" it's a Kaplan teacher who says it. I wonder if they actually tell the teachers that during training, or is that just what happens when you pay less than half what competitors pay and you don't even strictly require instructors to take a real LSAT.

Ok, whether or not my teachers were correct about how the LSAT is curved, excuse me, EQUATED... They definitely taught me a lot. And there's a possibility that I misunderstood... And I'm not sure about other teachers, but mine had all been accepted to T10 schools (I had 3 different teachers).

Without the course, I wouldn't have seen a 15-point increase in my score. Not sure how others feel, but just because the course isn't as expensive doesn't mean it isn't as effective.

[loomstate]

"Curve" is set before the test. Process is similar to a curve, but based on the results from the individual sections when they were experimentals.

incompetentia - is your avatar a self-portrait? if not, can i keep on pretending it is?

[FantasticMrFox]

"Curve" is set before the test. Process is similar to a curve, but based on the results from the individual sections when they were experimentals.

In fact, this is why LSAC refers to the process as "equating", and not as a "curve".

yup this is all explained in lsatblog as well

[d34d9823]

"Curve" is set before the test. Process is similar to a curve, but based on the results from the individual sections when they were experimentals.

incompetentia - is your avatar a self-portrait? if not, can i keep on pretending it is?

http://en.wikipedia.org/wiki/Emilia_Clarke

[loomstate]

[suspicious android]

Ok, whether or not my teachers were correct about how the LSAT is curved, excuse me, EQUATED... They definitely taught me a lot. . . Not sure how others feel, but just because the course isn't as expensive doesn't mean it isn't as effective.

I'm glad you had a good experience, but I wasn't saying that Kaplan is bad because they're cheaper. They're actually not significantly cheaper. They just pay way less than other companies. That's, as you noted, not to say that their product is necessarily inferior. Just an funny thing I had noticed after seeing this debate play out a million times.

[d34d9823]

Don't be embarrassed, man. I'm just giving you google image ammunition.