to the top

3 curiosities regarding PAST LSATs and score variance

lathlee
LSAT Destroyer
 
Posts: 523
Joined: Fri Apr 01, 2016 1:53 pm
Points: 505

Dear Dave and other experts of LSAT.

as you guys said many times before and Now I see it as well, in the past LSATs , generally more ancient, the LSAT test Questions in LSAT individually and overall WERE a lot more difficult; there can be a strong positive relationship made of as in more ancient LSAT Prep is more difficult starting from LSAT Prep 45-50ish.

As I have been doing all Past Prep LSAT and doing the difficult questions in the past, I became curious; I am pretty confident that some of these difficult LR Qs LG Games and RC passage, more than 50% test takers would not have solved correctly within reasonable time limit; even Powerscore Stat supports this phenom. Was there bigger variance in Percentiles in peoples scores than this generation (Starting from Prep test 60ish, would be fair definition) as in people who got in 180 vs 170 these days are usually 2-6 Incorrect Question answered difference These days. but in the past, let's say with these really difficult LSAT questions, in the era of prep test 10-20, the gap was wider as in average of 10-20 questions incorrectly answered between the test takers who scored 180 vs 170?

TWO, why do you guys think LSAT questions overall level has been gotten easier as the years passed by?

Three , do you think whatever LSAC intended to do making Questions generally more easier, has resulted a success regarding whatever the motive that LSAC had when they decided to make LSAT Questions generally easier than Prior versions?
Francis O'Rourke
PowerScore Staff
PowerScore Staff
 
Posts: 460
Joined: Fri Mar 10, 2017 2:44 pm
Points: 455

Hi Lathlee,

Can you tell me where you are getting those numbers from? The number of questions needed to achieve a 170 or a 180 has not changed significantly in recent years. You are still likely to need somewhere between 88 and 91 questions correct to get a 170 and between 98 and 100 questions to get a 180.

More specifically, the average number of required correct answers on tests 11 to 20 (June 1994 to October 1996) needed to get a 170 and a 180, respectively were 89.1 and 99.

From tests 70 to 83 (October 2013 to December 2017), the average number of correct answers required to get a 170 and a 180, respectively were 89.35 and 99.

You can view the data for June 2005 to present on the following page: https://www.powerscore.com/lsat/help/co ... rgeted.cfm

The percentile of students achieving higher scores has increased slightly over the past decade. This however may be explained best by the amount of preparation test-takers today do before taking the exam. As for the difficulty of the individual questions, many people feel that the Reading Comprehension has become slightly harder over the years, but this is debatable. Jonathan Evans has provided a good answer to this topic, which you can read on the following page: viewtopic.php?t=14356

Let me know if this helps! :-D
lathlee
LSAT Destroyer
 
Posts: 523
Joined: Fri Apr 01, 2016 1:53 pm
Points: 505

Hi. thx for the answer. It will take sometime to answer consdiering to quote some of the experts.
lathlee
LSAT Destroyer
 
Posts: 523
Joined: Fri Apr 01, 2016 1:53 pm
Points: 505

Hi. the reasons I said the past prep test area prior to 56 was more difficult and suspected more wide gap between 160-169 vs 170-180 was these 3 reasons:

Reason 1: I kept checking the Powerscore's student's statistics of past prep performances which is available in the web and especially prep 1 's LR (most of the questions getting correct percentage are less than 50%), the percentage of getting question correctly tends to be lower than the prep tests after prep test series 56. I Consider most of the students in Powerscore forum tends to be somewhat above average considering so many of Powerscore LSAT students always share how they increased the score after studying Powerscore bibles AND the reputation among LSAT prep students that so many say that they experienced score increase after studying Powerscore Bible.

The reason number 2 is according to Powerscore's publication of hardest questions of all three sections, they are heavily located prior to prep test 56 which has the statistics according to the Powerscore stat, usually less than 25% getting correct first time.

The reason number 3 is as I was trying to collect the difficult questions in the past preps to prepare myself for any out of blue situation in LSAT in the best manner possible, I saw many more frequent numbers of less than 30% correctly answered questions in each individual past prep tests prior prep tests 56.

There are the reasons which made me think that maybe past test takers prior to 56, there was a more significant gap between correctly answered question gaps of 170~180 vs 160~169 than the generation after prep test 55.
Francis O'Rourke
PowerScore Staff
PowerScore Staff
 
Posts: 460
Joined: Fri Mar 10, 2017 2:44 pm
Points: 455

Hi Lathlee,

Powerscore's student's statistics of past prep performances which is available in the web and especially prep 1 's LR
Can you link to this? I believe that I know what you are referring to, but I want to be sure before I respond to your first point.


As for points two and three, first let me know which publication you are referring to. There may be some articles that we published in 2011 that you can still read online, which would not take the tests 60+ into account.

It may still be the case that the most difficult questions come from a narrow band of tests. If this were true, we would need to look at the relative difficulty of the other questions on this test before assessing the difficulty of these exams as a whole.

For example, you may see a test with the hardest Reading passage or Logic Game ever given on it. If however the three other Readings or Logic Games were all easier than average, the overall difficulty of the test may even out. I would much rather have a Logical Reasoning section with one question that had s 10% correct response rate on it (in reality, I do not believe the LSAC would ever include this sort of question), so long as every other question on that section had a correct response rate of greater than 50%.