Monday, August 24, 2009

LSAT Statistics, Week Two - What Happens When People Retake the LSAT?

Welcome to the second week where Zen of 180 examines LSAT statistics to answer common questions people have while prepping.

Tomorrow we'll get into the question of whether you, specifically, should retake an LSAT, but today
we'll be offering an in-depth answer to how the average re-taker's scores change the second, or third, time they take an LSAT.

Some information from LSAC:
Data shows that scores for repeat test takers often rise slightly. Most people take the test only once; last year 70.2% of the total number of test takers took the LSAT just one time; 24.5% took the test twice; and approximately 5.3% took the LSAT more than twice.
While that information is somewhat helpful, it's really not an accurate glimpse of the whole picture. When we disaggregated the data into how each person performed on the retake by their initial scale score, some illuminating trends developed.

Please note that in this graph we have excluded the repeat test takers who made the same scale score or only marginal improvements. We chose each standard deviation's cut-off point as the definition for "significant gains" because of the stress caused by preparing for the LSAT: if someone decides to retake the test, they need to have a solid return on their investment.

For each scale score along the x-axis, if the blue "significant gains" bar is visible above the red "harmed" bar, then more people scored in a higher standard deviation than scored lower on the retake.
Conversely, if the blue bar is hidden by the red, then more people with that scale score actually lost points on the LSAT retake.

What this graph clearly shows is that people who initially score below the cutoff of one of the LSAT scale score's standard deviations at ~158 or ~170 have the most likelihood of benefiting from retaking the LSAT. While this could be an artifact of our
definition of "significant gains" as moving into a new standard deviation bracket--that is, they only have to improve a few points to be counted--it could also be because those test takers knew how much they had to gain by improving their score only a few points. Regardless, the trend is striking and has held across modern LSATs.

Tomorrow we'll discuss why you should ignore the implications of this data when you have a solid plan of attacking your weaknesses on the LSAT during "re-studying," using my example of moving from a 172--a dead zone according to the overall LSAC data--to a 180. Sometimes it's worth it to buck the trend if you can get your Zen on.