Why the new digital format makes simulating an LSAT score much more difficult

vandbytheriver

A Reddit post on the impact of reading speed on the digital LSAT got me thinking.

Some LSAT preparation services, such as 7 Sage, have begun converting LSAT practice tests into the new digital LSAT format. At first glance, this appears to compensate for the new digital format: people preparing for the LSAT will still be able to take practice tests in the actual format of the LSAT.

The problem is that past LSATS converted to the new digital format may provide an inaccurate assessment of preparation for the LSAT because of research indicating that reading on a screen is about 25% slower than reading on paper. I haven't taken a deep dive into where this 25% figure comes from, but it seems credible given some of the sources that have cited it.

Assuming that the 25% decreased digital reading speed is correct, the average person taking a practice LSAT converted to a digital format will read 25% slower than the original paper curve setting test takers. The decreased reading speed in the digital format will significantly impact how practice tests taken in the digital format compare to the curve composed of paper test takers. Additionally, it is likely that other aspects of the LSAT being on paper rather than digital, such as annotations and underlining, may also result in a curve divergence.

In other words, until the release of an LSAT curved to digital test takers, it has likely become much more difficult, if not impossible, to simulate an actual LSAT score due to the absence of a curve that accounts for the significant differences between paper and digital testing.

Any thoughts?

BinghamtonDave

Interesting points. Have we heard confirmation that lsac will scale the actual digital exams for both the relative difficulty of the questions presented and the idea that people might read slower on a digital screen? If not, if the lsac is not going to account for that variable when scaling the exam, and essentially (some would say unfortunately) say: “deal with it” then previous exams would be an accurate representation, would they not? They would be graded on the same criteria as digital exams.

I think your critique is strongest if we assume that the lsac will have some sort of calculus that accounts for something beyond the test difficulty when scaling the digital exams, hence a difference between the old and the new. I’m not 100% they will build this consideration into their digital scales. I’m open to being swayed by the evidence here.

A few other thoughts: just off the top of my head.
Can those who read slower on digital improve their reading speed through focused practice?

What percentage of the people in the study were smart phone/internet savvy people? How many were not?

BlindReviewer

I really think that this is the reason for the pilot July exam, where they'll collect all the data to see what the differences are between paper and digital test takers?