by StaceyKoprince Tue Jul 29, 2014 5:27 pm
They don't correspond to particular scoring levels, no - the only way we could figure that out would be if we put these into an adaptive test set-up and tested the problems on thousands of people. (ie, what the actual GMAT test makers did when they first created this problems and used them as experimental questions on their tests. :)
In general, we just tracked the percentage correct over a large number of our students and then divided the problems into these four categories. The labels are accurate in a relative sense - so something labeled medium is, on average, answered correctly at a lower rate than something labeled easier, and so on.
I know you're asking because you're thinking that this will help you know which problems to study - but you shouldn't actually base what you study on the difficulty level of the problems. It's important to study the easier and medium ones, too! :)
Stacey Koprince
Instructor
Director, Content & Curriculum
ManhattanPrep