Checkouts Vs. GPA?

Cindy Harper, Systems Librarian at Colgate University, posted to the IUG list with this notion today:

I’m clearing out a large group of expired student records, and wonder if anyone else has had the same idea that has occurred to me. [Our ILS] keeps track in the patron record of TOTCHKOUTs (total checkouts). At the expiration of the students’ record at the end of their four or so years, this represents a measure that is not perfect, but could distinguish heavy library users from non-users. Of course, it combines book chekouts, video and music checkouts, reserve checkouts, etc. And it misses the effect of electronic sources. I was thinking of trying to get GPA data for these graduates and use an ANCOVA (Analysis of Covariance) to see, once you’ve accounted for the effect of different majors and year-of-graduation effects, if there’s a correlation between library use and GPA?

Has anyone done this type of study? Do you analyze your TOTCHKOUT data in any way?

I’d second her question. Public libraries, I think, do better at correlating their statistics with other metrics in their communities. What do we know about academic libraries?