first thoughts on OCLC/Ohiolink study

OCLC and Ohiolink collaborated on a major study. Here's the report:

http://bit.ly/ohiolinkreport

I get excited when I see a big study like this done, because I think library staff are way too certain that they know what's going on in the library (in terms of collection use, patron wants/needs, etc). Better to try to find out a few facts, rather than compile a long list of things that we "know" (correct or otherwise), especially if you have to answer to a larger institution or community about what value they get in return for the money you're spending.

First thing I wanted to mention is on page 10:

"This study is limited to books, including manuscripts, as these materials typically circulate and circulation is the most significant quantifiable element in evaluating book collections."

This is a lame circular argument, which doesn't answer the question that came to my mind, "Why limit to books and manuscripts?" It is asserted that books typically circulate (true) and that circulation is the most significant quantifiable element in evaluating book collections (true, or at least arguably so). You're telling us why you chose the method you did to evaluate the things you evaluated. You're not telling us why you didn't evaluate other things. It would be worthwhile to explicitly say "we couldn't get decent data on non-book things," or "we don't see a big role for electronic resources in consortial collection building," or "we didn't have enough money and time to slog through all the various usage statistics available for non-book formats and try to normalize them." Ultimately, in 2011, I don't think librarians should be producing a huge report like this and ignoring electronic and other non-book formats. It makes me think of a big ship with iceberg-spotters on one side of the boat, yelling out that we're all clear on that side, and nobody looking the other way. (I don't want to be too negative here, but I see this
as a trap that we in libraries have fallen into before.)

Ok, so on to page 11.

"OCLC's work-set algorithm (Hickey and Toves 2005) was employed to identify works permitting FRBR (Functional Requirements for Bibliographic Records) analysis at the work, manifestation, and item levels (IFLA 2009, 17- 24)."

Are we excluding works that don't permit FRBR analysis? (Further down the page, we're explicitly excluding works that don't have standard numbers.)

Pg. 22 - they mention North American Title Count, which I'd never heard of before. Interesting stuff.

Anyway, on reaching the bottom of the report, it seems like all the most interesting stuff is in the data itself (no surprise there). I'm downloading it now, and I suspect I'll have more to say once I see it.