Speakers: Michael Levine-Clark & Christopher C. Brown
In Dec 2008, they added all the print and ebooks for university press publisher A, and the duplication is primarily in the frontlist. They did the same for an STM publisher B in Jan 2009.
The have gathered circ data that is compiled annually. Comparing ebooks and print books is like comparing apples and oranges. pBook checkouts are for an extended period of time, and we have no way of knowing how many times they view a chapter or copy a page, the measures of ebook uses.
What a cataloger thinks a title is and what a vendor thinks a title is are two different things. How do you merge use and circ data if the comparison points may vary? Solution was to create an ISBN9, by stripping away the first three numbers of ISBN13 and the last number of the ISBN10, which worked pretty well.
Publisher A had about a third of the ebook titles used, but publisher B had only about 2% used. For print, two thirds of Publisher A titles were used and a little over a third of Publisher B were used.
The two most heavily used ebooks from the UP were probably used for a course. The print books were only checked out a few times, but there were thousands of uses for the e. For publisher B, they were used less but with no print circ. For the top two print, the UP ebook was used some, but not even as many as the checkouts, and for the STM print books, the e wasn’t touched. Overall, there was a high rate of use for both formats of a single title, I think (need to study the slides a little more).
They saw increased checkouts of print books over the time period, but it is inconclusive and could be related to the volume of titles purchased. There isn’t a clear impact of e on p or p on e use, but there does seem to be a connection, since when both formats are used, the rate is higher.
Might there be differences by subject or date? What sort of measure of time in a book can we look at? How does discovery play in?
Questions/Comments:
Date & time of usage rather than one year might tell more of a story.
Agree about the discoverability challenge, and have encouraged them to put in chapter level data in their catalog records to create their own discovery with the MARC record. Full text searching in eBrary is great, but get it in the catalog along with the pbooks.
Did you consider the format of the ebook? Some publishers give PDF chapter downloads, which may account for lower use of Publisher B.
What about ILLs? They get included in the circ stats and aren’t separated.
How much of Chris’s time was spent on this? Hard to tell. Was ongoing over time as other things took priority.
How will this affect your collection development practices moving forward? Trying to give users a choice of format.