Comments on: The impact of book suggestions/recommendations?

By: The impact of serendipity (part 2) » "Self-plagiarism is style"

The impact of serendipity (part 2) » "Self-plagiarism is style" — Sun, 05 Apr 2009 14:53:04 +0000

[...] promised I'd dig a bit deeper into the book data, so here [...]

By: Dave Pattern

Dave Pattern — Wed, 01 Apr 2009 21:38:04 +0000

Hi Lizbeth!

I'm wondering if there's any way of using the Amazon Web Services to help locate gaps for a given author?

Alternatively, if circulation data from multiple different libraries could be aggregated together, that would allow you to identify which would be the best of the missing titles to purchase (i.e. out of the missing titles, which are the ones that have the highest circulation at the other libraries).

By: Lizbeth Jones

Lizbeth Jones — Wed, 01 Apr 2009 13:06:49 +0000

I'm curious at to whether data can specifically target authors that are popular and recommended so that the collection of that particular author's work can be deepened. Many librries don't have "all" of a particular author's works, yet don't always fill in missing titles. Having more titles of the popular and recommended authors, might help increase circulation as well.

By: Matthew Phillips

Matthew Phillips — Tue, 31 Mar 2009 13:59:48 +0000

Well the introduction of similar features at Dundee doesn't appear to have had the same dramatic effect. The number of distinct titles borrowed has been declining slowly since 2005 (as we moved systems I do not have any figures for earlier than this). The decline is perhaps slightly less since we introduced the recommendations. Unique titles per user sounds a better approach, but looking at the real-terms book spend (excluding serials) or disposals of material could also be revealing.

By: Dave Pattern

Dave Pattern — Sun, 29 Mar 2009 18:46:13 +0000

I'll see what other data I can pull together, although I might have to rely on the SCONUL stats for some of it. Using the circ data, I should be able to calculate the average number of unique bibs borrowed per user since 2000.

By: Owen Stephens

Owen Stephens — Sun, 29 Mar 2009 15:58:48 +0000

Great stuff. As Joy says, I guess to really understand behaviour you'd need to ask the users the question. With self issue machines the idea of impromptu questionnaires a la Co-op sounds quite interesting.

I wonder if there are some other factors you could plot against the number of bibs in circulation – e.g. average # copies per bib, registered users – this would help see if other factors are varying (and therefor related)

By: Joy Palmer

Joy Palmer — Sun, 29 Mar 2009 15:22:34 +0000

I'd hazard a bet that it has made a difference. Would be interesting to compare this graph to other similar libraries. It would also be nice if you could do some sort of meaningful survey.
(Do you shop at the Co-op ever? I've been impressed by their ways of doing 'impromptu' surveys via the cash card console. Not necessarily do-able, but has me thinking beyond 'Amazon' incentives;-))

By: OPAC Stats Showing the Impact of “People Who Borrowed This, Also Borrowed…” Feature…03.29.09 « The Proverbial Lone Wolf Librarian’s Weblog

Sun, 29 Mar 2009 13:38:27 +0000

[...] Self-Plagarism is Style blog inThe impact of book suggestions/recommendations? has some interesting multi-year statistical results to share on the impact of a "People Who [...]