circulation data visualized

There’s some interesting data hiding inside of our Integrated Library Systems! Here are four days worth of search terms, about 17,000, that were entered into DCPL’s OPAC earlier in the month.

I’m convinced that there’s more useful data inside of ILSs than generally gets used, and that there are more appealing way to use it than bar graphs. I’m not even close to being an expert about data visualization, but my little video above tells more of a story than a number in a spreadsheet. I used Wordle, ScreenFlow and some zooming tricks to make it so it wasn’t difficult but having the ILS spit out data an and automating the process would be great.

And of course, having to sit here and be passive while watching this is a drag. It would be great to be able to interact with this data and do stuff like compare it to items checked out or where the searches came from. We’ll see what comes of this experimentation.

Here’s the first video I made today which presents pretty much the same data in a slightly different way. I don’t like it quite as much, but I should include it to prove to you that I know about more colors than my beloved grey.

The most interesting stuff going on with circulation data exposure I’ve seen recently is from Dave Pattern who just helped the University of Huddersfield release 13 (!) years of circulation data under a CC0 (again !) protocol. That would make one heck of a word cloud. He spoke about it during the last Talk with Talis that I have yet to listen to but will soon.

19 thoughts on “circulation data visualized”

  1. Hey Aaron — great to see you’re playing with usage data :-)

    You can find some interesting stuff in OPAC searches. For example, we discovered that there are a sizeable number of people searching for “renew” on our OPAC:

    …presumably someone told them they could use the OPAC to renew their books?

    I’ll have to have another play around with our OPAC keyword data, as we’ve now got about 2 and half years worth of searches.

    I think it would be fun to compare clouds from different libraries. How much commonality would there be?

    Have you thought about creating some posters using the Wordle cloud?

    One really good use for the keyword data is to generate search suggestions. For example, if you search for “business” you’ll probably get too many results, so you could run a check through all the previous keyword searches to see what the words most commonly paired with “business” are and suggest those to the user (e.g. “business law”, “small business”, “business studies”, etc).

  2. Hi. I absolutely love this data cloud. I agree with the comment just before this, that it would be interesting to see libraries compared. It would be a fascinating useful thing to see our patron’s ‘thoughts’ captured.

  3. @rochelle – “to see our patron’s ‘thoughts’ captured.” Agreed! I wonder if many libraries would rather live in ignorance tho, if they had the choice. :)

  4. @scot – I don’t know whether to thank you or be mad at you. That seems really really fun and something I’d play with for a long time, but they’re in beta so my registration will probably take a few days to process! Gah! Cheers :)

  5. That’s awesome, Aaron!

    I have to say I agree that we should be able to accomplish much more with our usage data. Our electronic resource stats have been on the decline, and I’m convinced that there’s a new way we can examine these numbers to find out just what makes our services “work” for the people who use them regularly.

    It’s not like I’m Bill James or anything, which makes the work all the more difficult. But I’m hoping I can stumble onto something. Eventually.

  6. @kevin – yes. i think when we have some things that we’re really proud of to advertise, like our forthcoming website, we’ll use some stuff like this to do it. also, these things probably have some value on their own, so in the mean time we might include a gallery of images (once we have enough images to make a gallery). the stuff i really want to do requires knowledge of adobe flex which i find quite difficult. we’ll see!

  7. Dood. I got an invite in just a few days. You just have to know how to sell it. Say you are an “information professional” and emphasize that you “reach thousands of IP’s daily”. It works.

    If it doesn’t, email me and you can use my beta account. It’s way too fun.

  8. Our usage data (#s and search terms) are a gold mine! Kudos for looking at some of it and trying to make it understandable.

    My favorite view into search data means so much more work… but I also like to look at *failed* OPAC / database searches. the ones with only 1 count, like my favorite in a database search engine “Hamlet insane.” (when the right search term would be “MLA International Bibliography.” Stuff like that gives you an idea of cognitive mismatches between what the search WILL do and what they THINK it will do – like “renew” – that you wouldn’t otherwise know about.

  9. Nice! This looks to me like a very cool take on a similar idea I talked about in the podcast about using library data as a recruitment tool for universities. Here, with the videos animating what people are searching for, you’ve done a great thing by making libraries and what people do in them lots more interesting and dynamic. I hope it gets lots of people through your doors!

  10. Hey Aaron!! This is fantastic. I have inherited a lot of ILS data and am generating my own daily it seems. This is inspiration to see what if I can uncover what is lurking behind my own III catalog. I’ve been so curious about how our academic patrons are finding (or not) what they need.

Leave a Reply