Cousins: The Bookworm and Wikignome

As we all know, the best you can hope of a meeting is not a conclusion, but a chuckle at a statistical oddity. When OCLC’s Top Library Loans List came out, such a positive meeting was had. Upon glancing the pulp fiction (see chart below) I wondered if Wikipedia editors were also driven by such trivia? I turned to Python, R, and article edit histories to find out.

The top 10 list is such:

  1. Hunger Games by Suzanne Collins
  2. Catching Fire by Suzanne Collins
  3. Mockingjay by Suzanne Collins
  4. Fifty Shades of Grey by E.L. James
  5. Game of Thrones by George R. R. Martin
  6. The Help by Kathryn Stockett
  7. Thinking, Fast and Slow by Daniel Kahneman
  8. Steve Jobs by Walter Isaacson
  9. Quiet: The Power of Introverts in a World That Can’t Stop Thinking by Susan Cain
  10. Dance with Dragons by George R.R. Martin

Now let’s take a tour through Wikipedia’s history for a feeling for the editors affinities towards these monographs:

We can tell that there isn’t a lot of similarity between the novels, except that they’ve all experience small peaks within the last year or so. That isn’t surprising, because the list in question is for the most requested inter-library loans for the period of a year starting July 11th 2011. So let’s take a look at how actively edited these books were in that time frame.

Besides the fact that the relationship here looks a bit exponential, as we’d expect of crowdsourced material, there is another curious correlation afoot. The ordering of the monographs by edits, is remarkably similar to the ranking by loan-requests. Keeping the by-edits ordering, then charting the loan positions we get something reassuringly linear.

In fact the the Top 6 are exactly predicted. If you were going use  quick-sort inversion counting analysis to compare the closeness of two lists, I believe you get the low count of 2 (correct me if I’m wrong). This indicates that there is a possible correlation between book demand in the library and wikipedia editor interest online. So librarians take note, when deciding on your stock, pre-empt the rush and look to Wikipedia – the Wikignomes have a psychic connection with the bookworms.

Not confusingly yours,

Max Klein, Wikipedian in Residence

P.S. The code to look at and graph Wikipedia articles is a small project I’ve open sourced, and is available on github. I’ve also built in the functionality to pull stats from a Wikipedia category, which allows for such fun as, looking at the entire edit histories of all the Pulitzer Prize Winners.