| 223 to 1 Somewhere between the odds that I will have a child prodigy (250:1), and being audited by the IRS (100:1), is the chance that I might meet a fellow Republican librarian. Knowing this, lightning strikes and meteor showers seemed too scary to investigate. 223 to 1, according to the latest Library Journal, (10/04) is the ratio of librarians that have contributed to John Kerry's campaign as opposed to Dubya. Again 223 to 1. Hardly a ringing endorsement for a profession hell bent on diversity. One can only imagine what this number may have been without the efforts of the American Library Association's Office of Diversity. I've always contended that librarians bring their own political baggage to work. Some more than others. I have also contended that collection bias is real, not imagined, and is just as insidious, or innocuous, as knee jerk reactionary censorship. The results are identical as well. Now that Library Journal has validated what many in our profession refuse to acknowledge, I thought I would do a little research to learn how well my liberal colleagues are able to separate their political leanings from professional responsibilities. A few notes on methodology. This is not intended to be a scientific study, rather a glimpse with some truth into how we collectively purchase popular books with political agendas. My sample was random, taking 20 liberally biased and 20 conservatively biased books published after January 2000 and that had at one time been ranked in the top 150 in sales. (No special reason why I stopped at 2000). The holdings numbers for WorldCat represent book format only. Because objective book sale numbers are nearly impossible to get, I chose USA Today's Best Selling Books Database of the top 150 best sellers. Ideally a service like Nielsen Bookscan would be the best source for sales however I wasn't willing to pay for this data. Republicans are also cheap. USA Today has a fairly objective reputation and its breadth made it an attractive resource to use. Titles were grouped by "liberal" and "conservative". (I doubt anyone reading this will take issue with my designations here) Each title has its publication date, OCLC WorldCat holdings number, weeks on the USA Today top 150 list, and highest rank on the list. To quantify my results, I established two formulas to arrive at a score or quotient.
number of OCLC holding libraries / weeks on the top 150 example: A title held by 1000 libraries that spent 10 weeks on the top 150 list would have a score of 100. number of OCLC holding libraries / (151 - highest rank) example: A title held by 1000 libraries that had a high ranking of 10 would have a score of 7.
Last note. There are no benchmarks with respect to scores. The intent here is to detect any trend that hints at collection bias. That said, we can assume that titles with lower scores are more likely to be underrepresented in WorldCat libraries (around 9000 total) with respect to sales and likely examples of collection bias. All data used was collected in September 2004.
Folks, I leave it for to you to decide whether collection bias is an illusion or real. As for me, I've never been confused for a magician. |
