New England Star Libraries
March 3rd, 2009
Library Journal recently released its list of America's Star Libraries. The rating were based on data collected in 2006, and, on the whole, was pretty interesting.
Jessamyn brought up some good points about the relevancy and awkwardness of it, though beyond her post and a cursory reading of the LJ article, at first I didn't pay much attention (other than to see how my library ranked in the Google spreadsheet version).
But then a coworker of mine pointed something out which got me to take a more in-depth look. She said:
Did anyone else think it was weird that LJ used per capita figures to rate a library and 8 of the 11 commended [Massachusetts] libraries were vacation destinations?
Her point was that per-capita numbers are unduly inflated in vacation spots, as tourists boost the stats when they come in to check email, catch a storytime or author talk on a rainy day, or check out beach-reads on guest cards.
I am a visual person (and like maps), so I plotted the New England star libraries in Google Maps. It turns out that an easy majority (but not all) can be considered vacation spots - near either the ocean or ski areas. I'm not trying to detract from these libraries, or to imply that they're not playing an important role by serving (my) vacationing patrons. It just seems that the data used for these ratings is a bit skewed. So double congratulations to landlocked flatlanders like Dover and Newton - you guys must definitely be doing something right.
The authors did touch on this, and the article's various online components also offer useful information. But if a library wants to improve its rating, the way to do that seems to be to pick up and relocate to either the coast or the mountains.
March 3rd, 2009 at 6:09 pm
Since Internet use per capita plays a part in these ratings, libraries in towns where residents use their home computers instead of library computers are rated lower. Just another bias of the study of which to be aware.
March 3rd, 2009 at 6:17 pm
Good point – also, the number of computers in the library is a large factor. We have 25+ public terminals, whereas a library with 5 or 10 will probably have lower per-capita usage because they can’t have as many simultaneous sessions.
Also: we have lots of laptops accessing our wireless network at any given time, and we don’t have a way to include those numbers in our overall internet usage.
I’m still not trying to poke holes in the study, I just want to identify the limitations of the data so the next report can be more comprehensive and accurate.
March 3rd, 2009 at 7:31 pm
[…] Brian Herzog has been looking at the data for these honors and has identified something questionable regarding […]
March 14th, 2009 at 3:42 pm
Hi there. I’m a public librarian at a 5-star library, and while I’d like to disclaim up front that I know I work at a good library (I moved here deliberately because I knew the library by reputation), I’m not wild about LJ’s star system. It’s impossible to control for the factors they chose, for one thing. Like you pointed out, touristy areas do better (and my town is a major tourist desination). Another thing: length of time on computers makes a difference. Libraries that limit usage to thirty minutes would, theoretically, cycle through twice as many sessions as libraries that limit to sixty minutes.
More generally, I’m uncomfortable with the idea of judging libraries by four impersonal criteria. Program attendance, for instance, does not speak to the quality of the program. On Thursday I offered a program on how to apply for jobs online. Only six people showed. The numbers were disappointing, but I’d like to think it was a good program. I’d also like to think that libraries that make a concerted effort to help the community during rough economic times are better serving their mission than libraries that aren’t doing anything special.
Or– sorry to be long-winded– what about things that you simply can’t measure statistically? People who find me at the reference desk and ask for a good book will get good service. This is because I read a lot, and because my library actively supports and promotes readers advisory. Other libraries, through lack of funding, or indifference, or whatever, don’t make RA a priority.
At any rate, I think the most important measures of a library’s worth are too abstract to reduce to simplistic statistics.