Ranking slightly below my dislike of headlines like this one is lazy reporting off some press release touting a simplistic proposition thinly supported by pseudo-statistical data.
Yes, I'm talking about Amazon's annual list of The Most Well-Read Cities in America.
All manner of media uncritically reproduce this crap, from Time to local news outlets in featured cities to media who should know better (Galleycat, I'm looking at you). None I saw were as careful as the Christian Science Monitor to add a disclaimer that the results were based on Amazon sales alone.
Somewhere, Amazon's PR flacks are high-fiving.
In the marketing biz, this is known as "earned media"—free advertising which has about as much a relationship to news as does paid advertising. In this case, it helps align the Amazon brand with the notion of being well-read as well as positioning it as an authoritative source on U.S. book sales.
Here are some reasons not to give much weight to the ranking.
1. It only counts Amazon sales. Overall sales data is tough to come by in the book business for a variety of reasons but we will assume Amazon has an accurate picture of its own sales. The question is whether Amazon sales are a good proxy for total book sales in specific locales. We don't know, but based on other research cited below, I'm going to say maybe not.
2. It's not about reading. The Amazon list presumes purchased reading material is read. I can attest that most of the volumes in my library have not been read, and I don't think I'm that unusual. It also neglects to count books from libraries and, based on its sketchy description, used books purchased via Amazon affiliates.
3. It's weighted toward smaller cities. Of the top 20 cities in Amazon sales per capita, only six have populations of 250,000 or more. (The Amazon list includes cities of 100,000 or more.) Four others on the list are suburbs of or adjoin larger cities. Do readers in smaller cities rely more on Amazon? If so, is it because their local access to books is not as great as rich as in larger cities? Without more data about book-buying by city, there's no way to tell if those smaller cities are the list are actually impoverished relative to cities with a more robust literary ecosystem.
[A side note: Central Connecticut State University does an annual study it calls America's Most Literate Cities. Among the factors it includes in its ranking—booksellers and libraries per capita. Unfortunately, CCSU doesn't include cities with less than 250,000 population, so it's not possible to equate its findings with Amazon's.
But what the hell. I've already blown a morning when I should've been working on my next book. Let's give it a shot.]
4. Best "Amazon cities" are really strong bookstore cities. All six of the 250,000+ cities on the Amazon list are also among the top 20 cities in booksellers per capita, as ranked by the CCSU study. Put another way, Amazon had no "well-read" major city on its list that was not already a top bookstore city.
5. Strong bookstores and libraries go together. Twelve of the top bookseller cities on the CCSU list also showed up on the libraries list. Seattle, Cincinnati, St. Louis and Pittsburgh made all three lists. Miami and Atlanta among the "Amazon cities" failed to rank as top library cities.
6. The numbers don't tell the whole tale. Who would leave New York City and Chicago off the list of great book towns? Not me. But neither city appears in any of the lists, so per-capita counts can't tell us everything.
Oh yeah, and five cities on the Amazon list also rank in the top 20 metro areas most dangerous for pedestrians. Maybe folks order online because they're afraid to go out.
I compiled the three top 20s in the table below, using colors to highlight cities that appear on more than one list. This helped my analysis. If you see some other patterns, please leave a comment.