Operationally, defining literacy and literate practices was one of the most difficult aspects of completing this study. Thus, it is important to clarify, but not defend, the operational definitions. There may be better definitions of literacy, and if so, those factors will be considered and may be added to future analyses. The variables measured here and combined to form the literacy rankings in this study include thirteen separate variables grouped to form five factors of literacy.
Three variables are combined to form the factor Booksellers. Members of the American Booksellers Association (2002), which is an organization that represents independently owned bookstores with storefront locations, were indexed by city. Those member booksellers were then divided by the city population in order to calculate a ratio of booksellers to population. Yellow Pages (2002) listings were used to form two other variables. The number of retail booksellers and rare and used booksellers were each indexed. They were then divided by city population to calculate ratios of booksellers to population.
Library Support, Holdings, and Utilization
Public school library support was assessed by indexing the number of library/school media specialists and the number of public school students in the school district that serves each city as reported in the U.S. Census 2000. A ratio of librarians to students was calculated. The total circulation, volumes held, and branches operated for the public libraries that serve the cities in the study and reported in the American Library Directory (2002) were indexed. These numbers were then divided by the city population in order to calculate ratios of library services and resources available to the population.
Education Attainment was indexed with two variables. The percentage of the adult (over 23 years of age) population having a high school diploma or greater as reported in the United States Census 2000 was indexed. The percentage of the adult population having a bachelor’s degree was also indexed.
The number of periodicals published in a city was indexed through analysis of the National Directory of Magazines (2002) database. Because of the high number of very small single publication publishers, it was deemed more appropriate to utilize the list of multi-periodical publishers of which there are approximately six hundred in the U.S. The number of multi-periodical publishers located in a city was indexed to form one variable. A ratio of those publishers to the population of the city was also calculated to form another variable.
Newspaper circulation was indexed as two variables. Newspaper circulation for the Sunday paper as reported in the Audit Bureau of Circulation (2003) official circulation averages for the six months ending 3/31/2003 was indexed. This number was then divided by population of the city to calculate a ratio of circulation population. The weekday circulation for the weekday variable, which included the Monday edition, was indexed and divided by the population to calculate a ratio of circulation to population.