The iMapLibraries project (http://www.imaplibraries.org) at Florida State University is a one-year grant funded by the Institute of Museum and Library Services (IMLS) in Washington, D.C. The goal of the project is to prototype the development and demonstrate the utility of displaying public libraries in relationship to other geographic data such as U.S. Census data. As part of this IMLS project, we had used a gravity modeling technique to define an initial geographic market area for every outlet in the U.S. The gravity modeling method assigns everyone to their geographically nearest library. This data was already on hand and can be used to generate reports for local libraries nationwide (http://hermes.freac.fsu.edu/freac/imls_libraries/report.html). Please consider the following limitations of the data so that you can assess whether these data are appropriate for your use:
1) The data are based on U.S. 2010 Census and the 2005 – 2010 American Community Survey five-year estimates. In areas of rapid population growth and turnover, the data being presented may be outdated and hence potentially inaccurate.
2) Most of the data used is summarized to the census block group level but poverty data is available only at the census tract level.
3) We estimated a library branch’s geographic market area (i.e., the geographic area from which each branch is drawing most of its users) using a gravity model but this may NOT be an accurate representation of some or even most branches’ geographic market areas. To understand the impact of closing a library branch demands this type of geographic market area estimation — you must start with at least some basic estimate of who each library serves, and where those people live.
Ideally, each library branch should have an estimated geographic market area developed by the library director in consultation with the library staff and relevant stakeholders and that all feel is reasonable for the branch. Even though such an estimate will not be 100% accurate, the population characteristics in this estimated market area forms the basis for library branch statistics such as market penetration, needed (or non-needed) library services for the nearby population, and, in this case, the potential impact of closing a branch on its users. Libraries often do not want to estimate population served by a particular branch because they may have policies in which “all are served” who walk through the door.” But this types of planning data is critical to assessing the impact of closing or opening a new location for travel-to type services such as public libraries and retail stores.
Because of the lack of a widely-accepted geographic market area for each Miami-Dade County library branch, we supplemented the gravity model we used for our nationwide map. We also defined a geographic market area for each Miami-Dade County library branch using both a one- and a two-mile radius around each branch and calculating the socioeconomic and demographic characteristics of the population within those additional market areas as well.
4) The demographic and socioeconomic data associated with a block group or census tract may be assigned to more than one library branch’s geographic market area. For instance, if a block group fell within a two-mile radius of three library branches, that block group’s data was used in the calculations for all three library branches. The way to avoid this would have been if there was an estimated geographic market area for each library branch developed by the Miami-Dade library system prior to this analysis.
5) We calculated average median statistics (e.g., average median per capita income) for a library’s geographic market area by summing up each of the median value for each block group within a library branch’s geographic market area and then dividing that summed value by the total number of block groups in the geographic market area. But remember that this calculated average value of a median statistic probably does not represent the actual median value for the entire market area.