Alternative Strategies for Mapping ACS Estimates and Error of Estimation

Abstract

As you may know, the American Community Survey (ACS) is the replacement of the census ``Long Form'' and the only source of detailed socioeconomic and demographic data for small areas, such as place and tract levels. This paper is about exploring ways to improve communication of the estimates and reliability of ACS estimates in map making and in our published map products, whether in static ``printable'' form (e.g. a pdf) or in web-based interactive delivery format (html, JavaScript, KML). First, we summarize geo-visualization developments over the past two decades on ways to present uncertain data. Second, we present selected works of others more directly related to the ACS situation---polygons with attributes derived from a continuous monthly sampling activity and updated periodically. Third we present some of our own work at the Cornell Program on Applied Demographics exploring how to communicate simultaneously both the ACS estimates and the margins of error for polygons at the county and sub-county levels of geography.