Authors: David Wong*, George Mason University
Topics: Cartography, Medical and Health Geography, Spatial Analysis & Modeling
Keywords: SEER data, class separability classification method, data reliability
Session Type: Paper
Start / End Time: 9:55 AM / 11:35 AM
Room: Marshall South, Marriott, Mezzanine Level
Presentation File: No File Uploaded
When mapping statistical estimates, including many health statistics, cartographic designs and practices prior to this decade have not considered the reliability of estimates. They either implicitly treat the estimates as highly reliable without error or simply ignore the quality information of estimates. Map patterns revealed by such maps can be erroneous or misleading, as statistically similar estimates may be assigned to different classes while statistically different estimates are signed to the same class. The class separability classification in 2015 (by Sun, Wong, Kronenfeld 2015) was the first serious attempt to incorporate estimate reliability information in the determination of class break values in choropleth mapping. Associated with this classification method is a new legend design. The legend includes the likelihood that values on two sides of a class break value are statistically different. Based on this new classification method, a choropleth mapping tool was developed. The tool was customized to create choropleth maps using data downloaded from the Surveillance, Epidemiology, and End Results Program (SEER) provided by the National Cancer Institute (NCI). In this paper, we will illustrate the mapping method, the tool, and provide an update on the recent enhancements of the program. The mapping tool can handle many survey data sets, including the American Community Survey data (ACS), as long as the data include either the margin of error or the standard error for each estimate.