D9b Mapping with Small Numbers: How to Protect Security and Confidentiality of GIS Data

Thursday, March 11, 2010: 8:50 AM
Dogwood A (M1) (Omni Hotel)
Kimberlee Elmore, PhD and Donna J. Helms, MPH, CPH, DHAP, CDC, Atlanta, GA

Confidentiality of data linked with spatial information has unique issues, as unintentional breaches can occur even without releasing sensitive data (such as names). Information mapped using GIS must be protected to the same standards as when the same data are presented in tables, reports, or other formats. It is also important to consider the statistical accuracy and precision of rates calculated for small geographic areas.

Several methods of geomasking may be used: data aggregation; data suppression; random perturbation; use rates rather than counts; create a kernel density surface (e.g., smoothing); and, remove contextual information. These methods will be presented.