Geographical Data Science (GIS) Applications for Critical Record Identification

• Activity Number: 517 - New Approaches for Correlated and Clustered Data
• Type: Contributed
• Date/Time: Thursday, August 11, 2022 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Turkan Kumbaraci Gardenier*
• Companies: Teka Trends, Inc.
• Keywords: Geographic Indormation Science; Trinomial; Health Service Area; Mortality rates; Record identification; Efficiency

Abstract

Geographic Information Science (GIS) and maps generated by GIS enable display and configuration of strategy for further analysis and hypothesis generation. This paper will provide summaries of correlational analyses of mortality rates, and socioeconomic indicators in the United States compiled at the Health Service Area (HSA) level with aims to derive geographical subsets (Atlas of United States Mortality, 1996). The identification strategy includes the use of Bivariate Trinomial approach developed by the author. A summary of the correlation profile, variables selected, and recommendations for further pursuit will be discussed. The "diagonal tilt" concept will be illustrated in order to provide a roadmap for record identification. Efficiency implicit in the approach will be discussed relative to its iterative, stepwise features.