Applied Spatial Statistics for Public Health Data
While mapped data provide a common ground for discussions between the public, the media, regulatory agencies, and public health researchers, the analysis of spatially referenced data has experienced a phenomenal growth over the last two decades, thanks in part to the development of geographical information systems (GISs). This is the first thorough overview to integrate spatial statistics with data management and the display capabilities of GIS. It describes methods for assessing the likelihood of observed patterns and quantifying the link between exposures and outcomes in spatially correlated data.
This introductory text is designed to serve as both an introduction for the novice and a reference for practitioners in the field
Requires only minimal background in public health and only some knowledge of statistics through multiple regression
Touches upon some advanced topics, such as random effects, hierarchical models and spatial point processes, but does not require prior exposure
Includes lavish use of figures/illustrations throughout the volume as well as analyses of several data sets (in the form of "data breaks")
Exercises based on data analyses reinforce concepts
From the Back Cover
An application-based introduction to the statistical analysis of spatially referenced health data
Sparked by the growing interest in statistical methods for the analysis of spatially referenced data in the field of public health, Applied Spatial Statistics for Public Health Data fills the need for an introductory, application-oriented text on this timely subject. Written for practicing public health researchers as well as graduate students in related fields, the text provides a thorough introduction to basic concepts and methods in applied spatial statistics as well as a detailed treatment of some of the more recent methods in spatial statistics useful for public health studies that have not been previously covered elsewhere.
Assuming minimal knowledge of spatial statistics, the authors provide important statistical approaches for assessing such questions as:
- Are newly occurring cases of a disease "clustered" in space?
- Do the cases cluster around suspected sources of increased risk, such as toxic waste sites or other environmental hazards?
- How do we take monitored pollution concentrations measured at specific locations and interpolate them to locations where no measurements were taken?
- How do we quantify associations between local disease rates and local exposures?
- After reviewing traditional statistical methods used in public health research, the text provides an overview of the basic features of spatial data, illustrates various geographic mapping and visualization tools, and describes the sources of publicly available spatial data that might be useful in public health applications.
About the Author(s)
LANCE A. WALLER, PhD, is an associate professor in the Department of Biostatistics at Emory University in Atlanta, Georgia. He received his PhD in Operations Research in 1992 from Cornell University. Dr. Waller was named Student Government Professor of the Year in 2003 by the Rollins School of Public Health, Emory University, and is a Fellow of the American Statistical Association.
CAROL A. GOTWAY, PhD, is a mathematical statistician for the National Center for Environmental Health and an adjunct associate professor in the Department of Biostatistics at Emory University. She received her PhD in Statistics from Iowa State University in 1989. Dr. Gotway was awarded a Distinguished Achievement Medal by the American Statistical Association’s Section on Statistics and the Environment for her contributions to environmental statistics and is also a Fellow of the American Statistical Association.
Outstanding!, August 12, 2004
This book is incredably well-written. Although I had not read a statistics text in 12 years, this book balanced theoretical development, explanation, and examples in such a way that it was actually easy to read. The organization of the book makes looking up a particular topic very easy. The reference list is extensive. The best thing about the book is the set of graphics and maps to illustrate the mathematics, modeling, and outcomes. This book is a must-have for anyone involved in spatial statistics, disease clustering, epidemiology.