Topic:
Exploratory Spatial Data Analysis
OVERVIEW and OBJECTIVES
Luc Anselin defined exploratory spatial data analysis (ESDA) as "techniques to describe and visualize spatial distributions, identify atypical locations (spatial outliers), discover patterns of spatial association (spatial clusters) and suggest different spatial regimes and other forms of spatial non-stationarity...." We are particularly concerned with the use of ESDA for quantifying spatial patterns in (1) health outcomes and in (2) the causes of disease. This lectures has three major parts:
- Objectives and goals of ESDA
- The role of ESDA in hypothesis generation and inference
- Multiple testing
OBJECTIVES:
Those who successfully complete the module should know
- What ESDA is, its goals and problems in the analysis of disease data.
- The relationships between spatial analysis and the scientific method.
- What the "Gee Whiz" effect is, and why it should be avoided.
- What multiple testing is, and how to adjust for it when assessing whether or not a spatial pattern is a statistically unusual outcome.
SCENARIOS FOR DISCUSSION
|