Topic:
Introduction to Spatial Randomization Tests
OVERVIEW and OBJECTIVES
Many spatial statistics can be expressed as randomization tests. Because of
their simplicity and generality, an understanding of randomization tests
provides insights to the construction and behavior of autocorrelation
statistics such as Geary's c and Moran's I; and to disease cluster tests
such as Cuzick and Edward's test, Mantel's test, the Knox text and many
others. By the end of this presentation you should know what a randomization
test is and how randomization tests are applied to spatial statistics. You
will know the advantages and disadvantages of the randomization approach,
and you will be able to identify the five components of a spatial
statistical test.
OBJECTIVES:
After completing this module you should be able to:
- Identify the five components necessary to construct a test for spatial
pattern;
- Contrast randomization tests with statistical tests based on
distribution theory;
- Construct a simple randomization test for assessing spatial
autocorrelation.
SCENARIOS FOR DISCUSSION
|