ABSTRACT
The spatial analysis of single diseases is often sufficient. However, in some applications there is a need to consider different scales of aggregation within an analysis. One such situation arises when it is of interest to consider a relationship at different aggregation levels. The modifiable areal unit problem can be considered to include a variety of special cases. One of these is ecological bias where the issue is whether inference can be made at a lower level of aggregation (individual level usually) from aggregate data. Multilevel modeling often addresses the issue of multiple levels within an analysis and these can include spatially-aggregated covariates. The classic situation where ecological bias arises is an example of this scaling down: trying to make inference at the individual level from aggregate level analysis. The county set is a unique subdivision of the district set, i.e. each county falls uniquely within one public health district.
