4. Nuisance autocorrelation. Most commonly this occurs when the underlying spatial process creates regions of attribute clustering that are much larger than the units of observation chosen by (or available to) the analyst. Figure 1 reveals that the areas of high and low poverty generally are considerably more extensive than is the particular lens (counties) through which we are viewing the process. When units of analysis are much smaller than the regions of high or low attribute values, spatial autocorrelation in the observations is inevitable. As with substantive autocorrelation, nuisance autocorrelation must somehow be recognized and eventually brought into the formal analysis. Anselin (1988, p. 15) differentiates among these types of autocorrelation using the terms ‘‘apparent contagion’’ (spatial heterogeneity) and ‘‘real contagion’’ (spatial dependence).