Estimating the number of cases of podoconiosis in Ethiopia using geostatistical methods

where d(xi) is a vector of explanatory variables with associated vectors of regression coefficients β. A plot of the logit-transformed prevalence against elevation, shows a positive association up to about 2000 meters and negative thereafter (Figure 2). For this reason elevation was included in the model using a quadratic polynomial. The spatial random effects S(x) can be interpreted as the cumulative effect of unmeasured risk factors for podoconiosis. Finally, the unstructured random effects Z(x) represents extra-binomial variation within villages.