Linear Models for Spatial Data

When data is measured at multiple spatial locations, it often follows that measurements that are close together are more similar than measurements that are further apart.  This puts us at risk of violating the linear model assumption of independent observations.  Researchers can try to detect violation of independence via visual inspection of maps of residuals and through the use of variograms.  If the variogram indicates that there is spatial autocorrelation, we can address this by adding a spatial correlation structure to the model.   This workshop covers tools to detect spatial autocorrelation in your models and ways to address it by adding spatial correlation structures to standard regression models, with worked examples in R.  This workshop assumes familiarity with linear mixed models.