Spatial statistics include a variety of robust tools that can be applied in agricultural research.  With spatial statistics agricultural researchers can: (i) estimate treatment effects in spatial experiments and estimate the autocorrelation structure, (ii) predict data at unsampled locations in a spatial domain, and (iii) adjust experimental design to account for location effects.  The latter could, for example, address where to take observations or how to arrange treatments in a spatial experiment. The goal of this presentation is to show how spatial statistics can be used to estimate spatial dependence and model spatial data in agricultural research. We will cover the basic types of spatial analysis: geostatistics, lattice data and spatial points, and how to account for spatial variability in agricultural trials.