In this section, we'll learn how to test spatial autocorrelation. In this test, we examine whether the assumption of the independence of observations from one another is true. In a normal distribution, we assume that observations are independent of each other, which is likely not to be true for spatial data, according to the first law of geography:
Everything is related to everything else, but near things are more related than distant things.
As such, it is important to test for spatial autocorrelation when dealing with spatial data. Unlike with point pattern data, we are considering data that is picked or the location of which is fixed by the observer (for example, survey data with location information).