Understanding the null hypothesis

All the pattern analysis tools that we examine in this chapter work on the premise that our features or the values associated with those features are randomly distributed. This is known as Complete Spatial Randomness (CSR). This is the null hypothesis used with all the ArcGIS spatial statistics tools.

The pattern analysis tools return z-scores and p-values. These scores tell us if we can reject the null hypothesis of CSR. If we're able to reject the null hypothesis, then we can say that our data is either clustered or dispersed in a statistically significant pattern, and this is an indicator of some sort of significant underlying process at work that has caused this pattern.