Spatial autocorrelation is the measure of what?

• A. The average value of a variable across space.
• B. The randomness of a spatial distribution.
• C. The degree to which nearby observations are similar; positives indicate clustering, negatives dispersion; it matters for statistical inference and modeling.
• D. A method for creating continuous surfaces from sample data.

Answer: (C)

Spatial autocorrelation describes how much nearby observations resemble each other. When values near one another are similar, there is positive spatial autocorrelation and you see clustering of similar values; when nearby values differ, there is negative spatial autocorrelation or dispersion. This concept matters for statistics and modeling because many methods assume independence among observations, and spatial autocorrelation can bias results if not accounted for. It’s not just the overall average across space, nor is it simply about randomness in the pattern, and it isn’t a method for creating continuous surfaces from samples.