0

I am new to geographical statistics so please pardon this question if it is a duplicate. I would like to test if points that are similar are more or less likely to be near each other. I am latitude and longitude data for several alligator nests. I need to know if a singly sired nest is more likely to be near another singly sired nest or not. What would be the most appropriate way to formally test this?

  • "if a singly sired nest is more likely to be near another singly sired nest or not". Are you interested only in whether the nearest neighbor is also singly sired...or in the median/mean distance from singly sired nests to different nest types (perhaps with some maximum search radius)...and is this all with straight line distance or with corridor distance? – Tom Jun 28 '18 at 18:13
  • I'm only interested in whether the nearest neighbor is also singly sired right now and this would be with straight line distance. – Josh Jun 29 '18 at 14:22
  • Try the R code from this post, it will return Ripley's-K (Besag's-L) statistic, showing spatial relationships at various distance lags. Even though the question appears to be quite different, the presented answer will most certainly address your problem. However, I would note that you really should transform your data into a distance-based projected system. gis.stackexchange.com/questions/42427/… – Jeffrey Evans Sep 6 '18 at 22:25
0

I recommend reading over ESRI's documentation on Patterns, Clusters, and Spatial Relationships. Even if you don't have access to ArcGIS, this is a clear, concise, and surprisingly thorough account of the available analyses.

The solution will depend on the specific structure of your data and your specific desired outcome.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.