No Replicability of Niche Similarity Test Results?

[Morgan Ladue]

Hello,

When running our species' Niche Overlap multiple times, we get drastically different results every time. When the same set of occurrences and environmental variables are used, but the background points are changed, the D varies from 0-0.18 and the p-value varies from 0.01-0.9. When the same background points are used, there is still variance in the p-values although the significance does not change.

How should we interpret the results if the p-value is sometimes significant and sometimes not? How do we decide which run of background points to use?

Any information would be helpful!

Best,

Morgan