I recently ran a set of four two-species interaction models with competing hypotheses of dependence/independence of detection/occupancy for muskrat and mink. Using the equation provided in Richmond et al. (2010), I calculated SIF to be 1.8 from psiA, psiBA, and psiBa. The equation is:
SIF= (ψ^A ψ^BA)/(ψ^A (ψ^A ψ^BA+(1-ψ^A ) ψ^Ba))
However, the derived parameters produced by PRESENCE for specific sites average 0.76 for the same model. Why would my calculated SIF be above 1 but site-specific SIFs be below 1? Secondly, how can I calculate a SE for my calculated SIF? Or do I need to average site-specific SIFs to get a SE?
Thanks in advance,
Caleb