tocorrelation in Country-Level Models of Species Imperilment

g e cated by the fact that factors that influence species imperilment may extend or rate beyond arbitrary political boundaries. Following McPherson and Nieswiadomy (2005), c the advisability of controlling for spatial autocorrelation in models focusing on e birds, mammals, reptiles, amphibians, and vascular plants. We mpare the o dif e s of the spatial dependency. Although our a priori e n was that measures that more accurately reflect the degree of spatial interaction een countries, such as the percentage of shared border, would be superior to a measure of ple adjacency, in fact we find that the simple adjacency measure outperforms the other s Spatial Au ma ope we imp perf exp betw sim mea M co od mp elin li the determinants of species’ cological fragility using country-specific data y be onfi rilm rm ctat rm e an io nt ce of of also co ferent d finition sures in mo t cases. 1 Do Scie Sust spat c ent and Professor of Economics and Policy, respectively, in the School of Forestry and Wildlife nces uburn University. This work has been supported by Auburn University’s Center for Forest ainability and by a McIntire-Stennis grant to Laband. We thank Mike Nieswiadomy for sharing a copy of his ial weight matrix and for helpful co nts; remaining errors are our respo ility. toral s at A tud