The use of logistic regression in modelling the distributions of bird species in Swaziland

The method of logistic regression was used to model the observed geographic al distribution patterns of bird species in Swaziland in relation to a set of environmental variables. Reporting rates derived from bird atlas data ar e used as an index of population densities. This is justified in part by th e success of the modelling process. For each species the variables which we re significantly related to its distribution were identified. Various metho ds for coding environmental variables from maps were investigated. A jack-k nifing technique was used to demonstrate the predictive power of the logist ic models. A criterion for assessing the goodness of fit of the logistic mo dels was derived.