Multiple regression model used for evaluating the relative importance of different factors as drivers of commission and omission errors.

The rate of false positive and negative occurrences was calculated by comparing the spatial distribution of species under each predictive model and the true distribution. Amount of data refers to each of the three different models tested (poor, intermediate and good quality data models), the Area Under the Curve (AUC) was measured for each species and model through a K-fold validation procedure. Standardised Beta coefficients, a t statistic (degrees of freedom between parentheses) and an associated p value are shown. The adjusted R2 is also given.