In a preliminary analysis of listing decisions under Canada's Species at Risk Act (SARA), Mooers et al. (2007)demonstrated an apparent bias against marine and northern species. As a follow-up, we expanded the set of potential explanatory variables, including information on jurisdictional and administrative elements of the listing process, and considered an additional 16 species recommended for listing by SARA's scientific advisory committee as of 15 August 2006. Logistic model selection based on Akaike differences suggested that species were less likely to be listed if they were harvested or had commercial or subsistence harvesting as an explicitly identified threat; had Department of Fisheries and Oceans (DFO) as a responsible authority (RA); were located in Canada's north generally, and especially in Nunavut; or were found mostly or entirely within Canada. Subsequent model validation with an independent set of 50 species for which a listing decision was handed down in December 2007 showed an overall misclassification rate of <0.10, indicating reasonable predictive power. In light of these results, we recommend that RAs under SARA adopt a two-track listing approach to address problems of delays arising from extended consultations and the inconsistent use by the RAs of socioeconomic analysis; consider revising SARA so that socioeconomic analysis occurs during decisions about protecting species and their habitats rather than at the listing stage; and maintain an integrated database with information on species' biology, threats, and agency actions to enable future evaluation of SARA's impact.