Geographical variation in the likelihood of biological invasions can be affected by propagule pressure and habitat suitability, which are driven by ecological and social processes. Past studies have empirically quantified the role of drivers by comparing geographical variation in numbers of invading species with variation in candidate factors; however, lack of data has limited empirical studies for individual species. Lymantria dispar (L.), a nonnative forest pest formerly known as gypsy moth, is an exemplar species for exploring invasion drivers because of extensive records on its spread. Since its establishment in eastern United States in 1869, it has been repeatedly introduced into outlying areas, prompting 325 eradication programs from 1972 to 2014. We used these eradication programs as proxies for new establishment events, with the assumption that populations would have established in the absence of eradication treatments. These proxy events were used to quantify the effects of socio-environmental factors on the probability of L. dispar arrival and establishment. Establishment probability was significantly affected by propagule pressure (distance to the previously invaded area, human population size, and the area of source outbreaks) and habitat suitability (climate and availability of host trees). The statistical model developed here can be used to predict invasions and inform surveillance strategies to more efficiently manage these invasions.