Roadways fragment landscape connectivity for wildlife, which often leads to negative consequences for the genetic and demographic characteristics of populations. Transportation structures, including culverts, bridges, and underpasses, help mitigate the fragmenting and barrier effects of roads by providing safer areas of wildlife movement beneath roads. However, the relative value of structures to species largely depends on the extent of animal movement around a given structure, which reflects the landscape conditions at multiple spatial scales. We used a new electrical circuit theory approach to evaluate the relative value of transportation structures across the road system in Vermont, which lies in an important region for broader connectivity in the northeastern US. Our objectives were to 1) model and map connectivity at two spatial scales (statewide and structure level) for eight species (moose, white-tailed deer, American black bear, eastern coyote, bobcat, red fox, raccoon, striped skunk) using a combination of occupancy models, expert-derived resistance surfaces, and coarse and fine scale land cover data (NLCD and Lidar-derived, respectively); and 2) use models to score the connectivity values for each structure. We used existing occupancy models and built resistance surfaces by surveying state and federal biologists to model and map connectivity. Maps depicted patterns of movement flow for each species at both scales, and we present three methods for combining data at both scales to score each of the 6,206 agency-managed transportation structures in the state. A linear programming decision-making framework was developed to rank transportation structures according to connectivity scores and additional landscape characteristics not included in modeling, but important to conservation efforts. The framework allows agency managers to quickly evaluate the relative value of a given structure for connectivity, which will inform decision-making related to mitigating the impacts of roadways on wildlife.