Integrating Data Across Misaligned Spatial Units

Abstract Theoretical units of interest often do not align with the spatial at which data are available. This problem is pervasive in political science, particularly subnational empirical research that requires integrating across incompatible geographic (e.g., administrative areas, electoral constituencies, and grid cells). Overcoming this challenge researchers only to scale theoretical units, but also understand consequences change support for measurement error statistical inference. We show how accuracy transformed values estimation regression coefficients depend on degree nesting (i.e., whether fall completely neatly inside each other) relative source destination aggregation, disaggregation, hybrid). introduce simple, nonparametric measures scale, as ex ante indicators transformation complexity susceptibility. Using election Monte Carlo simulations, we these strongly predictive quality multiple change-of-support methods. propose several validation procedures provide open-source software make options more accessible, customizable, intuitive.