Jake Bowers (University of Illinois at Urbana-Champaign), Mark Fredrickson (University of Illinois at Urbana-Champaign), and Costas Panagopoulos (Fordham University) Reasoning about Interference in Randomized Studies,. (Vol 21, number 1: 97-124)
Bowers, Fredrickson, and Panagopoulos's "Reasoning about Interference Between Units: A General Framework" exemplifies rigorous, creative, and useful work in political methodology. Using Fisher/Rosenbaum-style randomization inference, the article tackles an difficult and pervasive problem---interference among units---in a novel and compelling way. Rather than treating spillover effects as a nuisance to be marginalized over or, worse, ignored, Bowers et al. use them as an opportunity to test substantive questions regarding interference. In doing so, they push randomization inference further than any previous work, showcasing its capacity to assess models much more theoretically sophisticated and scientifically interesting than the sharp null hypothesis of no effects. Their work also brings together causal inference and network analysis in an innovative and compelling way, pointing the way to future convergence between these domains. Finally, they do it all with care, transparency, and rigor. We expect their work to receive wide attention both within and outside the discipline.