Understanding Differences Between Person and Place Based Neighborhood Interventions through Direct and Indirect Effects

Researchers have recently relied on person based randomized interventions to estimate the effects of a neighborhood on a variety of outcomes. An alternative strategy is to implement an intervention at the neighborhood level such that whole communities rather than individuals are randomized into better conditions. I explore the efficacy of each approach by decomposing the total intervention effect into natural direct and indirect effects. A comparison of these effects may provide insight into the precise mechanisms that person and place based interventions work through to impact individual well being. Delineating when each type of treatment yields larger effects provides policymakers insight into their appropriateness given overarching goals.