In settings with interference, researchers commonly define estimands using exposure mappings to summarize neighborhood variation in treatment assignments. This paper studies the causal interpretation of these estimands under weak restrictions on interference. We demonstrate that the estimands can exhibit unpalatable sign reversals under conventional identification conditions. This motivates the formulation of sign preservation criteria for causal interpretability. To satisfy preferred criteria, it is necessary to impose restrictions on interference, either in potential outcomes or selection into treatment. We provide sufficient conditions and show that they can be satisfied by nonparametric models with interference in both the outcome and selection stages.
翻译:暂无翻译