A new model and related empirical work explain how the Senate and President affect the timing of power-sharing rule changes in the US House. We argue that shifts in the Senate's or President's preferences (e.g., a new majority party in the Senate; a new president) reshape House members'...

Published reports of experiments with noncompliance often fail to report information necessary for recovering average potential outcomes for compliers. We derive sharp bounds on the average potential outcomes for compliers, when given only average outcomes for units assigned to treatment,...

One of the hallmarks of the 2004 presidential election was the unusual emphasis on face-to-face voter mobilization, particularly face-to-face mobilization conducted within neighborhoods or social networks. Unlike previous studies of face-to-face voter mobilization, which have focused largely on...

In recent years, social scientists have increasingly turned to matching as a method for drawing causal inferences from observational data. Matching compares those who receive a treatment to those with similar background attributes who do not receive a treatment. Researchers who use matching...