Why do states choose multilateralism? We develop an argument focused on the burden-sharing versus control dilemma of principal-agent (PA) models. We also present two alternative theoretical frames that could explain this choice: a normative logic of appropriateness and hegemonic self binding. We examine the political bases of support for sending foreign aid through multilateral versus bilateral channels. First, we clarify the concept of multilateralism. We show that the choices for internationalism and multilateralism are distinct. Second, we develop hypotheses from each of the three theories and examine how public opinion data allow us to shed light on these different theories about multilateralism. Finally, we provide evidence about the correlates of public support for multilateral engagement. We isolate how two competing rationales—burden sharing and control—dictate some of the politics around the choice between multilateral versus bilateral aid channels. The data support our claim that a principal-agent model can help us to understand the choice for multilateralism.

Experimentation is a powerful methodology that enables scientists to empirically establish causal claims. However, one important criticism is that experiments merely provide a black-box view of causality and fail to identify causal mechanisms. Critics argue that although experiments can identify average causal effects, they cannot explain how such effects come about. If true, this represents a serious limitation of experimentation, especially for social and medical science research whose primary goal is to identify causal mechanisms. In this paper, we consider several different experimental designs and compare their identification power. Some of these designs require the direct manipulation of mechanisms, while others can be used even when only imperfect manipulation is possible. We use recent social science experiments to illustrate the key ideas that underlie each design.

Empirical testing of competing theories lies at the heart of social science research. We demonstrate that a very general and well-known class of statistical models, called finite mixture models, provides an effective way of rival theory testing. In the proposed framework, each observation is assumed to be generated from a statistical model implied by one of the theories under consideration. Researchers can then estimate the probability that a specific observation is consistent with either of the competing theories. By directly modeling this probability with the characteristics of observations, one can also determine the conditions under which a particular theory applies. We discuss a principled way to identify a list of observations that are statistically significantly consistent with each theory. Finally, we propose several measures of the overall performance of a particular theory. We illustrate the advantages of our method by applying it to an influential study on trade policy preferences.