Dustin Tingley

A General Approach to Causal Mediation Analysis

Traditionally in the social sciences, causal mediation analysis has been formulated, understood, and implemented within the framework of linear structural equation models. We argue and demonstrate that this is problematic for 3 reasons: the lack of a general definition of causal mediation effects independent of a particular statistical model, the inability to specify the key identification assumption, and the difficulty of extending the framework to nonlinear models. In this article, we propose an alternative approach that overcomes these limitations. Our approach is general because it offers the definition, identification, estimation, and sensitivity analysis of causal mediation effects without reference to any specific statistical model. Further, our approach explicitly links these 4 elements closely together within a single framework. As a result, the proposed framework can accommodate linear and nonlinear relationships, parametric and nonparametric models, continuous and discrete mediators, and various types of outcome variables. The general definition and identification result also allow us to develop sensitivity analysis in the context of commonly used models, which enables applied researchers to formally assess the robustness of their empirical conclusions to violations of the key assumption. We illustrate our approach by applying it to the Job Search Intervention Study. We also offer easy-to-use software that implements all our proposed methods.

Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies

Identifying causal mechanisms is a fundamental goal of social science. Researchers seek to study not only whether one variable affects another but also how such a causal relationship arises. Yet commonly used statistical methods for identifying causal mechanisms rely upon untestable assumptions and are often inappropriate even under those assumptions. Randomizing treatment and intermediate variables is also insufficient. Despite these difficulties, the study of causal mechanisms is too important to abandon. We make three contributions to improve research on causal mechanisms. First, we present a minimum set of assumptions required under standard designs of experimental and observational studies and develop a general algorithm for estimating causal mediation effects. Second, we provide a method for assessing the sensitivity of conclusions to potential violations of a key assumption. Third, we offer alternative research designs for identifying causal mechanisms under weaker assumptions. The proposed approach is illustrated using media framing experiments and incumbency advantage studies.

Monoamine oxidase A gene (MAOA) predicts behavioral aggression following provocation

Monoamine oxidase A gene (MAOA) has earned the nickname “warrior gene” because it has been linked to aggression in observational and survey-based studies. However, no controlled experimental studies have tested whether the warrior gene actually drives behavioral manifestations of these tendencies. We report an experiment, synthesizing work in psychology and behavioral economics, which demonstrates that aggression occurs with greater intensity and frequency as provocation is experimentally manipulated upwards, especially among low activity MAOA (MAOA-L) subjects. In this study, subjects paid to punish those they believed had taken money from them by administering varying amounts of unpleasantly hot (spicy) sauce to their opponent. There is some evidence of a main effect for genotype and some evidence for a gene by environment interaction, such that MAOA is less associated with the occurrence of aggression in a low provocation condition, but significantly predicts such behavior in a high provocation situation. This new evidence for genetic influences on aggression and punishment behavior complicates characterizations of humans as “altruistic” punishers and supports theories of cooperation that propose mixed strategies in the population. It also suggests important implications for the role of individual variance in genetic factors contributing to everyday behaviors and decisions.

Who are these people? Evaluating the demographic characteristics and political preferences of MTurk survey respondents

As Amazon’s Mechanical Turk (MTurk) has surged in popularity throughout political science, scholars have increasingly challenged the external validity of inferences made drawing upon MTurk samples. At workshops and conferences experimental and survey-based researchers hear questions about the demographic characteristics, political preferences, occupation, and geographic location of MTurk respondents. In this paper we answer these questions and present a number of novel results. By introducing a new benchmark comparison for MTurk surveys, the Cooperative Congressional Election Survey, we compare the joint distributions of age, gender, and race among MTurk respondents within the United States. In addition, we compare political, occupational, and geographical information about respondents from MTurk and CCES. Throughout the paper we show several ways that political scientists can use the strengths of MTurk to attract respondents with specific characteristics of interest to best answer their substantive research questions.

Redefine Statistical Significance

We propose to change the default P-value threshold for statistical significance from 0.05 to 0.005 for claims of new discoveries.

Causal Mediation Analysis Using R

Causal mediation analysis is widely used across many disciplines to investigate possible causal mechanisms. Such an analysis allows researchers to explore various causal pathways, going beyond the estimation of simple causal effects. Recently, Imai et al. (2008) [3] and Imai et al. (2009) [2] developed general algorithms to estimate causal mediation effects with the variety of data types that are often encountered in practice. The new algorithms can estimate causal mediation effects for linear and nonlinear relationships, with parametric and nonparametric models, with continuous and discrete mediators, and various types of outcome variables. In this paper, we show how to implement these algorithms in the statistical computing language R. Our easy-to-use software, mediation, takes advantage of the object-oriented programming nature of the R language and allows researchers to estimate causal mediation effects in a straightforward manner. Finally, mediation also implements sensitivity analyses which can be used to formally assess the robustness of findings to the potential violations of the key identifying assumption. After describing the basic structure of the software, we illustrate its use with several empirical examples.

Who Supports Global Economic Engagement? The Sources of Preferences in American Foreign Economic Policy

In this article we bring together opposing international relations theories to better understand U.S. foreign policy, in particular foreign trade and aid. Using votes in the U.S. House of Representatives from 1979–2004, we explore different theoretical predictions about preferences for foreign economic policy. We assess the impact of domestic factors, namely political economy and ideological preferences, versus foreign policy pressures. Our three main results highlight the differential effect of these factors in the two issue areas. First, aid preferences are as affected by domestic political economy factors as are trade preferences. Second, trade preferences, but not economic aid ones, are shaped by the presidents foreign policy concerns; for economic aid, domestic political economy factors matter more than foreign policy ones. Third, aid preferences are shaped more by ideological factors than are trade ones, but ideology plays a different substantive role in each. Different constituencies support aid and trade. This finding has implications for foreign policy substitutability, “the internationalist coalition” in U.S. foreign policy, “statist” theories of foreign policy, and the connection between public opinion and legislative voting.

Conditional Cooperation and Climate Change

It is widely believed that international cooperation can arise through strategies of reciprocity. In this paper, we investigate whether citizens in the United States and 25 other countries support reciprocity to deal with climate change. We find little public enthusiasm for intrinsic reciprocity, in which countries restrain their consumption of fossil fuels if and only if other countries do the same. In contrast, we find significant support for extrinsic reciprocity, in which countries enforce cooperation by linking issues. Citizens support economic sanctions against polluters and are willing to shame them in international forums, especially when the polluters are violating a treaty. Cooperation could, therefore, emerge from efforts to link climate with other issues and to embed climate commitments in international law.

The Effect of Repeated Play on Reputation Building: An Experimental Approach

What effect does repeated play have on reputation building? The literature on international relations remains divided on whether, when, and how reputation matters in both interstate and intrastate conflict. We examine reputation building through a series of incentivized laboratory experiments. Using comparative statics from a repeated entry-deterrence game, we isolate how incentives for reputation building should change as the number of entrants changes. We find that subjects in our experiments generally build reputations and that those investments pay off, but we also find that some subjects did not react to incentives to build reputation in ways our model had predicted. In order to explain this, we focus on the heterogeneity of preferences and cognitive abilities that may exist in any population. Our research suggests that rational-choice scholars of international relations and those using more psychologically based explanations have more common ground than previously articulated.

Public Opinion and Foreign Aid: A Review Essay

The study of public opinion and foreign policy has a long history (Almond, 1950; Converse, 1964; Lippmann, 1955). This history includes a long-standing debate over the utility of studying public opinion when considering international affairs (Holsti, 1992; Mueller, 1971; Page and Shapiro, 1983; Page and Shapiro, 1992; Wittkopf, 1986). The dismissal of the importance of public opinion stems from the concern that the mass public knows little about foreign policy. Prominent theories about foreign policy and international relations give no role to publics (Krasner, 1978; Mearsheimer, 2001; Waltz, 1979). Very few theoretical perspectives in international relations give any weight to public attitudes; neorealism, neoliberalism, and institutionalism provide very little space for the mass public to affect foreign policy.