Marc Ratkovic

Department of Politics

Princeton University

Recent Papers of Interest

Legislative institutions structure and order the myriad topics addressed by legislators. Jointly considering both roll call votes and floor speech, we show that the contemporary US Senate is ordered around but two dimensions: one ideological and another capturing leadership.  We characterize both word and vote choice in terms of the exact same ideal points and policy dimensions.  These findings emerge from our method, Sparse Factor Analysis (SFA), designed to combine vote and textual data when estimating ideal points, word affect, and the underlying dimensionality. This contrasts with the single dimension that emerges from an analysis of votes alone, and with the more numerous dimensions that emerge from analyzing speech alone. We then show how SFA can leverage common speech in order to impute missing data, to estimate rank-and-file ideal points using only their words and the vote history of party leaders, and even to scale newspaper editorials.  

With In Song Kim and John Londregan.

Voting, Speechmaking, and the Dimensions of Conflict in the US Senate

My research interests include causal inference, machine learning methods, and text analysis.  

I teach courses in political methodology at the undergraduate and graduate level.

A brief background of myself and my life.  Sometiemes I even do things that aren't political science.



About Me

Matching and weighting methods are commonly used to reduce confounding bias in observational studies. Many existing methods are sensitive to user-provided inputs, provide little formal guidance in selecting these inputs, and do not necessarily return a balanced subset of the data. The proposed method adapts the support vector machine classifier in order to provide a fully automated, nonparametric procedure for identifying the largest balanced subset of the data. The method allows for a sensitivity analysis and an assessment of the common support assumption. Two applications, a simulation study and a benchmark dataset, illustrate the method's use and efficacy.

Balancing within the Margin: Causal Effect Estimation with

Support Vector Machines

Instrumental variable estimation is a long-established means of reducing endogeneity bias in regression coefficients.  Researchers commonly confront two problems when conducting an IV analysis: the instrument may be only weakly predictive of the endogenous variable, and the estimates are valid only for observations that comply with the instrument. We introduce Complier Instrumental Variable (CIV) estimation, a method for estimating who complies with the instrument. CIV uses these compliance probabilities to strengthen the instrument through up-weighting estimated compliers. As compliance is latent, we model each observation's density as a mixture between that of a complier and a non-complier. We derive a Gibbs sampler and Expectation Conditional Maximization algorithm for estimating the CIV model. A set of simulations shows that CIV performs favorably relative to several existing alternative methods, particularly in the presence of small sample sizes and  weak instruments. We then illustrate CIV on data from a prominent study  estimating the effect of property rights on growth.  We show how CIV can strengthen the instrument and generate more reliable results.  We also show how characterizing the compliers can help cast insight into the underlying political dynamic.  

With Yuki Shiraito.

Strengthening Weak Instruments by Modeling Compliance

035 Corwin Hall

Princeton,NJ 08544



Office Hours:

Tuesday, 1:30 - 3:30

035 Corwin Hall

Recent Papers of Interest