Justin Grimmer

General Purpose Computer-Assisted Clustering and Conceptualization

We develop a computer-assisted method for the discovery of insightful conceptualizations, in the form of clusterings (i.e., partitions) of input objects. Each of the numerous fully automated methods of cluster analysis proposed in statistics, computer science, and biology optimize a different objective function. Almost all are well defined, but how to determine before the fact which one, if any, will partition a given set of objects in an “insightful” or “useful” way for a given user is unknown and difficult, if not logically impossible. We develop a metric space of partitions from all existing cluster analysis methods applied to a given dataset (along with millions of other solutions we add based on combinations of existing clusterings) and enable a user to explore and interact with it and quickly reveal or prompt useful or insightful conceptualizations. In addition, although it is uncommon to do so in unsupervised learning problems, we offer and implement evaluation designs that make our computer-assisted approach vulnerable to being proven suboptimal in specific data types. We demonstrate that our approach facilitates more efficient and insightful discovery of useful information than expert human coders or many existing fully automated methods.

How Words and Money Cultivate a Personal Vote: The Effect of Legislator Credit Claiming on Constituent Credit Allocation

the district—affect how constituents allocate credit. Legislators use credit claiming messages to influence the expenditures they receive credit for and to affect how closely they are associated with spending in the district. Constituents are responsive to credit claiming messages—they build more support than other nonpartisan messages. But contrary to expectations from other studies, constituents are more responsive to the total number of messages sent rather than the amount claimed. Our results have broad implications for political representation, the personal vote, and the study of U.S. Congressional elections. P articularistic spending, a large literature argues, cultivates a personal vote for incumbents (Cain, Ferejohn, and Fiorina 1987; Ferejohn 1974;LazarusandReiley2010;LevittandSnyder1997; Mayhew 1974). To build this support, legislators are assumed to direct projects and programs to their districts. Constituents, in turn, are thought to reward their legislator for the level of federal spending in the district (Levitt and Snyder 1997; Str¨ omberg 2004) or the

Congressmen in Exile: The Politics and Consequences of Involuntary Committee Removal

We show how preferred committee assignments act as an electoral subsidy for members of Congress—empowering representatives’ legislative careers. When holding preferred assignments, legislators are free to focus on legislative activity in Washington, DC. But when the subsidy is removed, legislators are forced to direct attention to the district. To test our theory of legislative subsidy, we exploit committee exile—the involuntary removal of committee members after a party loses a sizable number of seats. Legislators are selected for exile using members’ rank on the committee, causing exiled and remaining legislators to appear strikingly similar. Using exile, we show that it has only limited electoral consequences, but this is partly due to compensatory efforts. Exiled legislators shift attention away from Washington and towards the district: they raise and spend more money for reelection, author less legislation, are absent for more days of voting, and vote with their party less often.

The Impression of Influence: Legislator Communication, Representation, and Democratic Accountability

List of Illustrations ix List of Tables xi Acknowledgments xiii 1 Representation, Spending, and the Personal Vote 1 2 Solving the Representatives Problem and Creating the Representatives Opportunity 15 3 How Legislators Create an Impression of Influence 32 4 Creating an Impression, Not Just Increasing Name Recognition 64 5 Cultivating an Impression of Influence with Actions and Small Expenditures 81 6 Credit, Deception, and Institutional Design 121 7 Criticism and Credit: How Deficit Implications Undermine Credit Allocation 148 8 Representation and the Impression of Influence 174 9 Text as Data: Methods Appendix 186 Bibliography 189 Index 203

Money in Exile: Campaign Contributions and Committee Access

Understanding how money influences the legislative process is essential for assessing American democracy, but problems of endogeneity, legality, and observational equivalence make it difficult to isolate the effect of contributions on policy. We seek to answer long-standing questions about the influence of money in Congress by exploiting a congressional procedure (committee exile) that exogenously varies a member’s influence over the policy-making process. We leverage exile as an identification strategy to show that business interests seek short-term access to influential legislators. Industries overseen by the committee decrease contributions to exiled legislators and instead direct their contributions to new committee members from the opposite party. Partisan interests, in contrast, attempt to influence electoral outcomes—boosting contributions to exiled members. Together, we provide evidence that corporations and business PACs use donations to acquire immediate access and favor—suggesting they at least anticipate that the donations will influence policy.

TopicCheck: Interactive Alignment for Assessing Topic Model Stability.

Content analysis, a widely-applied social science research method, is increasingly being supplemented by topic modeling. However, while the discourse on content analysis centers heavily on reproducibility, computer scientists often focus more on scalability and less on coding reliability, leading to growing skepticism on the usefulness of topic models for automated content analysis. In response, we introduce TopicCheck, an interactive tool for assessing topic model stability. Our contributions are threefold. First, from established guidelines on reproducible content analysis, we distill a set of design requirements on how to computationally assess the stability of an automated coding process. Second, we devise an interactive alignment algorithm for matching latent topics from multiple models, and enable sensitivity evaluation across a large number of models. Finally, we demonstrate that our tool enables social scientists to gain novel insights into three active research questions.

Discovery of Treatments from Text Corpora.

An extensive literature in computational social science examines how features of messages, advertisements, and other corpora affect individuals’ decisions, but these analyses must specify the relevant features of the text before the experiment. Automated text analysis methods are able to discover features of text, but these methods cannot be used to obtain the estimates of causal effects—the quantity of interest for applied researchers. We introduce a new experimental design and statistical model to simultaneously discover treatments in a corpora and estimate causal effects for these discovered treatments. We prove the conditions to identify the treatment effects of texts and introduce the supervised Indian Buffet process to discover those treatments. Our method enables us to discover treatments in a training set using a collection of texts and individuals’ responses to those texts, and then estimate the effects of these interventions in a test set of new texts and survey respondents. We apply the model to an experiment about candidate biographies, recovering intuitive features of voters’ decisions and revealing a penalty for lawyers and a bonus for military service.

Mirrors for Princes and Sultans: Advice on the Art of Governance in the Medieval Christian and Islamic Worlds

When did European modes of political thought diverge from those that existed in other world regions? We compare Muslim and Christian political advice texts from the medieval period using automated text analysis to identify four major and 60 granular themes common to Muslim and Christian polities, and examine how emphasis on these topics evolves over time. For Muslim texts, we identify an inflection point in political discourse between the eleventh to thirteenth centuries, a juncture that historians suggest is an ideational watershed brought about by the Turkic and Mongol invaders. For Christian texts, we identify a decline in the relevance of religious appeals from the Middle Ages to the Renaissance. Our findings also suggest that Machiavelli’s Prince was less a turn away from religious discourse on statecraft than the culmination of centuries-long developments in European advice literature.

Obstacles to Estimating Voter ID Laws’ Effect on Turnout

Widespread concern that voter identification laws suppress turnout among racial and ethnic minorities has made empirical evaluations of these laws crucial. But problems with administrative records and survey data impede such evaluations. We replicate and extend Hajnal, Lajevardi, and Nielson’s 2017 article, which concludes that voter ID laws decrease turnout among minorities, using validated turnout data from five national surveys conducted between 2006 and 2014. We show that the results of their article are a product of data inaccuracies, the presented evidence does not support the stated conclusion, and alternative model specifications produce highly variable results. When errors are corrected, one can recover positive, negative, or null estimates of the effect of voter ID laws on turnout, precluding firm conclusions. We highlight more general problems with available data for research on election administration, and we identify more appropriate data sources for research on state voting laws’ effects.

Creating and Destroying Party Brands

Party brands are central to theories of Congressional action. While previous work assumes that a party’s brand—its long run reputation—is a direct consequence of the content of legislation, in this presentation I show how partisans from both parties use public statements to craft their own party’s reputation and to undermine their opponents party. The incentive to craft and destroy brands varies across legislators, creating systematic distortions in who contributes to partisan branding efforts, what is said about a party’s brand, and when partisan criticism becomes salient. To demonstrate the construction of party brands I use new collections of newsletters from Congressional offices, along with press releases, floor speeches, and media broadcasts. Across the diverse sources, I show that ideologically extreme legislators are the most likely to explain their party’s work in Washington and the most likely to criticize the opposing party—particularly when their is an opposing party president. Extreme legislators also engage in more vitriolic criticism of the opposing party, particularly when opposing presidents are unpopular. The result is that parties in rhetoric appear even more combative and polarized in public debate outside Congress than inside Congress.