Dahyun Choi

Hello! I am a Ph.D. candidate in the Department of Politics at Princeton University, where I am affiliated with the Center for the Study of Democratic Politics, the Data-Driven Social Science Initiative, and the Program for Quantitative and Analytical Political Science. I work on questions in interest group and bureaucratic politics, political economy, and American political institutions, using computational methods that involve text analysis and large language models, along with causal inference techniques and formal modeling. I also develop methodologies for applications of machine learning. I am on the 2024-25 job market.

Manuscripts Under Review

Teaming Up Across Political Divides: Evidence from Climate Regulations

(Previously titled "Teaming up with the Enemy: Firms and the Information Environment of Climate Regulations.) [ | paper | ]

Why do interest groups with contrasting interests and policy goals work together? I present a theory of collaborative policy production and show that interest groups can achieve higher policy gains through collaboration, even though their ideal policy goals may diverge significantly. To test theoretical results, I introduce original measurement strategies that reveal systematic patterns in which firms and environmental groups invest in joint efforts to improve fine-grained details of policy to achieve greenhouse gas emissions targets. The analysis, using public comments spanning 2010-2020, demonstrates that comments written jointly by environmental groups and firms contain more information that can contribute to the quality of policy implementation than individual efforts alone, despite compromises on policy preferences. These findings highlight the hidden dynamics of regulatory politics, wherein divergent political goals are reconciled for high-quality policy implementation.

Political Economy of Climate and the Environment, University of Pennsylvania

Frontiers in Money in Politics Research, University of Southern California

2023 American Political Science Association, Los Angeles

2023 Money in Politics Conference, Copenhagen Business School

Climate Pipeline Project, Brown University

2023 Midwest Political Science Association, Chicago

2022 Environmental Politics and Governance, Virtual conference

2022 International Political Economy Society, University of Pittsburgh

Quantifying Informative Signals from Interest Groups

(with Brandon Stewart and Denis Peskoff) Revise & Resubmit, Political Science Research and Methods [ | paper | ]

Understanding the political process in the United States requires examining how information is provided to politicians and the general public. While theoretical models identify interest groups as strategic information providers, empirically studying this aspect has been challenging due to the subject knowledge and sophistication required to understand the contexts. To address the challenges of quantifying information—often characterized by technical complexity and specialization—we propose a method for retrieving expert coding using recent advancements in large language models. This method allows us to quantify two signals that are difficult to disentangle empirically: (1) informative signals that help agents make better political decisions, and (2) associative signals that influence preference formation but lack direct relevance to the substantive question of interest. We validate our approach using two applications where the method generalizes out of distribution. This study not only demonstrates the applicability of large language models for complex measurement tasks but also provides substantive insights, revealing that interest groups strategically adjust the composition of signals across different institutional settings.

Email for current draft

2023 American Political Science Association, Los Angeles

2022 Society for Political Methodology, Washington University in St. Louis

Learning from Noise: Applying Sample Complexity for Social Science Research

(with Perry Carter)

[ | paper | ]

While statistical learning bridges the gap between theoretical concepts and complex empirical realities, the question of what constitutes ``good enough" data for social scientists remains understudied. In this article, we introduce the Probably Approximately Correct model and present sample complexity bounds, which take advantage of researcher-specified estimates of labeling error to guarantee the sample size required for a minimum level of accuracy. We develop a simulation-based approach to demonstrate its feasibility and provide the scR R package, offering a computationally efficient way to implement the proposed methods. We aim to improve standard practice by providing a general-purpose tool to validate the quality of measures when fuzzy measurement boundaries make generating data with ground-truth labels infeasible.

2023 American Political Science Association, Los Angeles

2023 Society for Political Methodology, Stanford University

Diminishing Regulatory Capacity and Corporate Political Disengagement: Evidence from State-Level Workforce Shocks

(with Kyuwon Lee)

[ | paper | ]

Although there are public concerns about the declining capacity of regulatory agencies and its impact on regulatory outcomes, such decline could also lead regulated firms to disengage themselves from politics. We examine whether and how firms reduce their campaign contributions in response to decreases in state-level regulatory capacity. To do so, we collect original datasets on the workforce size of U.S. state environmental agencies and leverage variations in workforce shocks that arise from the gap between actual and appropriated workforce sizes. Our analysis reveals that state environmental agencies' workforce shocks decrease firms' donations to state legislators, particularly to those in the majority party and the Democratic party, but do not affect firms' contributions to their ideological allies. We also find that existing state-level restrictions on corporate donations do not moderate firms' political responsiveness. Overall, this article provides a nuanced picture of how diminishing regulatory capacity could shape corporate political activities.

2024 American Political Science Association, Philadelphia

2024 Money in Politics Conference, Copenhagen Business School

Conference on Bureaucratic and Interest Group Politics, Princeton University

2024 Midwest Political Science Association, Chicago

Selected Working Papers

How Do Bureaucrats Learn in Absence of Autonomous Sources of Expertise?

[ | ]

A large body of literature points to bureaucratic expertise as an important factor in policy outcomes, while the question of how bureaucrats obtain policy-relevant information is relatively understudied. I demonstrate that in a policy environment where policy-specific expertise is concentrated in firms, the bureaucratic acquisition of expertise is contingent upon the political influence of private interests. Using an original dataset revealing the bureaucratic monitoring behavior of the Greenhouse Gas Reporting Program with a new measure of partisan alignment between firms and bureaucracy, I examine the practices of how bureaucrats collect information from regulated entities. This paper highlights the underlying, but underappreciated, sources of bureaucratic expertise when establishing an autonomous source of information is impossible.

Email for current draft

2024 American Political Science Association, Philadelphia

Princeton Institutions Working Group

Scientification of Politics? Interest Group Influence on Bureaucratic Expertise in Climate Change

[ | ]

To what extent do interest groups influence the processes through which bureaucrats obtain scientific and policy-relevant information? While science has been considered the primary source of expertise that helps regulators improve policy outcomes, the use of scientific information is inevitably intertwined with political reasoning. To explore this dynamic, I construct an original dataset that reveals the Environmental Protection Agency's internal decision-making process and examine how the proximity of each scientific study's funders to the agency affects the study's chances of being cited by bureaucrats in policy documents. My analysis provides evidence that bureaucrats are better informed about the quality of scientific studies but are also biased in their favor. Overall, the effect of quality appears to outweigh the bias introduced by the proximity between bureaucrats and the interest groups that fund scientific studies. This paper not only provides suggestive evidence on the roles of expertise and bias in the policymaking process but also contributes to our understanding of interest group influence on bureaucratic expertise.

Email for current draft

Princeton Institutions Working Group

2023 International Political Economy Society, Georgetown University

Politics of Academic Experts

(with Nolan McCarty)

Political Cycle of Government Innovation

(with John M. de Figueiredo)

Sample Complexity For Open-Ended Survey Responses

(with Perry Carter and Narrelle Gilchrist)

[ poster | ]

2024 American Political Science Association, Philadelphia

2024 Society for Political Methodology, UC Riverside

A Dataset of Presidential Election Results at the Congressional District Level

(with Christian Baehr, Francesca Tang and Rocio Titiunik)