Jiawei Fu Develops New Tools to Unravel the Complexities of Political Decision-Making

Imagine a politician battling accusations of corruption leading up to an election, accusations so serious that they are likely to impact how people cast their ballots. How do you account for the wide range of factors that can determine who a person votes for — partisanship, age, gender, race, educational level — to gain insight into the influence of the corruption allegations on Election Day? 

That challenge is being taken up by Jiawei Fu, a new assistant professor of Political Science who  develops tools to help social scientists gain deeper insights into political behavior.  

In high school, Fu was drawn to biology and appreciated that within the natural sciences, you have the ability to build experiments and control variables to better understand causal relationships. “In the natural sciences, it is relatively straightforward to design experiments and then replicate those experiments,” he said. “In social science, we also view randomized experiments as the gold standard for establishing causality, but in practice it is often impossible to design experiments that address every question. We study human beings, and it’s very hard to conduct experiments about the interactions between humans and institutions.” 

Fu is interested in better understanding the complex decision-making processes people rely on when choosing how to interact with one another. He develops new methods of causal inference and, beyond purely statistical research, combines formal theories, such as game theory or decision theory, with statistical tools commonly used in statistics and computer science to uncover causal effects and mechanisms. Called Theoretical Implications of Empirical Methods (TIEM), this approach seeks to provide formal foundations for the empirical strategies that are used to study human behavior and society.  

Using “get out the vote” experiments as an example, Fu said statistical methods that are strengthened by theory can promote better understanding of not just the effectiveness of a certain treatment, such as a text message campaign, but why it’s effective. “In the future, this could help people identify the most effective treatment for a particular individual or group,” he said. 

For Fu, part of the attraction of exploring TIEM and causal relationships is the ability to connect with researchers outside his discipline. He views political science at its core as the study of human behavior, but within the realm of political institutions and decision-making. Other social scientists are focused on human behavior in different contexts, such as the sociologist who investigates familial relationships and interactions. He’s excited for the opportunities for collaboration across disciplines that are a hallmark of Duke’s academic community. 

“We use the same language, we use the same tools, but we use them to study different topics,” Fu said. “In political science, the emphasis is more on collective decision-making, because people often make choices as groups rather than as individuals.” 

Fu comes to Trinity after earning his doctorate from New York University and spending a year as a postdoctoral associate at Yale University, but he’s no stranger to Duke. He completed his master’s degree in political science here, an experience that he said will give him a unique perspective in the classroom and in assisting students as they explore potential career paths. “I know what the students need because I was here just like them as a master’s student here,” Fu said.  

Department Chair Pablo Baramendi said that the department is in the middle of a transition across all its fields, and the addition of Fu, along with fellow new faculty members Allison Anoll and Francisco Garfias, is a major step forward to secure generational replacement and intellectual continuity in core areas of strength. 

"Together they are an exceptionally talented group of scholars that will make us better and raise our profile in the years to come," Baramendi said. "Jiawei Fu brings to Duke a very strong footing on frontier issues in political methodology and causal inference, with already visible synergies on multiple applied fields across the department and the university. We look forward to seeing him develop his potential in the years to come."