My research focuses extremism in democracies. I describe my research on this topic in the paragraphs that follow. These projects use a variety of methods, including experiments, surveys, large-language models and AI, text-as-data, and other techniques. I consider the nature of extremism and factors that promote and discourage extremism.
Please contact me for materials relating to any of these papers or projects. For information on my projects under review or working papers, please contact me.
Publications:
Busby, Ethan C., Andrew Ifedapo Thompson, Tomo Vierbuchen, and Suzy Yi. 2025. “Unheard Voices: The Importance of Intersectionality in Responsiveness and the Systematic Ignoring of Black Men by Elected Officials” Political Behavior.
All citizens are not listened to equally, despite the importance of responsiveness and listening to different theories of democracy. We take an intersectional approach to make several novel predictions about how citizens’ identities, the topic of constituent messages, and the identities of elected officials combine to influence responsiveness. These theories lead us to expect (and empirically confirm) that Black men in particular – more than other racial and gender groups – are systematically ignored by elected officials. We implement a wide-scale experiment with U.S. local elected officials (N = 23,738) to test our predictions. Extending previous work, we vary the race, gender, and topic of the constituent’s message, and we observe if elected officials both open and reply to constituents’ messages. We find that Black men are systematically ignored, regardless of the message they send. In contrast, elected officials respond less to Black women when they discuss race and less to White women when they discuss gender. We discuss the implications of this study for work on responsiveness in democratic government.
Lyman, Alex, Bryce Hepner, Argyle, Lisa P., Ethan C. Busby, Joshua R. Gubler, and David Wingate. 2025. “Balancing Large Language Model Alignment and Algorithmic Fidelity in Social Science Research”. Sociological Methods and Research.
Generative artificial intelligence (AI) has the potential to revolutionize social science research. However, researchers face the difficult challenge of choosing a specific AI model, often without social science-specific guidance. To demonstrate the importance of this choice, we present an evaluation of the effect of alignment, or human-driven modification, on the ability of large language models (LLMs) to simulate the attitudes of human populations (sometimes called silicon sampling). We benchmark aligned and unaligned versions of six open-source LLMs against each other and compare them to similar responses by humans. Our results suggest that model alignment impacts output in predictable ways, with implications for prompting, task completion, and the substantive content of LLM-based results. We conclude that researchers must be aware of the complex ways in which model training affects their research and carefully consider model choice for each project. We discuss future steps to improve how social scientists work with generative AI tools.
Busby, Ethan C. Andrew Ifedapo Thompson, and Suzy Yi. 2025. “Do They Even Care? Empirical Evidence for Importance of Listening in Democracy” Political Communication.
What role does listening play in the minds of American citizens? We test key theoretical ideas about listening in contemporary US politics: that listening should be a part of a wide range of political attitudes and considerations, that listening should be defined by citizens in terms of fair consideration, and that listening should have a significant impact on evaluations of government officials. We begin with open-ended questions on salient political topics to present evidence that Americans often think about listening as they think about important features of democratic politics. This is true when discussing different topics (such as responsiveness, extremism, and political problems) and across different parts of the American polity (the public, experts, and politicians). We then document how Americans define listening through the use of open-ended data, finding that Americans frequently think of listening as fair consideration of others’ ideas. Extending these findings with a conjoint experiment, we demonstrate that citizens react to listening from elected officials with more favorability and support in ways that are more powerful than other standard predictors (like policy responsiveness, partisanship, and shared group membership). After these analyses, we end with a discussion of how political science and politics broadly can benefit from an increased focus on listening.
Argyle, Lisa P., Ethan C. Busby, Joshua R. Gubler, Alex Lyman*, Justin Olcott***, Jackson Pond***, and David Wingate. 2025. “Testing Theories of Political Persuasion Using Artificial Intelligence”. Proceedings of the National Academy of the Sciences.
Despite its importance to society and many decades of research, key questions about the social and psychological processes of political persuasion remain unanswered, often due to data limitations. We propose that AI tools, specifically generative large language models (LLMs), can be used to address these limitations, offering important advantages in the study of political persuasion. In two preregistered online survey experiments, we demonstrate the potential of generative AI as a tool to study persuasion and provide important insights about the psychological and communicative processes that lead to increased persuasion. Specifically, we test the effects of four AI-generated counterattitudinal persuasive strategies, designed to test the effectiveness of messages that include customization (writing messages based on a receiver’s personal traits and beliefs), and elaboration (increased psychological engagement with the argument through interaction). We find that all four types of persuasive AI produce significant attitude change relative to the control and shift vote support for candidates espousing views consistent with the treatments. However, we do not find evidence that message customization via microtargeting or cognitive elaboration through interaction with the AI have much more persuasive effect than a single generic message. These findings have implications for different theories of persuasion, which we discuss. Finally, we find that although persuasive messages are able to moderate some people’s attitudes, they have inconsistent and weaker effects on the democratic reciprocity people grant to their political opponents. This suggests that attitude moderation (ideological depolarization) does not necessarily lead to increased democratic tolerance or decreased affective polarization.
Diether, Elle, Suzy Yi, Lisa P. Argyle, and Ethan C. Busby. 2024. “The Political Psychology of Cancel Culture: Value Framing or Group Identity?” Political Research Quarterly.
“Cancel culture” has become a prominent phrase in US political commentary, with supporters and opponents relying on different value-based arguments to express their views. At the same time, these camps often fall along predictable partisan political lines. What, then, are the real motivations for promoting and opposing behaviors sometimes labeled as “canceling” in contemporary American politics? We explore this question through survey data from Pew in 2020 and two original survey experiments, conducted on the 2021 CES and a 2023 online sample. We examine how Americans define the term “cancel culture,” to what extent cancel culture is linked to both a range of core democratic values, and the role of partisanship in shaping support or opposition to specific behaviors. We observe a significant range of views about cancel culture and document connections to various political values. From our experiments, we find that partisanship is a potent driver of support for and understanding of cancel culture and that value-based framing has a weaker impact as compared to partisanship. Cancel culture provides an important case study of how Americans process conflicting norms and values, including free speech and political accountability, beyond the typical constraints of the formal political system.
Busby, Ethan C., Adam J. Howat, and C. Daniel Myers. 2024. “Changing Stereotypes of Partisans in the Trump Era“. Political Science Research and Methods. Online Appendix, Replication data
Stereotypes of the two parties play an important role in political cognition, and a range of recent studies have examined the content and effects of partisan stereotypes. However, little work has studied change in partisan stereotypes over time. We address this question by comparing data on stereotypes of partisans collected before and after the Trump presidency, a time when we might expect individuals’ images of the two parties to undergo significant change. Using a structural topic model, we compare responses to open-ended questions asking respondents to list words describing members of the two parties from 2016 and 2021. We find that partisan stereotypes in the 2021 sample are less group- and issue-based and focused more on personal traits. These results suggest that, during the Trump era, members of the mass public came to see the parties in more personalized, character-focused terms, potentially contributing to affective polarization.
Argyle, Lisa P., Christopher A. Bail, Ethan C. Busby, Joshua R. Gubler, Thomas Howe, Christopher Rytting, Taylor Sorensen, and David Wingate. 2023. “Leveraging AI for democratic discourse: Chat interventions can improve online political conversations at scale” Proceedings of the National Academy of the Sciences, 120(41): e2311627120. Online appendix, Replication data.
Political discourse is the soul of democracy, but misunderstanding and conflict can fester in divisive conversations. The widespread shift to online discourse exacerbates many of these problems and corrodes the capacity of diverse societies to cooperate in solving social problems. Scholars and civil society groups promote interventions that make conversations less divisive or more productive, but scaling these efforts to online discourse is challenging. We conduct a large-scale experiment that demonstrates how online conversations about divisive topics can be improved with AI tools. Specifically, we employ a large language model to make real-time, evidence-based recommendations intended to improve participants’ perception of feeling understood. These interventions improve reported conversation quality, promote democratic reciprocity, and improve the tone, without systematically changing the content of the conversation or moving people’s policy attitudes.
Argyle, Lisa P., Ethan C. Busby, Nancy Fulda, Joshua Gubler, Christopher Rytting, and David Wingate. 2023. “Out of One, Many: Using Language Models to Simulate Human Samples” Political Analysis.
We propose and explore the possibility that language models can be studied as effective proxies for specific human sub-populations in social science research. Practical and research applications of artificial intelligence tools have sometimes been limited by problematic biases (such as racism or sexism), which are often treated as uniform properties of the models. We show that the “algorithmic bias” within one such tool– the GPT-3 language model– is instead both fine-grained and demographically correlated, meaning that proper conditioning will cause it to accurately emulate response distributions from a wide variety of human subgroups. We term this property algorithmic fidelity and explore its extent in GPT-3. We create “silicon samples” by conditioning the model on thousands of socio-demographic backstories from real human participants in multiple large surveys conducted in the United States. We then compare the silicon and human samples to demonstrate that the information contained in GPT-3 goes far beyond surface similarity. It is nuanced, multifaceted, and reflects the complex interplay between ideas, attitudes, and socio-cultural context that characterize human attitudes. We suggest that language models with sufficient algorithmic fidelity thus constitute a novel and powerful tool to advance understanding of humans and society across a variety of disciplines.
“Perceptions of Extremism in the American Public and Elected Officials” 2022, Electoral Studies, Online Appendix, Replication data
Democracies face rising concerns about extremism. At the same time, citizens and political actors define extremism in different ways with different consequences. In light of this, what do American citizens and elected officials consider to be extreme, and what political penalties are associated with extreme behavior and beliefs? To answer these questions, I rely on three surveys, two of the American public and one with American elected officials. Respondents’ open-ended definitions of extremism and two conjoint experiments suggest that Americans do have ideas about extremism that
correspond to research and theorizing about democracy – focusing, for example, on spatial extremity and an unwillingness to listen to others. However, the use of these standards is slanted in favor of one’s own political position and does not indicate a robust recognition or rejection of many normatively troubling forms of extremism.
“Should You Stay away from Strangers? Conducting Experiments on the Political Consequences of Intergroup Contact” 2021, Cambridge University Press, Online access –DOI , Online Appendix
Many theories of democracy suggest that harmonious relationship between groups are critical for democratic societies, and intergroup contact presents an appealing way to encourage these relationships. However, what types of contact encourage harmony between groups? In this Element, I review existing studies of contact, propose a framework for productively studying the political consequences of intergroup contact through experiments, and discuss four experiments following these recommendations. These studies focus on the consequences of the difficulty of communicating in the contact experience and are conducted on different samples and contexts to present a broad picture of the role of different forms of contact. I find that easy-to-understand forms of contact do not promote more political support for racial and ethnic outgroups and can, in some conditions, provoke a backlash consistent with theories about social identity and group threat. I conclude by discussing the implications of the substantive findings and the proposed framework for studying contact.
Busby, Ethan C., Adam J. Howat, Richard Shafranek, and Jacob Rothschild. “The Partisan Next Door: Stereotypes of Party Supporters and Consequences for Polarization in America” 2021, Cambridge University Press, Online Access – DOI, Online Appendix
Partisan polarization in the United States has, in recent decades, risen to near-unprecedented levels. Supporters of the Democratic and Republican Parties increasingly express a deep, fundamental dislike of opposing partisans, regardless of whether they disagree on policy. This phenomenon, known as affective polarization (Iyengar et al. 2019), arises from partisanship’s growing role as a social identity in its own right, comparable in its nature and consequences to race, religion, or gender (Green, Palmquist, and Schickler 2002; Greene 1999, 2004). In other words, partisanship shapes not only people’s expressly political orientations and behaviors; it also constitutes, for many individuals, a key component of their self-concepts—and, by extension, how they socially categorize others.
In the context of such polarization, how, precisely, do everyday party supporters think about each other? What images come to mind when ordinary people think of these ordinary partisans? Moreover, what are the consequences for polarization of holding different images of mass partisans in one’s head? Using both observational and experimental data, our book provides answers to these questions.
Our findings show that a large portion of the American public hold clear, sensible ideas about party supporters. These stereotypes tend to cohere around traits, political issues, and groups—in line with the identity, instrumental, and group views of partisanship, respectively. Moreover, the specific images people hold of run-of-the-mill partisans—and not merely the affective evaluations attached to those images—matter when it comes to relations between the parties and partisans. Thinking of everyday partisans in terms of what they “are like,” particularly when it comes to individual character traits, tends to accentuate polarization of all kinds. However, if people can be induced to take a more instrumental view, focusing on the different issues that the parties consider important, this may help to reduce partisan animus, facilitate political compromise, and promote social harmony between supporters of the two parties.
Thompson, Andrew Ifedapo and Ethan C. Busby. 2021. “Defending the Dog Whistle: The Role of Justifications in Racial Messaging“. Political Behavior. Replication data
American politicians frequently evoke race in their messages to the public; at the same time, politicians often pay a price for racialized rhetoric. We propose that elites continue to use messages about race because they can mitigate the costs of doing so with justifications for their original statements. Integrating literatures on elite rhetorical tactics and framing, we predict that when justifications and implicit racial messages are combined, elites can mobilize the support for racially resentful Whites without alienating others. In a pair of survey experiments conducted in 2019 and 2020, we examine the effectiveness of justifications in swaying Whites’ attitudes. We find that two different elite justifications bolster support for their messages. Importantly, we also find these tactics do not incur political costs. This provides a compelling reason that political figures continue to use racial messages in politics despite recent social movements and possible shifts in Americans’ attitudes about race.
Busby, Ethan, Kirk A. Hawkins, and Joshua R. Gubler. 2019. “Framing and Blame Attribution in Populist Rhetoric“. Journal of Politics 81(2):616-630. Replication data
Scholars have long known that the rhetoric of populist politicians is an important part of their appeal; however, less is known about how that rhetoric operates. Drawing on data from two large experiments conducted with American adults, we show that survey questions encouraging individuals to consider political problems within a dispositional blame frame activates latent populist attitudes, while an encouragement to consider these same problems in a situational blame frame does not. In our second experiment, we connect this framing change to voting intentions and find that subjects exposed to dispositional frames are more likely to express support for Donald Trump and less likely to express support for Hillary Clinton than subjects exposed to situational frames. Importantly, the impact of framing is contingent on pre-existing populist attitudes; subjects with moderate levels of populist attitudes are much more likely to demonstrate an increase in expressed populism and support for Trump.
Rothschild, Jacob, Adam Howat, Richard Shafranek, and Ethan Busby. 2019. “Pigeonholing Partisans: Mass Stereotypes of Party Supporters and Partisan Polarization“. Political Behavior 41(2): 423-443. Replication data.
What comes to mind when people think about rank-and-file party supporters? What stereotypes do people hold regarding ordinary partisans, and are these views politically consequential? We utilize open-ended survey items and structural topic modeling to document stereotypes about rank-and-file Democrats and Republicans. Many subjects report stereotypes consistent with the parties’ actual composition, but individual differences in political knowledge, interest, and partisan affiliation predict their specific content. Respondents varied in their tendency to characterize partisans in terms of group memberships, issue preferences, or individual traits, lending support to both ideological and identity-based conceptions of partisanship. Most importantly, we show that partisan stereotype content is politically significant: individuals who think of partisans in a predominantly trait-based manner – that is, in a way consistent with partisanship as a social identity – display dramatically higher levels of both affective and ideological polarization.
Busby, Ethan and James N. Druckman. 2018. “Football and Public Opinion: A Partial Replication and Extension“. Journal of Experimental Political Science 5(1):4-10. Replication data.
Do events irrelevant to politics, such as the weather and sporting events, affect political opinions? A growing experimental literature suggests that such events can matter. However, extant experimental evidence may over-state irrelevant event effects; this could occur if these studies happen to focus on particular scenarios where irrelevant event effects are likely to occur. One way to address this possibility is through replication, which is what we do. Specifically, we replicate an experimental study that showed the outcome of a college football game can influence presidential approval. Our results partially replicate the previous study and suggest the impact is constrained to a limited set of outcome variables. The findings accentuate the need for scholars to identify the conditions under which irrelevant effects occur. While the effects clearly can occur, there relevance to politics remains unclear.
Busby, Ethan, James N. Druckman, and Alexandria Fredendall. 2017. “The Political Relevance of Irrelevant Events”. Journal of Politics 79(1): 346-350. Replication Data
Do events irrelevant to politics affect citizens’ political opinions? A growing literature suggests that such events (e.g., athletic competitions, shark attacks) do in fact shape political preferences. We present an experiment that largely replicates a widely noted irrelevant event effect. Specifically, we find that the outcome of a sporting event (i.e., a football game) affects presidential approval and likely does so by affecting individuals’ moods. We also show that the effect is short-lived.
Busby, Ethan, D.J. Flynn, and James N. Druckman. 2018. “Studying Framing Effects on Political Preferences: Existing Research and Lingering Questions.” Doing News Framing Analysis II (Paul D’Angelo, ed.). New York: Routledge, p. 27-50.
This chapter reviews the state of framing research, noting strengths, weaknesses, and unexplored areas of this literature.
Hawkins, Kirk, A., Levente Littvay, and Ethan C. Busby. 2019. “The Causes of Populism: Explaining the Victory” Contemporary US Populism in a Comparative Perspective (Kirk A. Hawkins and Levente Littvay). New York: Cambridge University Press.
This chapter, as part of a larger book on populism in the United States from a comparative perspective, explores the insights research on populism brings to understanding the 2016 elections and the victory of Donald Trump. The chapter relies on survey data from the United States in 2008, 2012, and 2016 to explore this topic. We show that populist attitudes is a key driver of support for populist movements and candidates, even controlling for authoritarianism, immigration attitudes, and racial resentment.
Busby, Ethan C., David Doyle, Kirk A. Hawkins, Nina Wiesehomeier. 2018. “Activating populist attitudes: the role of corruption.” The ideational approach to populism: Concept, theory, and method. (Kirk A. Hawkins, Ryan Carlin, Levi Littvay, and Cristóbal Rovira Kaltwasser, eds.). New York: Routledge.
Recent advances in the study of populism, particularly at the individual-level, have conclusively demonstrated the importance of elite framing for activating and directing latent populist attitudes among the electorate. Ideational work in this vein however, has yet to explain exactly how policy failure and the political context also incentivize populist behavior. We expect that policy failures that result from intentional misconduct are more likely to trigger a populist response than those resulting from negligence or external causes, and that such a response is likely to be greater when this type of corruption is endemic. We tested these theoretical expectations in two laboratory-based experiments: one on a student sample at Northwestern University in the United States, the other on a mixed student/resident sample at Oxford University in the United Kingdom.
Ethan C. Busby 