Did George Floyd’s murder and its ensuing protests produce a racial reckoning? Conventional social-science accounts, emphasizing the stability of racial attitudes, dismiss this possibility. In contrast, we theorize how these events may have altered Americans’ racial attitudes, in broadly progressive or in potentially countervailing ways across partisan and racial subgroups. An original content analysis of partisan media demonstrates how the information environment framed Black Americans before and after the summer of 2020. Then we examine temporal trends using three different attitude measures: most important problem judgments, explicit favorability towards Whites versus Blacks, and implicit associations. Challenging the conventional wisdom, our analyses demonstrate that racial attitudes changed following George Floyd’s murder, but in ways dependent upon attitude measure and population subgroup.

Restrictive voting laws are an increasingly salient feature of American politics. Yet estimating their direct impact on turnout is challenging, given the strategic actions political actors take to impose and mitigate the costs of these laws. Using individual-level data from Davidson County, Tennessee, we leverage variation induced by an early-morning tornado on Super Tuesday 2020 to estimate the direct causal effect of polling-site consolidations. We find moving to a new polling station decreases in-person turnout by 5.65 percentage points, on average, and that the variable cost—proxied by change in travel distance—drives almost all of this decline. Voting at a consolidated site only decreases turnout when the number of individuals assigned to a station increases by more than 100%.

Can observing opposing partisans engage in dialogue depolarize Americans at scale? Partisan animosity poses a challenge to democracy in the United States. Direct intergroup contact interventions have shown promise in reducing partisan polarization, but are costly, time-consuming, and sensitive to subtle changes in implementation. Vicarious intergroup contact—observing co-partisans engage with outparty members—offers a possible solution to the drawbacks of direct contact, and could potentially depolarize Americans quickly and at scale. We test this proposition using a pre-registered, placebo-controlled trial with a nationally representative sample of Americans. Using both attitudinal and behavioral measures, we find that a 50-minute documentary showing an intergroup contact workshop reduces polarization and increases interest but not investment in depolarization activities. While we find no evidence that the film mitigates anti-democratic attitudes, it does increase optimism about the survival of democratic institutions. Our findings suggest that vicarious intergroup contact delivered via mass media can be an effective, inexpensive, and scalable way to promote depolarization among Americans.

How can local governments in developing countries, constrained by limited resources, identify and respond to the most pressing public demands? This paper posits that public deliberative platforms, even those with controlled agendas, can be instrumental in this regard by facilitating communication between local elites and ordinary citizens, thereby leading to an observable uptick in political trust over time. Public deliberation serves two functions: firstly, it highlights shifting societal issues, incentivizing bureaucrats to respond more promptly; and secondly, it generates narratives that temporarily improve the public perception of local governments, even among individuals not directly benefiting from government actions. This study provides evidence consistent with these theoretical implications by examining Chinese topical debate programs, during which local officials engage with citizens and respond to their concerns. Empirical results based on a staggered difference-in-differences design suggest that broadcasting such programs in China produces a prompt increase in citizens’ trust in local officials. Our results demonstrate that public deliberation can yield noticeable outcomes in developing countries, even with controlled agendas and constrained resources.

Presidents are often viewed as national policy leaders. Yet, they increasingly use negative rhetoric to attack the opposition rather than forge legislative compromise, contrary to theories of going public. Why? I argue presidents facing congressional obstruction eschew short-term policy persuasion. They speak as negative partisans to mobilize co-partisans and shape the longer-term balance of power in Congress, improving future policy-making prospects. I collect all presidential speeches delivered between 1933 and 2024 and use transformer methods to measure how often, and how negatively, presidents reference the out-party. They do so when the policy-making environment is unfavorable: when majorities are tenuous, government is divided, and as elections approach. I provide additional support with a case study of Democrats’ 2009 filibuster-proof Senate majority. Finally, this rhetoric has behavioral impact: presidential negative partisanship decreases co-partisan approval of the opposition. This research alters our understanding of going public and reinforces the partisan dimension of modern presidential representation.

This article describes how and why propaganda affected recipients differently in two distinct situations, namely forced exposure and selective exposure, when they received propaganda during a series of six original survey experiments conducted in China. The prevailing view is that people are more likely to resist information they receive if their exposure to it is forced. But the study addressed in this article found that citizens who prefer not to view propaganda news, when given a choice, actually demonstrate higher average treatment effects on pro-regime attitudes compared to those who willingly read propaganda news (i.e. where participants in the control group were assigned a reading of non-propaganda news). Moreover, this study shows that participants who prefer not to read propaganda news exhibit higher average treatment effects when rating the issue presented in the news as more understandable and important—compared to those who willingly engage with the propaganda. That suggests a possible rationalization pathway in this phenomenon.1

While research shows that public preferences across policy domains tend to move in parallel, the mechanisms behind this dynamic remain unclear. We examine four explanations: (1) alignment in preferred policy levels; (2) parallel policy movement combined with domain-specific thermostatic feedback; (3) feedback to global policy across domains; and (4) responsiveness to presidential partisanship. These mechanisms matter for how we interpret public opinion change and policy responsiveness. We develop and test a theoretical model using data on four social spending domains in the USA. Our findings suggest that spending mood reflects both parallelism in preferred policy levels and responsiveness to overall social spending and presidential party affiliation.

When and why do wealthy individuals support redistribution? Under standard political economy models, preferences for redistribution are a function of material conditions. The partisanship literature, on the contrary, argues that partisan identification determines redistributive preferences. We move beyond this dichotomy to argue that the ideology of the government enacting redistribution is a key factor explaining support for redistribution among the wealthy. Through survey experiments during the 2022 Colombian election, we find that the wealthy are more likely to support redistribution under a right-wing government and expect redistribution under the Right to be more efficient and less economically disruptive. We find heterogeneous treatment effects across ideological groups. However, regardless of ideology, the wealthy do not expect macroeconomic instability from right-wing redistribution.

How did the COVID-19 outbreak affect citizens’ democratic preferences? Were the changes persistent or temporary? We track a representative sample of Spanish citizens before, during, and after the pandemic, with eight survey waves from January 2020 to January 2024. We compare democratic attitudes before and after the pandemic with individual fixed effects models. We identify a sharp increase in preferences for technical rather than ideological policy-making at the very onset of the pandemic, as well as significant changes in voters’ preferences for competent rather than honest politicians. These changes are sudden and persistent over 4 years. Using a set of repeated survey experiments, we also document a widespread willingness to sacrifice rights and freedoms to deal with the pandemic as compared to other global threats, such as international terrorism and climate change. But this effect quickly faded over time. Overall, we identify significant changes in democratic attitudes during the pandemic and a durable shift in technocratic preferences that outlived the pandemic, setting the conditions for the long-term legacies of COVID-19 on democracy.

Political professionals and scholars maintain that raising money early in the election season is critical to a successful campaign, having downstream consequences on a candidate’s future fundraising potential, the stiffness of competition she will face, and her likelihood of electoral victory. In spite of early money’s perceived importance, there is no common operationalization for money as “early.” Moreover, existing measures often fail to reflect definitional aspects of early money. In this paper, we first lay out a theoretical framework regarding the utility of early campaign fundraising for candidates. We argue that early fundraising can be expressed as two conceptually distinct quantities of interest centered on either a candidate’s own fundraising performance (candidate-centered) or her fundraising performance relative to her electoral competitors (election-centered). We next lay out steps for operationalizing candidate- and election-centered measures of early fundraising. Lastly, we demonstrate that both our proposed measures for early campaign fundraising are predictive of a candidate’s future fundraising and electoral success. By putting forward a set of best practices for early money measurement and, additionally, producing off-the-shelf measures for early fundraising in U.S. House elections, we fill an important gap in scholarly research on the measurement of money in politics.

This paper studies voter turnout and selective abstention on voting days with more than one election or referendum. We extend the rational choice model to a setting with multiple concurrent votes. The model is based on a voter’s net benefit, which includes a vote’s salience and information costs. It explains how the net benefit of different concurrent votes enters a voter’s utility function and thereby affects turnout and selective abstention, the tendency to vote in one but not all votes held on the same day. We test our theoretical predictions using data on concurrent propositions in Switzerland from 1988 to 2016. Our results suggest that the proposition with the highest net benefit and the sum of the net benefits of all concurrent propositions are relevant determinants of the individual turnout decision. We also find that a proposition’s net benefit explains variation in selective abstention.

Although much has been written on legislative reciprocity, rarely have scholars had an opportunity to leverage a randomly assigned asset to assess whether and how legislators reciprocate when their colleagues assist them. Using the lottery that allows Canadian Members of Parliament (MPs) to propose bills or motions, we examine whether MPs’ priority numbers affect their proclivity to second motions made by other MPs, which would be expected if MPs sought to build support for their own proposals by supporting proposals by others. Although MPs almost always make a proposal if their priority number allows them to do so, we find a weak relationship between MPs’ priority numbers and their probability of seconding others’ proposals. Moreover, when we look at successive parliaments, we see only faint indications that those who, by chance, won the right to propose in the previous session (and who therefore were eligible to attract seconds) are more likely to second others’ proposals in the current session. Although subject to a fair amount of statistical uncertainty that will gradually dissipate as future parliaments are examined, this pattern of evidence currently suggests that correlated seconding behavior among legislators is more the product of homophily than reciprocity.

This study examines whether women politicians address violence against women (VAW) more effectively than their male counterparts at the local level in Mexico. Using a regression discontinuity design that leverages close mayoral elections, we find that women mayors reduce some of the most egregious violent crimes committed against women, with estimates suggesting a 64.7% reduction in homicides of women over their 3-year terms. As evidence of potential mechanisms, we find that women mayors actively work to combat VAW, appoint more women to leadership and support roles, and expand specialized services for crime victims. These findings suggest that women’s representation in local politics may be an important factor in advancing women’s safety.

How does a ruler implement state-building at the local level? This paper examines state-building in late 16th-century Japan by focusing on Toyotomi Hideyoshi’s land surveys, which were crucial for establishing a centralized regime. We argue that Hideyoshi strengthened control over the locality via land surveys as a strategic response to the perceived threats emanating from Catholic missionaries. Using various empirical strategies including spatial econometrics, sensitivity analysis, and an instrumental variable approach, we find that the presence of Catholic churches significantly increased the likelihood of a locality being surveyed. These results highlight the importance of information-gathering beyond fiscal purposes for security objectives and emphasize the role of threats from foreign religious institutions in state formation processes.

The increasing multimodality (e.g., images, videos, links) of social media data presents opportunities and challenges. But text-as-data methods continue to dominate as modes of classification, as multimodal social media data are costly to collect and label. Researchers who face a budget constraint may need to make informed decisions regarding whether to collect and label only the textual content of social media data or their full multimodal content. In this article, we develop five measures and an experimental framework to assist with these decisions. We propose five performance metrics to measure the costs and benefits of multimodal labeling: average time per post, average time per valid response, valid response rate, intercoder agreement, and classifier’s predictive power. To estimate these measures, we introduce an experimental framework to evaluate coders’ performance under text-only and multimodal labeling conditions. We illustrate the method with a tweet labeling experiment.

Despite widespread adoption of international anti-money laundering standards over the last 30 years, their effectiveness remains poorly understood due to persistent data limitations. I address this gap in the scholarship by leveraging cryptocurrency transaction data to assess how specific regulatory design features shape compliance. Using bunching estimation, I demonstrate that customers strategically adjust transaction sizes to avoid threshold-based screening requirements, while exchanges fail to adequately address this behavior through risk-based monitoring. Analysis of British Virgin Islands exchanges using difference-in-differences estimation before and after regulatory changes provides additional evidence supporting these conclusions. The findings reveal how regulatory design features shape behavior in cryptocurrency markets and suggest specific improvements for regulatory frameworks.

As the use of computational text analysis in the social sciences has increased, topic modeling has emerged as a popular method for identifying latent themes in textual data. Nevertheless, concerns have been raised regarding the validity of the results produced by this method, given that it is largely automated and inductive in nature, and the lack of clear guidelines for validating topic models has been identified by scholars as an area of concern. In response, we conducted a comprehensive systematic review of 789 studies that employ topic modeling. Our goal is to investigate whether the field is moving toward a common framework for validating these models. The findings of our review indicate a notable absence of standardized validation practices and a lack of convergence toward specific methods of validation. This gap may be attributed to the inherent incompatibility between the inductive, qualitative approach of topic modeling and the deductive, quantitative tradition that favors standardized validation. To address this, we advocate for incorporating qualitative validation approaches, emphasizing transparency and detailed reporting to improve the credibility of findings in computational social science research when using topic modeling.

The boom in survey experiments in international relations has allowed researchers to make causal inferences on longstanding foreign policy debates such as democratic peace, and audience costs. However, most of these experiments rely on mass samples, whereas foreign policy is arguably more technocratically driven. We probe the validity of generalizing from mass to elite preferences by exploring preferences of ordinary U.S. citizens and foreign policy experts (employees of the U.S. Department of State) in two identical conjoint experiments on democratic peace. We find that experts are not only more opposed to military actions against other democracies than members of the public—but also that overall preferences about the matters of war and peace are stronger among foreign policy professionals.