Social scientists often classify text documents to use the resulting labels as an outcome or a predictor in empirical research. Automated text classification has become a standard tool since it requires less human coding. However, scholars still need many human-labeled documents for training. To reduce labeling costs, we propose a new algorithm for text classification that combines a probabilistic model with active learning. The probabilistic model uses both labeled and unlabeled data, and active learning concentrates labeling efforts on difficult documents to classify. Our validation study shows that with few labeled data, the classification performance of our algorithm is comparable to state-of-the-art methods at a fraction of the computational cost. We replicate the results of two published articles with only a small fraction of the original labeled data used in those studies and provide open-source software to implement our method.

Although a master category of contemporary social and political thought, the conceptual import of colonialism has long been contested. Turning to the political thought of Jawaharlal Nehru, this article reconstructs the surprising, and often surreptitious, intellectual transformations that rendered colonialism into a generic name for European rule over Asia and Africa. It demonstrates how the dual claims of colonialism—a historical reference for the global event of European expansion and a threadbare analytical definition for a particular form of rule—generated a powerful framework in the anticolonial age. While the expanded juridical uses of colonialism in the Cold War era undermined its historical claims, the priority would reverse with its postcolonial re-signification as a shorthand for studying the paradoxes of global modernity. Reframing these debates, this article argues that reflexive navigation between the dual claims of colonialism is key to a capacious appreciation of its historical and normative contentions.

Migrant labor exploitation is widespread in developed countries, which host growing populations of undocumented migrants. While denouncing by migrants is essential to prosecute exploitative employers, an undocumented community actively hiding from the state is unlikely to whistleblow. I consider an intervention providing migrant farmworkers in Italy information and incentives to report on their racketeers. I leverage the staggered rollout of the intervention to study its effects in a difference-in-differences framework. The intervention empowered migrants to whistleblow, increased the prosecution of criminal organizations responsible for racketeering migrants, and raised awareness among natives, who became more favorable toward immigration and parties supporting it. These findings highlight the conditions under which undocumented migrants can take political action for their socioeconomic advancement. Unlike other integration policies which have been shown to backlash, highlighting migrants’ vulnerability to exploitation might foster solidarity and more liberal immigration attitudes among natives.

Plea bargaining figures heavily in criminal justice systems in the United States and, increasingly, around the globe. Conventional wisdom holds that plea bargaining generates efficiency gains for all parties, while sorting the guilty from the innocent. We build a series of formal models to consider the relationship between a defendant’s guilt and her likelihood of pleading guilty. In an inversion of the conventional wisdom, we show that under a range of empirically plausible scenarios—for example, if criminals are more risk-seeking than the wrongfully accused, or if prosecutors derive a career benefit from trial wins—the innocent are more likely than the guilty to plea bargain.

Since Key and Allport, scholars have argued that racial context affects political behavior, with some finding out-group contact increases intergroup hostility and others showing the opposite. We argue that Americans exist in multiple racial contexts simultaneously that may overlap or conflict, helping to explain past discord. Using novel data, we document in-group embeddedness among the four largest U.S. ethnoracial groups for three kinds of racial context: geographic, social, and psychological. These three contexts are only weakly correlated, we find, with social ties exhibiting distinctly high rates of in-group segregation. We next examine the relationship between racial contexts and political attitudes, showing that individuals who are highly embedded across contexts express notably different views than those who experience cross-cutting pressures. Our results underscore a need for greater care and specificity when examining the relationship between “racial context” and political phenomena.

There is a large literature on the use of cases, hardly anything on examples. They are different: cases get analyzed, examples get deployed. Examples can perform clarifying, didactic, persuasive, universalizing, critical, and cogitative functions. These six functions all have their own logic, and a set of guidelines for how to perform each of them well is developed. However, compelling and persuasive examples can also mislead. Following Kahneman’s distinction between system 1 (fast) and system 2 (slow) thinking, good examples both resonate in system 1 terms and invite system 2 scrutiny. The best examples are good in some aspect, flawed in interesting ways in others. A perfect example is a logical impossibility. The author’s interest in convincing the reader and the discipline’s interest in effective inquiry can diverge, a problem overcome if reason in inquiry is seen as essentially dialogical.

Introducing an international relations perspective into the literature on anti-immigrant attitudes, we hypothesize that immigrants from rival countries will be shunned and immigrants from allied countries preferred, especially by respondents who identify more strongly with the nation. We fielded a forced-choice conjoint experiment in 22 countries, whereby respondents chose between applicants for permanent resident status with randomized attributes. We identified rival and allied countries of origin for each surveyed country, with one such pair sharing a similar racial and cultural make-up as the majority of respondents, and one pair being more dissimilar. We find that discrimination against immigrants from rival states is so pronounced that it results in a net preference for racially and culturally dissimilar immigrants. Since we fielded the surveys amidst the Russian invasion of Ukraine, we are able to leverage exogenous changes in the intensity of one rivalry, providing further evidence for the proposed mechanism.

Mainstream parties have often shifted from initially portraying new competitors as undemocratic pariahs (i.e., a delegitimizing strategy) to portraying the same parties later as democratic (i.e., a legitimizing strategy). I argue that voters follow mainstream parties’ legitimizing strategies in their legitimacy evaluations of these parties. I investigate this argument with two independent survey experiments and a quantitative media content analysis in two countries that differ sharply in the nature of party competition—from mainstream parties delegitimizing a far-right party (i.e., Germany) to mainstream parties legitimizing it (i.e., Sweden). I find strong evidence that (a) mainstream parties can effectively legitimize pariah parties in the eyes of voters, (b) turning back to delegitimization has little effect, (c) legitimization is no less effective in the face of a third party’s delegitimizing strategy, and (d) legitimization resonates beyond co-partisans. The results suggest that mainstream party legitimization of pariah parties has far-reaching consequences.

Political scientists have long viewed values as a source of constraint in political belief systems. More recently, scholars have argued that values—particularly moral values—contribute to polarization. Yet, there is little direct and systematic research on which values are perceived as moral values. We examine 21 values, including Schwartz’s values, political values, and moral foundations. Drawing on a broad literature on cooperation, we first develop theoretical expectations for the extent of value moralization both between and within value systems. We next argue that this moralization matters because it intensifies the effects of value disagreement on social polarization. Using a probability-based survey of the US and an embedded conjoint experiment, we find substantial variation in moralization across values, and that highly moralized values are more polarizing. Our research brings together competing literatures on values and shows how moral values differentially shape polarization.

A long-standing debate in political psychology considers whether individuals update their beliefs and attitudes in the direction of evidence or grow more confident in their convictions when confronted with counter-attitudinal arguments. Though recent studies have shown that instances of the latter tendency, which scholars have termed attitude polarization and “belief backfire,” are rarely observed in settings involving hot-button issues or viral misinformation, we know surprisingly little about how participants respond to information targeting deeply held attitudes, a key condition for triggering attitude polarization. We develop a tailored experimental design that measures participants’ core issue positions and exposes them to personalized counter-attitudinal information using the large language model GPT-3. We find credible evidence of attitude polarization, but only when arguments are contentious and vitriolic. For lower valence counter-attitudinal arguments, attitude polarization is not detected. We conclude by discussing implications for the study of political cognition and the measurement of attitudes.

Studies have found that women’s representation is more likely to spike after corruption scandals. However, the mechanism underlying this increase remains unclear: are parties more likely to nominate women after corruption scandals, are voters more likely to support women candidates, or is it a combination of both? Using an original dataset of audit results and the gender of 47,000 candidates running in over 10,000 mayoral elections in Mexico (2000–2019), we find that voters drive the effect. While political parties are not more likely to nominate women as candidates in municipalities with recent revelations of spending irregularities, women candidates are more likely to win elections after corruption is uncovered. In contrast to previous studies, which expect strategic parties to be behind the increases in women’s representation following corruption scandals, our findings underscore that increases in women’s representation can happen despite parties and not because of parties.

Many political scientists hold that vitriolic speech against high-profile women has only negligible effects on other women in politics. They also contend that the prevalence of such vitriol is consistent with gender bias having no significant negative impact on most women in politics. This article argues that these sanguine positions rest on inadequate and untested assumptions regarding misogyny, and the role it plays in politics. In the service of putting us in a position to test the relationship between gendered discourse in particular elections and the structural obstacles faced by women in politics, this article develops a conceptualization of political misogyny: nasty claim-making that instills repugnant connotations into women’s collective political identities (e.g., their partisan identities). Attention is also paid to how political misogyny can distribute hatred burdens disproportionately among different groups of women.

The apocalypse is frequently deployed by political movements, especially contemporary climate activists, to advance their causes. This article develops a framework for defending such invocations of the end of the world. With many other political theorists, I suggest that the apocalypse is a dangerous concept, not least because of its association with authoritarian accounts of history. However, we should not reject the apocalypse. I argue for a form of anti-anti-apocalypticism, using the criticisms directed against the concept as a launchpad to rethink it in viable terms. While acknowledging the value of different ways of defending the apocalypse, I highlight the importance of the causes of apocalyptic movements. Simply put, apocalypses from below are defensible because they have the capacity to clarify the political position of the oppressed and open new political possibilities for the group. By contrast, apocalypses from above, because they fail to fulfill these functions, are not.

In this research letter, we examine whether gender and racial bias affect interruption rates at one of the most visible events in American politics: US Supreme Court confirmation hearings. Using original data from 1939 to 2022, we find that male and white participants are more likely to interrupt women and person of color speakers, respectively, relative to male and white speakers. This finding holds for both senators and nominees as interrupters. Our results provide evidence that biased interruptive behavior occurs in even the most public and salient of political settings and that it can be mitigated (or intensified) by shared (or opposite) partisanship among speaking pairs. We also find interruption inequalities are not isolated to women as the interrupted, revealing that people of color in political and legal settings are subject to heightened rates of interruptions as well.

We combine newly collected election data with records of public denials of the results of the 2020 election to estimate the degree to which election-denying Republican candidates over- or underperformed other Republicans in 2022 in statewide and federal elections. We find that the average vote share of election-denying Republicans in statewide races was approximately 3.2 percentage points lower than their co-partisans after accounting for state-level partisanship. However, we find no such underperformance on aggregate for U.S. House elections, perhaps due to the more-partisan nature of many House districts. Together, the results suggest that the types of candidates in American elections who take more-extreme positions tend to underperform, but that these performance gaps are relatively small in the present, polarized political environment.

Expanding suffrage is critical for democratic inclusion. In the United States, noncitizen residents are the latest focus of such (re)enfranchising efforts. Public opinion plays a significant role in the passage of legislation expanding or restricting noncitizen access to local elections. Although elite support for noncitizen suffrage is well-documented, little is known about public opinion toward such noncitizen voter policies. What accounts for voter support for noncitizen electoral participation? We argue that the partisan alignment between noncitizens and U.S. voters shapes U.S. voters’ support for noncitizen voting rights. Evidence from two survey experiments suggests that U.S. voters are pragmatic in their enfranchising preferences: voters increase their support for co-partisan enfranchisement but oppose this same policy when considering out-partisans. These dynamics are present among both Republicans and Democrats, underscoring the societal implications of heightened partisanship on American democracy.

New evidence suggests that the world recently changed for the U.S. Supreme Court owing to its decision to abrogate the abortion rights first announced in Roe v. Wade. In contrast to the conventional wisdom that Court support is little undermined by unpopular rulings, the Dobbs decision generated a substantial knock on the Court’s legitimacy. Two crucial frailties limit these findings, however. First, no one has determined whether the lost legitimacy has persisted, since earlier research relied on a one-shot survey conducted shortly after the decision. Second, no analysis has addressed the “values-based regeneration” hypothesis—that support reemerges not long after a legitimacy hit is inflicted. Based on a nationally representative 2023 survey, my analysis finds that the lost legitimacy has lingered, but institutional support may be being rebuilt owing to its close connection with democratic values. Overall, I conclude that understanding persistence is more complicated than many may have assumed.

For a member of Congress to be a successful lawmaker, they must work collaboratively with their colleagues. Previous work has found that interpersonal relationships among lawmakers are a key predictor of legislative collaboration—particularly among members from opposing parties. In the wake of the events of January 6, 2021, many Democratic lawmakers claimed that their relationships with some of their Republican colleagues were irretrievably broken and they would be unable to continue collaborating with members who voted against certifying electoral votes from the 2020 presidential election. Using data on original cosponsorship and legislative effectiveness from the 115th to 117th Congresses (2017–22) we find that Republican lawmakers who voted against the certification of electoral votes lost bipartisan collaborators and saw a drastic decrease in their legislative effectiveness in the 117th Congress.