Mentoring programs are widely assumed to benefit women’s advancement in professional settings, including political science. However, causal evidence is scarce. We conducted a randomized evaluation of the American Political Science Association’s flagship women’s mentoring program for PhDs, the most rigorous evaluation in political science to date. The program consisted of a workshop followed by periodic small-group meetings. We randomized applicants to the program or a control group. We administered surveys pretreatment, immediately after the workshop, and two to seven years afterwards, collecting curriculum vitae and publication data during each wave. The program was rated positively by participants, increasing their sense of belonging in the profession at year 2, but otherwise had null effects. The results hold when we account for treatment uptake and strength; for various cohorts and time frames; and for a range of attitudes, behaviors, and publication metrics. More comprehensive reforms may be needed to make a long-term difference for women in academia.

Despite the centrality of tolerance and free expression to liberal democracy, little is known about the American public’s attitudes toward disruptive protest actions that suppress constitutionally protected speech. Drawing on a nationally representative survey, this article examines the acceptance of shouting down speakers, blocking audiences from attending events, and using violence to stop public speeches across two different question formats: (1) an abstract, “non-group” question; and (2) a “most-offensive-idea” question in which respondents evaluate tactics aimed at speech that they find personally offensive. Across both formats, Gen Z is significantly more accepting of shoutdowns, blockades, and violence than older cohorts. Ideological differences, however, depend heavily on the measurement approach, with liberals and conservatives diverging on the non-group questions but converging on the most-offensive-idea questions. Together, these results reveal a robust generational divide in permissiveness toward speech-suppressing protest and more conditional, context-dependent ideological differences.

Historically, scholarship in race, ethnicity, and politics has focused on voting and protest. It overlooks lobbying as a mode of racial representation. This article introduces a novel dataset of more than 250,000 California bill analyses (1997–2018) to examine lobbying by racial minority interest groups (RMIGs). The data provide rare insight into when RMIGs are present in policy making. Preliminary findings show that interest groups are more influential when embedded in large, diverse coalitions. RMIGs also are more likely to be part of these coalitions. The results reveal that whereas large and diverse coalitions generally increase the probability that bills pass, RMIGs do not benefit when they are in ad hoc or less-coordinated coalitions. These benefits appear for RMIGs only when they are embedded in more coordinated and durable coalitions. These findings challenge the assumption that lobbying is reserved for elite actors and show that coalition-based lobbying is an important and underexamined mechanism of political representation for marginalized groups.

How do professors’ identities, including gender and perceived race and ethnicity, affect students’ course-enrollment preferences? Political science scholars have illustrated clearly the role that identity bias plays in the political arena, but what about the students we teach? Previous research described how gender, race, and ethnicity can impact professors’ career development and teaching evaluations. However, there is little research in terms of the impact of bias on students’ course selection. This study used a choice-based conjoint experiment of first-year university students and found that an instructor’s gender and perceived race and ethnicity have an impact on the courses they select. These results have significant implications for universities evaluation for faculty career success.

Donald Trump and his supporters claimed an historic mandate to provide critical legitimacy for the extraordinary changes the president initiated immediately after taking office. Did he actually receive such a mandate? The data show that he did not. His electoral victory was modest by historical standards, and the public did not view his election as a mandate for sweeping change and did not desire to accord the president additional power. Moreover, Trump did not campaign on many of the issues on which he took action, and the public opposed most of the president’s major changes in policy. These actions included drastically cutting or dismembering congressionally authorized agencies and programs, deporting most undocumented aliens, raising tariffs, and gutting foreign aid.

In his second term, President Donald Trump has pushed beyond the limits that have constrained previous presidents, suggesting that he should not be bound by norms, statutes, or constitutional language. The logic of the US Constitution is that majorities are required for change but that it also is necessary to protect minority interests. The system of checks and balances forces most majorities to grow and broaden before they are empowered. The Constitution provides an incentive for public officials to negotiate and compromise. Donald Trump has ignored this constitutional imperative. Instead of persuasion, compromise, and negotiation, he has employed unilateral action.

Although all presidents pursue their agendas unilaterally, President Donald Trump’s early second-term actions shocked the political system for their scope and breadth. One of Trump’s boldest moves was a frontal assault on Congress’s constitutional power of the purse through unprecedented impoundments and unilateral tariffs. Despite widespread public opposition to Trump’s gambits and clear statutory violations, Congress has offered little resistance, marking a stark departure from historical precedent. This analysis situates Trump’s actions within broader debates over the scope of executive authority and the weakening of institutional checks and balances. Partisan incentives and Trump’s dominance of the Republican Party have muted congressional resistance, raising urgent questions about the future of the separation of powers in an era of unprecedented executive overreach.

As gauged by the factors usually believed to determine legislative success with Congress—party control, seat share, and presidential approval—President Trump was situated similarly during the first years of each of his terms. Yet, he achieved substantially greater legislative success in 2025 than in 2017. This contrast indicates that the conventional explanatory factors no longer fully capture what drives presidential success in an era of intense party polarization. Instead, intraparty dynamics and legislative strategy have an important role in shaping outcomes under unified government today. Understanding contemporary presidential effectiveness requires looking beyond seat shares and public opinion to the internal politics of the governing coalition.

President Donald Trump’s actions to assert control of the executive branch in his second term have been breathtaking in their speed, volume, and diversity. Trump’s actions raise the more general question of how to make sense of his presidency in historical perspective. Presidency scholars have contributed landmark research characterizing presidents based on personal characteristics, historic patterns, common incentives at work in different institutional configurations, and behaviors associated with populist leaders more generally. This article is a distillation of the theory behind Trump’s actions and describes the mechanics of how the president has asserted control over federal personnel in his second term. It concludes with the implications of these actions for governance and for our understanding of the Trump presidency in history more generally.

Donald J. Trump began his second term with a tidal wave of presidential directives. A simple count does not equate to consequence—but, in aggregate, Trump’s directives had an important impact. Many of their targets were consistent with past presidential practice, although even then expanding prior claims to executive authority. Others, which exacted retribution on specific individuals or entities or ordered subordinates to disregard the law, were new and potentially dangerous. The long-term failure of Congress to rein in discretion embedded in past statute has empowered this and future presidencies.

In 2025, President Donald Trump expanded his own powers through unprecedented interpretations of congressional statutes and Article II of the US Constitution. Ensuing waves of litigation and a record number of emergency-relief applications by the administration to the US Supreme Court placed extraordinary pressure on the federal judiciary. Although US district judges have delayed or halted a range of significant administrative actions, this article’s overview of Trump 47 in court highlights three different scholarly approaches that doubt that the US Supreme Court alone can or ultimately will reverse the administration’s agenda. First, the Roberts Court’s emergency docket decisions thus far comport with recent polarization trends in presidential-power cases. Second, the US Supreme Court lacks institutional capacity and consistent jurisprudence to challenge each area of alleged presidential overreach. Third, the administration’s use of broad authorities previously delegated by Congress serves as a reminder that constitutional interpretation and executive-branch powers are rooted in the broader political system. Congress cannot easily retract granted authority but curtailing presidential unilateralism requires more than litigation.

Unilateral activities permeate almost every aspect of Donald Trump’s second presidency. While his use of these tools bears some resemblance to that of past presidents, he departs from established practices in two critical ways. The first difference concerns the sheer audacity of his actions. By employing unilateral directives with little regard for long-standing legal constraints, Trump—more than any of his predecessors—routinely and brazenly defies constitutional and statutory boundaries. Second, his unilateral actions reorient the traditional relationship between power and policy. Whereas past presidents viewed power as a means to advance policy, Trump routinely treats policies as staging grounds for redefining power relations.

Using a citation network approach, this study investigates how the subfield of African politics has evolved since its emergence in the late 1950s by focusing on the influence of African and Africa-based scholars in the top 20 political science journals. We find that African and Africa-based authors are systematically underrepresented in our sample and among the most influential authors today. Starting from a low base, African and Africa-based scholars experienced a period of increasing influence between 2000 and 2010; however, their influence has declined substantially since then. This article highlights two key factors associated with this decline: (1) the rising competitiveness of top-tier political science journals, which increasingly are privileging particular quantitative methodologies that require substantial financial resources and training; and (2) the increasing citation rates of non-African and non-Africa-based scholars in leading political science journals. The article concludes with recommendations that promote greater inclusivity and pluralism, with broader implications for the political science discipline.

Appeals to national security play a central role in contemporary US trade politics. Who drives this security narrative and why? We argue that executive branch actors, regardless of political party affiliation, are more likely to frame trade policy in national-security terms. In Congress, however, we expect Republicans to rely more heavily than Democrats on a national-security narrative. We tested these expectations through a systematic analysis of trade-related discourse by congressional and executive actors from 2001 to early 2025. Using a large language model to examine a substantial corpus of speeches, press releases, and official statements, we find only partial support for our argument: the anticipated partisan difference appears, but security framing is more prevalent in Congress than in the executive branch. Overall, the evidence suggests that actors use security framing as a strategic tool to reinforce their role and confer legitimacy on particular trade policies.

The ability to discuss politics intelligently with others is a central activity in many undergraduate political science courses, often forming a key part of student assessments and contributing significantly to the development of civic competence. This study explores the use of Artificial Intelligence (AI) as a tool to facilitate deliberative discussions among students. Based on our research, we present evidence that student interactions with an AI for political discourse can yield benefits that are similar to human-to-human political deliberation—including an increased openness to sharing views, discovery of common ground, and incremental improvement in respect for differing political views. These findings suggest that AI may offer a promising, scalable, and accessible solution to enhance deliberative learning experiences while potentially overcoming barriers observed in traditional classroom settings.

What is the gender distribution among authors in top-cited articles? To answer this question, we examined the 20 most-cited articles for each of the 201 most frequently used concepts in political science during the past 10 years. Using a sample of 4,020 articles comprising approximately 8,500 authors, we confirmed the underrepresentation of female authors among top-cited articles. Women account for approximately 38% of authors within this canon of articles. On the one hand, this finding is encouraging considering that women’s proportion of authorship did not decline relative to their overall representation among published authors. On the other hand, it is discouraging considering that most recent influential research continues to be written primarily by men. To explain variation in the proportion of female authors per article, we find that women are more likely to engage in single-authored publications and to use qualitative methods. In contrast, we find no significant difference between male and female scholars in the rankings of the journals in which they publish.

The Gender Balance Assessment Tool (GBAT) was introduced in 2016 as a shortcut for researchers and instructors who wanted to quickly determine the gender balance of the authors in their bibliographies and syllabi. In the years since then, some journals and departments have encouraged its use. However, technology also has changed significantly during this period, and the emergence of generative AI models have introduced systems with enormous potential to evaluate the demographic balance of syllabi and bibliographies. By leveraging information on the Internet other than names, and by being less constrained in terms of formatting and name recognition, this article shows that generative AI systems are superior to the GBAT, in terms of both their accuracy and their ability to evaluate general demographic balance rather than only gender balance.

This article offers the first test with valid measures and rigorous multivariate methods of the hypothesis that significant early-career research achievements of academics predict late-career achievements of the same type. Considerable research investigated the correlates of late-career research success; however, this promising hypothesis that originated in psychological and educational research on adolescents has not been tested systematically for academic careers. In a sample of three annual temporal cohorts of political science faculty members, our study finds notable evidence in support of this hypothesis. Our findings also provide a foundation for future research on how innate abilities, the character of doctoral programs, and the advantages of different types of academic positions shape long-term career productivity.

As online platforms seek to improve content-moderation strategies, large language models (LLMs) may be a potential tool. This study examines opportunities and limitations of LLM-powered moderation through a unique lens: student projects for a Stanford University course titled Trust and Safety. In this course, students developed Discord bots using LLMs to moderate specific types of harmful content. Interviews with 16 of the students suggest that these models demonstrate high accuracy, often exceeding students’ expectations. Notably, in cases of disagreement between the student and the model, closer analysis frequently validated the model’s judgments. However, students also observed limitations: LLMs proved unhelpfully sensitive to prompt phrasing and exhibited many contextual interpretation challenges common to human moderators and traditional machine-learning classifiers.

Congress is portrayed as compliant with President Donald J. Trump’s agenda because he is intimidating its members. This neglects an alternative explanation that focuses on the increased congruence of presidential and congressional electoral bases. Trump is the beneficiary of a geographical realignment that took decades and has created a high degree of overlap of the two bases. This analysis tracks that process from 1952 to 2024. It has produced a situation in which policy concerns overlap and encourage congressional compliance.