Research indicates that increasing diversity in doctoral programs can positively affect students’ academic success. However, little research examines students’ responses to female scholars’ representation. The two studies presented here examine how students’ exposure to female academic role models shapes students’ attitudes toward their own academic success (i.e. self-efficacy). Such attitudes are critical because they predict student retention rates. In our first study, we randomly exposed 297 Ph.D. students in one academic discipline to either a gender-diverse (i.e. 30% female authors) or non-diverse syllabus in research methods (i.e. 10% female authors). We examined the effect of the intervention on students’ perceived likelihood of succeeding in the hypothetical course. Contrary to expectations derived from the literature, we found that increasing women’s representation in syllabi did not affect female students’ self-efficacy. Rather, male students expressed lower self-efficacy when evaluating the more gender-diverse syllabus. We also found that students’ attitudes toward diversity in academia predicted their reactions more strongly than did their own gender: gender-diverse syllabi reduced self-efficacy among those students unsupportive of diversity. In our second study, we analyzed non-interventional survey questions to examine the relationship between female role models and long-term academic self-efficacy. Analysis was observational and thus did not assess causality. We found that students with more role models have higher academic self-efficacy, irrespective of student and role model gender. Nonetheless, results also suggested that some students actively seek female role models: namely, female students, and particularly those valuing diversity. Our results ultimately suggest that exposure to female role models relates in surprising ways to Ph.D. students’ self-efficacy. Having more female role models correlates with greater expectations of academic success among certain groups of students, but with diminished expectations of academic success among other groups.

No abstract

Most research on diversity within political methodology focuses on gender while overlooking racial and ethnic gaps. Our study investigates how race/ethnicity and gender relate to political science PhD students’ methodological self-efficacy, as well as their general academic self-efficacy. By analyzing a survey of 300 students from the top 50 US-based political science PhD programs, we find that race and ethnicity correlate with quantitative self-efficacy: students identifying as Black/African American and as Middle Eastern/North African express lower confidence in their abilities than white students. These gaps persist after accounting for heterogeneity among PhD programs, professional and socioeconomic status, and preferred methodological approach. However, small bivariate gender gaps disappear in multivariate analysis. Furthermore, gaps in quantitative self-efficacy may explain racial/ethnic disparities in students’ broader academic self-efficacy. We argue that the documented patterns likely lead to continued underrepresentation of marginalized groups in the political methodology student body and professoriate.

Why do criminal actors publicly display threatening messages? Studies of organized crime emphasize that criminal actors rely on clandestine networks of influence. Subtle or coded threats are an effective means of extending that influence, but publicizing these threats appears to undermine their chief advantage. We argue that publicized threats broadcast an imagined order, delineating who has a place in society under criminal control, and who does not. To demonstrate this argument, we construct a “grammar of threat” and use this to analyze public threats broadcast by four criminal actors: two groups in Colombia and two in Mexico. The analysis demonstrates that every group projects an order through their threats, but that the order imagined varies by group. Some orders are more clearly ideological; some are more localized or more expansive. These findings highlight the important role of communication—distinct from but often combined with violence—in criminal governance.

Do legacies of politically motivated violence influence future or current electoral behaviour? How so? This article considers the question of the impact of violence on voter behaviour, specifically on elections that centred on issues of peace in contexts of long-running civil conflict. This study theorises the ways in which decades of violence, and continued contexts of unevenly distributed violence during elections, impacts current electoral behaviour. This article explores whether continued exposure to violence makes voters more or less conciliatory in their political preferences as expressed through electoral institutions. To do this, the article utilises the second round of voting in the 2014 and 2018 Colombian presidential elections and the 2016 plebiscite vote on the peace accords with the leftist guerrilla group Revolutionary Armed Forces of Colombia, along with a data set that records politically motivated violent events perpetrated by insurgents, counterinsurgents, and the state forces at a municipal level from 1991 to 2012.