Interdisciplinary Workshops on Politics and Policy

Why is gender such an unreliable predictor of support for policies that promote gender equality? Women and men alike are generally raised in households and local contexts where men play an outsized role in shaping norms and wield a disproportionate amount of physical, economic, and social power, influencing the development of attitudes toward gendered policies. Across two national surveys, we look at the influence of various local institutions on attitudes toward gendered policies and find that family norms play a uniquely important role. In a follow-up survey experiment conducted on a national probability sample of American teenagers, we show that thinking of a single family member with liberal or conservative attitudes towards gender can shape teenage support for gender-related policies, particularly among boys. Together our findings highlight the powerful role that a single family member can play in shaping public opinion on gender equality. 

What determines women’s political influence? While the literature on political engagement focuses on individual traits, attitudes, and participation, we argue that how these factors translate to political influence is fundamentally a social process that requires recognition from the broader community, with important implications for understanding women’s political engagement. Using new data on networks of political influence in Philippine villages, we show that even after controlling for socioeconomic status or political participation, women are still markedly less likely to be recognized as influential. Furthermore, we show that engagement in politics through traditional means–running for office, participating in councils, or joining parties–are only associated with political influence for men. The determinants of influence are more complex for women: embeddedness in the community and participation in community activities are more important than traditional modes of political participation.

What is the relationship between trust in religious leaders and voluntary compliance with policies that are costly to the individual? Religious leaders have the moral authority to affect individuals’ willingness to adopt pro-social behaviors across many societies. Less clear is whether that influence will be positive or negative. It cannot be assumed ex ante that religious leaders will uniformly support social distancing guidelines both because they may be reluctant to discourage congregants from attending services and because their leadership within a country is often decentralized. We investigate how trust in religious leaders affects compliance in countries where religious authority is centralized and state aligned. We argue that, under these conditions, greater trust in religious leaders will be associated with more voluntary compliance, but that this effect will be limited to religious celebrations and rituals. Using novel data from surveys fielded in Kazakhstan and Kyrgyzstan during the COVID-19 pandemic, we find support for both hypotheses but only in Kazakhstan, where religious leaders consistently offered adherents substitutes that enabled them to practice their faith while social distancing. The influence of religious leaders on voluntary compliance, therefore, may depend as much on the content of the message as it does on the source.

Building upon the term “ethnographic sensibility,” this article makes the case for an “emotional sensibility” that seeks to glean the emotional experiences of the people whom we study and who participate in our research. I argue that an emotional sensibility is relevant for any human subjects research, including interviews, surveys, and experiments, even when emotions are not a direct variable under study. Methodologically, attention to research participants’ emotions can improve research because these emotions are themselves data, influence the validity of other data, and affect our ability to collect data in the future. Ethically, it balances ethics committees’ rationalist emphasis on elements such as information and cognitive capacity with appreciation for how emotions infuse consent, vulnerability, and experiences of risk and benefit. Against the tendency to narrow concern about participants’ emotions to a medicalized notion of retraumatization, an emotional sensibility encourages deeper thinking about research participants’ emotional experiences beyond harm and about emotional harms that do not qualify as trauma. Upon presenting these and other claims about how an emotional sensibility can make political science research more empirically valid and ethically sound, the article outlines strategies for cultivating and upholding an emotional sensibility in practice.

Dissidents face a tension when coordinating mobilization against a repressive regime. Dissidents benefit from making coordination public: this increases the spread of the call, and thus the potential size of mobilization. But public calls are heard by the regime, giving the security apparatus advanced knowledge of, and improved ability to, stifle mobilization. How do dissidents coordinate collective action under repressive environments? Using interview and ethnographic data, I examine mobilization during Sudan’s 2018-19 uprising. I find that, on the surface, it appeared that mobilization was publicly coordinated by a social movement organization through internet and communicative technologies, consistent with common channels of coordination identified by existing literature. Yet some dissidents independently used these widespread calls to secretly organize simultaneous collective action away from main protest sites, intentionally deviating from centrally organized calls to confuse the security forces. By looking deeply within the internal workings of a protest movement, this paper suggests that existing literature may be mis-attributing successful coordination in repressive environments to public-facing channels of mobilization. 

Presidents possess vast authority over policies and outcomes. Recent studies suggest the public checks this unilateralism through expressive opinions and political participation. We examine this accountability link by investigating the empirical implications of a model of presidential leadership. We present findings from a pre-registered panel survey prior to the 2020 presidential election and a dataset of news coverage of unilateral actions from 1993-2020. We find that presidents pay no net public cost — and may even benefit— by acting. Though respondents penalized presidents for failing to achieve their goals, our analysis of news coverage reveals that most action receives asymmetric media coverage. News stories are positive, assign credit for actions at their announcement, and rarely follow-up on realized outcomes. On balance, our theoretical and empirical evidence suggests that unilateral directives are credit-claiming devices most often unrestrained by public opinion and that presidents’ accountability relationship can incentivize actions that are welfare-reducing. 

This is joint work with Benjamin Goehring.

Debate persists as to whether, and under what conditions, civic education increases political knowledge and political participation, particularly among younger voters. This project contributes new evidence to these debates through a community-collaborative field experiment in Zambia in the lead-up to its 2021 general elections. The study was conducted in collaboration with major religious coalitions, which are the most prevalent providers of civic education in Zambia and prevalent providers of civic education worldwide. The study randomly assigned youth (ages 18-35) participants in a WhatsApp-based civic education program into one of three conditions: civic information only (with no motivational messages), or civic information accompanied by either religious or non-religious motivational messages.  We then measured political knowledge, political attitudes, and participation in the lead-up to and during the 2021 elections. The study found that the information-only course and the non-religiously motivated course both had compensatory effects on political participation, raising political knowledge and participation most robustly among young women and often reducing the ex-ante gender gap. However, the religiously motivated course increased participation most robustly among those who already occupied church leadership positions and often led to a retreat from participation among those without leadership positions (an exacerbating effect). The study shows that one should not take for granted that religious civic education will have large positive and compensatory effects on political participation, even in highly religious contexts. More generally, the paper underscores that the content of civic education courses can have consequences not just for the courses’ influence on aggregate political participation but also for the distribution of political involvement across groups.

Identifying the foundations of political polarization is a pressing issue across the social sciences. Our study lends insight to this matter through a nationally representative online experiment exploring how (partisan) group identity shapes the process of political opinion formation. In the week prior to the 2020 US presidential election, we incentivized participants to predict how policy-sensitive statistics develop over the next year, conditional on which candidate becomes president. Participants can update their initial predictions after selecting or exogenously receiving factually similar articles on the respective topics curated from news sources with a right- or left-leaning slant. Our results show that those individuals whose behavior is responsive to group identities are both initially more politically polarized and exhibit a stronger partisan bias during the process of opinion formation: (i) their initial predictions about the development of policy-sensitive statistics more strongly depend on whether their candidate becomes president or not; (ii) the partisan gap between these predictions increases more strongly after reading two curated articles with a right- and left-leaning slant, respectively; and (iii) they exhibit a stronger taste for articles curated from politically congenial than politically opposing news sources. Further, in a treatment manipulation we find that reducing the salience of group identity by de-labeling the source of information decreases partisan bias in information demand, especially for participants sensitive to group identity. Overall, our results suggest that through its role in the process of opinion formation, the group identity roots of partisanship are a key factor contributing to political polarization.