Sponsored by the Department of Politics, Center for the Study of Democratic Politics (CSDP), and Program for Quantitative and Analytical Political Science (Q-APS)
Tuesday, January 12, 2016
300 Wallace Hall
12:00 pm -2:00 pm
Quinn M. Albaugh and Harris Doshay | "The Better Angels of Our Nature" Reconsidered: A Finite Mixture Modelling Approach for Dual-Process Theories of Anti-Immigrant Political Behavior
Abstract: In a recent paper, Blinder et al. (2013) argue that citizens follow a dual-process model in their immigration attitudes, in which the cognition behind anti-immigrant political behaviors has an automatic prejudice-based process and (at least in some circumstances) a more deliberative norms-based process. We re-analyze their observational results in three ways. First, we verify that anti-immigrant policy opinions do not load on the motivation to control prejudice against immigrants scale (MCP). Second, we improve upon their model estimation by taking advantage of political socialization theories and panel data to identify more clearly which variables are ``pre-treatment" and then conduct matching analyses to estimate the causal impact of MCP on vote choice. Third, we examine which individuals rely on the prejudice- or norms-based processes using a finite mixture model. Results contribute to debates on norms in anti-immigrant political behavior and on dual-process theories more generally.
Meir Alkon and Erik H. Wang | Pollution and Regime Support: Quasi-Experimental Evidence from Beijing
Abstract: Using an eight-week long original survey conducted day-by-day in 2015, we leverage daily variation in air quality to estimate the causal effects of pollution on regime support. Our results cast doubt on theoretical assertions that environmental problems only become politically salient when countries reach higher income levels. Pollution substantially increases demand for government oversight, decreases overall satisfaction at both central and local levels, and negatively affects perceptions of economic performance. Since our results reflect daily attitudinal change, they are conservative estimates of pollution's effect. Additionally, because part of our survey coincides with a period during which the regime intentionally reduced air pollution, we exploit this unique experiment in authoritarian responsiveness to show that government efforts to reduce pollution do successfully improve citizen assessments of the regime. Our results causally estimate the importance of environmental challenges in developing countries and shed light on specific facets of political perceptions affected by pollution.
Clark Bernier | Performance and Structure of Hierarchy in Organizations
Many authors have rung the death knell for organizational bureaucracies and hierarchies. That many "flat organizations" ultimately incorporate hierarchies questions whether novel structures are sufficient to overcome Michel's "Iron Law of Oligarchy." However, hierarchies are not merely structures. They are enacted in interactions between individuals--performances of control and deference that shape and are shaped by the formal and informal structures of an organization. Understanding the evolution of nominally "flat" organizations thus proves elusive without understanding how structural hierarchies and performances of hierarchy relate to one other; an interaction largely ignored in the literature. This project uses two billion metadata log entries from a popular task and project management system to differentiate hierarchical performances--measured using control actions in the software--from organizational structures measured using Exponential Random Graph Models of interpersonal communication. We are thus able to characterize the co-evolution of performance and structure in 600 organizations over a multi-year time period.
John M. Chambers and Karis Yi | Leaders in Political Rhetoric: Media vs. Government
Abstract: Media and government appear to influence each other in the United States with both entities shaping the national agenda and helping determine the framing of its constituent issues. To determine which of the two is the primary leader, we take an observational approach, comparing the daily usage of key phrases in the Congressional Record and the New York Times over 2005. Our phrases are those shown by Gentzkow and Shapiro (2010) to be the most indicative of party in congressional floor speeches made during that year. For each phrase, we use cross-correlation to determine whether congressional usage is a leading or lagging indicator of media usage. Tallying across phrases, we find rhetorical leadership fairly evenly split between Congress and media. Further, we find that the leader may vary across phrases within the same topic.
Munji Choi and Alexandra Mayorga | When do we want revenge?
Abstract: Within the radical right literature one of the most potent theories to have emerged recently by Lenka Bustikova in her paper, “Revenge of the Radical Right” (2014), emphasizes its reactive nature and views it as a backlash against the political successes of minorities and concessions extracted on their behalf. We find this theory compelling and seek to bolster her arguments put forth by identifying the causal mechanism at play. Specifically, we are interested in analyzing party success at the district level to explore how different confounding covariates interact with the inclusion of ethno-liberal parties in coalitions to better understand its causal influence. Identifying the mechanisms at play is critical to understanding in what context radical right parties can successfully and powerfully mobilize as a backlash against ethno-liberal success versus when their message evokes a lukewarm response.
Denis Cohen and Naoki Egami | Heterogeneity in Coalition-Directed Voting. How Extreme Voters Shape the Vote Calculus in Multiparty Democracies
Abstract: Political preferences may enter the vote calculus of instrumental voters in different ways. While proximity theory argues that voters vote primarily based on their evaluation of parties' policy platforms, recent research shows that outcome-oriented voters prioritize parties' ability to affect policy into their desired direction. In multiparty systems, this often means evaluating parties' chances of entering a coalition government after the election. In this paper, we study the prevalence of coalition-directed voting within and across national electorates. We argue that the share of extreme voters and the supply they face in a party system play an important role. We expect that coalition-directed voting is less widespread at the fringes of the political system and particularly so when relevant far-right and far-left parties compete in an election. Consequently, characteristics of electorates and party systems constitute an important source of cross-national variation in the distribution of voters' decision rules.
Rebecca Johnson | Monetizing Disability: Do State-level Fiscal Incentives Shape Child Placement into Special Education?
Abstract: Do district-level financial incentives shape a child’s risk of special education placement? One thesis, inflexible uniformity, suggests that rigid federal requirements governing special education placement preclude a role for district-level financial incentives. Another thesis, district latitude, suggests that districts use ambiguity in federal requirements to shift children into and out of special education depending on financial incentives. Investigating these theses, the present paper exploits shifts between two sets of formulas states use to allocate money to school districts to finance special education: shifts from weighting financing—where a district receives additional state funding for each additional special education student the district reports serving—to capitation financing: the allocation is no longer sensitive to reported numbers. Preliminary results from a DID analysis and synthetic control suggest that a state’s shift to capitation financing causes a significant reduction in the proportion of students placed in special education, supporting the district latitude thesis.
Korhan Kocak, Daniel Tavana, Killian Clarke | Mechanisms of Diffusion in the Arab Spring
Abstract: The Arab Spring of 2011 began in Tunisia with mass protests that toppled the dictator Zine el-Abideen Ben Ali, and then quickly spread across the Arab World. But protesting diffused unevenly through the region. In five seemingly dissimilar countries – Egypt, Libya, Yemen, Bahrain, and Syria – protesting reached unprecedented levels, and fundamentally reshaped the political equilibrium in what had otherwise been considered stable authoritarian states. Other countries saw sporadic protesting, while still others remained relatively quiet in spite of activist efforts. Why did protest diffuse to some Arab countries but not others? And in general what can we learn from the Arab Spring about mechanisms of protest diffusion? We test three hypotheses regarding protest diffusion – a relational hypothesis, a cognitive hypothesis, and a structural mediation hypothesis – using two data sources: a survey conducted among Egyptian activists shortly after the January 25 uprising and Twitter data associated with the Arab Spring hashtags #sidibouzid and #jan25.
Asya Magazinnik and Sepehr Shahshahani | Strategic Abstention, Missing Data, and Ideal Point Estimation
Abstract: We replicate Clinton-Jackman-Rivers (2004) and Lauderdale-Clark (2014), then use our replication code to investigate a potential problem in the treatment of missing data. Efforts at ideal-point estimation using legislative roll-call votes typically assume that abstentions are random. This assumption is unjustified when abstention is a strategic decision that benefits the bill's opponents at the expense of its proponents. In the California and Minnesota legislatures, for example, passage of a bill requires the affirmative vote of a certain portion of the entire legislature, not of those present and voting, which implies that abstaining is practically equivalent to voting no. We apply our model to roll-call votes in California (and maybe Minnesota) to explore how ideal-point estimates change when we recode abstentions from missing to "no." Preliminary results indicate that the change is significant.
Rachael McLellan and Sondre U. Solstad | The Politics of Ethnic Inequality: The Long-Term Impacts of Artificial Borders and Initial Soil Inequality
Abstract: We provide a causal account which traces contemporary ethnic inequality and economic development to underlying differences in soil quality between ethnic homelands. Forthcoming research by Alesina, Michalopoulos, and Papaioannou establishes a relationship between contemporary ethnic inequality and economic development, but do not provide a causal account. We show that soil quality differences are deleterious for development, and that this underlying ethnic inequality is related to lower quality political institutions. This effect is particularly strong in states with borders imposed by colonial powers, emerging in an international environment where country borders were very difficult to change. Using the borders decided at the Berlin Conference, we leverage cases in Africa as natural experiments for the effects of ethnic inequality on subsequent economic and political development. We then explore the channels by which the effects of this ‘artificial inequality’ are created and persist.
Andrew Proctor and Tanika Raychaudhuri | Post-Racial America: Support for Racial Policies and the Absence of Group Specific Theories?
Abstract: Chong and Kim (2006) find that individual-level experiences of discrimination, economic status, and perceptions of opportunities predict attitudes towards racial policies in the United States. However, their analyses test within-racial group attitudes rather than general perceptions of opportunities available to all minority groups. First, we replicate their findings, confirming the results. We argue that the absence of group-specific theories cannot be tested using group-specific empirical analyses. Accordingly, we extend their analysis to include general perceptions of opportunities available to minority groups rather than within-group perceptions. Additionally, we vary controlling for race to gain leverage on the role race plays in shaping racial-policy attitudes. Finally, we run a genetic matching analysis, matching the closest respondents across minority groups on all other covariates. We think that this is a stronger test of the claim that attitudes towards racial policies are specific to individual-level experiences.
Patrick Signoret and Cassandra Emmons | Crime Victimization, Political Participation, and Political Attitudes
Abstract: Regina Bateson (2012) analyzes barometer survey data from five continents, finding that “individuals who report recent crime victimization participate in politics more than comparable nonvictims.” Victimization remains significant under nonviolent and violent crime across most regions of the world, and the effect survives a battery of tests to rule out reverse causation and omitted variable bias. Bateson also finds that victims are less supportive of democracy and more supportive of authoritarianism, vigilantism, and harsh policing tactics. Our analysis improves upon Bateson’s by using alternative matching methods and imputing missing data. We explore more deeply how crime victimization shapes political behavior and attitudes in two ways. First, we broaden the scope by turning toward the World Values Survey, which allows us to include regions and questions not available in Bateson’s surveys. Second, we narrow in on one country, Mexico, analyzing two independent surveys with a wealth of information on crime victimization and political attitudes.
Diana Stanescu and Audrye Wong | Chinese Aid and Civil Conflict in Africa
Abstract: We reexamine the nexus between foreign aid shocks and civil war along with the role of Chinese aid. This builds on Nielsen et al (2011), which finds that aid shocks increase conflict occurrence, as well as Strange et al (2015), which finds that Chinese aid provides an exit option that mitigates traditional aid shocks. We investigate if Chinese aid has a true substitution effect by looking at the sectoral and geographical allocations of aid. Understanding where aid is allocated also sheds light on the different ways resources are distributed to regime insiders, rebels, and citizens, and hence the causal mechanisms linking aid withdrawals and conflict. In addition, we vary the time lag between aid shock and conflict onset, and distinguish between Chinese aid already present during the shock and Chinese aid given subsequently. We also conduct a parallel analysis of the effects of Chinese aid shocks on the likelihood of conflict.
Hannah Waight | Informal Finance and Kinship Networks in Rural China
Abstract: In this project I use a nationally-representative, longitudinal data set to examine the relationship between informal lending and borrowing and kinship networks in rural China. Specifically, I look at how changes in the composition of rural villages, measured by the percent of villagers in the same lineage group, relates to changes in informal financial activity in that village. This project brings together the social network and informal finance literature, but is the first to use a longitudinal data set to look at the connection between the two phenomenon.