Interdisciplinary Workshops on Politics and Policy

In many post-election surveys, the proportion of respondents who claim to have voted is substantially greater than the turnout proportions reported in official statistics. Many scholars speculate that respondent lying is the main cause of such turnout overestimation. Consequently, some researchers replace respondent self-reports of their registration status and turnout behavior with data on those individuals from government records. Because this approach can yield survey respondent turnout rates that are closer to official government turnout rates, some scholars believe that validated turnout measures weed out liars and are more accurate than self-reports.

This paper describes findings from the American National Election Studies’ recent turnout validation exercise. It suggests that validated turnout measures may not be more accurate than self-reports. Indeed, the seeming accuracy between “validated” turnout estimates and official government statistics appears to be the result of two countervailing biases. A downward bias comes from failures to find survey respondents’ government records that generate implausibly low registration rate estimates. An upwards bias comes from survey respondents turning out at a higher rate than non-respondents and telling the truth about their behavior to interviewers. Indeed, we find little evidence of lying amongst survey respondents whose turnout records can be located.

This project taps one of the richest sources of political data in all history—the political blogosphere—to better understand political participation and communication. Blending techniques from social and computer science, I study blogging as an act of political participation. I test existing theories of participation in the context of the blogosphere, and discuss how to break new theoretical ground by studying attention, hyperlinking, and tone of discourse in the political blogosphere. As a form of high-bandwidth activism, blogging can provide unique insights about the intersection of information, communication, and participation in politics. Questions asked and answered in this practice job talk include:

• How can we construct representative samples online?
• Why do people blog about politics?
• Is the voice of the blogosphere representative of the electorate?
• Do bloggers' motivations affect the content of their blogs?
• How does the "bandwidth" of political participation matter for democracy?

Politicians on the campaign trail regularly use metaphors of fighting and war, yet little is known about how citizens respond to such appeals. I field two nationally-representative survey experiments preceding the 2010 midterm elections in which subjects are randomly assigned to different forms of the same political advertisements. I find that violent language mobilizes participation in citizens with aggressive personalities who trust the electoral process, but demobilizes those with doubts. More troubling, violent rhetoric fuels support for political violence among aggressive citizens regardless of political trust. These results uncover new mechanisms of campaign mobilization, but the same processes also unsettle democratic norms against violence in politics.

We estimate the effect of electoral quotas in India’s state legislatures on the provision of public services to villages. Roughly one out of six seats are reserved for Scheduled Caste candidates, but since reservation is not randomly assigned, it is difficult to identify its true impacts on welfare outcomes. To address endogeneity and selection concerns, we use a geographic discontinuity design, focusing on villages just inside and outside borders that delineate reserved and unreserved constituencies. By comparing villages that are similar except for reservation status, we estimate the effect of reservation on improving the provision of electricity from 1992 to 2009. Looking nationally, the average effect of reservation across India is negligible. Yet unpacking this result, we show that the effects in individual states is highly variable, swinging from positive to negative. We argue that the effect of reservation depends on whether reservation decreases political competition or not within state level politics.

IR theory suggests that "resolve," or government's willingness to fight a war, is one of two primary components that determine which party wins during crisis negotiations. The more willing to fight a government is, the more aggressive they can raise the stakes; their less-resolute adversary will eventually reach the limits of their own willingness to engage first and concede the dispute, at least under certain circumstances. For democratic governments, the concept of resolve must include both the state's ability to pay for a conflict as well as their voters' acceptance of the stakes at hand. Studies of domestic political behavior, however, indicate that voters decide whether or not to support a policy by responding to messages and other cues from political elites.

In this paper, I design a survey experiment that tests the ability of a government to persuade its citizenry of the danger of a potential rival state in a pre-crisis period. Following recent propaganda campaigns in the United States, I examine two specific persuasive mechanisms: whether military experience makes a speaker more persuasive, and whether incendiary, anxiety-provoking appeals make persuasive appeals more effective. In two early applications, I find that persuasion is less contingent on these two factors than the political psychology literature would suggest. However, I also find suggestive evidence that the broadly-accepted relationship between partisan identity and foreign policy opinion breaks down for prospective, as opposed to retrospective, evaluations of foreign policy performance.

Most political scientists argue that explicit racial appeals are no longer effective in contemporary American politics. According to this view, such messages are rejected because they are perceived as violating the almost universally accepted norm of racial equality. I reexamine this question with an experimental design, embedded in a nationally representative Internet survey in which respondents were randomly assigned to one of eight treatments. The race of the messenger (black or white) and the type of appeal were manipulated, depending on treatment condition. Contrary to the conventional wisdom, I argue that explicit racial appeals are not necessarily counterproductive, as black candidates who use explicit appeals generate more support among racial conservatives relative to black candidates who use implicit appeals. The results also suggest that White Americans' response to the norm of equality is contingent on the race of the messenger. That is, when the messenger is also white, respondents are more likely to reject the appeal, but when the messenger is black (with an identical message), they do not find the message offensive. Since previous research related to the implicit and explicit racial appeals has typically focused on white messengers, to date we have been unable to detect that the rejection of explicit appeals is moderated by the race of the messenger.

How do insecure property rights affect electoral competition and the level of violence? To answer this question, I explore original empirical evidence from Dagestan, Russia’s most turbulent North Caucasian republic. The exploration is based on a statistical analysis of district-level data with special emphasis on chronological validity. Studying the relationship between land titles of the Soviet period and post-Soviet amounts of tenured land, the research demonstrates that the amount of unregistered land in each district has a profound effect on local electoral competition and indices of violence. A higher percentage of untenured land at the district level leads to less electoral competition and more intense violence. Consequently, the study finds that the insecurity of property rights creates an opportunity structure for electoral patronage and violent expression of conflicts and grievances. In theoretical perspective this study sheds light upon a relatively unexplored institutional factor that drives electoral process and violence in predominantly agrarian societies.

Party polarization in the U.S. in the past few decades has made parties more distinct, thus facilitating the partisan sorting of American voters. However, party alignment along moral and economic issues has also made it more difficult for certain socio-demographic profiles to define their political allegiance: Will a wealthy, non-religious voter identify with the Republican party for its economic policies, or with the Democratic party’s moral views? Using a network-based method for detecting heterogeneity in collective patterns of opinion, I show that when people’s interests are incompatible with the traditional liberal-conservative polarity, individuals gravitate toward alternative ways of conceptualizing the political debate. Moreover, when faced with seemingly competing opinions on economic and moral issues, Americans are more likely to privilege their conservative views, and identify with the Republican Party.

Why do U.S. presidents allow the unionization of federal employees, given that unionization weakens bureaucratic control? We argue that presidents selectively use unionization to “lock-in” ideologically like-minded agencies’ current composition, thus preventing future presidents from drastically changing these agencies’ ideological direction. The intuition behind this theory is that unions protect the job security of employees, thereby reducing bureaucratic turnover and stabilizing an agency’s current workforce. Hence, the president strategically supports unionization only in agencies sharing the president’s political leanings. Corroborating the predictions of our formal model, we find that agencies with fewer unionized employees experience more frequent personnel turnover and greater ideological volatility when the President’s partisanship changes. Agencies with a greater proportion of unionized employees, by contrast, experience less personnel turnover during presidential transitions, and they remain more ideologically stable across presidencies.

The levels of government funding for public broadcasters is a flashpoint in policy debates in both the United States and other democracies. Past research has suggested that public broadcasting has a positive effect on levels of political knowledge and participation. However, my research suggests that government subsidies of public broadcasters are the important link to increases in knowledge, not merely the existence of a public broadcaster.

In our increasingly diverse society, many Americans identify with more than one social or political group. These identities, at times, align with opposing policy choices. When it comes to making political decisions, respondents are thus torn between two interests. In this experimental study, I examine how competing identities influence public opinion on contentious issues. I administered a survey experiment on a large sample of adults who simultaneously identify both as Democrats and as parents. In a 3-by-3 factorial design, I prime none, one, or both of these identities using three different types of identity appeals: threatening, high status, and straightforward messages. I find that each identity appeal in isolation influences an individual's opinion. However, when appeals are simultaneously used against two identities with competing interests, the identity facing the greatest perceived threat exerts the most influence. When the same type of appeal is simultaneously used on both identities, each cancels out the other and preference remains neutral. Existing studies show the effect of appealing to only one identity, but here I demonstrate that such appeals are only effective in isolation or in situations of low competing threat.

Since 9-11, the news media and political campaigns have frequently referenced the possibility of another terrorist attack on U.S. soil. Despite the multitude of ways in which these threats might differ, research on Americans’ reactions to the threat of terror often focuses on threats generally. Using data from a nationally representative survey experiment, we examine the effects that both the type of terrorist threat and the ethno-religious identity of the terrorists have on Americans.

The paper argues that religious nationalism is a coherent worldview in American public life and that it can explain seemingly incoherent patterns in foreign policy attitudes among American citizens. Evidence from an original survey conducted through Knowledge Networks in October 2010 supports this argument.

Scholars of American political behavior have generally found that gender plays a relatively modest role in shaping public opinion, especially compared with the role of race. This stems in part from a limited conceptualization of gender. I argue that gender, properly understood, has structured political competition and citizens' understanding of it while at the same time itself evolving. After sketching the contours of this structuring, I present two analyses that begin to unpack these dynamics. The first uses experimental data to explore the effects on citizens' impressions of candidate sex, gendered traits, and party identification. The second uses ANES data to explore the ways that sex and partisanship condition the role of gender in shaping political views among the American public.

Why do mainstream political parties attempt to incorporate immigrants into their electoral coalitions at some times and not others? I argue that immigration presents both challenges and opportunities to mainstream parties of the center-left and center-right in Western Europe. I propose a general framework to explain variation in these parties’ appeals toward immigrants and the groups to which these are targeted. From this framework, I derive and test hypotheses using Comparative Manifesto Data.

Political scientists have yet to provide a compelling account of public support for government-led economic downward redistribution in the United States. Since public opinion about redistribution can affect public policy, and concomitantly distributional outcomes, our failure to understand the determinants of public support for downward redistribution is a failure to come to terms with a central component of “who gets what, when, and how” in American politics. I specify conditions under which resentment for the rich and sympathy for the poor - factors that are generally overlooked by political scientists - can undergird support for downward redistribution.

Drawing from a survey of donors to federal campaigns in 2007-08 at all giving levels we examine the question of why people contribute. Prior studies have applied Clark and Wilson's organizational incentive schema (material, solidary, and purposive) which we replicate but find lacking. We develop and test for motives that include candidate and campaign centered motivations. The motivation work draws from a book length manuscript that examines the changing demographics of donors in 2008, the surge in new donors, female donors, small donors and Internet donors. We also include an analysis of when people contributed and an assessment of possible incentives to encourage broader participation in donating.

This paper examines the "culture war"' hypothesis by examining American citizens' choices among a set of core values. Specifically, a geometric model is developed to represent differences in the ways that individuals rank-order seven important values: freedom, equality, economic security, social order, morality, individualism, and patriotism. The model is fitted to data on value choices from the 2006 Comparative Congressional Election Survey. The empirical results show that there is an enormous amount of heterogeneity among individual value choices; the model estimates directly contradict any notion that there is a consensus on fundamental principles within the mass public. Further, the differences break down along political lines, indicating that it probably is reasonable to describe the American public's feelings about values as a culture war.

Why do leaders who routinely abuse their citizens, violate human rights, and use violence at home work to maintain peace and protect civilians abroad? Extant explanations on civilian protection focus on democratic norms and humanitarian pressures from within the state.

These arguments clearly cannot explain why repressive states, who lack domestic humanitarian pressures and peaceful democratic practices, would work to protect foreign civilians abroad. I argue that repressive regimes take these actions to stop civil unrest in other states for fear that such conflict may diffuse to their state through refugee and arms flows. This motivation has serious implications for how repressive states conduct peacekeeping and peace building missions.

Many studies of sub-state political violence rely on political or macroeconomic explanations, often relegating social variables to the margins. Social factors, specifically societal-level social capital or trust developed through healthy commercial and cooperative interaction, play an important role in explaining patterns of terrorist activity. The systemic relationship between societal-level trust and terrorism, however, is hidden at first glance and only emerges through the lens of complexity where macro-level implications arise from micro-level forces. The impact of trust or social capital, theoretically cast as a neutral force, becomes conditional on breadth of violent elements within the population. The talk will explore the underlying foundations of the theory linking social capital to political violence and provide supporting empirical assessments.

Scott Helfstein is the Director of Research for the Combating Terrorism Center and an Assistant Professor in the Department of Social Science at the United States Military Academy. His research has appeared in numerous academic journals and addresses transnational issues such as terrorism, nonproliferation, interstate conflict, behavioral economics and economic interdependence. In his government capacity, Scott has consulted for or worked with numerous government offices such as the National Security Staff, Office of the Secretary of Defense, Department of State Policy Planning, and the Joint Staff. He holds a Joint Doctorate in Public Policy and Political Science from the University of Michigan focused on world politics. Scott also degrees in finance and war studies and worked in the financial sector for five years. He is a Term Member in the Council of Foreign Relations and a member of National Defense University's Combating WMD Program for Emerging Leaders.

By many accounts, the United States is in the midst of a period where comity between political foes is scarce and, despite impending policy problems, compromises are elusive. It has not always been the case, as evinced, for instance, by the Tax Reform Act of 1986 -- monumental legislation crafted under divided government that garnered overwhelming bipartisan support. What explains the difference? I argue that standard explanations, such as elite polarization, are not fully satisfying. Borrowing new insights from cognitive psychology, I suggest that the changing nature of political campaigns causes politicians to adopt policies positions that, even if not more extreme, are more intransigent. I present results from a randomized experiment supporting this idea.

Does the thermostatic model of responsiveness characterize the relationship between opinion and policy in the states? Using a unique dataset on state spending preferences on education and welfare, I find evidence of dynamic policy representation and dynamic public. As state support for spending on education or welfare increases, so too does actual state expenditures. As state expenditures on education or welfare increase, state preferences for additional spending decreases.

Furthermore, the relationship between opinion and policy is conditional. States that are highly professionalized are more responsive compared to those that are less professionalized. And, the public is more likely to respond to policy actions on issues that are salient. The results provide a more nuanced understanding about the degree of dynamic representation and responsiveness in the states and the thermostatic model, more generally.

The effects of corruption on individual behaviour are by no means a new research topic. Most of the previous studies focused on its consequences on shaping political behaviour, regime survival, trust and democratic values. However, electoral behaviour scholars have not come to an agreement on whether corruption affects vote decision. This paper seeks to address this question by examining the contingency of party and leader effects on corruption, and thus contribute to the new developing literature on contextual effects on voting behaviour.

Based on the previous literature one may foresee the changes in ideological and economic vote as corruption increases, but it is less obvious how leader and party effects are adjusted, especially because scholars of voting behaviour still disagree about the roots and importance of leader effects on vote decision, relative (or in an absolute sense) to party effects. Building upon the psychological theory of trust, this paper argues that leaders who engage in a trust-repair process and look competent and reliable in corrupted countries outplay the party effects on vote choice.

These propositions are tested using the CSES Module 3 dataset and, World Bank and Transparency International macro indicators of perceived corruption. Preliminary results show that the repercussion of corruption in alienating and disentangling citizens from electoral politics is indeed reflected in the decreasing intensity of party effects and ideological vote, but leaders can partially offset this loss. These effects are statistically significant, but not strong enough to change the electoral outcome.

This research considers how partisan control and the institutional capacity of American state governments shape the volume and content of proposed regulations over time. We study rulemaking across state governments to understand how the organization of democratic government affects the policy responsiveness of unelected policy-makers. Drawing upon a novel data set of over 370,000 proposed and adopted regulations issued by state agencies from 1992 through 2009, we evaluate how changes in party control, professionalism, and institutional capacity of state governments shape the volume and content of proposed regulations.

Scholars have long debated whether political science experiments that rely on undergraduate student subjects can produce valid, generalizable results. Starting with Sears’ (1986) argument about the problem of “college sophomores in the laboratory,” political science scholarship has retained a “near obsession” with the question of who is studied (McDermott 2002). While some scholars have suggested that experiments that rely on undergraduate students do not lead to useful findings (Brady 2000), others have argued that in some cases undergraduate students can be an effective subject population (Druckman and Kam 2011). In this paper we address this debate using a series of parallel experiments with different subject populations. Specifically, we specify and test conditions under which relying on undergraduate student subjects undermines the generalizability of experimental research in political science. In total, we present the results of 12 experiments on three different types of samples – a student convenience sample, an adult convenience sample, and a nationally-representative sample. Unlike previous research on the topic – which puts forth sweeping arguments about the effectiveness of certain populations in political science experiments – our work identifies the specific steps researchers can take to ensure generalizability when relying on undergraduate student populations for experimental research.

Many scholars have explored the “democratic advantage” in international relations: the empirical regularity that democracies tend to win militarized disputes, wars, and interstate rivalries. Schultz and Weingast (2003) argue that the democratic advantage originates from competitive representative institutions lowering the cost of borrowing. Building from the work of Broz (1998, 1999), I argue that central banks, not representative institutions, are responsible for enhancing a country's financial health and, consequently, war fighting prowess. Because the exigencies of representative government do little to placate concerns over government default, only the presence of a central bank can ease creditor fears. Data covering the 19th and early 20th centuries reveals that possessing a central bank increases the probability of victory, improves the terms of credit for a sovereign, and enables the country to spend more on the military. This relationship holds whether the country does or does not possess representative institutions.
Full Abstract

The mass media devote a great deal of attention to high profile elections, but in American political life, such elections are the exception, not the rule. The majority of electoral contests feature candidates who are relative unknowns. In such situations, does name recognition breed contempt, indifference, or affection? Existing work presents modest theory and mixed evidence. Using three laboratory experiments, we provide conclusive evidence that name recognition can affect candidate support; and, we offer strong evidence that a key mechanism underlying this relationship is inferences about candidate viability and not inferences about traits or experience. We further show that the influence of name recognition breaks down in the face of more a more germane cue, incumbency. We conclude with a field experiment that demonstrates the robustness of the name recognition effect to a real world political context, that of yard signs and a county election.