Winning and losing
Conflict and Candidate Selection: Game Framing Voter Choice
Lori Cox Han & Brian Robert Calfano
American Politics Research, forthcoming
Political campaigns are often likened to a game typified by conflict. We consider whether using a conflict frame visually emphasizing the contested aspect of partisanship affects candidate support in the 2016 presidential election. Using a nationwide survey experiment (N = 975) that randomly assigns participants to different visual frames depicting politics as conflictual or process-based, we find that participants exposed to the conflict frame show significantly higher odds of supporting Donald Trump and Bernie Sanders, while rejecting Hillary Clinton. The conflicting frame also increases self-reported participant anger, which decomposition analysis shows increases support for Trump and Sanders while decreasing it for Clinton (and that we offer as a preliminary finding). Avenues for future research are then considered.
Political Polarization along the Rural-Urban Continuum? The Geography of the Presidential Vote, 2000-2016
Dante Scala & Kenneth Johnson
ANNALS of the American Academy of Political and Social Science, July 2017, Pages 162-184
This article documents the diversity of political attitudes and voting patterns along the urban-rural continuum of the United States. We find that America's rural and urban interface, in terms of political attitudes and voting patterns, is just beyond the outer edges of large urban areas and through the suburban counties of smaller metropolitan areas. Both Barack Obama and Hillary Clinton performed well in densely populated areas on the urban side of the interface, but they faced increasingly difficult political climates and sharply diminished voter support on the rural side of the interface. The reduction in support for Clinton in 2016 in rural areas was particularly pronounced. Even after controlling for demographic, social, and economic factors (including geographic region, education, income, age, race, and religious affiliation) in a spatial regression, we find that a county's position in the urban-rural continuum remained statistically significant in the estimation of voting patterns in presidential elections.
Manufacturing and the 2016 Election: An Analysis of US Presidential Election Data
Caroline Freund & Dario Sidhu
Peterson Institute for International Economics Working Paper, May 2017
Much of the public discourse and media analysis of the surprise outcome of the 2016 US presidential election has emphasized the role of manufacturing workers. This paper examines the importance of manufacturing jobs and job loss as determinants of voting patterns using county-level voting data from recent presidential elections. The share of employment in the manufacturing sector and long-run manufacturing job loss at the county level are not statistically significant in explaining the change in Republican vote shares from 2012 to 2016, when controlling for standard voting determinants. However, the change in the Republican vote share is positively correlated with manufacturing in predominantly white counties and negatively correlated with manufacturing in ethnically diverse counties, with these effects roughly offsetting each other. The paper further shows that this polarization between white and nonwhite manufacturing counties is more closely associated with polarizing candidates than a polarized electorate.
Taking Corrections Literally But Not Seriously? The Effects of Information on Factual Beliefs and Candidate Favorability
Brendan Nyhan et al.
Dartmouth College Working Paper, June 2017
Are citizens willing to accept fact-checks of false or unsupported claims of candidates they support in the heat of a political campaign? Previous studies have reached conflicting conclusions about people's willingness to update their factual beliefs in response to counter-attitudinal information. To discriminate between these findings, we conducted two experiments during the 2016 presidential campaign. Our results indicate that correcting misleading claims that Donald Trump made during his convention speech and in the first general election debate reduced belief in the claims in question even among his supporters. However, attitudes toward Trump were not affected. These results suggest that corrective information can reduce misperceptions, but will often have minimal effects on candidate evaluations or vote choice.
Party and Gender Stereotypes in Campaign Attacks
Erin Cassese & Mirya Holman
Political Behavior, forthcoming
Research on negative campaigning has largely overlooked the role of stereotypes. In this study, we argue that the gender and partisan stereotypes associated with traits and policy issues interact with a candidate's gender and partisanship to shape the effectiveness of campaign attacks. We draw on expectancy-violation theory to argue that candidates may be evaluated more harshly when attacks suggest the candidate has violated stereotypic assumptions about their group. Thus, attacks on a candidate's "home turf," or those traits or issues traditionally associated with their party or gender, may be more effective in reducing support for the attacked candidate. We use two survey experiments to examine the effects of stereotype-based attacks - a Trait Attack Study and an Issue Attack Study. The results suggest that female candidates are particularly vulnerable to trait based attacks that challenge stereotypically feminine strengths. Both male and female candidates proved vulnerable to attacks on policy issues stereotypically associated with their party and gender, but the negative effects of all forms of stereotype-based attacks were especially large for democratic women. Our results offer new insights into the use of stereotypes in negative campaigning and their consequences for the electoral fortunes of political candidates.
Attribution Errors in Federalist Systems: When Voters Punish the President for Local Tax Increases
Journal of Politics, forthcoming
How do voters attribute blame when policy responsibility is shared? While central to accountability, this question is difficult to answer because "who does what" is often ambiguous. This article exploits a case where policy responsibility is unambiguous: local tax referendums. Although presidents have no control over property taxes or the decision to raise local rates, I find that voters punish the president's party for tax increases enacted via direct democracy. This effect is robust to adjusting for population-based measures of the local economy, as well as panel and discontinuity designs to account for unobserved factors. The effect varies with the magnitude of the tax increase but not with local economic performance, suggesting that voters react to the change in spending money, as opposed to being "primed" to consider national issues. Thus, voters punish officials not only for events that no one controls but also for policies that voters themselves enact.
Republicans Should Vote: Partisan Conceptions of Electoral Participation in Campaign 2016
Sharon Jarvis & Jay Jennings
American Behavioral Scientist, forthcoming
Research has identified differences in partisan communication styles, cognitive processing, values, and cultures. This article assesses if there are also differences in how Democrats and Republicans conceive of electoral participation. We submitted 1,730 open-ended responses from a 2016 survey prompt soliciting thoughts and feelings about voting to computerized content analysis. Findings show that Republicans employed more confident and less negative language than Democrats in their responses. Additionally, a close read of the Republican statements reveals how they expressed that their electoral participation matters. Our conclusion addresses how the traditional ideals often associated with the Grand Old Party may make them a more duty-based constituency valuing voting more than their less conservative peers.
Empowering the Party-Crasher: Donald J. Trump, the First 2016 GOP Presidential Debate, and the Twitter Marketplace for Political Campaigns
Journal of Political Marketing, forthcoming
This article argues that an important political marketplace of keywords expands in social media around campaign events such as a debate; that rhetorical efforts to define the situation in which a campaign event occurs are met in this marketplace by user responses that more or less echo the keywords, thereby enhancing or diminishing the political power of their "caller" or speaker; and that social media monitoring platforms can enhance our understanding of public opinion influence competitions among candidates through the careful selection, tabulation, and inspection of words and phrases being voiced. In the case at hand, an analysis of Twitter volume data and a reading of a sample of 1200 tweets between July 30 and August 15, 2015, a period enveloping the first 2016 Republican presidential candidate debate on August 6, 2015, helps us understand how Donald J. Trump escaped political punishment from party and media elites for subverting Republican and U.S. norms of candidate behavior. Elite voices greatly disapproved of Trump's debate performance and conduct, a traditional augury of declining public support. But the presence of social media voices enhanced Trump's capacity to succeed with an insurgent marketing strategy, one he would continue into his election as president fifteen months later. Specifically, comparatively high user volume on a debate-oriented section of Twitter (i.e., posts with the hashtag #GOPDebate) for Trump's name, slogan, and Twitter address, and for such advantageous keywords as "political correctness," "Megyn Kelly," and "illegal immigration" relative to terms and phrases favoring other candidates and Republicans as a whole indicates the presence of heavy and active popular support for Trump. The contents of the corresponding tweet sample exhibit Twitter-savvy techniques and populist stances by which the Trump campaign solicited that support: celebrity feuding, callouts to legacy media allies, featured fan comments, a blunt vernacular, and confrontational branding. The contents also illustrate ways in which users manifested their support: from the aforementioned high keyword volume to imitative behavior and the supplying of evidence to verify Trump's contested claims during the debate.
Did Shy Trump Supporters Bias the 2016 Polls? Evidence from a Nationally-representative List Experiment
Statistics, Politics and Policy, forthcoming
Explanations for the failure to predict Donald Trump's win in the 2016 Presidential election sometimes include the "Shy Trump Supporter" hypothesis, according to which some Trump supporters succumb to social desirability bias and hide their vote preference from pollsters. I evaluate this hypothesis by comparing direct question and list experimental estimates of Trump support in a nationally representative survey of 5290 American adults fielded from September 2 to September 13, 2016. Of these, 32.5% report supporting Trump's candidacy. A list experiment conducted on the same respondents yields an estimate 29.6%, suggesting that Trump's poll numbers were not artificially deflated by social desirability bias as the list experiment estimate is actually lower than direct question estimate. I further investigate differences across measurement modes for relevant demographic and political subgroups and find no evidence in support of the "Shy Trump Supporter" hypothesis.
The Emerging Constitutional Law of Prison Gerrymandering
Stanford Law Review, May 2017, Pages 1473-1539
"Most prisoners in the United States are counted where they are incarcerated for the purposes of legislative redistricting. This practice - which critics label 'prison gerrymandering' - inflates the representation of mostly white, rural prison host communities at the expense of the urban and minority communities from which prisoners disproportionately hail. A battle to reform the practice has intensified in recent years, with federal courts on the front lines; the first federal court to invalidate a prison gerrymander did so in 2016, invoking the Equal Protection Clause's one-person, one-vote principle, and a division of authority has since emerged. As things stand, courts can expect a wave of these claims after the 2020 Census, but they are divided and ill equipped to resolve them. This Note undertakes an in-depth analysis of one-person, one-vote challenges to prison gerrymanders and is the first scholarly work to analyze this emerging body of law. It argues that the Equal Protection Clause does limit prison gerrymandering, advocating a novel approach for adjudicating these claims - one that looks principally to community ties (or the absence thereof) between prisoners and the localities that house them."
Does Incarceration Reduce Voting? Evidence about the Political Consequences of Spending Time in Prison
Alan Gerber et al.
Journal of Politics, forthcoming
The rise in mass incarceration provides a growing impetus to understand the effect that interactions with the criminal justice system have on political participation. While a substantial body of prior research studies the political consequences of criminal disenfranchisement, less work examines why eligible ex-felons vote at very low rates. We use administrative data on voting and interactions with the criminal justice system from Pennsylvania to assess whether the association between incarceration and reduced voting is causal. Using administrative records that reduce the possibility of measurement error, we employ several different research designs to investigate the possibility that the observed negative correlation between incarceration and voting might result from differences across individuals that lead both to incarceration and to low participation. As this selection bias issue is addressed, we find that the estimated effect of serving time in prison on voting falls dramatically and for some research designs vanishes entirely.
Competing loyalties in electoral reform: An analysis of the U.S. electoral college
Electoral Studies, forthcoming
A central tenet in the electoral systems subfield is that parties, when in power and motivated by partisan interest, seek desired outcomes via the strategic adoption of electoral rules. Such a focus, however, omits a key point: electoral rules also distribute power among geographic units. If, within a party, the partisan and geographic interests of some members conflict, then the canonical relationship between partisanship and rule choice may be conditional. The U.S. electoral college provides an opportunity to test for such intra-party variation, because it advantages some states over others and thus makes salient geographic allegiances. Using an original dataset on one reform proposal - the National Popular Vote Interstate Compact (NPVIC) - I find evidence of competing loyalties. Although NPVIC advances furthest when Democrats control state lawmaking, a state's status as a swing - but not as an overrepresented - state weakens the relationship to the point where even Democrats are unlikely to aid NPVIC.
Socioemotional Selectivity Theory and Vote Choice
Costas Panagopoulos & Charles Prysby
Presidential Studies Quarterly, forthcoming
Socioemotional selectivity theory posits that individuals invest more selectively in goals and activities that are emotionally meaningful as they age and time horizons gradually shrink. We extend socioemotional selectivity theory to the domain of voting in elections. We use data from the 2012 American National Election Study to test the hypothesis that older voters would place greater emphasis on emotional reactions to the candidates in their presidential voting, relative to younger voters. The empirical evidence suggests support for this contention, implying socioemotional selectivity extends to voting.
Legislative Term Limits and Voter Turnout
Robynn Kuhlmann & Daniel Lewis
State Politics & Policy Quarterly, forthcoming
According to reformers, legislative term limits should increase voter turnout by enhancing electoral competitiveness for legislative seats. However, this claim has been largely untested. The only existing study of the effect of legislative term limits on voter turnout, to date, finds that turnout in California did not increase after the imposition of term limits and may have decreased turnout. Yet, it is unclear whether this result generalizes to other states. This study employs a comparative state analysis of both aggregate turnout and district-level turnout rates in state legislative elections. We find that term limits significantly increase voting rates in state legislative elections.
Networks and Innovation in the Production of Communication: Explaining Innovations in U.S. Electoral Campaigning From 2004 to 2012
Daniel Kreiss & Adam Saffer
Journal of Communication, forthcoming
We outline a network analytic framework for analyzing the production of communication. In our framework, individuals - in part the products of the history of their social and professional ties - merge from various fields and a medley of prior production experiences within organizations to produce communicative innovations. Organizations with individuals who have diverse backgrounds and significant overlap in work experiences will be more innovative. We demonstrate this through a network analysis of the professional biographies of 629 staffers on U.S. presidential campaigns from 2004 to 2012. Democratic staffers came from more diverse organizations and shared significant overlap in prior experiences than their Republican counterparts. Through interview data, we argue that this in part explains Democratic innovativeness in technology during this period.
Social Desirability Bias and Polling Errors in the 2016 Presidential Election
Andy Brownback & Aaron Novotny
University of Arkansas Working Paper, July 2017
Social scientists have observed that socially desirable responding (SDR) often biases unincentivized surveys. During the 2016 presidential campaign, we conducted three list experiments to test the effect SDR has on polls of agreement with presidential candidates. We elicit a subject's agreement with either Hillary Clinton or Donald Trump using explicit questioning or an implicit elicitation that allows subjects to conceal their individual responses. We find evidence that explicit polling overstates agreement with Clinton relative to Trump. Dividing subjects by party identification, we find that SDR significantly diminishes explicit statements of agreement with the opposing party's candidate. Democrats are significantly less likely to explicitly state agreement with Trump. This threatens the predictive validity of polling, negatively impacts the ability of markets to accurately price assets, and exaggerates disagreements between Democrats and Republicans. We measure economic policy preferences and find no evidence that ideological agreement drives SDR. We find suggestive evidence that SDR correlates with county-level voting patterns.
The Effects of Election Festivals on Voter Turnout: A Field Experiment Conducted During a Presidential Election
Donald Green & Oliver McClellan
Columbia University Working Paper, July 2017
During the final days of the 2016 general election, Civic Nation coordinated a series of non-partisan election festivals designed to encourage voter turnout in targeted precincts across the country. Previous experimental research (Addonizio, Green, and Glaser 2007) indicates that festivals held at polling sites significantly increase voter turnout. Prior research, however, focused solely on low-salience elections, such as municipal or primary elections. The study reported here is the first to assess the effectiveness of festivals held in the context of a high-salience election. Festivals appear to increase turnout substantially and cost-effectively, but further research in high-salience elections is needed to pin down these effects with more statistical precision.
The ballot order effect is huge: Evidence from Texas
Public Choice, forthcoming
Primary and runoff elections in Texas provide an ideal test of the ballot order hypothesis, because ballot order is randomized within each county and the state offers many counties and contests to analyze. Doing so for all statewide offices contested in the 2014 Democratic and Republican primaries and runoffs yields precise estimates of the ballot order effect across 24 different contests, including several not studied previously. Except for a few high-profile, high-information races, the ballot order effect is large, especially in down-ballot races for judicial positions. There, the empirical results indicate that going from last to first on the ballot raises a candidate's vote share by nearly ten percentage points. The magnitude of this effect is not sensitive to demographic and economic factors.
The effects of electability on US primary voters
Journal of Elections, Public Opinion and Parties, Summer 2017, Pages 274-290
The general consensus of the research on US primary contests is that voters consider candidates' potential for a general election victory when choosing their party's nominee. Yet, at the individual level, this literature has failed to (1) clearly isolate the effects of electability from the money and media attention that they generate; and (2) properly control for the potential effects of ideology. Using an original experimental design, I find that electability can increase the likelihood of a voter supporting a more ideologically distant candidate. I also show that when faced with a tradeoff, a large percentage of subjects from both parties choose electability over ideology. The resulting implication is that there is potential for moderates to be successful in primaries, as even ideologically extreme voters appear to be willing to compromise on policy representation when confronted with a more distant but electable candidate.