To address the challenge of declining trust in American democracy in the wake of the 2020 presidential contest, election officials across the nation have undertaken innovative public information campaigns. Academic studies demonstrate that exposure to these messages can increase public confidence but do not show which types of messages are most effective. We report a set of three experimental studies that harness research-practice partnerships with these officials to vary one key aspect of an informational message while holding other features constant. The pre-registered experiments (accessible at: https://osf.io/y38sp; https://osf.io/fya69): (a) compare the impact of messages conveyed through earned versus paid media; (b) ask whether Americans are more responsive to messages from federal or from state election officials; (c) explore the impact of videos and static visuals. Taken together, this set of collaborative experiments demonstrates the unique opportunity that research-practice partnerships allow to test real-world messages through strong causal inference techniques, providing rigorous evidence that can inform practice on the front lines of American democracy.

Hundreds of US cities have switched from single choice to ranked choice voting (RCV) systems. In allowing voters to rank multiple candidates and eliminating fears of wasting votes on less viable candidates, RCV purports to improve both political expression and representation. Realizing this potential, however, depends on voters' willingness and ability to use their rankings. We assess these outcomes by conducting a survey with an embedded experiment during real-world elections for multiple municipal offices using RCV. We examine whether citizens' propensity to rank multiple candidates and choose like-minded candidates (spatial voting) vary with the prestige of local offices and in response to endorsements. Political expression and spatial voting are substantially lower in elections for comptroller and city council relative to mayor. Endorsements from unions and issue advocacy groups improve both outcomes for these lower-level offices. Our results highlight differences in the efficacy of RCV across local elections and illustrate how endorsements can help citizens navigate these challenging settings.

Can messages targeted to underrepresented groups increase their participation in politics? We predict that mobilization messages highlighting a politically relevant group identity will increase participation among racial/ethnic minorities. However, responsiveness to such appeals should vary depending on the strength of one’s identity with that group. We test these predictions by conducting a two-wave panel survey that randomly assigns get-out-the-vote (GOTV) messages that appeal to either racial/ethnic identities or a less politicized identity (state identity). We compare the effects of these messages to a standard GOTV message or one that encourages recycling. We find that racial/ethnic identity appeals increase participation among minorities with weaker group identities, who are initially less inclined to participate. Therefore, such appeals help close the gap in participation between those with weak versus strong identities. These findings indicate that messages highlighting a politically relevant group identity can sometimes be effective at increasing participation among underrepresented groups.

In states and localities with direct democracy, citizens can advance their policy interests without the aid of elected officials. Research documenting citizens’ lack of political knowledge raises questions about their ability to do so. We conduct three studies during real-world direct democracy elections to determine whether citizens choose alternatives (the ballot proposal or status quo) that are closest to their ideological positions and whether political information improves this outcome. Using original surveys, our first two studies estimate citizens’ ideological positions and show that citizens regularly choose alternatives that are closest to these positions. Using a survey experiment, our third study indicates that political information (party cues, policy information, and spatial maps) further improves such spatial voting. These results demonstrate citizens’ capacity to advance their policy interests in direct democracy elections and identify conditions under which political information strengthens the relationship between citizens’ policy interests and choices about ballot propositions.

AbstractPolitical parties and civic organizations disseminate information to improve citizen decision making in local elections. Do citizens choose to consume such information and, if so, how does it affect their decisions? We conduct a survey experiment during a real-world local election that randomly assigns 1) political party endorsements, 2) a voter guide, 3) no information, or 4) a choice among these options. Respondents assigned to receive party endorsements and a voter guide are more likely than respondents receiving no information to choose candidates who share their policy views. When given a choice, a majority opts to receive information (including many with low levels of political interest), with most respondents preferring a voter guide. Using an instrumental variable approach, we show that the effect of information on those who choose to receive it is substantial. These results offer hope that voter education efforts can succeed despite widespread political disinterest.

Abstract Income inequality has increased dramatically in the United States, yet public support for redistribution has not. The authors of this chapter theorize that mistaken beliefs about income inequality contribute to citizens’ lack of demand for greater redistribution, and that correcting these beliefs can help citizens connect their value for greater equality to their preferences for redistribution. To test their theory, they conduct a survey that measures citizens’ beliefs about income inequality and then randomly assigns them to receive factual or partisan information before expressing their support for redistributive policies. The survey found that citizens believe that income is distributed much more equitably than it actually is and that, in the absence of information, their preferences for redistribution are unrelated to whether or not they value greater equality. When citizens receive accurate information about income inequality, however, they connect their value for greater equality to their preferences for redistribution, even when their partisan allegiances recommend otherwise.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

What factors shape public opinion about government solutions to address police violence? We address this question by conducting a survey in which respondents express their opinions about actual proposals to reform police practices. Within the survey, we randomly assign respondents to receive the positions of traditional advocates (Black lawmakers) and/or opponents (law enforcement) of police reform efforts. Our results reveal broad bipartisan support for the proposals, but that information about groups that support or oppose these proposals polarizes partisans’ opinions. However, Democrats and even Republicans who support Black Lives Matter (BLM) express high levels of support for the proposals regardless of the information they receive. These results suggest that partisanship in the mass public is not necessarily a barrier to police reform efforts. A bipartisan majority of the public supports meaningful reforms, and any polarizing effects of elite signals are muted by Democrats’ and Republicans’ support for BLM.

In recent years, the credibility of social science has been tarnished by widely discussed replication failures and a lack of reporting about what exactly researchers did when conducting their studies. In response, scholars, policymakers, and the public have called for greater transparency in social science research. In this chapter, I emphasize that transparency is an important public good. However, because individual researchers lack incentives to contribute to this public good, institutional solutions are needed. I discuss three institutions that facilitate transparency in experimental research: 1) pre-registration, 2) reporting guidelines, and 3) the Data Access and Research Transparency (DA-RT) initiative. I also offer recommendations for what kinds of information researchers should pre-register and report in their published articles and appendices. I conclude with a discussion of how researchers might be incentivized to make greater use of these institutions when designing, conducting, and publishing their experiments.

Citizens are typically uninformed about politics and know little about issues at stake in direct democracy elections. Government efforts to inform electorates include requiring donors to initiative campaigns to report their activities and then circulating such donor information to citizens. What effects does donor information have on citizens’ opinions? We conduct a survey experiment where respondents express opinions about initiatives in a real-world election. We manipulate whether they receive donor information, party cues, policy information from a nonpartisan expert, or no additional information. We find that donor information influences citizens’ opinions in the aggregate, with effects comparable to those of party cues and policy information. However, donor information has negligible effects on uninformed citizens, who have difficulty inferring donors’ policy interests and connecting them to their own. These results underscore the potential benefits of efforts to inform electorates via disclosure laws and highlight disparities in their effectiveness for informed and uninformed citizens.

The implementation of ranked-choice voting (RCV) systems across the United States anticipates several benefits. For citizens, the ability to rank multiple candidates yields increased opportunities for political expression. Taking advantage of these opportunities, however, requires both more effort by citizens and more information about candidates. What types of citizens will benefit from the greater opportunity for political expression afforded by RCV systems? We address this question by conducting an experiment in the context of the 2020 general election in San Francisco. In the experiment, we randomly assign the electoral rules (single-choice versus ranked-choice) that govern citizen decision-making and the provision of political information (a voter guide summarizing candidates’ policy positions versus no information). We assess whether respondents use the choices available to them and the extent to which they choose / rank candidates whose policy views are similar to their own. Our results indicate that a majority of respondents in the ranked-choice setting take advantage of the increased opportunities for political expression. Nonetheless, uninformed respondents (i.e., those who know relatively little about local politics) are significantly less likely than informed respondents to rank multiple candidates and use all of their allotted rankings. The rankings of uninformed respondents are also more weakly related to their own policy views. Providing an issue-oriented voter guide increases the share of uninformed respondents who use their allotted rankings, and strengthens the relationship between respondents’ candidate rankings and their policy views. Together, these results indicate that the benefits of RCV elections may not be fully realized by all types of citizens. Providing citizens with accessible forms of political information, such as an issue-oriented voter guide, can reduce these disparities and help citizens make more and potentially better informed choices.

Can racial and ethnic minorities be mobilized to participate in politics at greater rates? We theorize that mobilization messages providing information about a group’s underrepresentation in government may increase participation among racial/ethnic minorities. However, responsiveness to such messages should vary depending on individuals’ prior awareness of their group’s underrepresentation. Using a two-wave panel survey that randomly assigned different get out the vote messages, we find that messages highlighting a racial/ethnic group’s underrepresentation in government do not increase Latinos’, Blacks’, or Asians’ likelihood of voting. We also find that such messages can decrease other forms of political participation among Asians and Latinos who were previously unaware of their group’s underrepresentation. These findings indicate that information about underrepresentation can actually demobilize certain segments of the electorate. Thus, practical efforts to boost participation among underrepresented groups should either communicate information about underrepresentation in other ways or provide a different type of message altogether.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

Incidents of police violence can undermine trust in legal authorities. Whether such incidents have this effect will depend on how citizens evaluate victims, the police, and public officials. Citizens’ evaluations may be shaped by information about (1) a pattern of police violence and (2) government responses. We study citizens’ reactions to police violence by randomly assigning these two types of information in the context of the Stephon Clark shooting in Sacramento. We find that information influences levels of blame for and trust in the police, but the effects depend on citizens’ race and whether they live in the community where the violence occurred. In contrast, information does not alter citizens’ perceptions of local police officer organizations and, in turn, their willingness to follow police endorsements in elections. These results suggest a catch-22 whereby police violence can diminish the standing of police personnel, but favorable local opinion preserves their political influence.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

Political endorsements (recommendations about which candidate or policy to support) are ubiquitous in political contexts. They may come from political parties, interest groups, politicians, or even celebrities. Can uninformed citizens identify endorsers who share their interests and use their recommendations as substitutes for detailed political information? This chapter surveys the literature on the persuasion effects of political endorsements. It first provides an overview of theoretical models that examine how political endorsements affect citizens’ choices. Then, it describes a seminal observational study of how endorsements affect political persuasion. It next discusses experimental research that assesses the effects of endorsements. It emphasizes that experiments are particularly useful for identifying when political endorsements will persuade citizens because they allow scholars to manipulate the conditions that theoretical models identify in a carefully controlled environment. It concludes by discussing open questions about the effects of political endorsements and describing how ongoing research addresses them.

An abstract is not available for this content so a preview has been provided. Please use the Get access link above for information on how to access this content.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.