Politics and Public Affairs

The Devil is in the Design: Testing the Effectiveness of Campaign Web Design

Faculty Researcher: Kathleen Searles

Student Researcher: Robyn Stiles

For a Complete Report of this Research, See: 

Searles, Kathleen, Robyn Stiles, and Levi Bankston. 2017. “The Devil is in the Design: Testing the Effectiveness of Campaign Website Design.” Prepared for The Midwestern Political Science Association Annual Meeting, Chicago, IL.


Despite an increasing acceptance of experimentation and data-driven processes, most campaign decisions are not subject to empirical testing. This truism extends to political website design.  While marketing and user experience research surveys best practices in web design for commercial and corporate organizations, we know little about what motivates political web design.  First, drawing from the universe of campaign websites of major presidential candidates from the 2008, 2012, and 2016 election, we first distinguish a set of characteristics common to political site design. Second, to test whether or not these practices are effective, we utilize eye-tracking technology to discern whether commonly used design elements garner visual attention and engagement. Overall, we find little support for the current design practices. These findings suggest that when it comes to political web design, the intuitions of campaign consultants should not rule.

Left to Our Own Devices: Political News Attention and Engagement in a Mobile Era

Faculty Researcher:

Kathleen Searles

For a Complete Report of this Research, See: 

Dunaway, Johanna, Kathleen Searles, Mingxiao Sui, and Newly Paul. (Revise and Resubmit). “Left to Our Own Devices: Political News Attention and Engagement in a Mobile Era.” Journal of Computer-Mediated Communication.


Mobile access to the Internet is changing the way people consume information, yet we know little about the effects of this shift on news consumption.  Consuming news is key to democratic citizenship, but is attention to news the same in a mobile environment?  We argue that attention to news on mobile devices is not the same as attention to news for those on computers.  Our research uses eye tracking in two lab experiments to uncover the consequences of mobile access for news attention and engagement.  We also conduct a large-scale study of web traffic data to provide further evidence that news attention and engagement are significantly different across computers and mobile devices.

The Knowledge of Human Trafficking in Baton Rouge

Student Researchers: Hillary Guidry, Rowan Knight, Lacy Miller, Kelsey King & Monica Raymond

Faculty Supervisors: Meghan Sanders

For a Complete Report of this Research, See: Guidry, H., Knight, R., Miller, L., King K., & Raymond, M. (2011), The Knowledge of Human Trafficking in Baton Rouge

Abstract: This research was charged with conducting research to determine what the residents of Baton Rouge know about human trafficking. Through secondary research, focus groups and a survey this research was able to meet the research objectives: 1) determine what individuals know about human trafficking and the laws in place; 2) inform people in the Baton Rouge area of local human trafficking; 3) determine the facts or information that will provoke high levels of inspiration or support against human trafficking.

Secondary research findings revealed technology and the Internet were used as an aid for human trafficking. Human trafficking is not only a global issue, but also affect Louisiana locally, especially schools. For the victims, the effects of trafficking are devastating and must be given the proper attention and treatment once a victim has been rescued.

Survey results showed that majority of participants had some prior knowledge of human trafficking which was obtained from entertainment, news, religion or school sources. Most of the terms provided show people are aware of the sexual abuse aspect of human trafficking, but are unaware of other issues involved in human trafficking. Also, only a small portion of respondents are aware of any type of human trafficking laws Louisiana currently has in place. After watching a video on local human trafficking, half of the respondents provided a factual response while one-fourth gave an emotional response, and the final one-fourth gave a factual and emotional response.

Focus group results showed that most participants collectively felt the definition consisted of selling or trading humans who were forced into human trafficking, and had no consent. In the perceptual to rank crimes, human trafficking was listed as very seri- ous and more of an individual problem; trafficking was listed among rape and attempted murder, as well as child/domestic abuse. After watching a testimonial video, a few participants said they would be likely to get involved because they didn’t realize how widespread and common of an issue human trafficking was, and they felt more people should be aware.

It’s All in the Name: Source Cue Ambiguityand the Persuasive Appeal of Campaign Ads

Faculty Researchers: Christopher Weber, Johanna Dunaway, and Tyler Johnson

For a Complete Report of this Research, See: Weber, C. R., Dunnaway, J., Johnson, T. (2011), It’s All in the Name: Source Cue Ambiguityand the Persuasive Appeal of Campaign Ads. Political Behavior. doi: 10.1007/s11109-011-9172-y

Abstract: As strategies for campaign political advertising become more complex,there remains much to learn about how ad characteristics shape voter reactions topolitical messages. Drawing from existing literature on source credibility, we expectad sponsorship will have meaningful effects on voter reactions to political advertisements.We test this by using an original experiment, where we expose a sampleof student and non-student participants to equivalent ads and vary only the paidsponsor disclaimer at the end of the message. The only thing that differs acrossstimuli is whether a political candidate, a known interest group, or an unknowninterest group sponsors the advertisement. Following exposure to one of these ads,participants complete a posttest battery of questions measuring the persuasivenessof the message, source credibility, and message legitimacy. We find that adssponsored by unknown interest groups are more persuasive than those sponsored bycandidates or known interest groups, and persuasion is mediated by perceived credibility of the source. We conclude by discussing our findings and their implications for our understanding of contemporary campaigns.

Moral Foundations and Heterogeneity in Ideological Preferences

Faculty Researchers: Christopher R. Weber, and Christopher M. Federico

For a Complete Report of this Research, See: Weber, C. R. and Federico, C. M. (2013), Moral Foundations and Heterogeneity in Ideological Preferences. Political Psychology, 34: 107–126. doi: 10.1111/j.1467-9221.2012.00922.x

Abstract: Scholars have documented numerous examples of how liberals and conservatives differ in considering public policy. Recent work in political psychology has sought to understand these differences by detailing the ways in which liberals and conservatives approach political and social issues. In their moral foundations theory, Haidt and Joseph contend the divisions between liberals and conservatives are rooted in different views of morality. They demonstrate that humans consistently rely on five moral foundations. Two of these foundations—harm and fairness—are often labeled the individualizing foundations, as they deal with the role of individuals within social groups; the remaining three foundations—authority, ingroup loyalty, and purity—are the binding foundations as they pertain to the formation and maintenance of group bonds. Graham, Haidt, and Nosek demonstrate that liberals tend to disproportionately value the individualizing foundations, whereas conservatives value all five foundations equally. We extend this line of inquiry by examining whether different types of liberals and conservatives value the moral foundations to varying degrees. Using survey data (n = 745), we rely on a mixed-mode latent class analysis and identify six ideological classes that favor unique social and fiscal policy positions. While most of the respondents belonging to these classes self-identify as conservative, they endorse the moral foundations in varying degrees. Since our findings demonstrate considerable heterogeneity with respect to ideology and moral preferences, we conclude by encouraging scholars to consider this heterogeneity in detailing the motivational and psychological foundations of ideological belief.

Political knowledge and exposure to the 2012 U.S. presidential debates. Does debate format matter?

Student Researcher: Jason Turcotte (PhD Candidate)

Faculty Supervisor: Kirby Goidel

For a Complete Report of this Research, See: Turcotte, J. (2012). Political knowledge and exposure to the 2012 U.S. presidential debates. Does debate format matter?

Abstract: Political participation in the U.S. continues to lag comparable democracies around the world (Fishkin, 1991; Putnam, 2000; Patterson, 2003). A strong antecedent of political engagement is knowledge (Dolan & Holbrook, 2001), yet democracy faces further obstacles from low levels of public knowledge and government trust (Delli Carpini & Keeter, 1996; Cappella & Jamieson, 1997). And election news coverage leaves much to be desired. While the highly fragmented and more competitive media environment of today provides citizens with more avenues for political learning these avenues, arguably, offer less substantive information as content yields to profit-driven pressures (Hamilton, 2004; Dunaway, 2008; Patterson, 1994; Cohen, 2008; Farnsworth & Lichter, 2011). Since the advent of cable and internet, the news media increasingly focus on soft news, negativity, and horse-race coverage (Cohen, 2008; Patterson, 1994; Cappella & Jamieson, 1997). Mediated presidential debates serve as one of the few remaining campaign events with a mass audience, and earlier research indicates exposure to these televised debates improves public knowledge (Miller & MacKuen, 1979; Pfau, 1988; Drew & Weaver, 1991, Zhu, Milavsky & Biswas, 1994; Holbrook, 1999; Benoit, Hansen & Verser 2003). Less clear, however, is whether these earlier knowledge studies hold up in a more competitive and fragmented media environment that has undergone considerable content change in the tone and frames used to cover elections. The influence of profit-driven pressures on news values suggests these changes mean most citizens receive less substantive campaign information today. It is possible that similar trends dominating campaign coverage spillover into debates, which provide voters with an opportunity to learn about policy issues and candidates. It is also possible that some debate formats are more substantive than others. Despite the addition of the town hall format in 1992, which places the agenda-setting function of debates in the hands of the public, scholars have seldom accounted for format when studying debate effects. Previous debate scholarship too often treats effects as uniform or limit scope to variables of persuasion (McKinney & Rill, 2009; Pingree, Scholl & Quenette, 2012). Moreover, Prior’s (2012) research raises validity questions concerning studies examining debate effects through survey methods, since respondents are known to overestimate debate exposure. In fact, his data show that for every person accurately reporting debate exposure, there’s another who falsely claims to have watched the debates (Prior, 2012, p. 361). Therefore, extant research in this area may wrongly attribute effects to debate exposure. This study calls for a more nuanced approach in testing the relationship between debate exposure and knowledge within the context of format. Using experimental methods to test the effects of exposure to the 2012 U.S. presidential debates, this study explores 1) whether debate exposure in the fragmented media environment still leads to political learning and 2) the extent to which learning varies among debate formats. After controlling for other known predictors of political knowledge I find that debate effects are no longer uniform; citizens can learn about issues and candidates but such learning is contextual, meaning format may influence debate effects.

The Effects of Empowerment on Self-Efficacy: How Different Models of Empowerment in the Media Can Impact Personal Empowerment

Student Researcher: Roussell, Stephanie O. (MMC)

Faculty Supervisor: Rosanne Scholl

For a Complete Report of this Research, See:   Roussell, S. (2012). The Effects of Empowerment on Self-Efficacy: How Different Models of Empowerment in the Media Can Impact Personal Empowerment. This paper was based on a a project as as part of “Public Opinion” graduate course.

Abstract: This study explores the concept of individual-level empowerment and the postfeminist shift from collective gain to personal empowerment. Specifically, the study utilized an experiment to examine the relationship between different models of empowerment and their effect on self efficacy and perceived empowerment of female media characters. Results indicate that womenfeel more empowered after viewing sexual media characters, yet do not assign the title of empowerment to those same media characters. These results suggest respondents personally believe sexuality is empowering, but are unwilling to publicly endorse sexual media characters as empowered women.

Volunteer based organizations: Bridging the gap between political efficacy and civic engagement

Student Researcher: Kristin Nicole Marks (MMC)

Faculty Supervisor: Amy Reynolds

For a Complete Report of this Research, See: Marks, K. (2012). Volunteer based organizations: Bridging the gap between political efficacy and civic engagement. This paper was the manuscript of the author’s thesis.

Abstract: With the declining participation in civic engagement behaviors such as voting, people are turning to other means to demonstrate civic participation. This study examined the role of volunteering as a means to establish civic engagement. Because of this civic participation behavioral shift, the relationship between civic engagement and the communication tactics used by volunteer based organizations was investigated.

This study investigated the relationship between civic engagement and political efficacy. Additionally, this study examined the interplay between organizational characteristics (trust, control mutuality, exchange relationship, and commitment) and civic engagement. Lastly, perceptions of new and traditional methods of communications to encourage civic engagement and build trust of volunteer based organizations were explored.

To address these relationships, this study used on online survey with 245 adults and the data were analyzed using a linear regression analysis and SPSS Textual Analysis for Surveys. This study used convenience and snowball sampling. This survey used various civic engagement scales and James Grunig’s PR Measurement Scale for organizational factors. Findings show higher levels of participation of political interest and trust lead to higher levels of political efficacy. Additionally, stronger control mutuality beliefs of an organization lead to higher levels of civic engagement. Participants reported email and direct mail to be the methods of communication they are most familiar with from their volunteer based organizations that encourage engagement. Furthermore, ease and access to information are crucial for volunteers regarding engagement and trust.

Splitting a Pair: Playing the Gender Card and the Race Card in American Politics

Student Researcher: Amy Ladley (PhD)

Faculty Supervisor: Regina Lawrence

For a Complete Report of this Research, See: Ladley, A. (2012). Splitting a Pair: Playing the Gender Card and the Race Card in American Politics. This paper is the author’s dissertation.

Abstract: More than any election before, the 2008 Presidential race revealed a persistent discussion of “race cards” and “gender cards.” In spite of the reported consensus that these alleged cards were everywhere, we know relatively little about those situations where the “card” label was applied, and even less about how this label influenced voters. In fact, among key electoral sources – political elites who use identity as a campaign tool, the journalists who cover and narrate elections, and researchers who make sense of elections-based behavior – there is no consensus regarding what a card is, how or when they are played, or who does the playing. This project seeks to begin to fill the gap in our knowledge of cards in campaigns by asking how were race and gender cards reported in news coverage of the 2008 presidential election, and how does labeling an appeal a “card” matter? Using content analysis and a two-part experiment, this study succeeds in drawing a much clearer picture of how card coverage, as an essential tool of narrating an election where women and racial minorities are present, affects American politics. While much of the research on cards defines their application and effects in terms of public policy issues, an examination of card coverage during the 2008 election reveals that much of the alleged cards were character-based. Moreover, the “card” label was not just used to categorize an appeal; cards were also invoked to maintain the identity narrative, even when identity was not a campaign issue. Using some of the most commonly reported cards from the 2008 race, the progressive experiments here revealed that, while the card label itself has little effect on how voters evaluate candidates, the addition of contextual information – for those with higher levels of racism and sexism – predicted increased support for white and male candidates, respectively. In short, these results show that how cards are covered defies our existing understanding of what a race card or a gender card is; moreover, in card coverage, the “card” label itself matters less than traditional cues like candidate sex and race in informing evaluations.

Preference and Recognition: the Interaction of Political Identity, Media Preference, Memory, and Social Judgment

Student Researchers: Arien Hossain, Cade Cypriano, Hillary Hall, Kate Royals

Faculty Supervisor: Meghan Sanders

For a Complete Report of this Research, See: Hossain, A., Cypriano, C., Hall, H., and Royals, K. (2012). Preference and Recognition: the Interaction of Political Identity, Media Preference, Memory, and Social Judgment. This paper is the author’s dissertation. This paper was based on a project as part of “the Introduction to Research Method in Mass Communication” graduate course .

Abstract: This study assessed the interaction of recognition, media preference, social identity, and social judgment. A participant group of eighty-three persons read two article sets containing a pair of politically divergent opinion-editorial articles on recent news events. The first editorial set used in the study pertains to the political conflict in Libya. In 2011, there was a wave of revolutionary uprisings, protests and demonstrations in the Arab world, known as the Arab Spring, to establish democracy. The second editorial set pertains to the case of Troy Davis, an American man executed in 2011 for the 1989 murder of a police officer. After reading the articles, the participants engaged in filler tasks and answered a questionnaire intended to measure recognition-based memory of details from the readings. The participants also answered questions about their political identity, their perception of the political identity for each article’s author, their preference for content from each author, and their preexisting familiarity with the subject of each article. The results were analyzed to determine the significance of any interaction between the variables representing the concepts studied. The study found that no statistically significant relationship existed between political identity and recognition, though a statistically significant relationship exists in some cases between political identity and media preference and political identity and social judgment.

Effective Direct Mail and Television Ad Design

Faculty Researcher: Kathleen Searles

The practice of using direct mail and television ads to persuade voters is a long-standing practice in American politics. Although electoral advertising is a billion dollar industry, it is still difficult to pinpoint exactly what visual elements make a mailer or ad more persuasive than others.  This is particularly true of negative advertisements, for which voter reactions have long run against what those in the industry and in academia known to be true: they work!  The tendency of individuals to say that they do not like negative political ads presents practitioners and academics alike with a measurement problem: how do we assess whether a negative mailer or television ad is effective when people are likely to tell us otherwise?  The answer is often either unreliable focus group testing or expensive in-cycle testing. However, recent technological advancements, namely eye-tracking technology, enable researchers to evaluate the processes underlying judgments about visual stimuli.  Specifically, eye-tracking technology allows us to isolate the elements of mail or an ad that capture attention and influence voter attitudes outside of individual awareness.   In this way, we can assess when a negative ad is effective without relying on individual responses.  We can also better understand which sort of visual stimuli capture participant attention across mediums in both television and mail, in an effort to sketch out best practices for television ad design.

A Question’s Still a Question: The Susceptibility of Values Scales to Order Effects

Faculty Researcher: Kathleen Searles

The purpose of the proposed research is to ascertain whether questionnaires designed to elicit the abstract values commitments of research subjects suffer from the known vulnerabilities of survey instruments more broadly. Since the pioneering work of Rokeach (1973), Schwartz (1992), and others (Haidt and Joseph 2004), social scientists have explored the nature of these abstract values which include concepts such as egalitarianism, universalism, authority, and traditionalism. Survey respondents are frequently asked to fill out standard questionnaires determining their relative placement of these various values. These value orientations are subsequently applied as moderators in analyses.

Most of this work (Bem 1970; Braithwaite 1994; Jacoby 2006) views values as stable, higher-order predispositions that are relatively stable in nature. As such, to these scholars, eliciting value orientations is key to understanding lower-order attitudes such as the evaluation of a specific governmental institution (Gibson and Nelson 2014). Yet others, (Zaller 1992; Converse 1960; Kertzer and Powers 2015) conceive of attitudes not as hierarchical, in which values influence other attitudes from the top-down, but rather values exist in a flat network with lower-order orientations such as party ID (Goren et al. 2009).

In sum, the tension between these two literatures motivates a puzzle: are values stable, higher-order dispositions that influence lower-order evaluations, or are they fluid orientations that vary depending on context? This theoretical puzzle motivates a methodological inquiry:  are values question subject to order effects?  To answer these questions we conduct a pair of survey experiments, one using a convenience sample and the second using a weighted nationally representative sample.  In so doing, we endeavor to offer methodological solutions for scholar and practitioners who use values questions in their survey instruments.  Our results will carry implications for the nature of preference formation.