Promoting Interdisciplinary Connections and Exchange

Experts discuss political influence and participation, polling, and causality coverage

James Druckman

More than 100 faculty and graduate students gathered at Northwestern on May 9 for a workshop that featured talks by some of the nation’s leading political and social scientists.

“The original motivation for the workshop was to have regular meetings with scholars from different institutions and disciplines come together to interact with one another, and the conference has well exceeded my expectations,” said IPR political scientist and associate director James Druckman, who organized the eighth annual workshop. “The success reflects the general environment of Northwestern, which truly creates a great center of interdisciplinary collaboration.”

Increasing the Political Influence of Women

Why do men chiefly occupy the nation’s most powerful political positions despite a rising proportion of female politicians? According to Princeton’s Tali Mendelberg, it is not just a numbers game.

Tali Mendelberg

She posits that rules governing group discussion could hold sway. To test this relationship, she randomly assigned 470 participants to groups of five, either all-male, all-female, or mixed. After being introduced to four principles of income redistribution, each group held an open discussion to choose the “most just” principle using either unanimous or majority rule. In tracking the discussions and ensuing decisions, Mendelberg finds that women participants talked the most in all-female groups. The only times both men and women spoke for equal amounts of time were in majority-female groups where decision making was based on majority rule. Conversely, if women were in the group’s minority, only unanimous rule increased their participation and influence. In terms of topics, women in majority-female groups tended to focus more on issues more significant to women, such as families, children, and the poor. Mendelberg pointed to these findings as evidence that institutions can play an important role in addressing gender inequality in discussions. The research is part of her forthcoming book, The Silent Sex: Gender, Inequality, and Institutions.

Conveying the Human Costs of War

Scott Althaus

Little research has examined how information about war casualties is reported and framed by the news media, and thus how such information might affect domestic support for wars. Scott Althaus, a political scientist at the University of Illinois at Urbana-Champaign, is the first to analyze casualty coverage across several major wars, comparing news reporting and framing across five large American conflicts. To establish a baseline, Althaus and his co-authors analyzed every war-related story from randomly sampled days of New York Times coverage during both world wars and the Korean, Vietnam, and Iraq wars. Results showed that wartime news coverage rarely includes casualty information. When it is reported, it is often framed in ways that minimize or downplay the human costs of war. Despite differences in the wars’ duration, scale of casualties, and types of warfare, the frequency and framing of casualty reporting varied little from World War I through the Iraq War. This similarity suggests that major news media rarely convey basic facts about the human costs of war to ordinary Americans.

Probabilistic Polling

Pollsters and political scientists generally assume that voters cannot accurately respond to percent-chance questions about future events. For example, what is the percent chance that they will vote for a particular candidate? Presenting research challenging this status-quo thinking, IPR economist Charles F. Manski encouraged the room full of political scientists to reconsider using such questions. Manski and his colleague Adeline Delavande of the University of Essex worked with the RAND American Life Panel to ask probabilistic as well as the traditional verbal polling questions of respondents in the lead up to the 2008 presidential election. They found similar results for both verbal and probabilistic responses and could predict whether a respondent who vote and for whom. Yet they also find differences in timing: Probabilistic responses did a better job of predicting behavior earlier, a couple months before the election, while traditional verbal responses performed better in the few weeks up to election. The 2012 RAND Continuous Poll, praised by many poll-watchers for its strong performance in the 2012 presidential election, included probabilistic questions designed by Manski and Delavande. An IPR working paper, the research was published in Public Opinion Quarterly.

Hypotheticals and Vignettes

Charles F. Manski (left) and James Gibson

Continuing the conversation about improving behavioral research methods, political scientist James Gibson of Washington University in St. Louis joined Manski for a panel on the use of hypothetical scenarios. In another IPR working paper, Manski and Delavande used percent-chance questions to ask respondents about their likelihood of voting in a series of hypothetical election scenarios. While the timing of voting and the closeness of an election seemed key in decisions to vote, candidate preference did not. The researchers corroborated responses using data on respondents’ actual voting behavior in the 2012 presidential election.  With “no canonical theory of voting,” political scientists need richer data to understand why people vote, Manski said, and introducing hypotheticals can help.

Gibson also discussed using hypothetical scenarios—which he calls vignettes—because he finds them better for investigating the influence of contexts and for being able to create counterfactuals. He shared his work in post-apartheid South Africa, where vignettes were used to gauge attitudes on justice when granting amnesty to human rights violators. In all cases, most respondents judged amnesty as unfair. The majority was smallest when distributive justice—where the victims’ families received financial compensation—was implemented. However, the combined effect of procedural justice—where the victim’s family was able to tell their story—and restorative justice—where the family received an apology—almost equaled the effect of monetary compensation. Thus, the vignettes showed that there are forms of compensatory justice that make up for the retributive justice deficit created by granting amnesty to gross human rights perpetrators. 

Motivational Underpinnings of Political Participation

Joanne Miller

While past research on political participation has primarily focused on the role of resources and mobilization, such as income, education, and GOTV efforts, Joanne Miller of the University of Minnesota argues that such models overemphasize the ability dimension of participation at the expense of the motivation dimension. Using time-series data from two panel studies, she first examines whether changes over time in traditional resource variables predict changes in participation. She finds that changes in resources, recruitment, and political engagement are not consistently, or significantly, related to micro-level changes in participation. Next, in a survey experiment of a representative sample of more than 4,400 adults, she assesses participants’ willingness to participate in political activities, as well as the importance and impact of group identity on participation, under six experimental conditions. The results provide evidence that strong motives correlate with certain political behaviors, implying that psychological motivation is important. Miller proposes that incorporating the concept of motivation will lead to a better understanding of individual-level participation, at both the cross-sectional and over-time levels.

James Druckman is Payson S. Wild Professor of Political Science and an IPR fellow and associate director at Northwestern University. Tali Mendelberg is Professor of Politics at Princeton University. Scott Althaus is professor of political science and communication at the University of Illinois at Urbana-Champaign. Charles F. Manski is Board of Trustees Professor in Economics and an IPR fellow at Northwestern. James Gibson is Sidney W. Souers Professor of Government at Washington University in St. Louis. Joanne Miller is associate professor of political science at the University of Minnesota. The next CAB Workshop will be held on May 8, 2015, in Evanston, Ill.