Recent Research: Politics, Institutions, and Public Policy


Public Opinion and Political Deliberation

Chicago Area Behavior Workshop

More than 100 faculty and doctoral students gathered at Northwestern on May 9 for the eighth annual Chicago Area Behavior (CAB) workshop, organized by IPR political scientist James Druckman. Princeton University’s Tali Mendelberg discussed increasing the political influence of women and an experiment in which she found that women tended to speak up more in majority-women groups. Scott Althaus of the University of Illinois at Urbana–Champaign spoke about conveying the human costs of war in news coverage. Comparing news reporting and framing of human costs across five American conflicts,he showed that wartime news coverage rarely includes casualty information, and downplays the cost of war when it is covered. IPR economist Charles F. Manski encouraged researchers to utilize probabilistic polling in evaluating whether voters will elect a particular candidate,as probabilistic responses did a better job of predicting voting behavior than traditional polling questions under certain circumstances. James Gibson, of Washington University in St. Louis, joined Manski on a panel discussing the use of hypothetical election scenarios in voting research. The University of Minnesota’s Joanne Miller spoke about how models of political participation focus on people’s ability to participate at the expense of their motivation to do so. Her work on this issue provides evidence that strong motives correlate with certain political behaviors. 

Approving and Mandating Vaccines

When faced with an outbreak of a communicable disease, such as the eruption of measles that started in California in December, or the recent outbreak of Ebola in West Africa, should federal and state governments require vaccinations to protect the health of their citizens—or should such decisions be voluntary? Deciding on the right course of action when it comes to vaccines is difficult because officials are working in an environment of “partial knowledge”—that is, they can point to some of a vaccine’s effects, but not all. IPR economist Charles F. Manski, Board of Trustees Professor in Economics, addresses this topic in an IPR working paper. While randomized trials have been important in the evaluation of noninfectious disease treatments, they do not shed light on population- wide disease transmission; in particular, they miss the effect of “herd immunity,” or how vaccinating a certain portion of the population can break the chain of disease transmission because the people vaccinated no longer spread the disease. Manski set out to address ways for the U.S. Food and Drug Administration and other federal bodies to operate in an environment of partial uncertainty by providing quantifiable criteria. He poses several criteria for decision making—expected utility, minimax, and minimax-regret—and derives the policies they yield. His findings point to formal decision analysis as a means to improve upon prevailing vaccine approval and mandating procedures. 

Communicating Uncertainty in Official Statistics

U.S. federal statistical agencies, such as the Census Bureau and Bureau of Labor, often report official economic statistics as point estimates, without accompanying measures of error. As a result, many news articles that use official statistics, such as gross domestic product, household income, and employment, present the estimates with little, if any, mention of the possibility of error—which might encourage policymakers and the public to believe that errors are inconsequential. In an IPR working paperManski urges agencies to better communicate uncertainty to the public in order to mitigate misinterpretation of official statistics. He offers different ways to communicate transitory uncertainty, which occurs when agencies initially have incomplete or insufficient data, but later revise estimates when new data become available; permanent uncertainty, or uncertainty that cannot be fixed over time due to issues such as nonresponse or finite sample sizes; and conceptual uncertainty, which arises from unclear definitions of concepts being measured. Manski also encourages future researchers to study how policymaking would change if agencies communicated uncertainty “regularly and transparently.” 

Elites, Citizens, and Policy Decisions

Who influences policy decisions in the United States? Political scientist and IPR associate Benjamin Page and Martin Gilens of Princeton University examine the role of average citizens, wealthy Americans, and interest groups in shaping American policy decisions. After considering four theories of American government that predict which actors influence American policy the most, they then conduct multivariate analyses on public opinion data on more than 1,700 policy issues and data on income levels. Their results demonstrate that economic elites and business-oriented interest groups have “substantial independent impacts” on American policies, consistent with theories of economic-elite domination and biased pluralism. Moreover, the researchers conclude that average citizens and other interest groups have little or no independent influence on policymaking. Page is Gordon Scott Fulcher Professor of Decision Making.

Political Views of the Wealthiest Americans

Page and his colleagues, including Northwestern’s Jason SeawrightFay Lomax Cook, and IPR graduate research assistant Rachel Moskowitz, continue to investigate how America’s wealthiest citizens think about issues and engage in politics. Their pilot study of a random sample of 104 Chicagoans with a median income of $7.5 million, as reported in Perspectives on Politics, indicates that the wealthiest Americans are far more active in politics than the average citizen: They are twice as likely to pay attention to politics and to volunteer for political organizations. Most contribute money to political causes, and one-fifth have “bundled” contributions by others. Many initiate contacts with public officials, especially members of Congress. The wealthiest Americans’ views on social welfare policies also tend to be much more conservative than the average American’s. Page and Seawright have recently begun using “web-scraping” techniques to extract information from websites in order to learn what U.S. billionaires say they want from government. The researchers note that many are silent about politics even while spending large sums of money on it.

Methods in American Political Development

What explains the decline of organized labor in the United States? What factors can account for the rise of modern interest-group politics? Questions such as these can be answered by American Political Development (APD) scholarship, a subfield dedicated to explaining substantial changes in the American political system over time. But, as Galvin points out in a forthcoming article in the Oxford Handbook of American Political Development, this type of scholarship often does not fully discuss analytical choices and methodological decisions, making it difficult for subsequent researchers to expand or challenge research. Galvin examines the varied ways APD researchers explore historical data, including historical narrative studies, causal narratives, and process tracing. While all of these methods have their benefits and drawbacks, as Galvin notes, researchers should be more upfront about the contributions their works make, and the limitations inherent in each type of analysis. With more discussion of what such studies accomplish—and what they do not accomplish—future researchers will be enabled to test existing APD hypotheses in other states, nations, or other settings. Furthermore, by elucidating the empirical limitations and limitations of scope, researchers might identify stronger evidence to test a hypothesis more easily.

Feminism and Psychology

Beginning in the late 1960s, many feminist psychologists argued that psychologists had, by and large, neglected the study of women and gender—and misrepresented women in the meager efforts they put forth. While psychological research on gender is now a sizeable specialization, Eagly and Stephanie Riger of the University of Illinois at Chicago point out that several methodological and epistemological issues raised by feminists are still not resolved. Using sources that include the PsycINFO database, the Publication Manual of the American Psychological Association, and popular psychology textbooks, the researchers discover that there is progress toward several of the goals advocated by feminist psychologists, namely eliminating the underrepresentation of women as researchers and study participants, as well as the tendency for researchers to draw unfounded conclusions from comparisons between men and women. Progress on other feminist criticisms, however, such as psychology’s reliance on laboratory experiments and quantitative methods, is less apparent. Furthermore, within the sample of psychological textbooks on research methods, the researchers note that little attention is given to epistemology— that is, what psychological scientists consider as knowledge about psychological phenomena. Although many, if not most, feminist psychologists work comfortably within the post- positivist epistemology that dominates psychological science, this neglect of epistemology offers little room to those who expand and challenge these dominant perspectives with alternative feminist epistemologies.

Fertility and Public Opinion

Does family size matter when it comes to changing public attitudes on social issues? According to a preliminary study by IPR sociologist Jeremy Freese and doctoral student and former IPR graduate research assistant Alex Kevern, it does. Past research has shown that collective opinions can shift as a nation’s demographics change and that children are more likely to share their parents’ cultural beliefs and political attitudes. Abortion attitudes are strongly linked to these two findings, which is why Freese and Kevern chose to study their “curious pattern.” They detail how Americans’ beliefs on abortion have taken a U-turn, becoming more liberal just after Roe v. Wade, then going flat and more recently seeming to turn more conservative. Using survey data from between 1977 (four years after Roe) and 2010, they then document a “liberalizing trend” on several correlated attitudes—including gay marriage, which has experienced an atypical, dramatically large increase—yet this trend has not extended to abortion. Freese and Kevern propose that part of the reason abortion has not liberalized might be due to a “fertility gap.” Pro-life parents have 27 percent more children on average than their pro-choice counterparts. Their higher fertility rate—perhaps due to having children at a younger age—means that children born to pro-life parents are more likely to hold pro-life views, eventually replacing those in the population who hold pro-choice views. While more research is needed, their results suggest that public opinion researchers need to take demography into account for a more complete understanding of changes in public opinion over time.

Surveying Socially Sensitive Issues

The survey measures researchers use to estimate socially sensitive issues, such as U.S. college student-athletes’ drug and alcohol use, might be biased, suggests a Social Science Quarterly article, which was previously an IPR working paper, by Druckman. Druckman and his colleagues, all former Northwestern doctoral students, used a survey technique known as a list experiment to investigate 1,300 student-athletes’ use of alcohol and banned substances; they then compared the results of the list experiment with the results of a traditional, self-reported survey of the same population. The list experiment asked whether students had participated in certain activities, but not which ones, thereby eliminating the fear inherent to traditional self-reporting methods, because respondents did not have to directly admit to certain activities. The differences between the list experiment and self-reported survey numbers were stark: 37 percent of list experiment respondents had knowingly taken banned substances, compared with only 4.9 percent who admitted to it when directly asked in the survey, and 46 percent of list experiment respondents consumed more than five drinks a week, compared with 2.4 percent who self-reported doing so. Though Druckman cautions against jumping to conclusions about student-athletes’ drug and alcohol use based on these results—for instance, a small amount of caffeine qualifies as a controlled substance under National Collegiate Athletic Association (NCAA) rules. The experiment demonstrates that overreliance on self-reported data on a sensitive topic like this can be problematic.

Labeling Social Interventions: The Case of Bail Bonds

In 1999, Cook County instituted a new system for bail bond hearings, in which those for less serious offenses were conducted via video, instead of face-to-face with a judge. A previous comparative interrupted time series study showed that the intervention had the unintended side effect of increasing bail bond amounts, threatening offenders’ right to fair treatment under the law. In part due to this finding, the new hearing system ended in 2008. Analyzing court data from 1991–2007 (eight years before and after the system’s implementation), a trio of IPR researchers confirmed that average bail amounts did increase significantly after it was enacted, but not for the reasons the previous study originally posited—poor video quality and the physical client-attorney separation. Rather, their analysis pointed to managerial inefficiencies stemming from how judges were assigned to the video bail hearings. Conducted by IPR social psychologist and research methodologist Thomas D. Cook, IPR postdoctoral fellow Yang Tang, and law professor, psychologist, and IPR associate Shari Seidman Diamond, the study points to the importance of a cause’s construct validity—or why social scientists must take care to construct meaningful labels that are empirically consistent with their data. Cook is Joan and Sarepta Harrison Chair in Ethics and Justice, and Seidman Diamond is Howard J.Trienens Professor of Law.

Redefining the Link Between Taxes and Sociology

In the Annual Review of Sociology, IPR sociologist Monica Prasad and Isaac William Martin of the University of California, San Diego argue that sociologists—especially those who study poverty and inequality—should pay closer attention to taxation. While taxation has been fingered as a critical factor in shaping inequality by thinkers such as Karl Marx and Émile Durkheim, current sociological research has largely failed to effectively study the links between tax policy, poverty, and income disparity. Prasad and Martin outline several key developments in past research, specifically examining taxation’s sociological impact in poor and rich countries via economic, sociological, and political science studies. One example they investigate is that of consumption taxes, such as sales taxes. Many economists argue that they are better at producing growth than income taxes, but a debate over whether high or low tax rates are better for development persists. The two researchers discover evidence that high taxes might help developing nations if the revenue is spent on “productive investments,” such as education, health, infrastructure, and efforts to reduce public debt.

Partisanship and Cooperation in Congress

Partisan Conflict and Legislative Compromise

As partisan conflict increases, public support for Congress tends to decrease, with approval for the highly polarized 113th Congress hovering around 14 percent in the weeks leading up to the 2014 midterm election. Yet survey evidence suggests that citizens do support partisanship if it advances their party’s positions. IPR political scientist Laurel Harbridge, in a working paper with Northwestern doctoral student D.J. Flynn, seeks to understand how partisan conflict affects Americans’ evaluation of how Congress is doing. First, the researchers test when party conflict is seen positively by comparing citizens’ evaluations of Congress across different legislative strategies—partisanship versus compromise, and outcomes of partisanship—a win for one’s own party, a win for the opposing party, and gridlock. Second, they investigate whether the effect of gridlock on public opinion differs depending on whether gridlock is attributed to ideological disagreements or to party strategy. Third, they consider whether issue differences, and party agreement over end goals, shape citizens’ reactions to partisan conflict. Their results indicate that citizens disapprove when partisan conflict prevents Congress from acting on important, national issues. In fact, partisans are more approving of Congress’ handling of policymaking when a policy debate results in a win for the other party than when the debate ends in stalemate, but only on consensus issues (where the parties agree over end goals). Citizens are also significantly more accepting of legislative inaction when it is characterized as the result of genuine ideological disagreements between the two parties.

Assessing Congressional Bipartisanship

According to conventional wisdom, the U.S. Congress has become increasingly polarized in recent years, with bipartisan cooperation falling as a result. At the same time, argues Harbridge in her new book, Is Bipartisanship Dead? Policy Agreement and Agenda-Setting in the House of Representatives (Cambridge University Press, 2015), research on bipartisanship in Congress has focused mainly on trends in roll-call voting— which has become more partisan over time—and overlooked cosponsorship coalitions, which have remained more bipartisan. Harbridge looks beyond—and considers the time before— roll-call voting to examine the extent to which bipartisan agreement in the House of Representatives has declined since the 1970s. In analyzing cosponsorship and voting patterns from the last 45 years, she determines that party leaders in the House changed from prioritizing legislation with bipartisan agreement in the 1970s to prioritizing legislation with partisan disagreement by the 1990s. Harbridge then examines the factors that “manufacture” these increasingly partisan roll-call agendas, including district sorting—or the extent to which a legislator represents a politically homogenous area, and public opinion, despite bipartisan agreement on some issues. Bipartisanship, she concludes, is not dead, but legislative and electoral processes encourage members of Congress to prioritize partisan disagreement.

Legislative Holdouts

High levels of gridlock might be the defining characteristic of the 112th and 113th Congresses. While widening policy differences between Democrats and Republicans are often pinpointed as the reason behind political stalemates, Harbridge’s research suggests that legislative opposition stems from reasons beyond policy disagreement. In collaboration with Sarah Anderson of the University of California, Santa Barbara and Daniel Butler of Washington University in St. Louis, Harbridge and her colleagues examined “legislative holdouts”—or when a legislator votes against a policy that is closer to their ideal policy point than the status quo. The researchers leveraged a unique survey of state legislators that elicited their views on their states’ gas tax and their willingness to vote for a compromise policy—midway between their most preferred policy and the status quo. The researchers found that 28 percent of legislators were holdouts. Furthermore, Harbridge and her colleagues discovered holdout was most common among Republicans, members of the majority party, and those legislators who feared voter retribution for compromise, patterns that suggest that legislative holdouts might contribute to legislative gridlock at the national level in recent years. The researchers conclude in their IPR working paper that Congress’ current political constellation might be “the perfect storm for legislative holdouts,” and in turn, this might be a factor contributing to policy gridlock.

Gendered Incentives for Legislative Compromise

Legislative compromise is generally well received, or seen as beneficial. To this end, political commentators often suggest that Congress could be more bipartisan if more women were in the legislature, using existing research on legislative behavior and gender stereotypes as evidence. In an IPR working paper with Nichole Bauer of Davidson College and Yanna Krupnikov of Stony Brook University, Harbridge reconsiders these arguments about gendered compromise. The researchers’ results indicate that while gender can affect evaluations of legislators who fail to compromise, its impact largely depends on two factors: whether legislators are co-partisans or members of the opposing party, and whether the compromise is about a “women’s issue,” such as education, childcare, healthcare, or other social welfare issues. While conventional media wisdom argues that female lawmakers will compromise more than their male counterparts, this is not necessarily an expectation held among the electorate as female lawmakers face backlash for not compromising on issues identified as male-owned but not on female-owned ones.

Understanding Group Stereotypes

How do stereotypes come to be, and can they be changed? In the Journal of Personality and Social PsychologyEagly uses social role theory to test how stereotypes can develop from people’s observations. Eagly and the University of San Diego’s Anne Koenig surveyed study participants on the current stereotypes of many social groups, like black women, white men, millionaires, high school dropouts, and college graduates. Other participants indicated the social roles in which they believed that members of these groups were overrepresented—for example, white men as business professionals and lawyers. These beliefs about groups’ representations in occupations proved to be quite accurate in relation to data from the U.S. Bureau of Labor Statistics. Eagly and Koenig also collected data indicating the stereotypes of these occupations, for example, lawyer, teacher, and fast food worker, which are typical of various social groups. The key finding is that groups’ overall stereotypes are strongly predicted from stereotypes about the occupations in which the groups are overrepresented. Also, when the researchers described a social group’s typical social roles as changing to different social roles in the future, participants’ group stereotypes changed as well. In fact, descriptions of a group’s future roles had a greater impact on each group stereotype than did that group’s present stereotype. This research has important implications for the workplace because movement away from current group stereotypes should arise from observing members of particular groups in new roles that have different demands than their old roles.

Political Communication and Issue Frames

Politicization of Science and Technology

Does the politicization of science influence support for scientific innovations? Can it render appeals to evidence inconsequential? In a series of studies, IPR political scientist James Druckman, IPR social policy professor Fay Lomax Cook, and Toby Bolsen of Georgia State University, a former IPR graduate research assistant, use experiments and survey data to examine public opinion related to energy policy. In an IPR working paper now published in Public Opinion Quarterly, they randomly present varying informational conditions, or “frames,” to a nationally representative sample of 1,600 participants.The frames included information on benefits and drawbacks of nuclear energy and various references to the politicization of science.Their results show that politicizing science undermines arguments about the environmental benefits of nuclear energy, regardless of whether the arguments do or do not cite supportive scientific evidence. It even serves to reduce support for using evidence in the first place. A second study reveals that references to the potential health risks associated with using nuclear power also decrease support for nuclear energy use, despite additional frames highlighting the benefits of the technology or the politicization of science. This research demonstrates that the politicization of science has created a “status-quo bias,” which future research should focus on overcoming to gain public support for emerging technologies. The Initiative for Sustainability and Energy at Northwestern (ISEN) provided funding for this work. Druckman is Payson S. Wild Professor of Political Science and IPR associate director. Lomax Cook heads the Directorate for Social, Behavioral, and Economic Sciences in the National Science Foundation. 

Partisan Motivated Reasoning

Does party identification make us support policies we would otherwise reject, or oppose ones we would otherwise support? In Political BehaviorDruckmanCook, and Bolsen wrote about another ISEN-supported experiment to test this question, for which they used the bipartisan example of the 2007 Energy Act. Participants were assigned to one of three motivational conditions: no motivation, accuracy motivation (where participants were told they would have to explain their reasons for opinions about the policy), and directional motivation (where participants were told they would be asked about their party affiliation after the policy questions). They were also either given no endorsement information, were told one party had endorsed the Energy Act, were told there was consensus, bipartisan support, or were told some, but not all, members of either party supported it. When individuals were primed to defend their partisan identity, they shifted their evaluations of the Energy Act toward their party’s position—away from positions endorsed by the other party. The researchers also found that both telling the participants they would have to explain their attitudes and being told of bipartisan support eliminates partisan-motivated reasoning. Support for the Energy Act increased across the board when both parties were shown to support it. 

News, Technology, and Online Behavior

Online “Turbulence”

In a world where social network sites allow us to share information with hundreds of people in mere seconds, it is not surprising that some Internet users experience “turbulence”— times when their personal information is distributed beyond their desired or intended social circles with negative consequences. In Computers in Human Behavior, communication studies researcher and IPR associate Eszter Hargittai and Northwestern doctoral student Eden Litt surveyed 547 young adults to determine who is more likely to encounter turbulence. The researchers asked students about several “privacy- enhancing” behaviors to determine if they knew, for example, how to “untag” themselves from content posted by another, about online “self-monitoring,” which involves using context and social cues to tailor online presentation, and about students’ own Internet skills. They discovered that more than one-third of survey respondents had experienced some form of online turbulence, or online content that embarrassed them, caused a fight, ended a friendship, etc. There was no relationship between a person’s gender or socioeconomic status and whether they experienced turbulence. Understanding that those who were more Internet savvy were less likely to experience turbulence offers a potential point of intervention: If people are more informed Internet users, they might be less likely to encounter such outcomes. Hargittai is April McClain-Delaney and John Delaney Research Professor.

Mobile Media Use in the Middle East

Twitter, Facebook, Instagram, and others have played an undeniable role in shaping the current sociopolitical landscape of the Middle East, yet most mobile media research to date has focused on the business aspects of its use—not, for instance, how its content might influence public education and engagement. Mersey is helping to launch a new study that will seek to understand the development and diffusion of mobile media content in the Arab world. Working with John Pavlik of Rutgers University and Everette Dennis, dean of Northwestern’s campus in Qatar (NU-Q), the trio will deploy computer science methodology and data analysis with an eye toward creating a model of innovative mobile content designed to foster learning and engagement in Arab countries. With funding from the Qatar National Research Program, the study will examine data from social networking sites, including location-based data, and will pair the data with field surveys and interviews with participants in Qatar and the United Arab Emirates. The project marks a continuation of Mersey’s research on Middle East media and collaboration with NU-Q, where she was a research fellow.

Monitoring the Economy

Are people more likely to base how they think the economy will do on their past evaluations or forward-looking information? A case for the former (past evaluations) emphasizes the high, and arguably prohibitive, costs of prospective information. An argument for the latter (forward-looking information) asserts simply that rational expectations agents will consume the information that provides them with the most accurate forecasts. Rather than contrasting retrospective with prospective, Kernell’s project allows the value and costs of prospective information to vary over time, as well as across individuals, with results showing that heterogeneity is preserved and even extended during periods when economic news is plentiful.

Analytics and Loyalty Among News Consumers

Though researchers and members of the media generally agree about the rising influence of media consumers, little research has been done to understand these new relationships and their implications for audiences, the media, and democracy. IPR mass communication scholar Rachel Davis Mersey and media scholar and IPR associate Stephanie Edgerly are developing two instruments to shed light on the relationship between media consumers and news organizations. With seed grant funding from IPR, the researchers are creating a survey for journalists working at various levels throughout news organizations to determine their perceptions of media analytics and the usefulness of analytics in their routine decision making. Mersey and Edgerly have also generated a new research program to understand consumers’ paths to media loyalty, taking stock of the complex media landscape and offering insight into the paths news organizations can take to build audience loyalty.

Tone, Topic Influence Video Comments

While video-sharing site YouTube has provided opportunities for political engagement, it also raises questions about whether informed democratic deliberation takes place on the site. In the Journal of Information Technology & PoliticsEdgerly and her colleagues examine the relationship between the tone and focus of a political video and its online comments, asking why an online comment thread might veer from civility. They review nearly 46,000 comments attached to 207 YouTube videos about California’s 2008 Proposition 8, which sought to legally define marriage as being between a man and a woman. While videos with an uncivil tone generated more impolite comments, the researchers discovered, surprisingly, that there was no relationship between humor—such as making fun of people—and uncivil comments. They also reveal that the topics in a video can serve as a springboard for comments. For example, if a video discussed religion, its comments touched on religion, too. The research displays the psychological mechanisms behind information processing and thought activation on YouTube, confirms the effects of the framing and priming process— how exposure to certain stimuli affects responses to a later stimulus—and indicates “exchange” between a video’s topic and its comments.

Examining Patents

In light of the controversy surrounding “patent trolls”—people who buy up a portfolio of patents for inventions they did not create, then threaten lawsuits against inventors who might infringe on them—the U.S. Patent and Trademark Office (PTO) has been criticized for excessive patenting. But is this actually occurring? IPR health and law scholar Michael Frakes is conducting research on how the structure of the PTO encourages over-granting. He and Melissa Wasserman of the University of Illinois at Urbana–Champaign examine if how long a patent examiner has to review a patent affects the number of patents the examiner accepts or rejects. They studied patent examiners over time as they received promotions within the PTO. Each promotion carried with it the expectation that the examiner would spend 10 percent less time evaluating each patent application before making a decision to accept or reject it. After sorting through 1.4 million patent applications from 2001–12, the researchers conclude that grant rates rose overall as examiners were promoted. Examiners also searched less for prior art—information related to the patent that was already available to the public—and rejected fewer patents when promoted.

Political Parties and Participation

Presidents as Agents of Change

Presidents are usually seen as operating in a political environment that is highly resistant to change. Though powerful actors, presidents are depicted as having limited capacities to alter the institutional and organizational arrangements that surround them. Building on recent historical-institutional research, IPR political scientist Daniel Galvin challenges this assumption. He shows that under many conditions, presidents can, in fact, alter their structural confines and reshape their political environment in historically significant ways. In an IPR working paperlater published in Presidential Studies Quarterly, Galvin develops this conceptual framework and offers methodological suggestions for conducting historically oriented research. Reconsidering some recent research into the relationship between presidential action and party development, he reveals that presidents contributed to several critical party developments in American history: Late 19th-century presidents contributed to the gradual nationalization of the party system, mid-20th-century ones helped build the modern “service” party, and more recent presidents fostered greater partisan polarization among interest group networks. Motivated to bring inherited party structures into closer alignment with their goals, these presidents summoned powerful resources to reshape their par ties. Rather than leave their structural environment undisturbed, as leading theories might predict, they reconfigured their parties and altered their trajectories.

Presidents, Public Opinion, and Manipulation

Democratic representation rests on the premise that elected officials respond to citizens’ opinions. In the book Who Governs? Presidents, Public Opinion and Manipulation (University of Chicago Press, 2015), Druckman and his co-author Larry Jacobs of the University of Minnesota reveal a system of governance in the United States that is geared toward the opposite—the diminishing or obstruction of responsiveness to broad public opinion. Relying on public opinion data from U.S. presidential archives, the authors show that presidents expand their ability to make policy by influencing which issues are salient—largely ignoring the views of citizens on those issues that are not—and turning the public’s focus to their personality, rather than policy issues. Melding archival and quantitative research with democratic theory, Who Governs raises broader issues about contemporary debates concerning renewed shared American prosperity and the advancement of national interests abroad. It challenges this conversation by questioning the records and motivations of American governing elites. Correcting America’s core, the authors write, requires more than new agendas; it requires a new form of governance that scrutinizes governing elites and their claims to serve the national interest when, in fact, they advance their own interests and those of their supporters.

Rust Belt Democrats

Since the 1970s, left-leaning parties around the world have been under pressure to adapt to changing economic and political conditions. With globalization and deindustrialization shrinking organized labor’s membership base and undermining the credibility of traditional social-democratic policy agendas, these parties have faced incentives to develop new policy initiatives and court new electoral constituencies. The U.S. Democratic Party is usually thought to have responded to these incentives slowly, poorly, or not at all, and this is presumed to help explain their electoral difficulties since President Ronald Reagan. To investigate this, Galvin turns to the Rust Belt—the region hit hardest by globalization-related trends—in his latest book project, to be published by Oxford University Press. He uncovers surprising variation in the adaptive capacities of Democratic parties in four of the heaviest manufacturing states—Michigan, Ohio, Indiana, and Wisconsin. Drawing upon extensive primary- source research, interviews with participants, and massive data collection of state party platforms, Galvin finds that these parties’ historical ties to organized labor, urban machines, and liberal interest groups (in different proportions in each state) had important consequences for their downstream activities. This project has already turned up some surprising findings. In an IPRworking paper, Galvin demonstrates that the relationship between the Michigan Democratic Party and the United Auto Workers Union remained unusually strong between 1970 and 2010, yet Democratic politicians frequently promoted “third- way” policies that clashed with labor’s longstanding priorities. This odd coincidence—party adaptation despite strong party- union linkages—can be explained by the simple fact that union leaders deliberately supported adaptation by Democratic politicians. Contrary to the expectation that interest groups will always push party politicians to take more extreme positions, UAW leaders adopted a highly strategic approach to party politics as they sought to build a broader Democratic coalition.

Workers' Rights, State Laws, and Enforcement

Galvin has launched a major project on workers’ rights, state laws, and policy enforcement in the United States. Across the country, there are enormous variations in wage and hour laws and dramatic differences in state capacities to enforce those laws, and Galvin seeks to examine the political causes and consequences of this variation. He considers both sides of the equation: the factors producing changes in employment regulatory policies and the effects of such policies on workers’ rights and protections, as well as their subsequent political mobilization.

Party Systems and Decentralization in Africa

What factors determine when and how governments implement decentralization reforms? Political scientist and IPR associate Rachel Beatty Riedl and J.Tyler Dickovick of Washington and Lee University tackle this question byexploring recent efforts to devolve power to subnational units in Ethiopia, Botswana, Ghana, and Benin. They examine the regime type, the degree of competitiveness, and the coherence—or the level of volatility and fragmentation—of the party system. Contrary to conventional wisdom, the researchers reveal authoritarian regimes have incentives to decentralize further than democratic regimes, in order to extend government leverage to the local level. They also determine that decentralization is likely limited in democracies with dominant political parties, where it could threaten dominant incumbent parties by empowering opposition parties to gain a foothold subnationally. In democracies with competitive and coherent parties, however, both incumbent and opposition parties might benefit from decentralization by establishing local footholds and increasing patronage opportunities. By tracking the logic of power holders, the authors suggest that the policy implications of decentralization for local autonomy vary according to the strategies of implementation.

Inside Political Parties

In her book manuscript, “Inside Parties: Organizations, Electoral Success, and Comparative Political Behavior,” IPR political scientist Georgia Kernell investigates the internal party rules that shape representation, participation, and electoral outcomes. In a study of 66 parties from 20 parliamentary democracies, Kernell examines intraparty rules that determine candidate and leadership selection, party platforms, and the allocation of resources. She discovers that decentralized parties, or those that allow local party members to participate in important decisions, are more likely to better represent their core supporters at the expense of adopting less competitive positions in the general electorate. In contrast, centralized parties—where party elites are responsible for candidate selection, setting the platform, and other activities—are better able to choose candidates and position the party platform to reflect the interests of the entire electorate. Kernell also argues that voters are less likely to identify with a party or politically participate if the party they support is decentralized—partially due to the ambiguity of policy appeals in decentralized parties.

Strategic Party Heterogeneity

Candidates from the same party often adopt different policy positions, yet existing models of electoral competition typically assume that each party is made up of a homogeneous set of like-minded politicians. Kernell’s research models the strategic decision by political par ties to field a more or less heterogeneous set of candidates. She uncovers that in two-party competition, there is a unique level of party heterogeneity that maximizes a party’s chance of winning the election. Parties with platforms positioned closer to the median voter should field a more cohesive set of candidates; those that are farther away should be more heterogeneous. The model has important implications for party discipline in legislatures, and when dealing with polarization and voter uncertainty. This articleis forthcoming in the Journal of Theoretical Politics.

Voting Patterns in New York

After 20 years of center-right and conservative mayors, New York City voters elected a far-left candidate, Bill de Blasio, in 2013. At an IPR colloquium talk, political scientist and IPR associate Thomas Ogorzalek examined the factors that led to de Blasio’s victory in a field of many viable opponents, such as former New York City Comptroller Bill Thompson and Council Speaker Christine Quinn. While some point to de Blasio’s strong stance against New York City’s “stop-and-frisk” policy as the reason he won the election, Ogorzalek does not see any link between candidates’ stances on stop and frisk and election results. Instead, he cites de Blasio’s backing by New York City’s “Working Families Party” as the basis for his success. Running on that party’s ticket, in addition to the Democratic ticket, resulted in a turnout boost among de Blasio voters in areas where the Working Families Party was strongest. Ironically, its stronghold during the 2013 election was Park Slope, an affluent neighborhood in northwest Brooklyn, residents of which would ostensibly identify less with its platform of paid sick leave, affordable housing, and free preschool and higher education. This research is the beginning of a multicity study Ogorzalek is launching to investigate the core of “blue America.”

Women and Leadership

At least one of the current candidates for president or vice president in the 2016 election is a woman, and several high- profile companies have recently named women as their heads. Are women finally breaking through the “glass ceiling”? In an ongoing project, IPR social psychologist Alice Eagly is uncovering subtle but important differences in how men and women lead— with women displaying slightly more of the leader behaviors found effective by researchers. The public also seems to favor seeing more women in leadership roles. At the same time, long-held stereotypes of gender and leadership continue to reinforce perceptions that women do not lead as effectively as men, portraying men as “take charge” and women as sensitive. Thus, gender stereotypes of men match well with those held for leaders, but not gender stereotypes of women; moreover, both men and women tend to equate leadership mainly with male characteristics. To ameliorate the issue, Eagly suggests that people would have to regard leadership as requiring feminine and masculine traits, and women would have to add assertive, culturally masculine qualities to their repertoire. Women, she notes, already manifest more ambition and assertiveness than in the past, but only time will tell if the differences in the values that women and men promote as leaders will persist as women achieve more equality in society. Eagly is James Padilla Professor of Arts and Sciences.