Search

This paper explains the recent eletoral rise of anti-establishment parties in Eastern Europe in terms of voter rejection of mainstream political elites in third-generation post-communist elections. Using survey data from the 2000 and 2004 Romanian parliamentary elections, the paper identifies the circumstances under which prospective or retrospective considerations dominate electoral choices of voters. [ABSTRACT FROM AUTHOR]

The purpose of this paper is to test the plausibility of the homogeneity or parallel publics hypothesis (Page & Shapiro, 1992) using the agenda-setting model as a framework. One of the implicit assumptions of the agenda-setting model is that in the absence of a news story about a particular topic, such as stories about air quality, public concern about the topic is at equilibrium. In response to a media story or external events such as a public information campaign asking drivers to alter their driving behavior, concern about the topic of air quality may quickly diverge from the long-term equilibrium level of concern. This equilibrium state is similar to Downs’ (1972) pre-problem state or Baumgartner and Jones’ (1991; 1993) values consensus state. A focusing event or system shock punctuates the equilibrium state causing concern about air quality to rise. Eventually, after the media or the public becomes bored with the issue of air quality, concern will return to an equilibrium level of concern (Downs, 1972; Henry & Gordon, 2001; Neuman, 1990). This change in the level of concern about air quality in response to a system shock such as a news story or focusing event such as a public information campaign raises a fundamental question; do all members of the public respond similarly or in the parallel fashion predicted by Page and Shapiro to the system shock? That is, do men and women or do people who are college-educated or not college-educated respond in the same way at the same time to news stories or a public information campaign about air quality? The homogeneity or parallel publics hypothesis suggests that all groups respond similarly. The heterogeneity hypothesis suggests that groups may differ in their response to a system shock with the public either converging or diverging. The results from the models tested in this paper provide strong evidence that the homogeneity or parallel publics hypothesis is untenable. The initial analysis, correlating the aggregate means and variances gave strong indication that there were different processes at work. The results of the vector autoregression analysis show significant differences in response patterns to the media coverage and public information campaign based on sex and education. The results provided support for what we will call the convergent heterogeneity hypothesis, though the sex series was more consistent with what we called the divergent heterogeneity hypothesis. In an attempt to explain the disparate findings between the people and group models, we simulated the effects of a public information campaign to demonstrate that different groups responded in systematically different ways to the system shock with the overall effect washing out. [ABSTRACT FROM AUTHOR]

In this proposed paper we expand on the earlier research by Garand and Procopio (1995), who used data from the 1992 American National Election Study (ANES) to model popular support for term limits as a function of a variety of variables, many of which are found in the scholarly literature. However, unlike previous work Garand and Procopio include independent variables representing the effect of characteristics of U.S. House members, especially seniority. Garand and Procopio suggest that, controlling for the effects of other variables, citizens will respond to the characteristics of their representatives and adjust their support for term limits accordingly. For instance, they find that citizens represented by House members with high seniority are less supportive of term limits than citizens represented by junior members. We propose to extend this general model to data from the 1994 and 1998 ANES surveys. We also propose to consider whether public opinion on congressional term limits is shaped by the configuration of partisan identification and partisan control of Congress. With rare exception, previous research on support for term limits in the mass public has focused on the pre-1994 period in Congress, during which time both the U.S. House and U.S. Senate were controlled by the Democrats. Such a focus does not allow for a full longitudinal test of the connection between partisan attitudes and attitudes toward term limits. Specifically, previous research has found that Republican identifiers are much more likely than Democrats to support term limits, but it is unclear if this is an integral part of being a Republican or if support for term limits would change once the Republicans are in control of the U.S. House and U.S. Senate. We speculate that Republican support for term limits in 1992 was based in part on self-interest, insofar as Republicans did not have majority status in the Congress at that time; on the other hand, we speculate that, once control of Congress shifted to the Republicans with the 1994 congressional elections, support for term limits among Republican citizens should have declined and support for term limits among Democrats should have increased. In this proposed paper we will utilize data from the 1992, 1994, and 1998 ANES surveys, which are the only ANES surveys that include items on support for term limits. We estimate a series of models to explain variation in individuals’ support for congressional term limits. Our core model includes demographic variables, as well as variables measuring political attitudes, system support, legislative support, and political knowledge. We also intend to include characteristics of respondents’ House members in our term limit model; these characteristics will include members’ seniority, the ideological distance between respondents and their House members, and estimates of grant funds brought to each district, among others. Moreover, we consider the mediating effect of political knowledge on the effects of these variables on support for term limits. Following Garand and Procopio, we suggest that high- and low-knowledge citizens use different kinds of information in shaping their attitudes toward term limits. Finally, we focus particular attention on how the effects of partisanship on support for term limits varies under periods of Democratic control of Congress (before 1994) and Republican control of Congress (after 1998). [ABSTRACT FROM AUTHOR]

The September 11, 2001 terrorist attacks impacted the United States as a nation, but also had implications for local government. In an earlier paper, I analyzed the impact of the terrorist attacks on the 2001 New York City Mayoral election. While proximity to the terrorist attacks increased the probability of voters expressing uneasiness about new mayoral leadership, voters were not willing to overturn term-limits and continue with Giuliani as mayor. Partisan identification remained an important indicator analyzing favorability for Giuliani and the new mayoral candidates. Many questions remain to be answered in the realm of public opinion on local government, particularly during a time of crisis. In 2001, despite heavy Democratic Party identification in New York City, the Republican candidate for mayor, Michael Bloomberg won with just over 50 percent of the vote. Public support for Republican Mayor Rudolph Giuliani was very high following the September 11 terrorist attacks. As a co-partisan of Giuliani, did Bloomberg benefit from the rally of support for the Republican incumbent? In this paper, I ask did the September 11 terrorist attacks increase the probability of cross-over voting in the 2001 New York City mayoral election? Alternately, did individual-level perceptions of the ideological positions of the candidates influence cross-over? Are Democratic Party identifiers who perceive Bloomberg as a moderate more likely to offer him their vote? Are candidate-specific characteristics, such as a “business” versus “political” background important in the analysis of cross-over? To study cross-over voting and the impact of the terrorist attacks on local elections, New York Times survey data on deposit with the ICPSR will be used. A poll of New York City residents conducted from October 27 through 31, 2001 provides the data necessary for this inquiry. I anticipate that concern about terrorist attacks and security will have a positive impact on the probability of voting for Bloomberg. Democratic voters impacted by the terrorist attacks will have a higher probability of issuing a cross-over vote. Further, Democratic Party identifiers with a history of voting support for Mayor Giuliani will also be more likely to support Bloomberg. In other words, in the context of a large city local election, cross-over should be a function of prior cross-over voting behavior. The presence of crisis should have an added effect. The co-partisan of the incumbent mayor will gain support from those impacted by crisis and by those concerned about additional attacks. This study will advance our understanding of how crisis impacts public opinion about local government and how local electorates behave immediately following a terrorist attack. [ABSTRACT FROM AUTHOR]

This paper studies the impact of public opinion on presidential decision-making in foreign policy. Although a good deal of research looks at this general question, beyond the broad distinction between crisis and non-crisis decisions, the literature is largely inconclusive. We begin from this central distinction, utilizing the International Crisis Behavior Project’s criteria as the starting point for case selection. To observe in detail how differing patterns of public attentiveness influence foreign policymaking, we compare four cases: the 1991 Gulf War; the 1999 U.S. intervention in Kosovo; the U.S. response to Ethiopian famine in the mid-1980s; and U.S.-Japanese economic relations between 1988-1992. To move beyond the broad question of “public opinion and foreign policy,” we propose a more precisely specified analytical model, conceptualizing foreign policy decision making as a five-stage process: problem representation, option generation, policy selection, implementation, and policy review. At each decision stage, the question is asked: How influential was public opinion on a president’s decision? The model generates several testable hypotheses. For instance, foreign policy crises tend to produce a public that is highly attentive throughout all stages of the decision process. While strategic imperatives often overwhelm domestic political considerations in the early stages of a crisis, the relative importance of public opinion as a decision premise tends to increase during later stages, with presidents apt to sacrifice strategic effectiveness to pacify a highly attentive domestic audience during the implementation phase. Conversely, salient non-crisis foreign policies produce a selectively attentive public, whose greatest influence is likely to be on the president’s selection of policy rather than the manner in which that policy is implemented. Freed from the constraints of an attentive public, presidents are afforded considerable autonomy in implementing non-crisis policy in a manner consistent with their vision of the national interest, even if the general public does not share that same vision. This paper outlines the model and provides a brief overview of each of the cases, then assembles data on aggregate public opinion from a wide range of surveys and polls, and analyzes the resulting time series to compare the pattern of public involvement to the model’s predictions. [ABSTRACT FROM AUTHOR]

What makes people care about issues? Why do some issues act as agents of change in the electorate while many others are swallowed up by existing cleavages or just fade away altogether? Is there any way to predict which issues will stick, which will die, and which might instigate gradual, mass partisan transformation? In their model of partisan change, Carmines and Stimson (1989) note that elites establish and shape the reputation of political parties on issues. But what mechanisms do elites use to do that? A great deal of work has established the media’s role as an agenda setter (Iyengar and Kinder 1987), but less is known about the impact of second level agenda-setting, or framing (Nelson and Kinder 1996, McCombs, Weaver, and Shaw 1997). I demonstrate that the framing of issues impacts the public’s belief that an issue is important. Additionally, partisan framing appears to impact opinion formation and change on political issues. After analyzing coverage of President Bush’s tax cut, the California energy crisis, and coverage preceding the latest war in Iraq, this paper demonstrates that one answer to the question of what mechanisms do elites use to shape the reputation of political parties on issues lies in the way that political issues are framed in mass media coverage. This paper uses content analysis of four major newspapers to show that at times of low mass salience, issues receive coverage that include several, often disparate frames. As the number of frames for issues are winnowed down in mass media coverage, the salience of the issue increases. Further, opinion change appears to be linked to the winnowing down of frames into two dominant, partisan ways of characterizing a particular issue. [ABSTRACT FROM AUTHOR]

Existing research shows that the public knows the least about the Supreme Court, but affords the Court more favorable ratings than it does to the legislative and executive branches. We ask: Is the Supreme Court?s legitimacy, in part, a function of public uncertainty about its decisions? This paper examines the effects of increased information about the Supreme Court on public confidence in the Court. Specifically, the paper seeks to understand how, if at all, public confidence in the Court changes after individuals are presented with information and asked their opinions about multiple Court decisions on salient issues. The paper analyzes the results of a June 2003 split-ballot national telephone survey. Employing an experimental design that measures public confidence in the Court after exposing a randomly assigned group of respondents to information about six Court decisions across a range of salient issues, compared to a control group not given such information, allows us to see how political information affects public opinion toward the Court in the aggregate and across subgroups. While overall it seems, not inconsistent with prior research, that public confidence in the Court is largely unaffected by the new information, we do find noticeable differences among certain subgroups that indicate that public opinion toward the Court is more complex than has been shown thus far but in ways that are readily explicable. In particular, confidence in the Court may decline as the public acquires information on the Court’s rulings that contradict the preferences of a very large and longstanding majority of the public. [ABSTRACT FROM AUTHOR]

This paper evaluates the impact of individual attitudes on the relationship between presidential rhetoric and the public agenda. Extant research on presidential agenda setting relies almost exclusively on data collected at the aggregate level and, as a result, is incapable of capturing much of the variation in individual reactions to presidential speeches. The available evidence indicates that individual attitudes, such as presidential approval and partisan identification, can play a critical role in determining whether presidential speeches succeed in influencing the public’s policy preferences (e.g. Meernik and Ault 2001; Page, Shapiro, and Dempsey 1987; Page and Shapiro 1992). But it is much less clear whether such attitudes influence the dynamics of presidential agenda setting. Relying on individual-level survey data collected immediately following four State of the Union Addresses delivered by four different presidents, this paper estimates the relationship between presidential rhetoric and the public agenda as a function of exposure to the president’s speech, presidential support, political predispositions, and demographic characteristics. The results indicate that presidential rhetoric is an important determinant of the public agenda across all respondents, but an even more powerful determinant among those who voted for the president and approve of his performance, among those who share partisan and ideological predispositions consistent with the president’s, and among those who belong to demographic groups traditionally supportive of the president’s party. [ABSTRACT FROM AUTHOR]

The structure of beliefs, the role of ideology as an influence on policy preferences, and the connections between foreign and domestic beliefs have motivated analyses of public and elite opinion. To this point the findings of empirical research in this area have been limited largely to American survey data. There has been little cross-country and cross-region comparative research to examine the generality of American findings. In this paper, we will report the findings of an analysis of foreign policy beliefs of technically sophisticated elite communities drawn from the member countries of the European Union and the United States. To what extent do elites in other countries display a similar organization of belief systems and arrive at similar policy preferences? Are policy preferences influenced by ideology and situational differences, even among a group internationalized by training, professional socialization, and to a large extent by language, and even within technically sophisticated issue areas? To answer these questions we administered an identical survey in sixteen different national settings (translated into French, German, and Italian) to scientific elite subsets of the population in 2002. The paper will explore the role of ideological constraints and geopolitical context in models of foreign and domestic policy preferences toward nuclear weapons and energy respectively. We examine the Atlantic divide and compare superpower United States with European Union responses. Second, we examine national samples for the United States, United Kingdom, France, and Germany, and “other EU” countries. Preliminary analysis of the data suggests strong support for the general hierarchical pattern of beliefs, with important cross-country differences within a hierarchical pattern. [ABSTRACT FROM AUTHOR]

Through the 2008 elections, the majority of Americans have favored “the establishment of an independent Palestinian state on the West Bank and the Gaza strip.” Reflecting such public opinion, after the “roadmap,” US foreign policy will continue the pursuit of that end, calling for “establishing two independent states, Israel and Palestine, living side by side within secure and recognized borders.” Consequently, the main question asked by this paper is why the US has shifted from an “Israel only” position toward the Israeli-Palestinian conflict to supporting an “Israel and Palestine” formula for peace. In order to address this question, the pro-Arab lobby was examined for interactions with US decision-making toward the Israeli-Palestinian conflict before and after the end of the Cold War, specifically from the Camp David period of the late 1970s until the invasion and occupation of Iraq during the mid 2000s. Paper concludes that while international factors such as the end of the Cold War cannot be overlooked, changes in domestic public opinion and lobbying had significant impacts on the US led international community’s efforts to resolve the Israeli-Palestinian conflict with two states. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper augments existing theoretical treatments of executive compensation by developing a rationale for political factors to influence corporate decision-making. It assess whether either changes in public trust in corporate America or in the SEC’s enforcement activity affect the salaries and benefits awarded to CEOs, while controlling for sociological and management theories. Political science has recently emphasized questions of inequality, and this paper contributes by demonstrating that due to institutional constraints, public forces have a limited ability to check inequality at its root. I analyzed the CEO salaries provided by the S&P 500 from 2001 to 2005 using a panel analysis. The data were gathered from Gallup polling, the SEC’s enforcement division, and firms' statements. The results demonstrate no effect on compensation for shifts in public opinion. This finding is attributed to two institutional roadblocks. In contrast, regulatory threat in the form of increased SEC activity does appear to lead relatively more firms to limit compensation. Nevertheless, the paper concludes with a discussion of how few firms limit the compensation provided to their CEOs (less than one-quarter) and provides reform proscriptions. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper analyses how European level deliberations affect individual opinions, and how this translates into domestic public policies. It looks at deliberation as a way to investigate non-coercive modes of governance in the European Union, in particular the Open Method of Coordination and uses the Bologna Process, designed to create a European Higher Education Area, as an example. This paper suggests a theoretical framework inspired by political philosophy to understand the impact of deliberation on individual attitudes and policies. It also uses multiple lenses of analyses, relying on participants' surveys, a case study of deliberations leading to the Sorbonne declaration of 25 May 1998 and an analysis of domestic implementation to understand how deliberation affects many levels of policy-making. Results show that individuals change opinions in Bologna Process deliberations toward a more economically liberal perspective. This change transfers into domestic reforms because European and domestic deliberations are connected: the arguments learnt in European deliberations are used in domestic deliberations. But the extent of change depends on the level of entrenchment of the issue and personal interests. Those results lead to inferences on the way to achieve consequential deliberative decision-making in the Open Method of Coordination in conclusion. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper addresses the organization of Chinese nationalist movements in 2008. I focus on online facilitation of offline activism for Chinese nationalists. The paper articulates the framing, organizing, financing, recruiting as well as reconfirming effect of cyberspace on social movement. First, the Internet plays a remarkable role in mobilizing collective action across boarders. The BBS forums broadcasts protest appeals and boycotting actions, provides follow-ups and feedbacks, and financially supports upcoming protests. Second, Cyberspace has recruiting effect among the politically inactive population, having engaged a large number of female Chinese. Finally, high technology diffuses nationalist emotions at a global base. The imagined “patriotic community” reconsolidates the nationalist identity in the overseas Chinese community. The data was collected through web-based surveys and content analysis of postings. The research enriches academic understanding of how virtual communities transform public opinion into a new form of globalized cultural trend for a cross-sectional ethnic community. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

The paper uses survey data from 2004 from eight democracies in Central America, Mexico and Colombia from the Latin American Public Opinion Project to explore political legitimacy and its links to political participation. Theory argues that political legitimacy is multidimensional; this proves empirically correct, producing six legitimacy dimensions. Six modes of political participation are also identified empirically. Prior research on Costa Rica had suggested that some legitimacy-participation functions are U-shaped, such that citizens with very high and very low legitimacy norms would take part in politics more than those with intermediate legitimacy norms. This paper employs OLS regression analysis to test for the effects of legitimacy upon participation, modeling the effect of both linear and quadratic forms of political legitimacy to test for the curvilinear relationships, as well as the influence of multiple intervening and control variables. By far the predominant form of relationship revealed is indeed U-shaped. The paper explores the implications of this finding for both the legitimacy research and for the potential stability of democracies. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

The paper presents a quantitative content analysis of media coverage of ethanol appearing in the New York Times from 1998 to 2008, paying particular attention to changes overtime in the quantity, tone, and framing of coverage. The analysis also takes into account key political and economic events such as elections, economic downturns, and energy crises, and the role of various political actors. Lastly, the paper includes a discussion of the public opinion and policy implications of the trends in media coverage. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Oliver Wendell Holmes famously said '[e]very idea is an incitement,' and that, "[t]he only difference between the expression of an opinion and an incitement in the narrower sense is the speaker's enthusiasm for the result." Building off of Holmes, we suggest that the Court move away from incitement-based tests, including the Clear and Present Danger and Brandenburg standards, and create a new and more narrowly-tailored ex ante category to be beyond the reach of First Amendment protection. In order to do this the Court must move away from analyzing speech that is 'close to' crime as a matter of proximity, degree, or close temporal connection. Instead, the Court must focus on speech that is substantially and integrally related to the successful completion of a criminal enterprise. This category of speech is defined by UCLA Professor of Law, Eugene Volokh as crime-facilitating speech. In this paper, we argue that in order for speech to be a type of concrete advocacy that can be appropriately considered beyond the pale of First Amendment protection it must supply more than an underlying motivation for the commission of a crime. This requires that the speech take on a conduct-like character that facilitates or aids in a criminal enterprise.There is growing evidence that the courts have been heading in the direction of crime-facilitative speech standards. This paper draws from these experimentations in setting aside the Brandenburg standard, and suggests that the Court adopt the following standard: (1)The speech must rise to the level of facilitation/instruction of a crime serious enough to be categorized as a felony. Such facilitative/instructive speech must provide some 'essential' knowledge, beyond general knowledge widely accessible to the general public. (2)By providing this essential knowledge, the facilitative information provides the 'inextricable link' between the information that aids and abets the crime and the successful completion of the criminal enterprise, and; (3)The speaker/publisher must willfully provide the facilitative/instructional information to the principal with the purposeful intent of aiding and abetting a successful criminal enterprise. In the final section of this paper, we will address how our standard will stand up against potential challenges, including the special challenges related to published material, dual-use material, facilitative information that has entertainment value, severe harms related to terrorism and nuclear attacks, and finally the internet and 'dissemination-once-removed.' ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

In his Discourses Machiavelli claims that one of the best aspects of the Roman republic was that conflicts between the nobles and the plebs resulting from policies such as the agrarian laws helped to sustain the republic, and he recommends pitting different factions in society against one another as a method that should be employed by contemporary republics. This suggestion was not one shared by Machiavelli's Florentine contemporaries such as Savonarola, who asserted the opposite: that faction was the cause of all of the republic's ills. While scholars have given much attention to Machiavelli's analysis of the Roman republic, very little attention has been given to how Machiavelli's recommendations comported with the founders of the republic in which he lived, one of the most influential of whom was Savonarola. In this paper, I will compare and contrast Savonarola and Machiavelli's views on the causes and consequences of faction and argue that the distinct understanding that Savonarola and Machiavelli have of factions is deeply rooted in their different views of the interplay between politics and morality. This paper is part of a larger dissertation examining religion and republicanism in thought of Savonarola and Machiavelli. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Black and mainstream media sources tend to cover political events and issues in a different fashion. On the one hand, the Black press emphasizes advocacy and correcting the inaccuracies and stereotypes presented in the mainstream media. On the other hand, the mainstream papers value objectivity and balance, which often leads them to ignore important stories and sources within the Black community. Black citizens turn to Black papers for critical information, especially when events of special interest occur. Thus, I hypothesize that exposure to Black newspapers will have an influence on Black issue attitudes. Because Blacks are likely to filter information from mainstream papers and because mainstream papers tend to ignore stories of interest to the Black community, I hypothesize that exposure to mainstream papers will have no effect on Black public opinion. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Against the backdrop of the specific claim that the events of September 11th have transformed and/or is largely driving politics and policy within the major immigration-receiving countries this paper poses two related questions. First, has European public opinion objectively become more illiberal on immigration-related questions since September 11, 2001? Specifically, is it significantly less receptive to new immigration and/or less accommodating towards settled immigrants than previously? Second and more subjectively, are Europe's political elites under unusual pressure to align public policy with the preferences of an increasingly illiberal electorate? Are the parameters of immigration policy making in Europe in the post-September 11th era more circumscribed by public opinion than previously? In addressing these questions this paper pushes as far back as possible in the respective national public opinion records, with special attention paid to the pattern of British, French, and Spanish public attitudes before and after September 11. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This research identifies sources of interpersonal trust in Latin America by exploring why the most developed countries in the Americas, namely Canada and the United States, show higher levels of interpersonal trust than their neighboring countries. The findings of this paper suggest that Latin America's high economic inequality poses a challenge to democracy because it triggers interpersonal mistrust. This paper also finds that in Latin America interpersonal trust, rather than civic participation, ideology or trust in government, promotes citizens' support for the implementation of public policies aimed to close the gap between rich and poor. Taken together, the results suggest a vicious circle between Latin America's historically high economic inequality, low interpersonal trust, and low support for redistributive policies within certain segments of the population, especially the well-off. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper argues that public opinion research would benefit from shifting its focus on the latent distribution of survey items, not point estimates thereof. The paper shows why this is desirable and how it can be done. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

The relationship between government policy and public preferences is a core concern of democratic theorists. One particularly powerful method of relating policy to opinion is the 'dynamic representation' approach. Scholars in this tradition test to what extent current policy changes are a function of past public preferences. This paper derives hypotheses from the dynamic representation approach and tests them for the United Kingdom in a left-right context from 1976-2006. First, it is shown that government policy on the left-right scale changes as a consequence of changing public preferences (the direct mechanism of 'rational anticipation'). Second, a right-wing public results in the election of the Conservative Party, which consequently pursues right-wing policies in office (the indirect mechanism of 'electoral turnover'). Third, government responsiveness is found to be conditional on electoral vulnerability. Popular incumbents are less likely than unpopular incumbents to adjust their policy position to the public. While the Westminster system has received much criticism for its failure to reliably link rulers to the ruled, this paper finds that dynamic representation on the left-right scale in the United Kingdom functions quite admirably. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Public opinion and foreign policy in Latin America has been sparsely studied in its own right. This paper seeks to test hypotheses drawn from the study of American and comparativist politics. Since the 1920s the impact of public opinion on American foreign policy has been a point of both optimism and dismissal. Liberals have argued that public opinion has the potential to promote peace while pessimists have questioned the stability of public opinion and the ability of the public to make decisions toward foreign policy. Within Latin America, the study of public opinion and foreign policy has been largely overlooked as the emergence of the scientific study of public opinion occurred during a period of authoritarian rule. The rebirth of democracy and liberal economics in Latin America began within the "lost decade" of economic crises of the 1980s. Democratic and structural adjustment reforms since the 1990s have changed the region dramatically making issues of trade a potential touchstone for every citizen. This makes the present the opportune moment to study the public opinion and foreign policy nexus within the Latin American context. Specifically, this paper asks whether public opinion toward economic integration can be predicted by established hypothesis. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper is concerned with the political communication of opinion through social networks. Attention focuses on opinion variance within populations and networks, and how this variance conditions communication among individuals. [ABSTRACT FROM AUTHOR]

This paper examines the impact of perceptions of threat on trade-offs between civil liberties and national security, and the role of these trade-offs in shaping public policy preferences. [ABSTRACT FROM AUTHOR]

This paper examines how the mass public makes decisions on foreign policy during the post-Cold War era. It will show that humanitarianism and strategic considerations play a central role in shaping the public?s foreign policy choices. [ABSTRACT FROM AUTHOR]

This paper examines the effects of public opinion and interest groups on spending priorities in the American states, using data that cover the years 1990 through 2000. Results indicate that interest groups have a stronger impact than public opinion. [ABSTRACT FROM AUTHOR]

This paper assesses the relationship between public opinion and American foreign policy. The interaction of mass preferences and issue salience can provide variation in presidential responsiveness to public opinion within the same policy case. [ABSTRACT FROM AUTHOR]

This paper proposes the best order for questions on abortion, based on a national survey that tests four possible orders. It examines the demographics of ideologically conflicted respondents to describe who is most likely to be confused on abortion. [ABSTRACT FROM AUTHOR]

This paper proposes that terror alerts keep the public's attention on foreign affairs and off of domestic issues, which is favorable to the president?s approval ratings. [ABSTRACT FROM AUTHOR]

The article focuses on a study which examined relations between income and presidential voting preference in rich and poor states in the U.S. at the individual, county and state levels. The aim of the paper is to provide a framework to understand Republican strength among richer votes and in poorer states. Survey data show a correlation between income and support for the Republican party, but at the aggregate level, it is the Democrats who do better in the richer states. The poor people red states tend to be Democrats; the rich people in blue states tend to be Republicans. Income matters within states and geography also matters. With the 2000 presidential election richer states voted for a Democrat.

For what kind of policies are elected officials more likely to be responsive to public opinion? The limited research in this area has found varying degrees of strength in the relationship between public opinion and policy on different policy domains. But scholarship about this important question has been handicapped by a lack of adequate measures and estimates of constituency opinion on policy issues. In this paper, I use the unprecedented statistical power and breadth of the 2000 National Annenberg Election Study (NAES) to explore the representation of constituency interests in Congress in greater detail than has previously been possible. I examine the relationship between congressional roll-call votes and constituency opinion on 20 different public policy issues. I find that in the House of Representatives between 1999 and 2000, the roll-call votes cast by members of Congress were responsive to public opinion was significant on a wide range of policy issues, including abortion, military spending, education, crime, taxes and the environment. Democratic and Republican lawmakers are responsive to public opinion on substantially different subsets of policies, suggesting a typology of issue responsiveness that is highly dependent on the varying levels of credibility that parties establish with voters on different issues. [ABSTRACT FROM AUTHOR]

This paper examines whether there has been a long-term change in public opinion towards national vs. supranational identities and how the structural transformation resulting from globalization affects citizens? territorial attachments. [ABSTRACT FROM AUTHOR]

Drawing on ANES panel data from 2000 and 2002, this paper shows how important individual-level factors and elite persuasion affected American opinion on the conflicts in Afghanistan and Iraq. [ABSTRACT FROM AUTHOR]

In this paper, we compare two cases, welfare and health care, in order to investigate how elite discourse shaped public opinion in ways that facilitated or discouraged significant social policy reforms during the Clinton Administration. [ABSTRACT FROM AUTHOR]

This paper prsents an approach to the study of urban politics that is distinct from regime theory by incorporating political culture and public opinion into the analysis. It is based on ongoing reserarch into the cases of Minneapolis and Gary. [ABSTRACT FROM AUTHOR]

This paper reports on an exploratory study of public opinion in the wake of 9/11. It is based on data from the initial two waves of a national survey fielded immediately following 9/11 and again several months later (How America Responds; ISR, University of Michigan). The interrelated concepts of ethnocentrism and authoritarianism are a central focus of the study. Some ethnocentric bias in the way Americans interpret the events of September 11 is revealed in an open-ended survey question. Furthermore, the survey data show indicators of ethnocentrism and authoritarianism strongly predict support for aggressive military action by the U.S. against countries thought to be involved in terrorism. [ABSTRACT FROM AUTHOR]

Public opinion research continues to have difficulty distinguishing between moderate, ambivalent, and indifferent policy attitudes. Krosnick (1991) notes that the number of respondents who appear to have moderate policy preferences is likely inflated because midpoint responses do not distinguish between those with meaningful moderate preferences and those who are simply indifferent towards the policy. Others may choose the midpoint because they are ambivalent: torn between good arguments on both sides of the policy debate. Distributions of policy preferences generally show an uptick at the midpoint, likely due to amalgamation of moderate policy preferences, policy ambivalence, and simply indifference. Even when respondents refuse to indicate a policy preference, scholars sometimes override this refusal and place them at the midpoint anyway. I argue, however, that this “muddle in the middle” (Converse 1995) masks important differences between these attitude attributes. Those with a true moderate policy preference favor a middle of the road approach. They want a specific policy enacted and their preference may be weak or strong. For example, they may feel strongly that abortion should be legal, but only in some circumstances (a strong preference for a moderate policy). Those with ambivalent policy preferences are internally conflicted and see the validity of arguments both for and against a policy. Faced with the choice between falsely claiming that they “haven’t thought much about it,” picking sides, or straddling the fence, the select the latter. Indifferent attitudes are those that are simply not that important or relevant to the respondent. Indifferent respondents simply don’t know enough or care enough about the issue to state meaningful preferences. Some will readily admit that they haven’t thought about the issue, while others will place themselves at the midpoint to avoid perceived negative stigma attached to indifference. This paper attempts to disentangle moderate, ambivalent, and indifferent policy attitudes. The above discussion suggests that moderate preferences are likely just as informed and likely to affect other attitudes and behaviors as more extreme preferences. While ambivalent attitudes are unlikely to affect other attitudes and behaviors, they should nonetheless be informed attitudes. Indifferent attitudes are likely uninformed and unlikely to influence other attitudes and behaviors. There are several design features built into the 2000 NES that are particularly helpful for this task. First, the policy preferences are measured using both traditional seven-point scales and a series of branching questions. The former allow respondents to give moderate policy preferences while the latter do not (by narrowing the permissible policy preferences to four). The branching format also measures the strength of a binary policy preference, rather than measuring a more precise location on a policy preference continuum. Second, two versions of these questions are included for some policies: one that explicitly allows respondents to say that they “haven’t thought much about it” and another that doesn’t. The first version allows respondents to express indifference, while the second discourages doing so. This research provides important insights into the substantive meaning of policy preferences and carries implications for their proper measurement. [ABSTRACT FROM AUTHOR]

Major changes are affecting how juries, prosecutors, state legislators, and governors think about death penalty issues, with increased concern to ensuring that no innocent individuals are wrongly executed. This change shifts the debate from one about the morality and effectiveness of the punishment to one about the fairness of and possibility of error in the criminal justice system. We have analyzed every New York Times article on capital punishment since 1960 (over 3,500 stories) and can demonstrate changes in media coverage corresponding to these shifting frames. Since 1996, there have been dramatic changes in media coverage towards the death penalty, with increasing numbers of stories and with an unprecedented rise in stories focusing on characteristics of the defendant and the possibility of executing the innocent. In this paper we focus on public opinion, asking whether it reflects these shifting elite and media issue-framing effects. Our analysis is two-fold. First, we construct a quarterly time-series model of attitudes toward the death penalty from 1973 to present, based on responses in over 270 surveys. We use Stimson’s (1999) dimensional analysis algorithm to extract the shared movement in attitudes across these survey questions and over time, producing a time series that is both richer in information and also covers a longer time period than any single indicator. At the aggregate level, we use these data to assess whether changes in media coverage have had a discernable impact on public opinion, controlling for the crime rate and other relevant variables. Second, at the individual level, we assess the correlates of support for the death penalty in a pooled time-series format. If the traditional issue-frame surrounding the issue of capital punishment related to moral / religious reasoning and to an issue of efficacy in fighting crime (e.g., deterrence), then an individual-level model of support for the death penalty should reflect these attitudinal predispositions. If this traditional frame has been replaced by a new one focusing on the possibility of errors in the system, then the same predictive model should no longer hold. We model this process with an analysis of over 43,000 survey interviews in 24 surveys from 1972 to 2002 using the General Social Survey. Our study will have broad implications concerning the links between elite-level discussion, media coverage, and public opinion. The death penalty appears to be undergoing significant change in its public understanding and the frames that are used to describe and understand it. Public policy is changing on capital punishment issues, and public opinion is changing as well. However, in contrast to typical models of representation, it appears that both public policy and public opinion are being affected by a similar set of elite-level changes occurring within professional communities of defense and prosecution attorneys. Thus, our detailed study of capital punishment in America allows an unusual opportunity to assess the reactions of public opinion to changing elite discourse. The fact that the issue is so charged morally also makes the implications of our study relevant to the study of other and divisive social issues more generally. [ABSTRACT FROM AUTHOR]

George W. Bush enjoyed an historic rally effect, in the aftermath of 9/11 that is widely regarded as the basis for public and elite support for the U.S. invasion of Iraq. This paper applies the affective intelligence model of political judgment to explore the impact of attitudes toward Bush upon support for the pending invasion, as of late February 2003. Three hypotheses are examined: first, that enthusiasm (E) for Bush was associated with receptivity to pro-war rhetoric and support for hawkish policies; second, that anxiety (A) about Bush was related with receptivity to anti-war rhetoric and support for dovish policy options; and third, that emotional responses to Bush on these dimensions interacted to yield four distinct orientations – Hawk (high E, low A), Dove (low E, high A), Ambivalent (high E, high A), and Apathetic (low E, low A) In the pretest phase of a quasi-experimental design, 157 undergraduate respondents were recruited from three locations. They completed questionnaires in late Feb. 2003, including items for evaluating: approval of presidential job performance, affective responses to Bush, four Iraq policy options (ranging from no use of force through unilateral invasion), and 16 items evaluating pro-war and anti-war rhetoric. Findings regarding attitudes towards George W. Bush include: the expected two dimensional structure of emotional responses to Bush, that emotional enthusiasm for Bush played a critical role in overall approval ratings of the president and was far more important than domain specific performance ratings, and that anxiousness played little role in presidential approval. With respect to the pending war, findings are that: (1) enthusiasm for Bush was strongly associated with hawkish attitudes, (2) that anxiousness about Bush, even among his supporters and those who approved of the invasion options, was associated with receptivity to anti-war rhetoric (e.g. concerns about the costs of war), and (3) that enthusiasm and anxiousness interacted in their effects on attitudes in a manner consistent with the hypotheses. These results hold with controls for the effects of ideology, partisan identification and gender. A principal finding is that feelings of anxiousness about Bush, regardless of partisan or ideological agreement, were significantly and substantially associated with consideration of arguments and policy options in opposition to unilateral invasion of Iraq. Unbridled enthusiasm, on the other hand, was associated with rejection of concerns about American casualties, Iraqi civilian deaths, domestic economic costs, or destabilization of the Middle East. These findings support the affective intelligence proposition that anxiety arousal stimulates attention to alternative choices and reflective political thought. [ABSTRACT FROM AUTHOR]

What variables determine the extent to which individuals feel anxious about presidential candidates, and what effect does this anxiety have on political participation? An emerging stream of research suggests that emotion plays a significant role in political judgment (Damasio, 1994; Damasio, 1999; Marcus et al., 2000; Marcus & MacKuen, 1993). Marcus et al.’s (2000) theory of affective intelligence suggests that anxiety in particular has a powerful effect on the level of attention that we pay to political information By this theory, new and/or threatening stimuli activate the brain’s surveillance system, causing feelings of anxiety and motivating the individual to seek more information. Thus far, evidence suggests that new and ominous political information – such as the NAFTA debate and the 1992 Gulf War – does indeed have this effect on those who receive it (Marcus et al., 2000). However, many questions remain. This paper seeks first to discover whether any particular issue domain causes more anxiety than others. Are certain types of messages more unique or threatening, either universally or during specific presidential campaigns? I examine the comparative impact of issues that reflect three fundamental antecedents of public opinion: self interest, inter-group relations, and values. Using American National Election Studies data from 1988, 1992 and 2000, I assess the degree to which concern with economic, racial and moral issues predicts anxiety about presidential candidates. In 1988, racial animosity produced by the “Willie Horton” advertisements may have been more powerfully implicated in creating public anxiety than any other type of issue. In 1992, by contrast, more citizens may have agreed with James Carville: “It’s the economy, stupid!” Finally, in the wake of the accusations that colored President Clinton’s second term, moral issues may have come to the fore in the 2000 election. I find conditional support for each of these hypotheses. However, economic issues appear to be more universally correlated with anxiety, regardless of the campaign specifics. I then examine the impact of anxiety on interest in the campaign, finding that indeed, those individuals most anxious about a given presidential candidate are most likely to report greater interest. I conclude that particular issue domains are more predictive of anxiety depending on the specifics of the presidential election campaign, although the state of the economy is universally important Further, I conclude that political participation and interest in the campaign is increased by anxiety about presidential candidates, motivated by concerns with these different issues. [ABSTRACT FROM AUTHOR]

As part of an effort to understand the changing political landscape of China, this study uses data from the most recent wave of World Values Surveys, conducted on a national representative sample of China in 2001, to examine citizen’s trust on the government. I find that, while thinking of government, citizens in China have two very distinctive sets of objects. Their trust on the national institutions (the Communist Party, the parliament, and the national government) is significantly higher than that on the institutions that they deal with in their daily life (mainly the police and the government bureaucracy). I contend that these two sets of institution represent the state in two forms: one abstract and the other concrete. I then examine the factors that affect individual’s trust on both forms of state. Most importantly, people with a democratic outlook----that is, people who value individual liberty and participation in governance, among others----tend to hold lower trust on the government. And this relationship holds firmly even after controlling other factors, such as level of education, interest in politics, and financial well-being as well as overall life satisfaction. On the contrary, people who follow news more frequently appear to hold higher level of trust on the national institutions (the abstract state) than others, but not on the concrete state. This finding has many important implications for the prospect of political change in China, which I discuss in the paper. (238 words) [ABSTRACT FROM AUTHOR]

In this paper, we seek to describe, explain, and compare attitudes toward abortion in the United States and Poland over time. While our preliminary results suggest that abortion attitudes have remained quite stable in both countries, this result is somewhat perplexiing. Numerous demographic and cultural changes have taken place in both countries which would normally be expected to push the population in a pro-choice direction. We plan to compare the actual distributions of abortion attitudes for both countries with a set of expected values. The latter will be computed by assuming constant relationships between the dependent variable and important predictors of abortion attitudes,while allowing the marginal distributions of such variables to change over time. Based on these comparisons, we hope to account fot the reasonf or the anomalous stability of abortion attitudes in the two nations. [ABSTRACT FROM AUTHOR]

On June 23, 2003, the U.S. Supreme Court issued landmark rulings on the University of Michigan’s use of affirmative action in its admissions policies. The Court’s decisions presented a unique opportunity to systematically examine the relationship between both the opinion of black Americans toward a policy articulated by the Court and the Court as an institution. In this paper, we report the results of a panel survey of black Americans’ reactions to the Court and its rulings. Specifically, we analyze public opinion data from a national representative sample of black Americans. This unique data set (funded by the National Science Foundation) includes measures of attitudes toward affirmative action policy, diffuse support for the Court, racial identification and group consciousness, as well as exposure to a variety of media sources. The inclusion of these measures and the survey’s panel design yields substantial intellectual payoffs. These data provide us with significant analytical leverage on the Court’s capacity to move public opinion with respect to a policy it has articulated. [ABSTRACT FROM AUTHOR]

Prior research on citizen support for European integration has primarily focused on individuals? evaluations of the process of integration or its institutions, with emphasis on the importance of direct benefits and costs integration can confer. Explanations focus on overall support for integration and little work has been done on explaining public opinion on specific policy areas, such as the development of the Common Foreign and Security Policy (CFSP). Prior work also does not consider individuals? evaluations of member states in models. This paper will fill this gap in the research by formulating and testing a political cohesion model, which can be considered complementary to preexisting models. The analysis synthesizes systems theory with social identity theory to produce a core claim that the probability of supporting the CFSP increases with greater levels of political trust in the European Union member-states. The development of political cohesion, as measured by the amount of trust in member-states, is assumed to be reflective of a positive perception. Positive perceptions of member-states are critical because integration?s development is influenced strongly by and dependent upon the resources of the relatively more powerful European member-states. Therefore, positive perceptions of the top EU powers, namely Germany and France, improves the probability of holding a European identity and support integration, more so than trusting the remaining members. The results hold even when controlling for demographic variables, political values, ideology, and the democratic deficit. Binary logistic regression analysis using pooled repeated cross-sectional data from the Eurobarometer surveys conducted in 1992 through 1997 among individuals of 11 member-states largely support these claims. [ABSTRACT FROM AUTHOR]

Although many studies have examined the economic status of immigrants, most studies of immigration have neglected immigrant opinions. When immigration is addressed at all using survey data, studies almost exclusively focus on native attitudes toward immigration and not on attitudes of immigrants themselves. In this paper, we investigate whether immigrants and natives have systematically different attitudes toward social spending and redistribution. We employ surveys from Germany and Great Britain that sample a relatively large number of immigrants compared to typical national election surveys. In addressing this potential “opinion gap,” we control for socioeconomic characteristics, such as income and education as well as immigrants’ ability to assimilate into their new country’s labor market by comparing an immigrant’s skill level with his or her actual occupation in the new country. We also include contextual covariates to account for potential network effects. Our statistical analyses use ordered logistical regressions and are fully interactive (following Franzese 1999), consistent with our theoretical expectation that the models’ independent variables affect natives differently than they do immigrants. Once these controls are included, we observe that immigrants are never more likely than natives to favor increases in social spending or to endorse redistributive measures. In instances where we do find a significant opinion gap between natives and immigrants, the latter tend to support more conservative positions, favoring tax cuts at the expense of social spending. While some of our findings are preliminary in nature, they nevertheless represent a significant contribution to the extant literature, which, to our knowledge, does not include comparative studies on the political attitudes of immigrants. [ABSTRACT FROM AUTHOR]

There are 14 million Medicare beneficiaries, mostly elderly, who lack prescription drug coverage, and their number is increasing. Living on fixed incomes, many of them are often forced to choose between food and life-saving medicines. Polls show that, across the board, Americans recognize the burden that expensive prescriptions represent for the most vulnerable groups. Survey results describe adding a prescription drug benefit to Medicare as the most important health policy issue the Administration and the Congress should address. However, despite numerous attempts over the last three Congresses, no legislation has been enacted to date. This paper provides a detailed analysis of the debate on Medicare prescription drugs over three Congresses (106th to 108th). Methodologically, a modified Kingdonian framework will be used to study the specifics of the policy formulation process and to predict the most likely policy outcome. In addition to the three ‘streams’ defined by Kingdon –problem, policy, and political streams—particular attention will be given to the influence of public opinion and media coverage throughout the process. The role played by the White House will also be considered. The analysis will rely on multiple data sources, from legislative history and CQ analyses to public opinion polls, media coverage indices, as well as public statements put forward by various interest groups involved in the debate. Prescription drug coverage for Medicare enrollees is not just another topic among the numerous health policy issues confronting the American society today. Due to the millions of people that could be affected by the outcome and the powerful interest groups involved in the debate, finding a solution to this crisis will play an important role in the 2004 presidential campaign, and cannot be ignored without serious political consequences. [ABSTRACT FROM AUTHOR]

“Presidential power in an era of divided government”. While this may sound paradoxical, the modern governmental context highlights the reality of this statement. Given the modern political environment where public opinion is the currency of individualized pluralism, Samuel Kernell (1997) claims that ‘going public’ represents the most effective strategy for presidential influence. Paul Light (1999) asserts that agenda setting can be a formidable source of presidential influence: “…control of the agenda becomes a primary tool for securing and extending power” (2). Although a rich agenda setting literature exists which is frequently utilized to examine congressional policy making and its subsequent implications for representation writ large (e.g. Cobb and Elder 1983; Kingdon 1984; Baumgartner and Jones 1993), studies of presidential agenda setting are more sparse (Light 1999; Cohen 1995). The presidencies of Bill Clinton and Richard Nixon present a unique opportunity to explore problem definition and agenda setting in the context of the domestic policy arena. Intriguing aspects of the policy process are overlooked when scholars myopically hone in on the legislative branch to the (relative) exclusion of the executive branch. This paper endeavors to illuminate the important role that two presidents played in setting the welfare reform agenda. As such, an exploration of Nixon’s Family Assistance Plan as a case of policy failure and Clinton’s Welfare Reform as a case of policy success will undoubtedly shed new theoretical insight and raise questions for future research in the areas of problem definition and presidential agenda setting. Indeed, these two policy cases amply demonstrate the inseparability of the concepts of problem definition and agenda setting. Finally, a ‘window of opportunity’ exists to test different theories of agenda setting for efficacy with regard to the president. A perusal of presidential rhetoric, in the form of speeches, press releases, and other public statements, will be utilized to determine the extent to which welfare reform permeated Nixon and Clinton’s public strategies. Furthermore, Deborah Stone’s theory of causal stories will serve as an analytical framework whereby an examination of competing efforts to define the problem both within and outside the presidential arena will be conducted. Since the initial framing and definition of a policy problem has a significant impact on the shape and tenor of a policy, presidential agenda setting has real implications for public opinion concerning and congressional estimations about public policies, particularly salient ones like welfare reform. [ABSTRACT FROM AUTHOR]

Since the 1970’s, the Benefit Cost Analysis (BCA) framework has been a critical component of the authoritative allocation of value, and has been the foundation for many of our public policy decisions. However, there has been some controversy over the appropriateness of this framework. Proponents of the benefit/cost framework have argued that it is a quantifiable method that makes it possible to include a full range of benefits and costs, it ensures that the values of the less politically powerful (or politically vocal) are included, it helps to distance policy decisions from the political process, and it puts the benefits and the costs into a common framework: dollars. When pointing out the shortcomings of the BCA framework, detractors usually describe the difficulty of comparing soft benefits and hard costs, missing data on the private valuation of policy proposals, and the enormous and expensive efforts required to try and fill some of these gaps. Others have questioned the appropriateness of using a method based on utilitarian assumptions for assisting in making policy choices, particularly when deontological questions are part of the issue. In trying to fill these data gaps, the field of economics has focused on two primarily utilitarian techniques, the first is hedonic analysis whereby actual market transactions are used to estimate the value of a particular "good"or "bad"; and contingent valuation (CV), an interview based method, which uses individuals’ stated preferences (as opposed to preferences that are revealed in market transactions) to estimate willingness to pay for an hypothetical public good. Contingent valuation is currently the only avenue available for the inclusion of ‘soft benefits’ in a BCA framework. This method has been prominently used in many public policy arenas particularly in environmental policymaking and damage assessments (like the Exxon Valdez case). Given that CV has evolved into a prominent, widely used policy tool for making collective decisions, I am interested in the following questions: Does the economic valuation of public goods carry "political" baggage? Does the nature of the economic valuation change the value of the good? Can we model the political foundations of economic valuation and come up with a "correction"? This paper applies the CV method to the issue of global climate change, where the "good" is the implementation of the Kyoto protocol as a means to reduce greenhouse gas emissions. I test the hypothesis that if the CV method is viewed as inappropriate, it will bias the estimated value of the good downward, resulting in a "tainted good". We also test for the effects of equity concerns, the appropriateness of reliance on public opinion for decision making, and deontological views on the estimated value of the public good. Preliminary results indicate that respondent perceptions of the appropriateness of CV has an effect on stated willingness to pay. Those individuals that see the method as inappropriate are, on average, willing to pay less. This has far reaching implications both for the BCA framework in general and the contingent valuation method in particular. [ABSTRACT FROM AUTHOR]

Previous research has documented that public attitudes toward the news media have become dramatically more negative over that past 35 years. At the same time that public confidence in the press has declined, political scientist have become increasingly adept at demonstrating the massive influence the media exert over public opinion. This paper examines the question of what effect public attitudes towards the press have on the news media?s ability to influence the public beliefs. To do this, I utilize a Bayesian model of persuasion to develop predictions of the consequences if source judgments do play an important role in media influence. I then test these predictions using three very different datasets: a cross-sectional survey, a panel survey, and an experiment. The results from all three datasets confirm that those with positive feelings towards the media are more influenced by media messages than those with negative feelings. I briefly discuss the implications of these findings for democratic accountability. [ABSTRACT FROM AUTHOR]