Trust_ Distrust and Political Skepticism by maqboolshahin

VIEWS: 123 PAGES: 26

									Trust, Distrust and Political Skepticism

Tim Emmert PhD Candidate, University of Virginia

Abstract: Political skepticism is a concept which receives little attention in political science but may better serve our understanding the role of public opinion and democracy than the well-researched concepts of trust and distrust. Both empirical and theoretical work on trust in government has provided strong arguments for the role of both trust and distrust. Building on existing scholarship an argument is made here for political skepticism as a concept, different from trust or distrust, which best describes the traits or attitudes of a model citizen in a healthy democratic system. The goal is to establish an unambiguous measure of the relationship between citizen and system of governance. Relying on empirical analysis of data from the 2000 National Election Survey this paper lays out the theoretical underpinnings of political skepticism, reports initial findings from the data and makes suggestions for further directions in this line of research.

I.

INTRODUCTION

Over the past three decades the National Election Survey (NES) trust index has been used extensively to evaluate attitudes toward executive leadership, government policies and political parties. While the trust in government index was intended to capture trust for government as a whole, some of the earliest work on trust by Miller (1974a, 1974b) and Citrin (1974) questioned the use of trust in government measures as a proxy for support of the government in this way. Citrin (1974) claimed that respondents were reacting to incumbents and officeholder performance rather than with a general affect toward the federal government. In 1986 Citrin and Donald Green supported this claim with research showing that the trust in government index varied in relation to positive images of the president and approval of his policy agenda. Work by Marc Hetherington (1998, 1999) builds on Citrin’s approach by asserting the trust in government index as a proxy for trust in the president. The scholarly treatment of the trust index has not been restricted to tests of support for incumbents but this type of approach comprises the lion’s share of research on the topic in the field of American politics.

Trust in government is often conceptualized as a trait exhibited by ideal democratic citizens although the argument establishing a positive normative basis trust can vary from author to author. A fundamental critique provided by Gamson is that trust enables governments to act without resorting to coercion or having to submit every governmental act to a vote of public confidence. (Gamson, 42) Others find a trusting public to be important to democratic institutions generally but to representative government

2

specifically. Citrin writes that “a high level of political trust implies that as much as disagreements and conflicts may arise over public policies, a basic respect for the system remains” (Citrin, 1999: 466) Similarly, Mishler and Rose argue that “it is trust which gives representatives the leeway to postpone short term constituency concerns while pursuing longer term national interests.” (Mishler, Rose: 419) Material outcomes such as “the willingness of the public to provide such crucial resources as tax dollars [and] the willingness of bright young people to go into government” are also compelling reasons for viewing confidence in government (diffuse trust) as an important aspect of any governmental structure. (Nye, 1997)

In contrast with the positive assessments of trust made here distrust might be viewed as something that undermines the stability and legitimacy of democratic systems. A number of authors have defended distrust, though, as a trait necessary to democratic citizens and one that should be particularly salient in the United States. “The United States was founded with a mistrust of government,” Nye writes, “The American Constitution was deliberately set up in such a way that a King George could never rule over us again. And some might add, ‘Nor anybody else’.” (Nye, 2) Under this interpretation, the survey data defined as distrust should be re-defined as a wariness of any governing institution. This interpretation has appeared elsewhere in different forms (Barber, 1983; Gibson, 1992; Warren, 1999b) but in each case distrust is characterized as a healthy characteristic of the democratic citizens’ ability to ensure accountability and defend oneself against tyranny. Some go so far as to say this sense of distrust is something that should be institutionalized in healthy democratic states. (Braithwaite, 1998)

3

Given the seemingly vital nature of distrust to the healthy operation of democracy it would seem odd to find that literature on trust/distrust often does not present distrust in a more positive light. Unfortunately, that appears to largely be the case in empirical studies in the field of American Political Science. Miller’s work on trust (1974a, b) found that distrust was most commonly a trait of extremists unhappy with the centrist policies and platforms provided by the centrist politics of the Republican and Democratic parties. This notion was later inverted by some (Craig 1996, King 1997) to state that party polarization had led to decreasing levels of trust among centrist Americans. When cast as a trait shared by an exclusive class of extremists upset with the political options available to them distrust sounds like an ingredient in a recipe for revolution. The latter notion of mistrusting centrist voters sounds almost positive in comparison and suggests that voters are using critical capacities to evaluate their institutions and actors. Craig interprets the decreasing levels of trust as a sign of ill health for democracy, though, and laments, “It is difficult to know exactly what might be done to prevent the confidence gap from becoming a permanent feature of the American political landscape.” (Craig, 1996: 63)

More recent work on American political processes and trust is more positive about the usefulness of distrust to democratic institutions; in fact, authors John Hibbing and Elizabeth Thiess-Morse find distrust of government to be a decisive factor in American political participation. In Stealth Democracy they argue that most Americans are not eager to participate in politics, they perceive political conflict as a breakdown of otherwise civil democratic procedures. Americans are motivated into political

4

participation only by distrust of government generally and public officials specifically. “[W]hen people are moved to involve themselves in politics, it is usually because they believe decision makers have found a way to take advantage of their positions. Consequently, political participation in the United States is often connected to resentment, dissatisfaction, and puzzlement rather than to legitimacy, trust and enlightenment.” (Hibbing, 10)

This radical distrust of government results in affects towards the government as a whole and representatives specifically. First, Americans only trust government to enact policies which apply to all citizens and will withhold their support from any action beyond that limited realm. Second, as a result of their distaste for political bickering and general apathy towards politics Americans desire a stealth democratic arrangement “in which decisions are made by neutral decision makers who do not require sustained input from the people in order to function.” (Hibbing, 7) The thesis put forward by Hibbing/ThiessMorse finds a crucial role for distrust in democracy and breaks new ground by uncovering an empirical relationship between distrust and the political process.

In the last decade some political scientists have started putting forth a new interpretation of the trust/distrust scale. These scholars, by focusing their efforts on studying the citizen’s ability to critically assess government performance, have added the term “skepticism” to the trust literature’s lexicon. William Mishler and Richard Rose surveyed post-Communist European societies regarding trust in the then-newly established governing institutions using a model with two determinants of trust –

5

perceptions of freedom and fairness and evaluations of economic performance. (Mishler/Rose, 446) The authors graded survey responses on a scale ranging from trust to distrust with a measure in-between the two that they deemed “skepticism”. One of their conclusions was that the middle measure, skepticism, “reflects trade-offs between public dissatisfaction with current economic performance, optimism about future economic performance, and satisfaction with the political performance of contemporary institutions in providing greater individual liberties than in the Communist past.” (Mishler/Rose, 418)

Mishler/Rose argues that the ideal democratic citizen is a skeptical one. “Democracy requires trust but also presupposes an active and vigilant citizenry with a healthy skepticism of government and a willingness, should the need arise, to suspend trust and assert control over the government.” (Mishler/Rose, 419) Healthy democratic citizens in the Mishler/Rose model require skepticism as a mediating factor between trust and distrust – this critical capacity ensures citizens will be alert to changes in governing attitude towards themselves or government performance in certain key areas of civil society.

The move towards focusing on the role of skepticism, rather than trust and distrust, in democratic states is an important one. One of the defects in the discussion of trust measures is that the results are easily subjected to ambivalent interpretation. In other words, it seems that democratic institutions benefit when citizen trust in government increases, at the same time distrust appears to be a necessary trait of the democratic

6

citizen. How is it possible to evaluate measures of trust in government when the outcome, either trust or distrust, can be interpreted as a positive one? For example, when we hear that citizen trust in government has declined significantly over the past 40 years should we be concerned that citizens are no longer supportive of their government? Or should we be optimistic that an increasingly well-educated population is exercising its critical faculties in assessing government performance? We lose the ability to accurately assess democratic institutions and citizen evaluations of those institutions when we are confronted with such ambivalence.

Rather than continuing to follow the existing paths of research and scholarship on trust I argue we should be looking for skeptical citizens, studying how they differ from other sorts of citizens, and better explicating their role in a democracy. While future researchers may be able to devise accurate measures of skeptical attitudes our present measures of attitudes towards government should yield some indicators of skeptical beliefs and attitudes. But what, exactly, are we looking for? How is skepticism defined and how can skeptical citizens be differentiated from the rest of the public?

One of the best available definitions for skepticism, provided by James Vice, describes it as “an attitude concerning the difficulty or impossibility of knowing certainly and a strategy for dealing with uncertainty.” (Vice, 11) By defining skepticism in this way, Vice elucidates a single characteristic that distinguishes it from the trust/distrust dichotomy. As a strategy, skepticism is also conceptually unique; while trust implies no strategy whatsoever and distrust implies radical reactionary engagement (or, possibly, a

7

return to self-sufficiency), skepticism suggests an individual who is engaged but wary. Similarly, skepticism is also a trait which might be most apparent in citizens of democracies; by empowering citizens with a voice in government democracies encourage citizens to pair their political attitudes with a strategic understanding of their role as citizens.

How do skeptics differ from the rest of the public? Given their interest in competitive elections and the election system generally we might expect skeptics to espouse a greater support for democracy. Similarly, skeptics should be more likely than others to believe that their vote matters or that they have a say in what government does. This sense of efficacy is particularly important for distinguishing political skeptics from political cynics (distrusters) who might be best characterized as withholding belief or support in government but lacking hope for a new outcome that differs from previous outcomes.

It is not as easy to distinguish political skeptics and those who trust in government. As noted above, trust in government seems to imply no inherent strategy with regard to ruling authorities. This seems to be borne out in the existing trust literature, Citrin and Muste summarize the trust literature in noting “attitude scales that putatively measure support for the political regime usually are positively correlated with conventional indicators of support for authorities, such as party affiliation, approval of the incumbent national leader, and agreement with his or her policies.” (Citrin and Muste referencing Citrin, McCloskey, Shanks and Sniderman, 1975) Such support, ranging across leaders

8

and their associated policies and affiliates, seems to obviate the possible use of critical faculties by citizens who trust.

In fact, many citizens who completely trust in government may be exhibiting authoritarian attitudes – this point has been made fairly recently by Mitchell Seligson and Julio Carrion. In their study of system support and support for military coups in Peru the authors ask “… what are we to make of citizens in such a country who express high system support? These are individuals who are highly supportive of a regime that has minimally democratic credentials, suggesting that their support may be high precisely because of the many non-democratic features of the regime rather than its democratic elements. Such individuals, then, could both support the regime and support military coups to suppress threats to order.” (Seligson and Carrion, 60) Here I am not suggesting that trusting Americans exhibit a greater willingness to support undemocratic means to cope with political conflict and uncertainty. Instead, I would argue that those who trust are much like those citizens described in Hibbing and Theiss-Morse’s Stealth Democracy. These citizens exhibit a deep distaste for political bickering and, when combined with a general apathy towards politics, these Americans desire a stealth democratic arrangement “in which decisions are made by neutral decision makers who do not require sustained input from the people in order to function.” (Hibbing, 7)

II.

MEASURES OF SKEPTICISM

It makes sound theoretical sense, then, to locate skepticism somewhere between trusting and distrusting attitudes toward government. The NES Trust Index is one measure

9

available to use in looking for skeptical citizens. Traditionally the index has been used as a dependent variable but recent work by Hetherington and Seligson have used the trust index and support for government as independent variables predicting other attitudes. Likewise, searching for skeptical citizens would necessarily involve using the trust index as an independent variable.

In predicting attitudes toward democracy and efficaciousness the trust index can be split into three sections – low, middle and high – representing distrusters, skeptics and trusters. If skeptics are really distinguishable from distrusters (cynics) and trusters then a parabolic relationship should emerge demonstrating trusters and distrusters, at either end of the index, to hold similar beliefs and skeptics, at the middle of the index, having significantly different attitudes. Dummy variables can also be created representing each section of the trust index, the partial regression coefficients should support the theorized parabolic relationship by demonstrating partial slope coefficients which are signed differently for trust and distrust.

The data explored here are from the 2000 NES Pre/Post-election survey. My trust index has been constructed according to the process described by Hetherington in Why Trust Matters. (Hetherington, 2000: 16-17) The two dependent variables I explore here are satisfaction with democracy (V001651) and whether or not the respondent has a say in what the government does (V001528).

10

III.

FINDINGS

A. Satisfaction with Democracy

Satisfaction with Democracy
4 1 0 2 3

.2

.4 Trust Index

.6

.8 Fitted values

1

Satisfaction with Democracy

In exploring these two dependent variables one of the easiest ways to demonstrate a parabolic relationship is by generating graphs of the relationship between the dependent variable and the trust index. The variable relating to the respondent’s satisfaction with democracy, when regressed onto the trust index, yielded a slight parabolic relationship. By squaring the trust index we have a more exaggerated demonstration of the parabolic effect (Figure 1, above). The satisfaction with democracy variable ranges from one (satisfied) to four (not at all satisfied), both trusters and distrusters register a response between fairly satisfied and not very satisfied with the way democracy works while

11

skeptics - those in the middle of the Trust Index – registered responses somewhere between fairly satisfied and satisfied with the way democracy works in America.

B. Respondent Say in Government

Does Respondent Have a Say in What Government Does?
5 1 0 2 3 4

.2

.4 Trust Index Have no say

.6 Fitted values

.8

1

The second dependent variable asks respondents to react to a sentence which reads, “People like me don't have any say about what the government does.” Answers range from one (agree strongly) to three (neither agree nor disagree) to five (disagree). As was done in the empirical relationship above the trust index was squared in this equation to exaggerate the quadratic fit. Here (Figure 2, above) we see that trusters and distrusters fall closer to being either ambivalent towards the statement regarding their own democratic efficaciousness or agreeing with it ‘somewhat’. On the other hand, skeptics –

12

defined as those found in the middle of the trust index – fall much closer to ‘disagreeing somewhat’ with the statement. In short, skeptics appear to have more faith in democratic systems than people who exhibit greater trust, or distrust, of government.

How much more do these parabolic effects explain than linear regression? One way to test whether the quadratic fit is superior to a linear fit is to compare the R2 value of a regression with another regression which includes both the base variables and the squared term.* If the quadratic fit is more appropriate then the R2 value will increase; for both my variables this appears to be the case. Starting with satisfaction with democracy, Figure 1a indicates that by adding the squared term for the trust index the R2 value increases by almost .01.

C. Test of Quadratic Fit: Satisfaction with Democracy
Test of Quadratic Fit (1) Satisfaction w/Democracy -0.98** (10.49) (2) Satisfaction w/Democracy -2.37** (6.58) 1.68** (4.00) 2.47** (38.30) 1414 0.0813

Trust Index Trust Index2

2.27** (57.42) Observations 1414 Adjusted R-squared 0.0716 Absolute value of t statistics in parentheses * significant at 5%; ** significant at 1%

Constant

The results are similar for the variable representing respondent attitudes regarding whether they have a say in what government does. Adding the squared variable to the
The technique described here is probably most appropriate for Stata users. I’ve been told that SPSS has some features not present in Stata which make it easier to diagnose and analyze quadratic fits such as those described here.
*

13

equation adds roughly .006 to the R2 value. In both instances all the independent variables – indexes and squared terms alike – are significant and the index and squared term are signed differently (as confirmation of the curvilinear nature of their relationship to the dependent variable.)

D. Test of Quadratic Fit: No Say in Government?
Test of Quadratic Fit (1) No Say in Government? 1.88** (12.75) (2) No Say in Government? 3.70** (6.48) -2.19** (3.30) 2.18** (21.25) 1546 0.1005

Trust Index Trust Index2 Constant

2.44** (39.22) Observations 1546 Adjusted R-squared 0.0947 Absolute value of t statistics in parentheses * significant at 5%; ** significant at 1%

These findings for political skepticism are not exactly as might be expected. For instance, there is a question of the exact shape of the quadratic fit between the trust index and the two dependent variables studied here. Calculating the peak of the curve for the two variables reveals that the curve peaks at different points along the trust index for each dependent variable:

E. Peak of Parabolic Curve Dependent Variable Satisfaction with Democracy No say in Government? Peak of Curve .70 .84

14

That the peak of the curvilinear relationship between satisfaction with democracy and the trust index occurs at .7 is not terribly problematic. While it is not a perfect curve in the sense that it does not peak at the middle of the trust index (.5) and fall evenly toward trust and distrust it does represent a significant difference between those who fall near of the middle of the trust scale versus those who fall at the ends. It is somewhat more problematic for the peak of the variable dealing with a say in government to occur at .84 on the trust index. Here there is a relationship that peaks late and seems to encompass a number of those whom we might consider “trusters” as opposed to an ‘ideal’ where the middle of the scale would be neatly differentiated from the ends. Where this is most problematic is the confusion it creates in attempting to separate the trust index into three dummy variables representing trust, skepticism and cynicism (distrust). There is a more extended discussion below on attempts to utilize these dummy variables in gaining a better understanding of what drives skepticism.

Ultimately only more research will determine how problematic the location question of these different peaks really is. It could be that the parabola rises quickly from the lower values and plateaus slightly across the middle values before dropping again as it reaches the higher values. If this is the case it would seem that determining a peak value to the curve is not as important as the cutoff for the rising and falling of the curve. Here it is important to note that the peak may only be important insomuch as comparing levels of skepticism across years. In other words, the peak of the satisfaction with democracy variable falls at .7 in the year 2000 but in 1996 it might peak at .4 or at .9 in 1976. This

15

paper works well to establish a snapshot of how a relationship might work but does not put skepticism in a larger context with other election years.

As noted above, dummy variables can be used to split the trust index into three parts – low, medium and high levels of trust. Here three variables were constructed based on the number of observations available across the trust scale. Since the scale is “bottom heavy”, having a much larger proportion of observations closer to zero than to one, the number of observations is likewise skewed. In this instance I have constructed dummy variables with cutoff points at .2 and .5 for distrusters (<.2), skeptics (>.2 and <.5) and trusters (>.5). These cutoff points do not reflect a normally distributed scale but, unfortunately, the observations across the trust index are not normally distributed.

Before proceeding further there is an additional, and important, problem encountered when applying the dummy variables. Given the parabolic nature of the relationship it should be expected that when we eliminate the dummy variable for skeptics (those in the middle of the trust index) that the remaining dummy variables for those who trust and distrust should be signed similarly (either both variables are signed negatively or both positively, depending on the direction of the curve.) Unfortunately this does not occur. Despite much manipulation of the range and cutoff points for each dummy variable it seems there is little that can be done to the dummy variables to get them to replicate in numbers what they appear to represent graphically. Manipulating the variables often results in a loss of significance for one or more of the dummy variables. This is a particularly intractable problem and one that appears to call these findings into question.

16

If there wasn’t strong counter evidence – both graphically and through the added knowledge of the dependent variables provided by the squared trust index terms – I would have little faith that the parabolic relationship described here does, in fact, exist. From here I will proceed with description of the findings from regression analysis but bear in mind the problems encountered with the dummy variables.

In addition to the dummy variables the model involving the satisfaction with democracy variable included independent variables theoretically relevant to skeptical beliefs including a number of variables related to knowledge like interviewer assessments of the respondent’s level of intelligence, knowledge of politics and years of formal education. A number of other variables related to information acquisition and integration were used including attentiveness to the 2000 presidential election and news consumption. The model with these variables predicting satisfaction with democracy (below) yields an R2 value of .11 and three independent variables with predictions above 10% level of significance.

Of these, two of the independent variables most closely related to information acquisition – attentiveness to elections and news consumption – both appear to lead to greater satisfaction with the way democracy works in the United States. Oddly enough this model finds that the more intelligent a person seems to be (as assessed by the interviewer) the more likely they are to be dissatisfied with the working of democracy.

17

Actual political knowledge, constructed from general political knowledge questions asked in the NES survey, was not significant and neither was years of education.†

F. Regression Coefficients: Satisfaction with Democracy
Satisfaction with Democracy Satisfaction with Democracy Trusters -0.24** (5.00) Distrusters 0.33** (6.54) R Intelligence (interviewer 0.07* assessed) (2.27) Attention to elections -0.38** (2.96) News consumption -0.15 (1.90) Years of education -0.01 (1.52) Political Knowledge (constructed -0.09 scale) (1.20) Constant 2.27** (12.03) Observations 1362 Adjusted R-squared 0.11 Absolute value of t statistics in parentheses * significant at 5%; ** significant at 1%

The dependent variable representing whether or not the respondent has a say in government includes some of the same variables with some new variables represented as well. As a whole this model (below) has a strong R2 value of .21, roughly half of which is accounted-for by the trust index variable. The model demonstrates several significant independent variables; years of education and attention to elections increase the likelihood of a person believing they have a say in what government does. On the other
The level of education variable could potentially be split into more discrete education levels and degrees but there appear to be some date entry errors in my dataset. This could be a product of translating the dataset from an .export (SPSS) format to a .data format (Stata). I hope to find some explanation for the errors and assess the role of education in future work on skepticism and support for democratic systems.
†

18

hand, the respondent’s perceived political knowledge and news consumption lead a person to believe they have less of a say in what government does. The variable for discussing politics with others is not significant here but is relevant to the model.

G. Regression Coefficients: No Say in Government
No Say in Government? No Say in Government Trusters 0.53** (6.76) Distrusters -0.46** (5.40) Years of education 0.10** (6.36) Attention to elections 0.70** (2.89) R political knowledge (interviewer -0.23** assessed) (5.76) News consumption -0.36** (2.63) Discusses politics with others -0.15 (1.33) Constant 2.22** (6.53) Observations 1227 Adjusted R-squared 0.21 Absolute value of t statistics in parentheses * significant at 5%; ** significant at 1%

To summarize briefly, there is a parabolic relationship here representing skeptics as a different animal than people who trust or distrust but there are also significant obstacles to overcome before this concept can be fully explored. One barrier to better understanding noted above is that involving creation of dummy variables. Another is that the creation of interactive variables would greatly aid in our understanding of skepticism yet my attempts at utilizing interactive terms have yielded variables that do not rise to an acceptable level of significance. These are significant barriers to better understanding

19

skeptical attitudes and point to a need for more research using NES data and, possibly, to new attitudinal measures.

IV.

DISCUSSION

What we do know is that the parabolic relationship revealed here notes an unusual similarity between trusters and distrusters in their attitudes toward democratic systems and their own personal political efficacy. It appears that the sorts of citizens who are most supportive of democratic systems, and therefore provide for greater stability in civil society and legitimacy for ruling elites, are those who do not entirely trust or distrust their government.

As noted in the introduction there are precedents for these findings. Unfortunately, existing work on skeptical beliefs has largely been conducted in emerging democracies. The findings from that work may be significantly different than those of consolidated democracies like the United States. Generally speaking it appears that there may be some attitudinal similarities between the two in terms of support for democracy and skeptical attitudes toward government. There is certainly strong theoretical support for such a relationship but only further research, perhaps better measures or new ways of constructing existing measures, will yield findings that can be discussed in greater detail and with more confidence.

20

Tentatively we can state that attentiveness to elections yields results that point to greater faith in the way democracy works in America and in personal opinion that one vote can help determine what government does. Those who pay most attention to elections are likely already predisposed to have some faith in the political system, so this may not be too surprising a finding. News consumption appears to be positively related to attitudes towards democracy in the United States but also points to a decrease in beliefs in individual efficacy. It is possible that news consumption leads people to feel that democracy works but only as a competition between elites and not as a citizen check on governing elites.

One surprising finding is that level of education doesn’t demonstrate a stronger influence on attitudes towards democracy or personal efficacy. While education is not significant in relation to satisfaction with democracy it is significant and does yield a small positive influence on a person’s sense of political efficacy. Previous work on skeptical attitudes and civic education in the Dominican Republic has revealed a strong correlation between the two. (Bevis, Finkel and Sabatini, 2000) It was impossible to disentangle types of education (e.g. liberal arts versus professional) and levels of education (e.g. high school versus college versus graduate degree) in my particular dataset so a stronger and more nuanced relationship may exist here. If this 2000 NES dataset is unworkable it is possible that data from other years are more workable.

Future work on skeptical attitudes should look to replicate these findings in NES datasets from other years. There are a number of ways to gain more confidence in these findings,

21

one is to locate them in other years. Another would be to find a way to segment the trust index into dummy variables which better reflect the demonstrated graphical relationship. To this end further work needs to be done on the causes of skepticism and research on the ‘tipping points’ where skepticism turns into distrust or trust. Finally, this year will bring attempts by this author to develop and explore new measures of skepticism and its relation to democratic systems and political efficacy.

22

APPENDIX

Question and response wordings from the 2000 NES used for this paper:

T6 Spec: 685__ V001651 On the whole, are you satisfied, fairly satisfied, not very satisfied, or not at all satisfied with the way democracy works in the United States? 1 SATISFIED 2 FAIRLY SATISFIED 3 NOT VERY SATISFIED 4 NOT AT ALL SATISFIED

Q1b. Spec: 549__ V001528 ' People like me don't have any say about what the government does. ' Do you agree strongly, agree somewhat, neither agree nor disagree, disagree somewhat, or disagree strongly with this statement? 1 AGREE STRONGLY 2 AGREE SOMEWHAT 3 NEITHER AGREE NOR DISAGREE 4 DISAGREE SOMEWHAT 5 DISAGREE

23

Bibliography

Barber, Bernard. 1983. The Logic and Limits of Trust. New Brunswick, NJ: Rutgers University Press Braithwaite, Valerie. 1998. Institutionalizing distrust, enculturating trust. See Braithwaite & Levi 1998, pp.343-75 Braithwaite, Valerie, Levi, Margaret, eds. 1998. Trust and Governance. New York: Russell Sage Foundation Citrin, Jack. 1974. Comment: the political relevance of trust in government. American Political Science Review. 68: 973-88 Citrin, J., McClosky, H., Shanks, J.M., Sniderman, P.M. 1975. Personal and Political Sources of Political Alienation. British Journal of Political Science. 5: 1-31 Citrin, Jack, Green, Donald. 1986. Presidential leadership and the resurgence of trust in government. British Journal of Political Science. 5: 1-31 Citrin, Jack, Muste, Christopher. 1999. Trust in government. See Robinson 1999, pp.465532 Cooper, Joseph. 1999a. Congress and the Decline of Public Trust. Boulder, CO: Westview Cooper, Joseph. 1999b. The puzzle of distrust. See Cooper 1999a, pp.1-26 Craig, Stephen. 1993. The Malevolent Leaders: Popular Discontent in America. Boulder, CO: Westview Craig, Stephen. 1996. The angry voter: politics and popular discontent in the 1990s. In Broken Contract: Changing Relationships Between Americans and their Government, ed. SC Craig, pp. 46-66. Boulder, CO: Westview Easton, David. 1965. A Systems Analysis of Political Life. New York: Wiley Easton, David. 1975. A reassessment of the concept of political support. British Journal of Political Science. 5: 435-57 Finkel, Steve, Sabatini, Christopher, Bevis, Gwendolyn. 2000. Civic education, civil society, and political mistrust in a developing democracy: The case of the Dominican Republic. World Development. Vol. 28 No. 11: 1851-1874

24

Foley, Michael, Edwards, Bob. 1996. The paradox of civil society. Journal of Democracy. 7.3: 38-52 Gamson, William A. 1968. Power and Discontent. Homewood, IL: Dorsey Gibson, James, Duch, Raymond, Tedin, Kent. 1992. Democratic values and the transformation of the Soviet Union. The Journal of Politics. Vol. 54 No. 2: 329-71 Hardin, Russell. 1998. Trust in Government. See Braithwaite & Levi 1998, pp.9-27 Hart, Vivien. 1978. Distrust and Democracy: Political Trust in America and Britain. New York: Cambridge University Press Hetherington, Marc. 1998. The political relevance of political trust. American Political Science Review. 92: 791-808 Hetherington, Marc. 1999. The effect of political trust on the presidential vote, 1968-96. American Political Science Review. 93: 311-26 Hetherington, Marc. 2005. Why Trust Matters: Declining Political Trust and the Demise of American Liberalism. Princeton: Princeton University Press Hibbing, John, Theiss-Morse, Elizabeth. 1995. Congress as Public Enemy: Public Attitudes Toward American Political Institutions. New York: Cambridge University Press Hibbing, John, Theiss-Morse, Elizabeth. 2002. Stealth Democracy: Americans’ Beliefs about How Government Should Work. New York: Cambridge University Press King, David. 1997. The polarization of American parties and mistrust of government. See Nye et al. 1997, pp. 155-78 Levi, Margaret. March 1996. Social and unsocial capital: A review essay of Robert Putnam's Making Democracy Work. Politics and Society 24: 45-55 Levi, Margaret. 1998. A state of trust. See Braithwaite & Levi 1998, pp. 77-101 Levi, Margaret, Stoker, Laura. 2000. Political trust and trustworthiness. Annual Review of Political Science. 3: 475-507 Miller, Arthur. 1974a. Political issues and trust in government: 1964-1970. American Political Science Review. 68: 951-72 Miller, Arthur. 1974b. Rejoinder to “comment” by Jack Citrin: political discontent or ritualism? American Political Science Review. 68: 989-1001

25

Mishler, William, Rose, Richard. 1997. Trust, distrust and skepticism: popular evaluations of civil and political institutions in post-communist societies. Journal of Politics. 59: 419-51 Nye, Joseph, Zelikow, Philip, King, David, eds. 1997. Why People Don’t Trust Government. Cambridge, MA: Harvard University Press Robinson, John, Shaver, Philip, Wrightsman, Lawrence, eds. 1999. Measures of Political Attitudes. New York: Academic Rosenstone, Stephen, Hansen, John. 1993. Mobilization, Participation, and Democracy in America. New York: Macmillan Uslaner, Eric. September, 1998. Social capital, television, and the ‘mean world’: Trust, optimism, and civic participation. Political Psychology. 19: 441-67 Uslaner, Eric. 1999. Democracy and social capital. See Warren 1999a, pp.121-150 Uslaner, Eric. 2002. The Moral Foundations of Trust. New York: Cambridge University Press Vice, James W. The Reopening of the American Mind: On Skepticism and Constitutionalism. Atlanta: Rodopi Warren, Mark, ed. 1999a. Democracy and Trust. New York: Cambridge University Press Warren, Mark. 1999b. Democratic theory and trust. See Warren 1999a, pp.310-45

26


								
To top