The Centrist Paradox

Political Correlates of the Democratic Disconnect

David Adler

Abstract
A wave of new research has documented the decline in citizen support for demo-
cracy. But the distribution of anti-democratic attitudes remains unclear. In this
article, I use the World Values Survey and European Values Survey to exam-
ine the relationship between democratic discontent and the left-right political
spectrum. I find that, contrary to much contemporary commentary, hostility
to democracy is strongest not at the political extremes, but in the center. Re-
spondents at the center of the political spectrum are the least supportive of
democracy, least committed to its institutions, and most supportive of author-
itarianism. I refer to this surprising finding as the ‘centrist paradox.’ Recent
research has advanced our understanding of extremist parties. We know far
less, however, about moderates. The centrist paradox points to this lacuna.

1
Introduction
Across Europe and North America, established democracies confront an array of
new challenges. Elite influence on policymaking has eroded support for mainstream
parties (Bartels 2009; Hacker and Pierson 2010; Gilens and Page 2014). Declining
attachment to mainstream parties has encouraged the rise of extremist candidates
(Ezrow et al. 2014). And the polarization of political parties has engendered cross-
partisan hostility (Iyengar and Westwood 2015; Sood and Iyengar 2016). These
trends have led some observers to worry about the possibility of democratic decon-
solidation, in which democracy would cease to be the ‘only game in town’ (Linz and
Stepan 1996; Foa and Mounk 2016).
Among these diverse challenges, the decline in support for democracy among
voters has attracted considerable attention. The evidence of disaffection with demo-
cracy is extensive (Armingeon and Guthmann 2014), but the drivers and distribution
of such anti-democratic preferences remain unclear. Recent scholarship has high-
lighted two sets of variables associated with democratic discontent. Some studies
have focused on economic factors like exposure to global trade, which generate eco-
nomic anxiety and breed political resentment, in turn (Autor et al. 2016; Colantone
and Stanig 2017; Giuso et al. 2017). Others have focused on sociological factors like
demographic change, which lead to a ‘cultural backlash’ against democratic pluralism
(Becker et al. 2016; Norris and Inglehart 2016).
Despite these multiple hypotheses, to the best of my knowledge, no one has
examined empirically the distribution of democratic attitudes across the political
spectrum. The implicit assumption of the existing scholarship seems to be that the
threat to democracy is coming from the political extremes, where ‘the fringes appear
eager to escalate’ political conflict (Prior 2013: 102). The intuition is that there is an
elective affinity between extreme political views and support for strongman politics
to implement them. From this premise, a large and growing literature has examined
the link between political polarization and democratic dysfunction (Heberlig et al.
2006; Mann and Ornstein 2012; Gordon and Landa 2017).
In this paper, I test this assumption in Western democracies, using data from the
most recent wave of the World Values Survey and the European Values Survey. I find
that — contrary to conventional wisdom — anti-democratic sentiment is strongest
at the center. Respondents who identify as centrists are the most frequently crit-
ical of democracy and most likely to support authoritarianism. These patterns are
generated, in part, by higher levels of apathy among self-identified centrists (Ab-
ramowitz 2011). However, even when I exclude politically apathetic respondents,
the relationship between centrism and hostility to democracy remains robust. I refer
to this surprising finding as the ‘centrist paradox.’ The rest of the article is organized
in three parts. In the first, I validate the left-right measurement in the WVS and
EVS. In the second, I examine the relationship between these political positions and
respondents’ attitudes toward democracy. Finally, I offer a partial explanation of the
centrist paradox and discuss directions for future research.

2
Data & Methods
To examine the relationship between political orientation and attitudes toward demo-
cracy, I use the most recent data from the World Values Survey (WVS), Wave 6
(2010-2014), and the European Values Survey (EVS), Wave 4 (2008). Although they
predate recent events that have drawn attention to democratic discontent, these data
are the most recent that allow for consistent cross-country analysis across a battery
of questions about respondents’ political attitudes.
The key independent variable is respondents’ left-right position. The survey asks
respondents to place themselves on the spectrum from 1 (“left”) to 10 (“right”). As
Table 1 shows, there is a balance in the survey between left, right, and center. I
recode the data with five groups: far left (1-2), center-left (3-4), center (5-6), center-
right (7-8), and far right (9-10).

Table 1: Distribution of Political Views in EVS

Left-Right Political View N Percent
Left 2,684 5.5
2 1,913 3.9
3 4.259 8.7
4 5.382 8.9
5 15,639 31.9
6 6,240 12.7
7 4,684 9.6
8 4,526 9.2
9 1,716 3.5
Right 3,002 6.1

Several studies have validated the use of left-right self-placements in cross-national

analysis (Huber 1989; Knutsen 1995; Evans et al. 1996). However, some reasonable
concerns remain. It is possible that respondents place themselves at the center when
they are confused, or when they do not care enough to take sides. I address this pos-
sibility in four ways. First, the survey includes a “Don’t Know” response to left-right
placement, which account for more than 10 percent of the sample. I exclude “Don’t
Know” responses from subsequent analysis. Second, I test the self-placement against
views of immigration (Appendix A). The EVS includes a question about whether
“Immigrants are a strain on the welfare system,” from 1 (“Strain”) to 10 (“Not a
strain”). Across models, political orientation strongly predicts views of immigration,
with large and statistically significant differences between groups. On the issue of
immigration, then, the centrist self-placement appears to align with the moderate
position. Third, I test self-placement against the question of whether equality or free-
dom is “more important” (Appendix A), with the expectation that respondents on the
left prefer equality while those on the right prefer freedom (Rokeach 1973). Again,
self-placement strongly predicts preferences over equality and freedom, providing

3
further validation of the left-right placement. Finally, I re-run all the analysis on
the sub-sample of politically engaged respondents (Appendix C), using two different
measures of engagement: interest in politics (“Very interested” or “Somewhat inter-
ested”) and participation in political activities (“Would vote tomorrow” in a general
election).
To examine the relationship between left-right placement and attitudes toward
democracy, I conduct Foa and Mounk’s (2016) three tests of the ‘democratic dis-
connect’, plotting the frequency of support for democracy, democratic institutions,
and authoritarianism across the political spectrum. To check for robustness, I run
a series of OLS regressions across the pooled sample (Table 2 and Appendix B) as
well as in single countries (Appendix D), using the center as the reference category
to measure differences with the rest of the political spectrum. To test for omitted
variable bias, I re-run all models with control for three key variables identified in
recent research on the rise of populism — income (Rodrik 2017), education (Norris
and Inglehart 2016), and age (Foa and Mounk 2016).

Analysis
Support for Democracy
Respondents who identify with the center of the political spectrum are the least
supportive of democracy, according to several measures from the EVS and WVS.
In Figures 1 and 2, I plot the relationship between left-right political views and
support for democracy in Europe and the United States, respectively. In Table 2,
I examine two EVS variables in greater depth: (i) view of democracy as the “Best
Political System,” from 1 (“Disagree Strongly”) to 4 (“Agree Strongly”); (ii) rating of
democracy, from 1 (“Very Bad”) to 4 (“Very Good”). Models 1 through 4 confirm large
and significantly negative relationships between centrism and support for democracy.
Perhaps the most striking data, however, are views of free and fair elections. The
WVS asks respondents to rate from 1 (“Not essential”) to 10 (“Essential”) that “People
choose their leaders in free elections.” I find that support drops at the political center
in all mature democracies in the sample (Appendix B). The size of the centrist gap
is striking. In the case of the United States, less than half of the political centrists
in the United States view free elections as essential to democracy – over 30 percent
less than their center-left neighbors.

4
Figure 1: Europe: “Agree Strongly” that Democracy is the “Best Political System”

Figure 2: United States: Democracy is a “Very Good” Political System

5
 Table 2: Left-Right Position & Attitudes Toward Democracy in the EVS
(1) (2) (3) (4) (5) (6)
Best Best Rating Rating Strong Strong

Far Left 0.025** 0.030** 0.082*** 0.096*** -0.096*** -0.130***

(0.011) (0.012) (0.0119) (0.013) (0.0171) (0.018)
Center Left 0.123*** 0.080*** 0.135*** 0.084*** -0.329*** -0.248***
(0.009) (0.009) (0.00930) (0.010) (0.013) (0.014)
Center Right 0.110*** 0.078*** 0.111*** 0.076*** -0.048*** 0.015
(0.009) (0.009) (0.00912) (0.010) (0.013) (0.014)
Far Right 0.102*** 0.130*** 0.105*** 0.142*** 0.188*** 0.120***
(0.011) (0.012) (0.0119) (0.013) (0.017) (0.0177)
Income 0.045*** 0.051*** -0.107***
(0.001) (0.001) (0.002)
Education 0.031*** 0.030*** -0.033***
(0.003) (0.003) (0.004)
Age 0.002*** 0.001*** -0.002***
(0.000) (0.000) (0.000)
Constant 3.264*** 2.865*** 1.738*** 2.127*** 2.815*** 2.084***
(0.005) (0.015) (0.005) (0.016) (0.007) (0.022)

Observations 46,224 38,707 46,199 38,644 45,863 38,463

R-squared 0.007 0.048 0.006 0.050 0.020 0.107
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

6
 Support, however, is not satisfaction. The EVS asks whether respondents are
“satisfied with democracy”, ranging from 1 (“Very Satisfied”) to 4 (“Not at all satis-
fied”). I find that dissatisfaction grows from right to left (Figure 3). The distinction
between support and satisfaction is crucial for our understanding of recent protest
movements. For many observers, the rise in resentment toward the establishment ap-
pears, in and of itself, as a threat to democracy. In contrast, I find those who identify
at the center of the spectrum — where we find moderate levels of satisfaction — are
the most hostile to democratic principles.

Figure 3: Europe: “Not Very Satisfied” or “Not At All Satisfied” with Democracy

7
 Support for Democratic Institutions
While the data in the previous section are striking, the concept of “support for
democracy” is somewhat abstract: respondents may interpret the question in differ-
ent ways (Voeten 2017). I therefore turn to measures of support for actually existing
democratic institutions. I begin with a WVS question about civil rights: “Civil rights
protect people’s liberty from state oppression.” Respondents rate this question from
1 (“Not an essential characteristic of democracy”) to 10 (“An essential characteristic
of democracy”). Support for civil rights falls at the political center in six out of eight
mature democracies in the sample.
If civil rights represent democratic institutions enforced by the state, political
engagement is central to the maintenance of democratic culture. According to the
WVS data, engagement is also weakest at the political center. In the United States,
for example, while 76 percent of the left and 77.5 percent of the right claim they
“always” vote in national elections, only 50 percent of the center vote in national
elections. Interest in politics also displays a V shape in 100 per cent of WVS cases.
Table 5 in Appendix B displays large and statistically negative relationships between
centrism and support for civil rights, as well as support for free and fair elections.

Figure 4: United States: Civil Rights are “Essential”

8
 Support for Authoritarian Alternatives
Of course, criticism of democracy does not necessarily imply support for authorit-
arian alternatives. Turning to a new set of survey questions, I find evidence of sub-
stantial support for authoritarianism in the center of the political spectrum. Both
EVS and WVS asks respondents to rate a system in which a “strong leader does not
have to bother with parliament,” from 1 (“Very bad”) to 4 (“Very good”). In Figure
5, I plot the percentage of respondents that answer “Very good” or “Fairly good” as
an indicator of support for authoritarianism. In Table 2, I regress all ratings in the
EVS on the left-right placement. Findings suggest that the only group with higher
support is the far-right.

Figure 5: Support for “Strong Leader” who ignores “Parliament” or “Congress”

9
Discussion
Across Europe and North America, citizen support for democracy is waning. To
explain this trend, conventional wisdom points to the rise of political extremes. My
analysis of data from the World Values Survey and European Values Survey, however,
suggests that hostility to democracy is strongest in the center. Centrists in the
sample are the least supportive of democratic institutions and the most supportive
of authoritarian rule, controlling for key covariates.
What explains these surprising patterns? One important component is apathy.
Centrists in this sample are less interested in politics and less likely to vote than other
parts of the political spectrum. Apathy is, however, an insufficient explanation. The
relationship between centrism and hostility to democracy is not confined to apathetic
respondents. Among the sub-sample of respondents who identify as “Very interested”
or “Somewhat interested” in politics, I find similarly large and significantly negative
relationships to the centrist position (Appendix C). Among the sub-sample that
“Would vote tomorrow” in a general election, the findings are similar (Appendix C).
The centrist paradox refers to this striking relationship between political moderation
and anti-democratic attitudes.
These findings call for a broader examination of centrist politics in mature demo-
cracies. Recent research has made significant progress in our understanding of ex-
tremist parties on both the left (March and Mudde 2005; March 2012; Gomez et al.
2016) and the right (Kitschelt and McGann 1997; Norris 2005; Mudde 2007). We
know far less, however, about centrist moderates. Modernization theory has long
highlighted the support for authoritarian transitions among a base of middle-class
moderates (O’Donnell 1973; Rueschemeyer et al. 1992), but this insight remains
largely unexplored in advanced democracies. Future research might usefully explore
varieties of political moderation — across national contexts and, building from these
data, over time.

10
References

Armingeon, K. and Guthmann, K., 2014. Democracy in crisis? The declining sup-
port for national democracy in European countries, 2007–2011. European Journal
of Political Research, 53(3), pp.423-442.

Autor, D., Dorn, D., Hanson, G. and Majlesi, K., 2016. Importing Political Po-
larization? Massachusetts Institute of Technology Manuscript.

Bartels, L.M., 2009. Unequal democracy: The political economy of the new gil-
ded age. Princeton University Press.

Becker, S. O., Thiemo Fetzer, and Dennis Novy, “Who Voted for Brexit? A Com-
prehensive District-Level Analysis,” University of Warwick, October 2016.

Castles, F.G. and Mair, P., 1984. Left–right political scales: Some ‘expert’judgments.
European Journal of Political Research, 12(1), pp.73-88.

Colantone, I., and Piero Stanig, “The Trade Origins of Economic Nationalism: Im-
port Competition and Voting Behavior in Western Europe,” BAFFI CAREFIN
Centre Research Paper Series No. 2017- 49, January 2017.

Evans, G., Heath, A. and Lalljee, M., 1996. Measuring left-right and libertarian-
authoritarian values in the British electorate. British Journal of Sociology, pp.93-
112.

Ezrow, L., Tavits, M. and Homola, J., 2014. Voter polarization, strength of par-
tisanship, and support for extremist parties. Comparative Political Studies, 47(11),
pp.1558-1583.

Foa, R.S. and Mounk, Y., 2016. The democratic disconnect. Journal of Demo-
cracy, 27(3), pp.5- 17.

Gilens, M. and Page, B.I., 2014. Testing theories of American politics: Elites, in-
terest groups, and average citizens. Perspectives on politics, 12(3), pp.564-581.

Gomez, R., Morales, L. and Ramiro, L., 2016. Varieties of radicalism: examining
the diversity of radical left parties and voters in Western Europe. West European
Politics, 39(2), pp.351- 379.

Gordon, S.C. and Landa, D., 2017. Common Problems (or, What’s Missing from
the Conventional Wisdom on Polarization and Gridlock). The Journal of Politics,

11
79(4), pp.1433-1437.

Guiso, L., Helios Herrera, Massimo Morelli, and Tommaso Sonno, “Demand and
Supply of Populism,” CEPR Discussion Paper DP11871, February 2017.

Hacker, J.S. and Pierson, P., 2010. Winner-take-all politics: How Washington made
the rich richer--and turned its back on the middle class. Simon and Schuster.

Heberlig, E., Hetherington, M. and Larson, B., 2006. The price of leadership: Cam-
paign money and the polarization of congressional parties. The Journal of Politics,
68(4), pp.992-1005.

Huber, J.D., 1989. Values and partisanship in left-right orientations: Measuring

ideology. European Journal of Political Research, 17(5), pp.599-621.

Inglehart, R. and Norris, P., 2016. Trump, Brexit, and the rise of populism: Eco-
nomic have-nots and cultural backlash. Harvard Faculty Research Working Paper
Series.

Iyengar, S. and Westwood, S.J., 2015. Fear and loathing across party lines: New evid-
ence on group polarization. American Journal of Political Science, 59(3), pp.690-707.

Kitschelt, H. and McGann, A.J., 1997. The radical right in Western Europe: A
comparative analysis. University of Michigan Press.

Knutsen, O., 1995. Value orientations, political conflicts and left-right identifica-
tion: A comparative study. European journal of political research, 28(1), pp.63-93.

Linz, J.J. and Stepan, A., 1996. Problems of democratic transition and consolid-
ation: Southern Europe, South America, and post-communist Europe. JHU Press.

Mann, T.E. and Ornstein, N.J., 2016. It’s even worse than it looks: How the Amer-
ican constitutional system collided with the new politics of extremism. Basic Books.

March, L., 2012. Radical left parties in Europe. Routledge.

March, L. and Mudde, C., 2005. What’s left of the radical left? The European
radical left after 1989: Decline and mutation. Comparative European Politics, 3(1),
pp.23-49.

Mudde, C., 2007. Populist radical right parties in Europe (Vol. 22, No. 8). Cam-
bridge: Cambridge University Press.

12
Norris, P., 2005. Radical right: Voters and parties in the electoral market. Cam-
bridge University Press.

O’Donnell, G., 1973. Modernization and Bureaucratic-Authoritarianism Institute

of International Studies. University of California, Berkeley.

Rokeach, M., 1973. The nature of human values. Free press.

Rueschemeyer, D., Stephens, E.H. and Stephens, J.D., 1992. Capitalist develop-
ment and democracy. Cambridge, UK.

Prior, M., 2013. Media and political polarization. Annual Review of Political Sci-
ence, 16, pp.101-127.

Rodrik, D., 2017. Populism and the Economics of Globalization (No. w23559).
National Bureau of Economic Research.

Sood, G. and Iyengar, S., 2016. Coming to Dislike Your Opponents: The Polar-
izing Impact of Political Campaigns.

13
Appendix A: Left-Right Validation

Figure 6: Left-Right Placement & Opposition to Immigration

Figure 7: Left-Right & “Equality More Important Than Freedom”

14
 Table 3: Validation of the Left-Right Spectrum in the EVS
(1) (2) (3) (4)
Immigration Immgration Equality–Freedom Equality–Freedom

Far Left 0.112*** 0.125*** -0.134*** -0.107***

(0.014) (0.015) (0.016) (0.018)
Center Left 0.106*** 0.104*** -0.144*** -0.149***
(0.011) (0.012) (0.013) (0.014)
Center Right -0.034*** -0.028** 0.119*** 0.124***
(0.010) (0.011) (0.012) (0.014)
Far Right -0.0304** -0.022 0.121*** 0.129***
(0.0135) (0.015) (0.016) (0.017)
Income -0.012*** -0.001
(0.002) (0.002)
Education 0.055*** 0.038***
(0.003) (0.004)
Age -0.003*** -0.002***
(0.000) (0.000)
Constant -0.295*** -0.258*** 0.107*** 0.077***
(0.006) (0.018) (0.007) (0.022)

Observations 47,188 39,410 47,328 39,530

R-squared 0.004 0.019 0.010 0.015
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

15
Appendix B: Analysis of WVS

Table 4: Support for Democracy & Authoritarianism in the WVS

(1) (2) (3) (4)
Rating Rating Strong Strong

Far Left 0.117*** 0.133*** -0.157*** -0.171***

(0.030) (0.031) (0.041) (0.041)
Center Left 0.147*** 0.155*** -0.202*** -0.193***
(0.023) (0.024) (0.032) (0.032)
Center Right 0.0556** 0.0231 -0.0371 0.009
(0.024) (0.024) (0.032) (0.032)
Far Right -0.092*** -0.152*** 0.032 0.087**
(0.031) (0.032) (0.042) (0.042)
Income 0.032*** -0.027***
(0.004) (0.0057)
Age 0.003*** -0.003***
(0.000) (0.001)
Education 0.021*** -0.091***
(0.004) (0.005)
Constant 3.422*** 2.956*** 1.941*** 2.836***
(0.018) (0.043) (0.025) (0.057)

Observations 7,318 6,753 7,169 6,630

R-squared 0.012 0.039 0.010 0.071
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Sample: Australia, Germany, Netherlands, New Zealand, Poland, Spain, Sweden, and the United States.

16
 Table 5: Support for Democratic Institutions in the WVS
(1) (2) (3) (4)
Civil Rights Civil Rights Free Elections Free Elections

Far Left 0.663*** 0.681*** 0.418*** 0.513***

(0.098) (0.100) (0.085) (0.085)
Center Left 0.611*** 0.598*** 0.481*** 0.459***
(0.077) (0.078) (0.0657) (0.066)
Center Right 0.147* 0.089 0.259*** 0.172**
(0.077) (0.079) (0.066) (0.067)
Far Right -0.235** -0.285*** 0.152* 0.074
(0.101) (0.104) (0.086) (0.088)
income 0.045*** 0.082***
(0.014) (0.012)
age 0.000 0.006***
(0.002) (0.001)
education 0.117*** 0.123***
(0.013) (0.011)
Constant 7.908*** 6.925*** 8.630*** 7.178***
(0.0605) (0.140) (0.0519) (0.118)

Observations 7,327 6,758 7,414 6,831

R-squared 0.019 0.039 0.008 0.047
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
Sample: Australia, Germany, Netherlands, New Zealand, Poland, Spain, Sweden, and the United States.

17
 Appendix C: Politically Engaged Respondents
Below, I focus on a sub-sample of the EVS that is politically engaged. In Figures 4-7
and Table 3, I examine respondents who identify as “Very interested” or “Somewhat
interested” in politics. In Table 4, I examine respondents who “would vote in a general
election tomorrow.” The results suggest that apathy is not the only explanation for
the centrist paradox.

Figure 8: United States: Support for “Strong Leader”

18
 Figure 9: United States: Elections are “Essential”

Figure 10: United Kingdom: Democracy is a “Very Good” Political System

19
Figure 11: Europe: Democracy is a “Very Good” Political System

20
 Table 6: Attitudes Among Non-Apathetic Respondents in the EVS
(1) (2) (3) (4) (5) (6)
Best Best Rating Rating Strong Strong

Far Left 0.023 0.026* 0.107*** 0.125*** -0.207*** -0.207***

(0.014) (0.015) (0.015) (0.016) (0.023) (0.023)
Center Left 0.119*** 0.071*** 0.145*** 0.093*** -0.274*** -0.274***
(0.012) (0.012) (0.012) (0.013) (0.019) (0.019)
Center Right 0.089*** 0.051*** 0.096*** 0.057*** 0.023 0.023
(0.011) (0.012) (0.012) (0.013) (0.018) (0.018)
Far Right 0.069*** 0.010*** 0.073*** 0.117*** 0.089*** 0.089***
(0.014) (0.015) (0.015) (0.016) (0.023) (0.023)
Income 0.050*** 0.057*** -0.117*** -0.117***
(0.002) (0.002) (0.002) (0.002)
Education 0.029*** 0.024*** -0.0240*** -0.024***
(0.003) (0.004) (0.00520) (0.005)
Age 0.002*** 0.000* -0.00162*** -0.002***
(0.000) (0.000) (0.000) (0.000)
Constant 3.342*** 2.921*** 3.335*** 2.948*** 2.897*** 2.897***
(0.00672) (0.020) (0.007) (0.021) (0.031) (0.031)

Observations 24,724 20,896 24,684 20,839 20,779 20,779

R-squared 0.005 0.059 0.007 0.066 0.131 0.131
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

21
 Table 7: Attitudes Among Voters in the EVS
(1) (2) (3) (4) (5) (6)
Best Best Rating Rating S. Leader S. Leader

Far Left 0.031** 0.029** 0.083*** 0.089*** -0.123*** -0.149***

(0.012) (0.013) (0.013) (0.014) (0.019) (0.020)
Center Left 0.120*** 0.074*** 0.139*** 0.088*** -0.342*** -0.247***
(0.010) (0.010) (0.010) (0.011) (0.015) (0.016)
Center Right 0.101*** 0.066*** 0.102*** 0.067*** -0.051*** 0.018
(0.009) (0.010) (0.010) (0.011) (0.015) (0.015)
Far Right 0.0820*** 0.112*** 0.0889*** 0.129*** 0.182*** 0.107***
(0.0120) (0.013) (0.013) (0.014) (0.019) (0.019)
Income 0.046*** 0.050*** -0.112***
(0.001) (0.002) (0.002)
Education 0.028*** 0.028*** -0.032***
(0.003) (0.003) (0.004)
Age 0.002*** 0.001*** -0.002***
(0.000) (0.000) (0.000)
Constant 3.302*** 2.891*** 3.307*** 2.916*** 2.162*** 2.917***
(0.005) (0.017) (0.006) (0.017) (0.008) (0.025)

Observations 36,221 30,605 36,201 30,556 35,883 30,365

R-squared 0.006 0.051 0.006 0.053 0.021 0.119
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

22
Appendix D: Country Case Studies

Table 8: Attitudes Toward Democracy in the United States

(1) (2) (3) (4) (5) (6)
Civil Rights C. R. Free Election F. E. Strong Strong

Far Left 1.808*** 1.541*** 1.159*** 0.864*** -0.562*** -0.470***

(0.236) (0.233) (0.218) (0.210) (0.091) (0.090)
Center Left 1.487*** 1.247*** 1.296*** 0.998*** -0.390*** -0.298***
(0.161) (0.160) (0.149) (0.145) (0.062) (0.061)
Center Right 0.496*** 0.324** 0.914*** 0.622*** -0.362*** -0.286***
(0.140) (0.141) (0.129) (0.126) (0.054) (0.053)
Far Right 0.462*** 0.401** 0.908*** 0.736*** -0.333*** -0.283***
(0.168) (0.169) (0.156) (0.152) (0.065) (0.064)
Income 0.0195 0.060** 0.023**
(0.029) (0.026) (0.011)
Age 0.002 0.022*** -0.010***
(0.003) (0.003) (0.001)
Education 0.378*** 0.398*** -0.150***
(0.043) (0.039) (0.016)
Constant 7.064*** 4.001*** 7.952*** 3.613*** 2.843*** 1.328***
(0.075) (0.361) (0.069) (0.323) (0.029) (0.137)

Observations 2,114 2,091 2,125 2,101 2,136 2,106

R-squared 0.056 0.094 0.054 0.134 0.043 0.110
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

23
 Table 9: Attitudes Toward Democracy in the United Kingdom
(1) (2) (3) (4) (5) (6)
Best Best Rating Rating Strong Strong

Far Left 0.179** 0.161* 0.216** 0.239** -0.287** -0.323**

(0.082) (0.092) (0.099) (0.113) (0.120) (0.136)
Center Left 0.148*** 0.113* 0.107* 0.0335 -0.262*** -0.204**
(0.051) (0.058) (0.061) (0.071) (0.074) (0.086)
Center Right 0.200*** 0.170*** 0.154** 0.0753 -0.222*** -0.177*
(0.054) (0.064) (0.063) (0.077) (0.077) (0.093)
Far Right 0.473*** 0.475*** 0.292*** 0.194 -0.145 -0.111
(0.0903) (0.109) (0.108) (0.129) (0.131) (0.159)
Income 0.039*** 0.057*** -0.069***
(0.010) (0.0120) (0.014)
Education 0.052*** 0.073*** -0.086***
(0.017) (0.020) (0.025)
Age 0.008*** 0.008*** -0.005***
(0.001) (0.002) (0.002)
Constant 3.159*** 2.315*** 3.261*** 2.265*** 2.037*** 3.040***
(0.026) (0.109) (0.031) (0.131) (0.037) (0.157)

Observations 1,176 813 1,178 815 1,174 813

R-squared 0.033 0.117 0.013 0.090 0.017 0.086
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

24