Public confidence in the judiciary: The interaction between political awareness and level of democracy

This article was downloaded by: [85.107.190.32]On: 18 April 2015, At: 04:02Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,UK

Click for updates

DemocratizationPublication details, including instructions for authorsand subscription information:http://www.tandfonline.com/loi/fdem20

Public confidence in thejudiciary: the interactionbetween political awareness andlevel of democracyAylin Aydın Çakıra & Eser Şekercioğlub

a Political Science and International RelationsDepartment, Yeditepe University, Istanbul, Turkeyb Business Administration Department, Izmir Universityof Economics, Izmir, TurkeyPublished online: 17 Apr 2015.

To cite this article: Aylin Aydın Çakır & Eser Şekercioğlu (2015): Public confidencein the judiciary: the interaction between political awareness and level of democracy,Democratization, DOI: 10.1080/13510347.2014.1000874

To link to this article: http://dx.doi.org/10.1080/13510347.2014.1000874

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all theinformation (the “Content”) contained in the publications on our platform.However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, orsuitability for any purpose of the Content. Any opinions and views expressedin this publication are the opinions and views of the authors, and are not theviews of or endorsed by Taylor & Francis. The accuracy of the Content shouldnot be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions,claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connectionwith, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Public confidence in the judiciary: the interaction betweenpolitical awareness and level of democracy

Aylin Aydın Cakıra∗ and Eser Sekercioglub

aPolitical Science and International Relations Department, Yeditepe University, Istanbul,Turkey; bBusiness Administration Department, Izmir University of Economics, Izmir, Turkey

(Received 18 June 2014; accepted 12 December 2014)

Apart from the studies that focus on public attitudes toward higher courts inadvanced democracies, we know little about the factors that can explainpublic confidence in the judiciary in a comparative setting. In this regard,the goal of this study is to explain whether, and to what extent, thecountry’s level of democracy moderates the impact of political awareness onpublic confidence in the judiciary. This study uses hierarchical linear modelsto analyse the interaction between individual and country level factors byusing the World Values Survey (2005–2009) data for 49 countries andvarious other data sources. Our empirical results show that in advanceddemocracies political awareness variables like education and politicalparticipation have a positive impact on public confidence in the judiciary,whereas in countries with weak levels of democracy higher politicalawareness leads to increased cynicism about the judiciary. These resultssuggest that a one-size-fits-all approach to explain confidence in thejudiciary is not possible when we are dealing with a wide range of societiesthat vary in terms of many characteristics, both institutional and cultural.

Keywords: public confidence in the judiciary; public trust in institutions;judicial legitimacy; political awareness; judicial performance; hierarchicalmodels

Introduction

Public confidence in democratic institutions is an important precondition for anefficient and sustainable democratic regime. Yet, not only publicly elected insti-tutions such as the legislature and executive but also non-elected institutionssuch as the judiciary require a deep reservoir of goodwill. Especially whenjudges make decisions contrary to the preferences of political majorities, thejudiciary faces a crisis of legitimacy and, if courts are not publicly supported,non-compliance and non-enforcement might become viable alternatives for theincumbent governments. However, if courts are highly supported by the citizens,

# 2015 Taylor & Francis

∗Corresponding author. Email: [email protected]

Democratization, 2015http://dx.doi.org/10.1080/13510347.2014.1000874

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

non-compliance with and non-enforcement of judicial decisions will not be a cred-ible threat.1 For this reason, public support is a crucial component of judicial legiti-macy without which the functioning of the legal system, the rule of law and abilityof the courts to manage public and political conflicts would be harmed.

The existing literature on judicial legitimacy presents public knowledge andawareness of the decisions and workings of the judiciary as one of the key deter-minants of public confidence in the judiciary. Many scholars have argued that thosewho know more about law and courts are more likely to support the judiciary andbelieve in its legitimacy. The mechanism seems clear enough. The more informedthe individual, the more likely that s/he will understand how the judicial processworks.2 As such, high levels of knowledge about the judiciary will create strongerattitudes about the impartiality, incorruptibility and trustworthiness of the judiciary.

Nevertheless, most of our working knowledge on the relationship between pol-itical savvy and confidence in the judiciary is based on the analyses of well-estab-lished democracies and mostly in the United States.3 According to these studiesconfidence in courts partly stems from an expectation that courts are an importantpart of a democratic system and that they mostly function properly. It is argued thatwell informed citizens, who are in a better position to understand the system,appreciate the judiciary more. But what happens in countries where the judiciaryis not fully independent and cannot efficiently enforce the rule of law? Does theinformed public still have high confidence in the judiciary? Or is the synthesisof heightened awareness and chronic systemic problems creating a cynical public?

Focusing on Latin American courts, Salzman and Ramsey find that individualswith higher levels of political knowledge have less confidence in the judicialsystem of their own country.4 Although this study focuses strictly on the regionof Latin America, its findings challenge the existing literature and reveal that theimpact of political awareness on judicial confidence may change across advancedand developing democracies. In order to take this study one step further, it is impor-tant to conduct a cross-national comparative research where a researcher will beable to compare and contrast the impact of political awareness on judicial confi-dence across countries with different democratization levels. This type of a com-parative study will not only serve as a generalizability test of the existingliterature but also empirically show whether the impact of political awareness onjudicial confidence significantly changes across advanced and developingdemocracies.

Arguing that individuals tend to evaluate political institutions as part of the pol-itical system, we suggest that the effect of individual level political awareness onpublic confidence in the judiciary should be modelled contingent on the country’slevel of democracy. Using the World Values Survey data (2005–2009) for 49countries and various other data sources, the current study tries to test theimpact of political awareness on public confidence in the judiciary across countrieswith different levels of democracy. In this regard our purpose is to (1) account forwhether, and to what extent, a country’s level of democracy moderates the impactof political awareness on public confidence in the judiciary and to (2) explain to

2 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

what extent individual-level and to what extent country-level determinants matterin shaping public confidence in the judiciary.

In the first section of the study we discuss the impact of political awareness onconfidence in judiciary and present this impact to be conditioned on a country’slevel of democracy. The subsequent section describes the data and variables thatwe use to test this micro-macro interactive model of public confidence in the judi-ciary. After the empirical section where we test our key hypothesis by using hier-archical models, we conclude the study by discussing the theoretical and practicalimplications of our findings.

Political awareness and public confidence in the judiciary: theoreticalframework

One of the most prominent approaches in the literature that is used to explain publicconfidence in institutions is formulated around performance based explanations.The studies using these explanations suggest that individuals have high levels oftrust in institutions only when they believe that the given institutions performwell.5 These explanations hypothesize that trust in political institutions stemsfrom how people perceive the respective political institutions work and to whatextent they are satisfied with their performance. The elements of the judiciary(courts, higher courts, justices) are no different. Citizens express high levels of con-fidence in the judiciary when courts and judges perform well or behave indepen-dently from any external political influences. As such, for instance, in theirrecent study Buhlmann and Kunz show that the more independent the judicialsystem is, the greater the probability that an individual living in that country hasconfidence in the judiciary.6 For this reason, it is natural to expect higher levelsof confidence in countries with high levels of judicial performance and quality.

One should note that individual level evaluation of judicial performance andquality requires accurate impressions of the institution and its workings. We donot argue that correct impressions require perfect information. The functioningof the judiciary is fairly technical and even a rudimentary working knowledgecannot be automatically assumed. Most of the time, regarding the workings ofthe judiciary, citizens can have direct experience with prolonged cases and judicialoutcomes where the individual rights are not protected. In this regard, personalexperience with courts is shown to affect the individual perceptions about the fair-ness of court procedures and this is claimed to affect the public’s support for thestate courts.7

Individual perceptions about the performance and quality of an institution arenot only shaped by personal experiences. Most citizens acquire what knowledgethey possess about the judicial system from sources such as media, including tele-vision and newspapers. One should note that through their influence on media cov-erage about a judicial decision and/or a legislative response, opinion leaders mayplay a central role in shaping public awareness about the judicial system. Yet, notall citizens are passive receivers of whatever media or opinion leaders make public.

Democratization 3

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Certain individual level traits may highly affect the information they receive andtheir willingness to accept persuasive influence.8 Education, interest in politicsand involvement in political processes are some of these key individual traitsthat closely determine a person’s awareness and knowledge about the politicalsystem and its institutions. In this regard, many scholars have argued that individ-uals with high levels of education, interest in politics, and political activism havehigh levels of trust in institutions.9

Contingent effect of political awareness: the mechanism and expectations

Public support for political institutions can take two different forms. While specificsupport refers to citizens’ satisfaction with particular policy outputs, diffusesupport refers to more general support for an institution divorced from its specificpolicy outputs.10 Although he acknowledges that it is difficult to identify underwhat conditions specific support converts to diffuse support over time, in his influ-ential study David Easton argues that confidence in institutions is a form of diffusesupport and reflects a general positive orientation towards the given set of insti-tutions.11 On the other hand, focusing on public confidence in the judicialsystem, Gibson, Caldeira and Spence argue that confidence in judiciary reflectsa blend of short term and long term judgments about the institution.12 Nevertheless,the formation of these judgments is far from simple.

Attitudes are not objects recalled from memory but they are constructed on thespot when an attitude needs to be revealed.13 During this process, the mind samplesfrom a plethora of negative and positive considerations related to the object. Yet,receiving new information about a memory object almost never influences thememory object in isolation but all memory objects that are connected to the originaltarget are influenced.14 Similarly, a considerable number of scholars who focus oninstitutional trust acknowledge that citizens have difficulty in distinguishingbetween political institutions.15 Thus, even when individuals do not receive or reg-ister specific information on the judiciary their evaluation of the judiciary are tied tothe cues they receive about the rest of the political system that is related to the judi-ciary. To be more specific, most of the relevant considerations that are sampled forattitude construction will be about the political system in general rather thanspecific news about the courts. For this reason, the effect of political awarenesson confidence in the judiciary should be modelled contingent on a country’ssystem performance (that is, how well the democratic system works).

The literature on judicial trust that mostly focuses on the analyses of well-estab-lished democracies suggests that those who know more about law and courts aremore likely to support the judiciary and believe in its legitimacy. In other words,well informed citizens, who are in a better position to understand the system, arestated to appreciate the judiciary more.16 For instance, Gibson, Caldeira andBaird argue that the individuals who are ignorant about politics and know littleabout courts do not believe that judges make decisions differently from other poli-ticians. This aspect is stated to jeopardize the legitimacy of the courts in those

4 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

people’s eyes.17 On the other hand, the individuals with greater awareness areexpected to differentiate judges from politicians and believe that they rely onlaw and not on values in making decisions.18 As such, the citizens’ tendency toview the court as protector of the constitution and champion of justices and civilliberties is argued to increase the citizens’ confidence in the judiciary.19

Although the logic of this argument is quite appealing, it assumes that in allcountries the judicial system functions properly. But what happens in countrieswhere the judiciary is not completely independent from political constraints andcannot efficiently act as the protector of the rule of law? Do high levels of individ-ual-level awareness still generate high levels of judicial confidence? Focusing onLatin American countries, for instance, Salzman and Ramsey argue that individ-uals with more political knowledge will recognize the inadequacies of their judici-aries and have lower levels of confidence in the judiciary compared to theindividuals who have low levels of political knowledge.20 Although the findingsof this study challenge the existing literature, a more comprehensive cross-nationalresearch in which the impact of political awareness on judicial confidence can betested across advanced and developing democracies will provide a more accurateanswer for the above-stated questions.

In research that examines whether and if so how the effect of education on insti-tutional trust varies cross-nationally, Hakverdian and Mayne suggest that educatedpeople will be better able to accurately identify practices that undermine the func-tioning of democratic institutions and will also be normatively troubled by suchpractices.21 As such, their empirical findings show that in countries with highlevels of corruption more educated citizens express lower institutional trust thanthe less educated; and in countries with low levels of corruption, more educatedcitizens have higher institutional trust than the less educated.22 Based on the theor-etical framework of this study we argue that the impact of political awareness onjudicial confidence will also change across countries with different levels ofdemocracy.

First, one should note that in developing democracies the issues covered andviews expressed in the media are heavily shaped by the power holders who canexercise influence on what is reported.23 In these types of regimes a judicialsystem that is under strong political influence – directly or through more subtleincentives and intimidation – might be portrayed by the ruling elite as highly inde-pendent. Yet, individuals with high levels of political awareness about the pro-blems related to the political system will be less susceptible to such rosydepictions of the judiciary. Being aware of the political pressure on media, forinstance, individuals with high levels of education and interest in politics willalso be able to follow international media channels to get objective informationabout the developments in their own country.

Second, while evaluating the performance and quality of political institutionscitizens do not only take into consideration short and medium term performanceevaluations but also long term political legacies and ideal structure of the politicalinstitutions.24 More knowledgeable citizens would know the ideal role and

Democratization 5

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

function of the courts in a democratic regime and would be able to evaluate theirown courts in comparison with the ones in advanced democracies. At the end ofthis comparison these citizens will realize that their country’s political system ingeneral and the legal system in particular is not living up to the standards ofadvanced democracies.

As a result, we argue that increased understanding of the political system and itsinstitutions leads to higher levels of confidence only when the system functions asit is supposed to. When there are chronic problems about the political system, highlevels of education and interest in politics allow individuals to see how bad thingsactually are and might lead to a cynical response. In other words:

Hypothesis: While in advanced democracies, political awareness has positiveimpact on confidence in the judiciary, in countries where the democratic systemhas chronic problems political awareness will have negative impact on public con-fidence in the judiciary.

Model, data and variables

In the literature, many different aspects are presented as possible determinants ofpublic confidence in the judiciary. These aspects range from individual-level deter-minants such as inter-personal trust, political ideology and political awareness tocountry-level determinants such as judicial performance. By focusing on theimpact of political awareness and controlling for other possible determinants, inthis section we develop and test a set of empirical models that aim to showwhether the impact of political awareness changes across advanced and developingdemocracies. In order to test whether confidence in judiciary is a function of theinteraction of political awareness and level of democracy, we use the WorldValues Survey (WVS) data (2005–2009) for 49 countries and combine thisdataset with several country level indicators compiled from different sources(see Appendix 1 for the countries included in the sample).25

Our dependent variable is confidence in judiciary and we use the WVS data tomeasure it. The WVS asks respondents a battery of questions about how much con-fidence they have in several institutions and other groups. The WVS uses a fourcategory question to measure how confident the respondents are in the judiciary.In the survey question, individuals are asked to rate their confidence in judiciaryon a scale that ranges from 1 ¼ “a great deal of confidence”, to 4 ¼ “no confidenceat all”. When we look at the percentages of survey respondents who have high con-fidence in the judiciary we see a considerable difference across 49 countries wherethe judicial system item was included in the survey. While in some countries only8% of the survey respondents had high confidence in the judiciary, in others thisnumber increased to 86%. This rough query on the data shows that the cross-country variations in confidence in judiciary might be systematic and can be mod-elled (see Appendix 1).

Although the standard question used in the WVS to measure confidence in thejudiciary is a four category ordinal item, we cannot take the item as it is and use it in

6 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

an ordered probit (logit) model. There are two main interrelated reasons. First,especially in a battery of questions asked in quick succession, the confidence ques-tions reveal information about two latent traits. The responses contain both ageneral propensity to be confident in an institution and a specific indication of con-fidence in the specific institution the item inquires about. Having answered severalconfidence questions, interviewees tend to be primed. Therefore the confidence inthe judiciary question, on its face value, may not be a clean measure of confidencein the judiciary. More importantly however, several attitudinal variables we areinterested in such as life satisfaction, democratic attitudes, and political interestare variables associated with confidence in institutions in general as well as confi-dence in the judiciary. Thus we need a measure that is relatively clean from thespurious relationships created by its correlation with other confidence items.

Second, in the literature it has been stated that individuals cannot easily differ-entiate the judiciary from other political institutions.26 As such, it has been arguedthat confidence questions measure an overall satisfaction with the whole systemand not specifically the confidence in one institution.27 For this reason, we needto find a way to clean the confidence in the judiciary item as much as we can.One approach could be to create a general confidence propensity measure by com-bining the information from the remaining confidence items and use that propen-sity measure as an independent variable in the model. However, because we alsohave several items that could predict the general confidence propensity as predic-tors in the main model as well, this approach would cause as many problems as itsolves. Causal relationship between the independent variables in the model wouldcause complex multicollinearity problems. Another approach would be to separatethe effects of other confidence items on the variable of interest. In order to do this,we regressed the justice item on three institutions (confidence in government,parties, and parliament) and obtained the model residuals. The resulting regressionexplained about 55% of the variation in the confidence in the judiciary. Theresiduals we obtained from the model are then used as the dependent variable inthe models we develop and test later. As a result, our dependent variable variesroughly in the range [225, 25].28

The first key independent variable of this study is political awareness.Although in the WVS there are no items that even indirectly measure the levelof knowledge about the judiciary or the political system in general, an individual’slevel of education, interest in politics and participation in politics can be used asindicators of political awareness (see Appendix 2 for the measurement of thesevariables). Arguing that education increases citizen’s knowledge about how thepolitical institutions work, many studies showed that educated people are morelikely to respond correctly to questions related to the key players and workingsof their political system.29 The amount and type of attention citizens pay to politicsand public affairs has been presented as another aspect to improve an individuals’ability to understand the political events.30 For instance, it has been argued thathigh levels of interest in politics will direct a person to increased news mediause that would increase people’s awareness of political processes and working

Democratization 7

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

of the political institutions including the judiciary.31 As such it has been recognizedthat court decisions on controversial issues do gain public notice especially amongthe individuals who are politically attentive.32 Repeated participation in variouspolitical processes such as signing petitions, attending protests/demonstrations,membership of civic organization, working for a political party, attending electionrallies or community meetings provides another opportunity for citizens to learnabout the performance of political institutions.33 Accordingly, it has been shownthat individuals who frequently participate in the political system display higherlevels of confidence in the judiciary.34

Arguing that the effect of political awareness on confidence in the judiciaryshould be modelled contingent on a country’s level of democracy, our secondkey independent variable is level of democracy. “Level of democracy” refers toa rough categorization of regimes that ranges from “advanced democracies”where democratic values are fully consolidated, political processes successfullyinstitutionalized and civil rights and liberties efficiently protected through “devel-oping democracies” that meet the procedural minima for democracy but lack con-solidation of democratic values, constitutional liberalism and institutionalization ofpolitical processes35 to “non-democracies”.

In our analysis we created “level of democracy” dummies based on theFreedom House (FH) categorization for the year 2006. FH categorizes all countriesin the world according to the institutionalization and protection of the politicalrights and civil liberties of citizens. Since the protection of civil rights and libertiesis one of the key aspects that differentiate advanced democracies from developingdemocracies36 and political rights differentiate democracies from non-democraticregimes, FH ratings provide a good summarized answer to the question of “Howwell does the political system function in one country?”. Each pair of politicalrights and civil liberties ratings is averaged to determine an overall status of“Free”, “Partly-Free”, or “Not-Free”. The countries whose ratings averagebetween 1.0 and 2.5 are named as “Free” and refer to advanced democracies.The countries whose ratings are between 3.0 and 5.0 are named as “Partly-Free”and refer to developing democracies. Finally, the countries whose FH scoreranges between 5.5 and 7.0 are named as “Not-Free” and refer to non-democraticregimes.

In our study we also expect that in societies where the performance and qualityof the judiciary is good, confidence in the judiciary will be higher even after con-trolling for other factors. We use Legal System and Property Rights Index tomeasure the overall performance and quality of the judicial system. This is an addi-tive index formulated by the country scores on judicial independence, impartialcourts, protection of property rights, legal enforcement contracts, and integrityof the legal system. The scores of this additive index are taken from EconomicFreedom of the World report published in 2006. The mean centred version ofthis variable is used in the analysis and it varies in the range [23, 3].

In the empirical model we also explore the robustness of our estimates to theinclusion of other individual-level variables that are presented in the literature as

8 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

potential determinants of confidence in political institutions. In this regard wecontrol for interpersonal trust, importance of democracy, religiosity, satisfactionwith life, ideological positioning and a standard set of demographic variables(see Appendix 2 for the measurement of these variables). For instance, in theliterature an important strand of research argues that the citizens with higherinter-personal trust tend to display more confidence in democratic institutions.37

The literature on public trust in the judiciary largely agrees with the broader lit-erature on institutional trust and asserts that individuals with higher commit-ments to democratic values have higher confidence in the judiciary.38 In theprevious studies on institutional trust, political ideology is also observed tohave a significant impact on public confidence in institutions.39 While individ-uals who situate themselves on the far left of the ideological scale appear to bemore distrustful of political institutions, individuals who situate themselves onthe far right are stated to have higher levels of confidence in political insti-tutions.40 As such, other things being equal, confidence in courts is expectedto be higher and more widespread among people who position themselves onthe right of the political scale. Another group of scholars assert that personallife satisfaction has positive impact on institutional trust. The idea is thatthere exists a positive spillover from personal life satisfaction to institutionssuch that dissatisfied individuals become less confident in political insti-tutions.41 Existing studies also show that women and elderly people developmore trust in institutions.42 In this regard we suggest that personal life satisfac-tion, age and gender will significantly affect individual’s confidence in thejudiciary.

Empirical analysis and discussion

In order to show the impact of individual and country level determinants on judicialconfidence and to delineate whether and to what extent the impact of politicalawareness on judicial confidence changes in line with a country’s level of democ-racy, we estimate three hierarchical models (see Table 1). In these models, thedependent variable is confidence in the judiciary and given that the variableranges between [225, 25] we use linear multilevel regression analysis.

Model 1 presented in Table 1 is a random intercept model and Model 2 is arandom coefficient model. In both models we include the key variables (judicialperformance, level of democracy and political awareness indicators) and controlfor inter-personal trust, religiosity, life satisfaction, importance of democracy, ideo-logical positioning, age and gender. In both models the judicial performance andquality index has a significant effect on confidence in the hypothesized direction.In Model 1, the effect of judicial performance and quality is estimated to be around1.91. The effect is not only statistically significant but also substantive. This meansthat when we compare two hypothetical countries with the minimum andmaximum values of judicial performance and quality, we see that the average con-fidence (country mean) of the country with the best judicial performance and

Democratization 9

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Table 1. Multilevel analysis of confidence in judiciary.

Model 1 Model 2 Model 3

Random-Intercept Model Random-Coefficient Model Random-Coefficient Model

Fixed Effects estimates s.e estimates s.e estimates s.e

Key VariablesJudicial Performance & Quality 1.908∗∗∗ (0.443) 1.833∗∗∗ (0.403) 1.815∗∗∗ (0.407)Political AwarenessPolitical Participation 0.058∗∗ (0.025) 0.066 (0.051) 0.082∗ (0.049)Education 0.023∗∗∗ (0.006) 0.017∗ (0.009) 0.018∗ (0.009)Political Interest 0.124∗∗ (0.045) 0.160 (0.077) 0.160 (0.077)Level of DemocracyPartly-Free 1.181 (0.750) 1.631∗∗ (0.693) 1.626∗∗ (0.762)Not-Free 0.177 (1.815) 24.998∗∗ (1.787) 22.091 (1.908)(Reference:Free)

Cross-Level InteractionsEducationXPartly-Free 0.013 (0.021)Political InterestXPartly-Free 20.428∗∗ (0.161)ParticipationXPartly-Free 20.177∗ (0.103)EducationXNot-Free 20.157∗∗ (0.052)ParticipationXNot-Free 20.929∗∗∗ (0.282)Political InterestXNot-Free 20.722∗ (0.398)

ControlsInter-personal Trust 0.490∗∗∗ (0.094) 0.465∗∗∗ (0.094) 0.475∗∗∗ (0.094)Religiosity 0.025 (0.017) 0.029∗ (0.017) 0.029∗ (0.017)Life Satisfaction 0.126∗∗∗ (0.019) 0.127∗∗∗ (0.020) 0.126∗∗∗ (0.020)Importance of Democracy 0.046∗∗ (0.022) 0.041∗ (0.022) 0.042∗ (0.022)Ideological Positioning 20.004 (0.018) 0.002 (0.018) 0.001 (0.018)

(Continued )

10A

.A

ydınC

akırand

E.

Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Table 1. Continued.

Model 1 Model 2 Model 3

Random-Intercept Model Random-Coefficient Model Random-Coefficient Model

Fixed Effects estimates s.e estimates s.e estimates s.e

Age 20.020∗∗∗ (0.003) 20.020∗∗∗ (0.003) 20.020∗∗∗ (0.003)Gender 0.073 (0.079) 0.062 (0.079) 0.062 (0.079)Constant 0.385 (0.387) 0.305 (0.396) 0.21 (0.395)

Random EffectsCountry Level (d2) 1.648∗∗∗ (0.192) 1.823∗∗∗ (0.250) 1.731∗∗∗ (0.211)Individual Level (d2) 6.867∗∗∗ (0.027) 6.839∗∗∗ (0.027) 6.839∗∗∗ (0.027)Participation 0.269∗∗∗ (0.042) 0.209∗∗∗ (0.039)Education 0.040∗∗∗ (0.010) 0.031∗∗∗ (0.010)Political interest 0.374∗∗∗ (0.070) 0.297∗∗∗ (0.070)22 X log Likelihood 2106,407.93 2106,333.43 2106,319.73N(number of groups) 31,783 (49) 31,783 (49) 31,783 (49)

Notes: Table entries are maximum likelihood estimates with estimated standard errors in parentheses. Centred explanatory variables are used. s.e, standard error.∗¼p , .10.∗∗¼p , .05.∗∗∗¼p , .01.

Dem

ocratization11

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

quality will be about 11.5 points higher than the country with the worst judicial per-formance and quality.

Since judicial performance and quality variable is measured at country-levelwe need to be careful about the inferences we make based on these results. Weneed to be guarded against the slippery slope of ecological fallacy. We do notnecessarily argue that individuals actually observe judicial performance andquality as the experts do and form a well-informed mental map of the system.However, the effects of the performance of the system trickle down to the livesof individuals affecting the way they evaluate their social and political environ-ment. Thus even in the absence of specific knowledge regarding the performanceof the judicial system, individual evaluations are expected to reflect the effects ofsystem performance on evaluations.

Obviously a more direct measurement of how individuals evaluate the systemperformance is preferable under most circumstances. However, even if we had areliable individual-level measurement of how individuals evaluated the perform-ance and quality of the judiciary we might still want to control the effects ofactual performance and quality which influence different aspects of the respon-dents’ lives. Respondents may even not attribute the freedoms they enjoy tospecific institutions. Therefore both conscious and informed evaluations of thejudicial performance and quality and the subtle effects of judicial quality absorbedthroughout the polity might have discernible and separate effects on confidence ininstitutions including the judiciary.

Regarding the impact of level of democracy on public confidence in the judi-ciary our empirical findings are quite interesting. When country specific slopes aresuppressed to be equal (Model 1), democracy status does not seem to have any sig-nificant effect on confidence. However when we allow random slopes for variablesof political awareness across countries (Model 2), the level of democracy seems tohave a significant impact on public confidence in the judiciary. The base category is“Free” countries. Therefore we compare “Partly-Free” and “Not-Free” countries to“Free” countries as categorized by the level of democracy variable. An interestingresult is that “Partly-Free” countries seem to have higher average confidence in thejudiciary than “Free” countries. The standard interpretation of the coefficient for“Partly-Free” is that on average, keeping everything else constant, “Partly-Free”countries have a higher (1.63 points) confidence in the judiciary than “Free”countries.

Given that in countries with poor democratic performance media is largelyinfluenced by the government, level of transparency in these countries would bequite low. In the literature it has been shown that low media freedom is stronglyassociated with poor political knowledge.43 As such, in these types of countriespublic awareness about the incumbent politicians’ wrongdoings would be quitelimited. Yet, not only politically controlled media but also low levels of education,low levels of public interest in politics and extremely low levels of social and pol-itical participation44 might explain the low levels of political awareness in devel-oping democracies and the citizens’ susceptibility to propaganda. For all these

12 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

reasons, it is no surprise to find that in developing democracies people have higherconfidence in judiciary compared to the individuals living in advanceddemocracies.

Finally, looking at the impact of control variables on judicial confidence wefind that what has been previously discussed in the literature is supported onceagain. For instance, we find that interpersonal trust and life satisfaction signifi-cantly increase an individual’s confidence in the judiciary. We also find thatolder respondents confess less confidence in the judiciary. The effect is significantbut substantially moderate at best. A 10 year age differential is estimated todecrease confidence by about 2 points. Finally, we observe that self-declaredimportance given to democratic government has a positive but marginal effecton confidence.

The varying impact of political awareness

Having shown that some of the most widely discussed effects on confidence alsoholds for our data, we can move towards a more nuanced investigation. Ourhypothesis states the expectations on the effects of political awareness on confi-dence and how these effects are moderated by the socio-political context. In thisregard, first we investigate whether the argument which states that political aware-ness causes more support for the judiciary survives a comparative cross-culturalscrutiny. The existing studies that have focused on advanced democracies toexplain public confidence in the judiciary have shown that political awarenesstends to bolster support for the judiciary.45 The mechanism of this argumentsuggests that awareness leads to a better understanding of the system and especiallywhen judicial processes are concerned awareness helps people to better evaluatethe system. In other words, in a political system that functions well, the betterinformed individuals are stated to trust and support the judiciary more than theless informed. However, should we expect a uniform effect across societies?Should the effect of political awareness be same in all countries? Or should weexpect systematic variations in the effectiveness of political awareness in increas-ing confidence in the judiciary?

There is sufficient reason to believe that when political awareness is coupledwith a political system with chronic problems, a cynical response is possible.Especially when a system fails to deliver what it promises on paper, the better edu-cated, informed and more interested people are in a better position to see the systemfailure. For example, when a political system fails to provide and protect individualrights and freedoms we should expect the better informed individuals to see thesituation as it is and reflect that disillusionment in their evaluations of thesystem including the judicial processes. In short, we might expect to see stronginteractive effects between political awareness and level of democracy.

The random coefficient model (Model 2) confirms the initial expectations thatpolitical awareness has a positive impact on judicial independence. To be morespecific, of the three political awareness items education has a significant, political

Democratization 13

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

interest has a marginally significant and political participation has an insignificanteffect on confidence. When we turn to the random part of the model we observe thatrandom effects parameters are significant across the board. Both the random inter-cept and random coefficients for education, political interest and political partici-pation are significant, implying that country specific slopes vary significantly. Inorder to investigate if the random variation in the slopes has any substantivebearing on our inference we construct confidence intervals around the randomslopes. The slope indicating the effect of political participation varies betweenabout 20.45 and 0.59. Similar ranges are found for the effects of political interestand education. In Table 2 we report the random coefficient ranges for randomeffects.

The interesting finding in support of our argument is that all three awarenessvariables are expected to have quite a wide range of values across the 49 countriesthat ended up in our analysis. It seems that all three variables of awareness couldhave positive and negative effects depending on the society where they areobserved. Now having shown that the effects of awareness are not uniformacross countries, we turn our attention to explaining the variation across nations.

As our hypothesis suggests we expect to find that effects of education, politicalinterest and political participation to be moderated by the democratic performanceof the regime. A rough and less rigorous test of our hypothesis is illustrated inFigure 1. For all countries in our sample, confidence in the judiciary is regressedsubsequently on political awareness indicators. The three charts plot the slope ofeach awareness factor against FH scores. Observations below the dotted line arecountries where the relationship between awareness and confidence are negative.In all three panels of the figure the fitted lines are sloping downwards implyingthat as FH scores increase (that is, less democratic), the regression coefficientsof political awareness indicators decrease.

For a more rigorous test, we specify cross-level interactions between the aware-ness variables and two dummy variables indicating the level of democracy. We for-mulate an interaction model where the political awareness indicators (education,interest in politics and political participation) interact with democracy level. The“Free” category is the baseline and Model 3 reports the results. While the obser-vations we reported before still hold, the interaction between the level of democ-racy and political awareness variables are at least marginally significant and inthe expected direction for all six interactions specified by the model. It appears

Table 2. Range (%95) of random coefficients.

Min Max

Political Participation 20.45979 0.59273Political Interest 20.57206 0.89206Education 20.0614 0.0954Constant 23.224 3.832

14 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

that compared to “Free” countries, in “Partly-Free” countries the effects of allthree political awareness variables are dampened. The same observation holdsfor “Not-Free” and “Free” comparison as well. The remaining important questionis whether the interactive effects are strong enough to change the sign of the overalleffect.

Figure 2 illustrates the moderated effects of our awareness variables. As wecan see, in all three cases the distinction between “Free” and “Not-Free” societiesis sharp and clear. When we compare “Partly-Free” countries to “Free” countrieswe observe that the slopes for participation and political interest actually cross.At lower levels of participation and interest it seems that residents of “Free”countries would be unable to give due credit to the system while in “Partly-Free” societies the relatively less informed would fail to see system failure andremain unduly confident. Regarding confidence in the judiciary, it appears thatwhile ignorance is bliss in “Partly-Free” societies, it actually is a curse in well-established democracies.

As a result, our empirical findings show that the impact of political awarenessis not uniform across countries. While low levels of political awareness lead tohigh levels of judicial confidence in developing democracies, low levels ofpolitical awareness lead to low levels of judicial confidence in advanced democra-cies. As the level of political awareness increases, however, judicial confidencelevels increase in advanced democracies whereas it falls in developingdemocracies.

Figure 1. The effects of political awareness conditional on level of democracy.

Democratization 15

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Figure 2. Moderated effects of political awareness on confidence in judiciary across differentgroups of countries.

16 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Conclusion

Public confidence in institutions is an important precondition for the consolidationof democracy and it is generally associated with how the public perceives theseinstitutions to be working. In this regard, the existing literature suggests thatthose who know more about law and courts are more likely to support the judiciaryand believe in its legitimacy. But what happens if the judicial system has certainproblems and is not functioning well?

In this paper we have tried to show that a systematic study of cross-culturalvariation in confidence in the judiciary is not only possible but quite informative.In line with the previous findings in the literature, our empirical findings showedthat performance and quality of the judiciary has a positive impact on public con-fidence in the judiciary. Yet, regarding the impact of level of democracy on publicconfidence in the judiciary, we find that in advanced democracies people tend tohave lower confidence in the judiciary compared to their counterparts in develop-ing democracies. In countries with low levels of democratic performance it is quitecommon to observe low levels of transparency, strong government pressure overthe media, and low levels of public awareness. For that reason, it is of no surpriseto observe high levels of public confidence in these types of polities.

Regarding the interaction between individual and country level determinantswe find that while in advanced democracies people with low levels of politicalawareness have low levels of confidence in the judiciary, in countries withcertain democracy problems less informed people would fail to see the deficienciesof the political system and remain unduly confident in the judiciary. On the otherhand, in countries with poor democratic performance, the more informed, inter-ested and educated people seem to be in a better position to see the systemfailure and this is well reflected in their low levels of confidence in the judiciary.

As a result, while we reiterated some of the earlier results and extended theresults established for the United States and some Latin American countries to awider range of societies we also demonstrated that the cross-cultural variationshave a systematic component as well. For that reason, the results we report inthis paper suggest that a one-size-fits-all approach to explain confidence in the judi-ciary is not possible when we are dealing with a wide range of societies that vary interms of many characteristics, both institutional and cultural.

As much as we are satisfied with the results we report here, we must acknowl-edge the shortcomings of the data and ensuing analyses. Many of our variables arenecessarily indirect measures and proxies. Although we are confident that noserious spurious relationships are rampant in the dataset, our results lead to conser-vative inferences. We believe that a data collection dedicated to cross-cultural judi-cial study would yield much more reliable data and stronger results. Finally, webelieve that the generic version of the main arguments of this study – thateffects of political awareness are moderated by the level of democracy – could(and should) be extended to cross-cultural studies dealing with institutions otherthan the judiciary.

Democratization 17

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Notes1. Staton, “Constitutional Review and Selective Promotion of Case Results”; Vanberg,

The Politics of Constitutional Review in Germany.2. Gibson et al., “On the Legitimacy of National High Courts”; Hibbing and Theiss-

Morse, Congress as Public Enemy; Casey, “The Supreme Court and Myth”;Benesh, “Understanding Public Confidence in American Courts.”

3. See Tanenhaus and Murphy, “Patterns of Public Support for the Supreme Court”;Adamany and Grossman, “Support for the Supreme Court”; Mondak and Smithey,“The Dynamics of Public Support for the Supreme Court”; Caldeira and Gibson,“The Etiology of Public Support for the Supreme Court”; Gibson et al., “MeasuringAttitudes toward the United States Supreme Court.”

4. Salzman and Ramsey, “Judging the Judiciary.”5. See Citrin and Green, “Presidential Leadership and Trust in Government”; Hethering-

ton, “The Political Relevance of Political Trust”; Mishler and Rose, “What are theOrigins of Political Trust?”

6. Buhlmann and Kunz, “Confidence in the Judiciary”, 334.7. Benesh, “Understanding Public Confidence in American Courts.”8. Zaller, The Nature and Origins of Mass Opinion.9. Galston, “Political Knowledge, Political Engagement, and Civic Education”; Finkel

et al., “Civic Education, Civil Society, and Political Mistrust.”10. Easton, “A Re-assessment of the Concept of Political Support.”11. Ibid.12. Gibson et al., “Measuring Attitudes toward the United States Supreme Court.”13. Zaller and Feldman, “A Simple Theory of the Survey Response.”14. Anderson, “A Spreading Activation Theory of Memory”; Anderson, Cognitive Psy-

chology and its Implications; Roelofs, “A Spreading-Activation Theory”; McNamaraand Diwadkar, “The Context of Memory Retrieval.”

15. Gibson, “Challenges to the Impartiality of State Supreme Courts”; Norris, “Insti-tutional Explanations for Political Support”; Caldeira, “Neither the Purse Nor theSword.”

16. See Note 2.17. Gibson et al., “On the Legitimacy of National High Courts”, 343.18. Ibid., 34519. Mondak and Smithey, “The Dynamics of Public Support for the Supreme Court”, 131.20. Salzman and Ramsey, “Judging the Judiciary.”21. Hakverdian and Mayne, “Institutional Trust, Education and Corruption”, 742.22. Ibid.23. Zaller, The Nature and Origins of Mass Opinion.24. Mattes and Bratton, “Learning about Democracy in Africa.”25. The 2005–2009 wave of the WVS data includes information for 58 countries but since

the necessary individual and country level variables were available for 49 of thosecountries, our sample is composed of individuals from 49 countries.

26. Gibson, “Challenges to the Impartiality of State Supreme Courts.”27. Caldeira, “Neither the Purse Nor the Sword”; Norris, “Institutional Explanations for

Political Support.”28. The 50 point range for rescaling is largely a cosmetic choice. Due to relatively low

coefficients a 0–1 scale on the dependent variable would have meant showing toomany zeroes on regression tables.

29. Hakverdian and Mayne, “Institutional Trust, Education and Corruption”, 741.30. Ibid.31. Mattes and Bratton, “Learning about Democracy in Africa.”

18 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

32. Casey, “The Supreme Court and Myth”; Adamany and Grossman, “Support for theSupreme Court.”

33. Hakverdian and Mayne, “Institutional Trust, Education and Corruption”; Mattes andBratton, “Learning about Democracy in Africa.”

34. See Caldeira, “Neither the Purse Nor the Sword”; Tanenhaus and Murphy, “Patterns ofPublic Support for the Supreme Court”; Adamany and Grossman, “Support for theSupreme Court”; Caldeira and Gibson, “The Etiology of Public Support for theSupreme Court”; Gibson et al., “On the Legitimacy of National High Courts.”

35. Zakaria, “The Rise of Illiberal Democracy.”36. Zakaria, “The Rise of Illiberal Democracy”; Levitsky and Way “Assessing the Quality

of Democracy.”37. Bjornskov, “Determinants of Generalized Trust”; Zmerli and Newton, “Social Trust

and Attitudes toward Democracy”; Mishler and Rose, “What are the Origins of Pol-itical Trust?”

38. Caldeira and Gibson, “The Etiology of Public Support for the Supreme Court”;Mondak and Smithey, “The Dynamics of Public Support for the Supreme Court.”

39. Miller and Listhaug, “Political Performance and Institutional Trust”; Listhaug andWiberg, “Confidence in Political and Private Institutions.”

40. Ibid.41. Listhaug and Wiberg, “Confidence in Political and Private Institutions.”42. Norris, “Institutional Explanations for Political Support”; Mishler and Rose, “What

are the Origins of Political Trust?”; Rohrschneider and Schmitt-Beck, “Trust in Demo-cratic Institutions in Germany”; Citrin and Green, “Presidential Leadership and Trustin Government.”

43. Leeson, “Media Freedom, Political Knowledge, and Participation.”44. Finkel, “Civic Education and the Mobilization of Political Participation.”45. See Note 2.

Notes on contributorsAylin Aydın Cakır is an Assistant Professor of Political Science and International Relationsat Yeditepe University in Turkey. Her research interests include judicial politics and com-parative political institutions. She is the author of Converging Dialogues: Interactive Con-flict Resolution and Turkey’s Headscarf Conflict (VDM Verlag, 2011). She has been alsopublished in International Political Science Review, Law & Society Review and PoliticalResearch Quarterly.

Eser Sekercioglu is a visiting professor at Izmir University of Economics in Turkey andowner and director of Data-ist Data Solutions. His main areas of interest are political meth-odology, voting behaviour, and game theory. He has been published in Turkish Studies.

ReferencesAdamany, David W., and Joel B. Grossman. “Support for the Supreme Court as a National

Policymaker.” Law & Policy Quarterly 5 (1983): 405–437.Anderson, John R. “A Spreading Activation Theory of Memory.” Journal of Verbal

Learning and Verbal Behavior 22 (1983): 261–295.Anderson, John R. Cognitive Psychology and its Implications. 4th ed. New York:

W. H. Freeman, 1995.Benesh, Sara C. “Understanding Public Confidence in American Courts.” The Journal of

Politics 68, no. 3 (2006): 697–707.

Democratization 19

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Bjornskov, Christian. “Determinants of Generalized Trust: A Cross-country Comparison.”Public Choice 130, no. 1 (2006): 1–21.

Buhlmann, Marc, and Kunz Ruth. “Confidence in the Judiciary: Comparing theIndependence and Legitimacy of Judicial Systems.” West European Politics 34, no. 2(2011): 317–345.

Caldeira, Gregory A. “Neither the Purse Nor the Sword: Dynamics of Public Confidence inthe U.S. Supreme Court.” American Political Science Review 80 (1986): 1209–1226.

Caldeira, Gregory A., and James L. Gibson. “The Etiology of Public Support for theSupreme Court.” American Journal of Political Science 36 (1992): 635–664.

Casey, Gregory. “The Supreme Court and Myth: An Empirical Investigation.” Law &Society Review 8 (1974): 385–419.

Citrin, Jack, and Donald Philip Green. “Presidential Leadership and the Resurgence of Trustin Government.” British Journal of Political Science 16, no. 4 (1986): 431–453.

Easton, David. “A Re-assessment of the Concept of Political Support.” British Journal ofPolitical Science 5, no. 4 (1975): 435–457.

Finkel, Steven. “Civic Education and the Mobilization of Political Participation inDeveloping Democracies.” Paper presented at Political Participation: Building aResearch Agenda Conference in Princeton University, New Jersey, October 12–14,2000.

Finkel, Steve E., Christopher A. Sabatini, and Gwendolyn G. Bevis. “Civic Education, CivilSociety, and Political Mistrust in a Developing Democracy: The Case of the DominicanRepublic.” World Development 28, no. 11 (2000): 1851–1874.

Galston, William A. “Political Knowledge, Political Engagement, and Civic Education.”Annual Review of Political Science 4 (2001): 217–234.

Gibson, James L. “Challenges to the Impartiality of State Supreme Courts: LegitimacyTheory and “New-Style” Judicial Campaigns.” American Political Science Review102, no. 1 (2008): 59–75.

Gibson, James L., Gregory A. Caldeira, and Vanessa A. Baird. “On the Legitimacy ofNational High Courts.” American Political Science Review 92 (1998): 343–358.

Gibson, James L., Gregory A. Caldeira, and Lester Kenyatta Spence. “Measuring Attitudestoward the United States Supreme Court.” American Journal of Political Science 47, no.2 (2003): 354–367.

Hakverdian, Armen, and Quinton Mayne. “Institutional Trust, Education and Corruption: AMicro-Macro Interactive Approach.” The Journal of Politics 74, no. 3 (2012): 739–750.

Hetherington, Marc J. “The Political Relevance of Political Trust.” The American PoliticalScience Review 92, no. 4 (1998): 791–808.

Hibbing, John R., and Elizabeth Theiss-Morse. Congress as Public Enemy: Public AttitudesToward American Political Institutions. Cambridge: Cambridge University Press, 1995.

Leeson, Peter T. “Media Freedom, Political Knowledge, and Participation.” Journal ofEconomic Perspectives 22, no. 2 (2008): 155–169.

Levitsky, Steven, and Lucan Way. “Assessing the Quality of Democracy.” Journal ofDemocracy 13, no. 2 (2002): 51–65.

Listhaug, Ola, and Matti Wiberg. “Confidence in Political and Private Institutions.” InCitizens and the State, edited by Hans-Dieter Klingemann and Dieter Fuchs. Oxford:Oxford University Press, 1995.

Mattes, Robert, and Michael Bratton. “Learning about Democracy in Africa: Awareness,Performance, and Experience.” American Journal of Political Science 51, no. 1(2007): 192–217.

McNamara, Timpothy P., and Vaibhav A. Diwadkar. “The Context of Memory Retrieval.”Journal of Memory and Language 35 (1996): 877–892.

20 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Miller, Arthur H., and Ola Listhaug. “Political Performance and Institutional Trust.” InCritical Citizens: Global Support for Democratic Governance, edited by PippaNorris, 204–216. Oxford: Oxford University Press, 1999.

Mishler, William, and Richard Rose. “What are the Origins of Political Trust? TestingInstitutional and Cultural Theories in Post-communist Societies.” ComparativePolitical Studies 34, no. 1 (2001): 30–59.

Mondak, Jeffrey J., and Shannon Ishiyama Smithey. “The Dynamics of Public Support forthe Supreme Court.” Journal of Politics 59 (1997): 1114–1142.

Norris, Pippa. “Institutional Explanations for Political Support.” In Critical Citizens: GlobalSupport for Democratic Governance, edited by Pippa Norris, 217–235. Oxford: OxfordUniversity Press, 1999.

Roelofs, Ardi. “A Spreading-Activation Theory of Lemma Retrieval in Speaking.”Cognition, 42 (1992): 107–142.

Rohrschneider, Robert, and Rudiger Schmitt-Beck. “Trust in Democratic Institutions inGermany: Theory and Evidence Ten Years After Unification.” Latin German Politics11, no. 3 (2002): 35–58.

Salzman, Ryan, and Adam Ramsey. “Judging the Judiciary: Understanding PublicConfidence in Latin American Courts.” Latin American Politics and Society 55, no. 1(2013): 73–95.

Staton, Jeffrey K. “Constitutional Review and the Selective Promotion of Case Results.”American Journal of Political Science 50 (2006): 98–112.

Tanenhaus, Joseph, and Walter F. Murphy. “Patterns of Public Support for the SupremeCourt: A Panel Study.” Journal of Politics 43 (1981): 24–39.

Vanberg, George. The Politics of Constitutional Review in Germany. New York: CambridgeUniversity Press, 2005.

Zakaria, Fareed. “The Rise of Illiberal Democracy.” Foreign Affairs 76, no. 6 (1997): 22–43.

Zaller, John. The Nature and Origins of Mass Opinion. Cambridge: Cambridge UniversityPress, 1992a.

Zaller, John, and Stanley Feldman. “A Simple Theory of the Survey Response: AnsweringQuestions versus Revealing Preferences.” American Journal of Political Science 36, no.3 (1992b): 579–616.

Zmerli, Sonja, and Kenneth Newton. “Social Trust and Attitudes toward Democracy.”Public Opinion Quarterly 72, no. 4 (2008): 706–724.

Democratization 21

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

Appendix 1. Public confidence in the judicial system across 49 countries

CountryConfidence injudiciary (%) Country

Confidence injudiciary (%) Country

Confidence injudiciary (%)

Vietnam 86.7 Great Britain 57.6 Russia 36.1Jordan 86.4 Germany 55.6 Andorra 35.5Finland 81.8 United States 55 Bulgaria 35.1Malaysia 77.7 Spain 53.8 Colombia 34.1Japan 77 Australia 51 Trinidad &

Tobago31.9

China 76.2 South Korea 50.8 Slovenia 31.4Rwanda 76.1 Italy 50.1 Poland 30.9Switzerland 74.9 Indonesia 50 Ukraine 29.3Turkey 73.8 Brazil 49.3 Moldova 29.3Sweden 73 Zambia 48.4 Chile 28.3Thailand 71.5 New Zealand 44.3 Serbia 27.6Cyprus 69.3 Netherlands 43.4 Romania 26.8South Africa 64.7 Iran 42.4 Ethiopia 24.8Ghana 62.3 Burkina Faso 42.3 Argentina 19.8Morocco 61.6 France 39.9 Peru 8.1Mali 58.8 Taiwan 39India 58 Mexico 36.9

Notes: This is the list of 49 countries covered in the World Values Survey (WVS) for the years 2005–2009. The WVS uses a four category question to measure how confident the respondents are in thejudiciary. In the survey question the individuals are asked to rate their confidence in judiciary on ascale that ranges from 1 ¼ “a great deal of confidence”, 2 ¼ “quite a lot of confidence”, 3 ¼ “notvery much confidence to 4 ¼ “no confidence at all”. The numbers presented in the table arecalculated by taking the sum of percentage of the respondents who have “a great deal” or “quite alot” of confidence in the judicial system.

Appendix 2

Independent variables: measurement

Political awareness indicators

Education: The WVS asks respondents to specify at what age they have completed theireducation (v239). This is a continuous variable and the mean centred version of it is usedin the analysis.

Political interest: The WVS asks respondents to designate how much they are interested inpolitics (v95). This variable is originally coded using a scale that varies from 1 ¼ “veryinterested” to 4 ¼ “not interested at all”. We re-coded this variable so that the highervalues refer to more interest in politics (1 ¼ “not interested at all” and 4 ¼ “very inter-ested”). The mean centred version of this variable is used in the analysis.

Political participation: The WVS asks the respondents a battery of questions on whether ornot they would participate in a series of political action which include signing a petition(v96), joining boycotts (v97), and attending peaceful demonstrations (v98). Each variableis originally coded using a scale: 1 ¼ “have done” to 2 ¼ “might do” and 3 ¼ “neverdo”. An additive index of political action is formulated by summing up the responsesgiven to the three items. The index is recoded so that its value changes between 0 and 1

22 A. Aydın Cakır and E. Sekercioglu

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015

as the value of the index goes from the minimum political action to maximum politicalaction. The mean centred version of this variable is used in the analysis.

Control variables

Interpersonal trust: In the WVS individuals are asked whether they think most people can betrusted or one need to be very careful when dealing with people (v23). Originally the vari-able is coded as 1 ¼ “most people can be trusted” and 2 ¼ “can’t be trusted”. We recodedthe variable so that if the respondent is trustful the variable is coded as 1 and 0 otherwise. Themean centred version of this variable is used in the analysis.

Religiosity: In the WVS individuals are asked to answer “how important is god in theirlives”. This is used as a proxy to measure religiosity levels of individuals (v192) and it iscoded on a 10 point scale where 1 refers to “not at all important” and 10 refers to “veryimportant”. The mean centred version of this variable is used in the analysis.

Satisfaction with life: In order to measure this variable in the WVS individuals are asked toindicate how satisfied they are with their lives in general (v22). This variable is originallycoded on a scale that ranges from 1 ¼ “not satisfied” to 10 ¼ “very satisfied”. The meancentred version of this variable is used in the analysis.

Ideological positioning: The WVS question that measures this variable asks individuals tosituate themselves on a political ideology scale where 1 refers to “far left” and 10 refers to“far right” (v114). The mean centred version of this variable is used in the analysis.

Importance of democracy: The WVS question asks individuals to indicate “how important isit for them to live in a country that is governed democratically” (v162). This variable iscoded along a 10 point scale where 1 ¼ “not at all important” and 10 ¼ “absolutely impor-tant”. The mean centred version of this variable is used in the analysis.

Sex: This variable is originally coded as 1 ¼ “male” and 2 ¼ “female” (v235). We recodedthis variable so that 1 refers to “male” and 0 refers to “female”.

Age: In the WVS individuals are asked to state their actual age (v237).

Democratization 23

Dow

nloa

ded

by [

85.1

07.1

90.3

2] a

t 04:

02 1

8 A

pril

2015