institutional trust, education, and corruption: a micro-macro interactive approach

Institutional Trust, Education, and Corruption:A Micro-Macro Interactive Approach

Armen Hakhverdian University of Amsterdam

Quinton Mayne Harvard University

This article examines how the effect of education on institutional trust varies cross-nationally as a function of thepervasiveness of public-sector corruption. We approach institutional trust as a performance-based evaluation ofpolitical institutions. Given their greater capacity to accurately assess the level of corruption coupled with theirstronger commitment to democratic values, we hypothesize that higher-educated citizens should react differently tocorruption from those with less education. Employing multilevel models we find that education has both aconditional and a conditioning effect on institutional trust. First, education is negatively related to institutionaltrust in corrupt societies and positively related to institutional trust in clean societies. Second, the corrosive effect ofcorruption on institutional trust worsens as education improves. The article ends with a discussion of theimplications of these findings for the functioning of contemporary democracies.

Why are citizens in some countries moretrusting of the actors and institutions ofgovernment than citizens in other coun-

tries? Over the past two decades this question has cometo occupy an important place in the field of com-parative political science (Dalton 2004; Norris 1999a;Pharr and Putnam 2000). In the large body of workthat now exists on this question of political trust, acontradictory set of findings regarding the effect ofeducation has gone largely unnoticed. Whereas someresearch shows that education boosts political trust(e.g., Anderson and Singer 2008), in other work theeffect of education is negative (e.g., Seligson 2002);and in a third group of studies, education failsaltogether to predict individual-level variation inpolitical trust (e.g., Mishler and Rose 2001). Interest-ingly, specialists of political behavior have puzzledover the same variability of the effect of education onelectoral participation (Gallego 2010; Powell 1986;Verba, Nie, and Kim 1978). These authors show thatthis inconsistency of the effect of education onturnout is attributable to variation at the macro level.In a similar vein, we argue that the direction andmagnitude of the effect of education on institutional

trust depends on a country’s political and institu-tional context.1

At the heart of our study lie the following twopropositions. First, citizens grant and withhold trustbased on their evaluation of the performance of theactors and institutions of government. Second, citi-zens with more education are not only more likely tobe better able to identify practices that undermine thesmooth functioning of democratic institutions, theyare also more likely to be normatively troubled by suchpractices. Based on these twin claims, we hypothesizethat political trust is determined by the interactiveeffect of a person’s level of education on the one handand the overall quality of their country’s system ofdemocratic governance on the other.

Using data from 21 European democracies, wheresystem performance is operationalized using a measureof public-sector corruption, the results of our analysesstrongly support a cross-level, interactive understand-ing of political trust. In countries with comparativelyhigh levels of corruption, we find that educationdampens political trust; in countries with low levelsof corruption, education actually boosts political trust.Moreover, our study also demonstrates that the effect

The Journal of Politics, Vol. 74, No. 3, July 2012, Pp. 739–750 doi:10.1017/S0022381612000412

� Southern Political Science Association, 2012 ISSN 0022-3816

1An online appendix with supplementary material is available at http://journals.cambridge.org/jop. The data used in this article areaccessible at http://ess.nsd.uib.no/ess/round4. Supporting materials necessary to reproduce the numerical results will be made availableat http://quintonmayne.net upon publication.

739

of corruption on political trust depends on educationalattainment. Specifically, we find that citizens with thelowest levels of education are unresponsive to theeffects of corruption; for all other citizens, the corro-sive effects of corruption on political trust increasewith education.

In contrast to existing work that examines theinteractive effects of education on public opinion andpolitical behavior, we go beyond a purely cognitiveconception of education (Huber, Kernell, and Leoni2005; Jacoby 1991; Krause 1997; Nagler 1991). Weposit that education not only facilitates the acquisitionand processing of information, it also fundamentallyinforms how citizens react to certain types of informa-tion. Thus, a key innovation of this study is that inspecifying our hypotheses we are sensitive as much tothe normative as to the knowledge-enhancing proper-ties and effects of education. This theoretical refine-ment challenges the common proposition that risinglevels of education have been responsible for declinesin a variety of forms of political support observed inmany advanced industrial democracies over the pasthalf century (Dalton 2004, 86–91; Inglehart 1999,245–56). What we show is that there is nothingabout education per se that should encourage institu-tional distrust. Ultimately the role played by theimproving educational profile of contemporary soci-eties is to bring into striking, and at times alarming,relief the overall democratic quality of a country’sprocesses of government.

In the next section we discuss the concept ofinstitutional trust as an evaluative, performance-basedorientation toward political actors and institutions.We then elaborate the accuracy- and norm-inducingrole played by education in this process of evaluation,which results in citizens granting or withholdingpolitical trust depending on the quality of theircountry’s system of governance. The following sectiondescribes the data and methods that we use to estimatethis micro-macro interactive model of political trust.After presenting the results of this model, we discussthe importance and implications of our findings bothfor the study of political trust and for the functioning ofdemocracy more generally.

Theory

Institutional Trust

In the course of their lives citizens are vulnerable tothe decisions and actions of a myriad of political

actors and institutions. Political trust refers to thefaith that citizens place in political actors and in-stitutions not to act in ways that will do them harm(Levi and Stoker 2000; Newton 1999). Objects ofpolitical trust can range from the abstract to the veryspecific (Norris 1999b). Forming a midrange objectof trust are political institutions, examples of whichinclude local government, the civil service, the legalsystem, parties, parliament, the police, judges, andpoliticians. Lying at the very heart of contemporarydemocratic life and following a long line of researchfrom comparative political behavior, we focus onpublic trust in this category of political institutionsand actors.

Over the years a number of compelling argu-ments and empirical findings have been put forwardin support of the study of institutional trust. Mostfundamentally, institutional trust has commonly beenconceived of as a democratic good in and of itselfas well as an important gauge of a democracy’s polit-ical health (Dalton 2004; Pharr and Putnam 2000).A number of studies have also argued that a public’strust in the actors and institutions of political author-ity facilitates democratic consolidation in that institu-tionally trusting individuals have been found to bemore supportive of democratic principles (Seligsonand Carrion 2002). In addition, existing researchunderscores the importance of institutional distrustas a determinant of political participation, especiallynonelectoral participation such as protest activityand political consumerism (Micheletti, Follesdal, andStolle 2003; Mishler and Rose 2005). More recentresearch has also shown how institutional trust isrelated to a range of key political and nonpoliticalattitudes, including public policy preferences (ChanleyRudolph, and Rahn 2000; Hetherington 2005) andtrust in fellow citizens (Zmerli and Newton 2008).Finally, several studies have drawn attention to thepositive relationship between institutional trust andcompliance with government regulation and civic duty(Letki 2006; Tyler 1990).

Building on existing theoretical and empiricalresearch from the field of comparative political behav-ior, we understand institutional trust within a ‘‘ration-alist’’ framework as a primarily evaluative orientation(Hetherington 1998; Mishler and Rose 2001). Thus, incontrast to ‘‘culturalist’’ understandings of institu-tional (dis)trust as deeply socioculturally embedded,we view institutional trust as a product of individual-level judgment regarding the performance of politicalinstitutions and actors. Borrowing the conceptualapproach developed with regard to the EuropeanUnion by Scharpf (1999), we propose that citizens

740 armen hakhverdian and quinton mayne

evaluate political institutions and actors based on twodifferent sets of performance criteria: those related to‘‘input’’ or procedural performance and those relatedto ‘‘output’’ or policy performance.

Procedural performance is synonymous with thelong-standing democratic adage of government bythe people. Citizens judge procedural performanceaccording to the extent to which the democraticpromise of political fairness and equality is made realby the rules, procedures, norms, and structures thatguide and constrain the functioning of political in-stitutions and the behavior of political actors. Policyperformance, on the other hand, is synonymous withthe notion of government for the people. As such,citizens evaluate policy performance based on sub-stantive considerations concerning how successfulpolitical institutions and actors are in implementingpolicies and providing services that are responsive tothe public’s preferences and priorities. Where politicalactors and institutions achieve high levels of proce-dural and policy performance, we anticipate citizenswill reward such positive performance with trust. Incontrast, we expect citizens to express low levels ofinstitutional trust where the actors and institutions ofdemocratic political authority perform poorly proce-durally and in terms of policy responsiveness.

This idea that institutional trust is fundamentallythe product of a performance-based evaluation findsa great deal of support in the sizeable body of workthat has examined the effects of public-sector corrup-tion on attitudes related to political support. Time andagain we see the corrosive effects of corruption onpeople’s trust in the actors and institutions of govern-ment (Anderson and Tverdova 2003; Chang and Chu2006; Mishler and Rose 2001; Seligson 2002). Whatthis literature highlights is that corruption underminespolitical trust in a number of different ways. On theone hand, it affects the procedural performance ofpolitical institutions; on the other hand, corruptionmakes it extremely difficult for governments to pro-duce policies and services that are responsive to thegeneral public (Rothstein and Uslaner 2005; Warren2004).

Accuracy- and Norm-Inducing Functionsof Education

In defining the concept of institutional trust anddescribing the process of performance evaluationbehind it, we have painted an undifferentiated pictureof citizens. When it comes to the process of perform-ance evaluation undergirding institutional trust, there

are sound theoretical and empirical reasons to thinkabout citizens in heterogeneous rather than homoge-neous terms (Anderson and Singer 2008). Specifi-cally, we propose that citizens will assess and react tothe general performance of their respective country’spolitical institutions differently depending on theirlevel of education. This idea that education plays apivotal role is based on an abundance of evidenceidentifying the deep imprint left by education bothon citizens’ capacity to acquire and process informa-tion and on the types of values and principles citizenshold dear. For simplicity’s sake, we will refer to theseas the accuracy-inducing and norm-inducing func-tions of education.2

Any act of evaluation requires a capacity toacquire and process information related to the objectbeing evaluated. For the present purposes the objectof evaluation is the performance of political institu-tions. Single-country and comparative research hasdemonstrated that, even after controlling for a host ofother variables, formal education produces a numberof different and oftentimes related changes in indi-viduals that facilitate the acquisition and processingof information necessary for arriving at an accurateassessment of the functioning of political institutions.For one, education has repeatedly been shown to havea positive effect on the amount and type of attentioncitizens pay to politics and public affairs. For example,the more educated consistently express more interestin politics than the less educated (Delli Carpini andKeeter 1996; Verba, Schlozman, and Brady 1995).Education has been shown to be positively related totime spent reading newspapers, itself an importantmeans of obtaining political information (Dee 2004;Elvestad and Blekesaune 2008; Shehata and Stromback2011). Moreover, citizens with higher levels of educa-tion have a greater tendency to undertake a range ofpolitical activities that provide opportunities for learn-ing about the quality of political institutions. Thisincludes, for instance, attending public meetings, mak-ing contact with public officials, and following elec-toral campaigns (Aars and Strømsnes 2007; Milligan,Moretti, and Oreopoulos 2004; Pattie, Seyd, andWhiteley 2004).

A large body of research also underscores theimportant direct and indirect role played by formaleducation in enhancing political knowledge. Themore educated are more likely to respond correctlyto questions of fact related to the key players andworkings of their political system (Delli Carpini and

2We thank one of our anonymous reviewers for suggesting thisphrasing.

institutional trust, education, and corruption 741

Keeter 1996; Hyman, Wright, and Reed 1975; Milner2002). The more educated have also been shown tobe better able to place parties on issues and ideologyand more likely to form attitudes consistent withtheir own ideological leanings (Aarts and Semetko2003; Gordon and Segura 1997; Jacoby 1991). Inter-estingly, there is also evidence to suggest that,compared to the less educated, the higher educatedare more likely to accurately assess their particularcountry’s overall respect for human rights (Andersonet al. 2005).

While education makes it easier for citizens toacquire and process information related to the qualityof the workings of democratic institutions, there isnothing intrinsic about the possession of such in-formation that should lead citizens to view under-performing or mismanaged institutions negatively. Inorder to hypothesize about this link between knowl-edge and evaluation, we need to establish how citizensreact to information about phenomena, such ascorruption, that undermine democratic performance.Education has long been shown to leave an indeliblemark both on people’s broad normative proclivitiesand on the store they set by democratic rule. Forexample, over the years research has repeatedly shownthat support for a range of liberal moral values,including equality and tolerance, grows with years ofschooling (Bobo and Licari 1989; Hyman and Wright1979; Nie, Junn, and Stehlik-Barry 1996; Vogt 1997).Scholars have also argued that in established democ-racies the more educated are more likely to supportand defend core democratic values and principles(Dalton 1994, 483; Hibbing and Theiss-Morse 2002,771; McClosky and Zaller 1984, 239–40). In the pastdecade, drawing in part on earlier waves of modern-ization research, a body of work has emerged that alsohighlights the positive links between years of schoolingand support for democracy in new democracies andnondemocracies (Evans and Rose 2007; Jamal 2006;Kotzian 2011). Thus, in addition to being better ableto accurately identify practices that undermine thesmooth functioning of democratic institutions, thesedifferent strands of research suggest that the bettereducated will also be normatively troubled by suchpractices.

As discussed earlier, corruption represents amajor challenge to the performance of political insti-tutions. As such, and in view of the accuracy- andnorm-inducing functions of education, when facedwith the absence or presence of corruption we expectcitizens to grant and withhold institutional trustdifferently depending on their level of education. Inother words, we hypothesize that corruption and

education interact to affect a person’s level of trust inpolitical institutions. In hypothesizing that such aninteractive effect exists, we are making two separate,symmetric claims (see Berry, Golder, and Milton 2012):one is about the conditional effect of education; theother is about the conditioning effect of education.

Hypothesis 1: Regarding the conditional effect ofeducation, we hypothesize that in countries withcomparatively high levels of corruption more edu-cated citizens will react more negatively to cor-ruption than the less educated. In contrast, incomparatively clean countries we hypothesize thatmore educated citizens will react more positively tothe absence of corruption than the less educated.

Hypothesis 2: As for the conditioning effect ofeducation on corruption, we envisage that whilecorruption will corrode institutional trust amongall citizens irrespective of their level of education,the size of this negative effect of corruption oninstitutional trust should increase with a person’slevel of education.

In testing the first hypothesis we are able to address along-standing, but overlooked, aspect of existingresearch that becomes clear from surveying impor-tant single-country and comparative studies regard-ing the effects of education on political trust. We canclassify existing studies according to whether theyexamine individual-level variations in political trustin one or more advanced industrial democracy, inone or more of the new and/or low-, middle-, andhigh-income democracies of Central and EasternEurope, Latin America, Africa, and Asia, or in amix of old and new, low-, middle-, and high-incomedemocracies (see Table A1 in the online appendix).Of the existing research that is based on data from theworld’s longest-standing, advanced industrial democ-racies, many studies show a positive effect of educa-tion (e.g., Anderson et al. 2005; Rohrschneider andSchmitt-Beck 2002), whereas only one study showseducation as having a negative effect on political trust(Cooke and Gronke 2005). In contrast, many studiesshow education as negatively related to political trustin the world’s newer and less affluent democracies(e.g., Canache and Allison 2005; Chang and Chu2006). Moreover, equally telling is the fact that theeffect of education on political trust fails to achievestatistical significance in a number of pieces ofresearch using data from both old and new democ-racies (e.g., Anderson and Tverdova 2003). Thesecontradictory findings suggest that that the effect ofeducation on institutional trust is context specific.The empirical analyses that follow demonstrate thatthis is indeed the case.


Data and Variables

In order to test whether institutional trust is afunction of the interaction of educational achieve-ment and corruption, we make use of Round 4 of theEuropean Social Survey (ESS) conducted in 2008–2009(ESS Round 4 2008). The necessary individual- andmacro-level variables were available for eight Easternand Central European and 13 Western Europeandemocracies: namely, Belgium, Croatia, the CzechRepublic, Denmark, Estonia, Finland, France, Germany,Great Britain, Greece, Hungary, Latvia, the Netherlands,Norway, Poland, Portugal, Romania, Slovenia, Spain,Sweden, and Switzerland. This section outlines the keyvariables and data sources of our study. The exactwording of the survey items can be found in the onlineappendix.

Dependent Variable

Institutional trust constitutes the dependent variableof this study. The ESS asks respondents about theirlevels of trust in their country’s parliament, legal sys-tem, police, politicians, political parties, the EuropeanParliament, and the United Nations. Since our sub-stantive focus is on domestic institutions and theirrespective performance, we chose to drop trust ininternational organizations such as the EuropeanParliament and the United Nations from our index.We combine the remaining five items to create anindex of institutional trust.3 For our sample of 21European democracies, Cronbach’s alpha, a wide-spread measure of reliability for scales, ranges from0.83 in Poland to 0.92 in Romania. These figures arewell above the conventional boundaries for scaleconstruction, so for our dependent variable, we simplycalculate the mean score of these five separate domes-tic trust items.4

Institutional trust has a theoretical range of 0(‘‘no trust at all’’) to 10 (‘‘completely trusting’’). Theempirical, aggregate-level range runs from 2.78 inLatvia to 6.58 in Denmark (see Figure A1 in the onlineappendix). We can divide the sample into roughlythree sets of countries based on levels of institutionaltrust. The first group of ‘‘highly trusting’’ countriesconsists of Sweden, the Netherlands, Switzerland,

Norway, Finland, and Denmark. A second group of‘‘moderately trusting’’ countries consists of Slovenia,Estonia, Spain, France, Great Britain, Belgium, andGermany. Finally, a third group of ‘‘low-trusting’’countries consists of Latvia, Hungary, Croatia, Poland,the Czech Republic, Portugal, Romania, and Greece.As this grouping makes clear, Central and EasternEuropean countries generally exhibit lower levels oftrust than their Western European counterparts (alsosee Mishler and Rose 1997). However, while control-ling for region is therefore important, there arenoticeable exceptions to this pattern. Portugal andGreece, for instance, are on average less trusting thanSlovenia and Estonia. Furthermore, there are starkwithin-region differences in institutional trust, so evenafter controlling for region sizeable variation in trustremains among the Western European and Centraland Eastern European countries in our sample.

Independent Variables

The two main independent variables are level ofeducation and corruption. For the former, we recodedthe original ESS education item into four categories.The lowest education category contains those respond-ents with no qualification or only primary education(16.0% of the entire sample). The next categorycontains those with a lower secondary level of education(19.4%). The third category consists of individuals withan upper-secondary or postsecondary (nonuniversity)education (40.9%). Finally, the highest category con-tains those with a university education (23.7%).5

The primary, macro-level, contextual variable ofinterest is corruption. We draw upon TransparencyInternational’s well-known Corruption PerceptionIndex (CPI) which gauges the perceived level ofpublic-sector corruption. The CPI is a compositemeasure, consisting of 13 items from 11 independentinstitutions, that taps perceptions of corruptionamong country experts and business leaders andhas been employed in numerous other comparativestudies of the causes and consequences of corruption(Anderson and Tverdova 2003; Gerring and Thacker2004; Treisman 2000).6 Higher scores on the CPIindicate a relative lack of corruption in the public

3Klingemann (1999) and Anderson and Singer (2008) adopt asimilar empirical strategy in their studies of political trust.

4An index consisting of all seven trust items is virtually identicalto the five-item index (Pearson’s r 5 0.96), and employing it asthe dependent variable does not alter any of the substantiveresults reported here.

5We also conduct the analysis using the original seven-categoryeducation item from the ESS to check whether our results aresensitive to configurations of the education categories. Thefindings turn out to be similar.

6For a detailed description of the construction of the CorruptionPerception Index, see http://www.transparency.org/policy_research/surveys_indices/cpi/2008.


sector and political sphere. Using data for 2008, theCPI ranges from 9.3 in Denmark and Sweden to 3.8in Romania. The fact that the CPI is a measure ofperceived corruption could raise questions of endo-geneity, but Anderson and Tverdova (2003) employ aHausman test to show that corruption is not endog-enous to system support. We therefore include theCPI as an explanatory variable with the knowledgethat the CPI is exogenous to mass attitudes towardsthe political system.

Control Variables

Previous comparative studies have identified a host ofindividual-level factors besides level of education thatpredict political trust (Anderson and Singer 2008;Chang and Chu 2006; Mishler and Rose 2001; VanDer Meer 2010). These include demographic andsocioeconomic variables such as age, gender, income,religious denomination, and religious attendance. Wealso control for preadult experiences and influencesby including paternal level of education (see Kam andPalmer 2008). Furthermore, we control for severalattitudinal and behavioral variables that have beenshown to be positively related to political trust,including social trust (Zmerli and Newton 2008),election winner or loser status (Anderson and Tver-dova 2003), and satisfaction with the present state ofthe national economy (Mishler and Rose 2001;Rohrschneider and Schmitt-Beck 2002).7

At the country level, the control variables includelevel of development as captured by GDP per capita,level of unemployment, and a region dummy forCentral and Eastern Europe (Anderson and Singer2008). Some scholars have made the argument thatthe inclusion of citizens into the political processinduces more positive attitudes toward the politicalsystem (Lijphart 1999). We therefore add the dis-proportionality of the electoral system, as calculatedby Gallagher’s least squares index (Gallagher 1991),to our country-level controls. We refrain fromincluding additional country-level controls due toour relatively small sample of countries.

Method

A study of the link between macropolitics andmicrobehavior calls for data collection at the level

of individuals as well as countries. Since individualsare nested within national contexts an explicitlymultilevel modeling strategy is in order (Snijdersand Bosker 1999). Steenbergen and Jones (2002)recommend, as a first step toward building appropri-ate multilevel models, checking whether and to whatextent the dependent variable in question varies acrossthe relevant levels of analysis. Applied to our case, weshould investigate whether institutional trust variesbetween individual respondents as well as betweencountries. As our earlier discussion of the dependentvariable made clear, large between-country differencesexist in institutional trust. What therefore remains tobe investigated is the extent to which this variation canbe accounted for by the characteristics of citizens inthese specific countries or by macro-level character-istics specific to the countries themselves.

Following the advice of Steenbergen and Jones(2002), we decompose the variance in institutionaltrust into individual-level variance and country-levelvariance such that

Institutional trustij ¼ g00 þ v0j þ eij ð1Þ

Here, institutional trust varies around a grand meang00, while n0j and eij capture deviations from thegrand mean for country j and individual i respectively.Since both variance components are statistically sig-nificant (see Table A2 in the online appendix), we canconclude that there appears to be significant variancein institutional trust at both levels providing statisticaljustification for adopting a multilevel approach. Inaddition, about 27% of the total variance in institu-tional trust occurs at the country-level, so by ignoringcontextual variance one is likely to miss out onimportant explanations of institutional trust.

We now specify a random-coefficient model thatassesses the interactive effect of education and cor-ruption on institutional trust, while controlling forconfounding factors at both the individual andcountry levels. At the individual level, educationtakes the form of three dummy variables with thelowest education cohort functioning as the referencecategory. The model contains a variance componentfor the intercept, accounting for mean differences ininstitutional trust across countries, and for theeducation dummies, effectively allowing the impactof education on political trust to vary across the 21sampled democracies. The variance estimates of theslopes are all statistically significant and increase inmagnitude as education level increases. The fact thateducation has a varying impact on institutional trustis of course the focus of this study. A cross-levelinteraction between corruption and the education

7We note that some of these control variables might beendogenous to our dependent variable. For example, Hether-ington (1998) shows that political trust is an important cause ofperceptions of presidential performance.


categories then estimates the extent to which, on theone hand, the effect of education on institutionaltrust is contingent upon corruption, and on theother, the effect of corruption on institutional trustis contingent upon a person’s education.8 The orig-inal intraclass correlation drops from 0.27 in theempty model to 0.17 when we allow the slopes to varyacross countries and add individual-level controls.This figure drops even further to 0.03 after introduc-ing the country-level controls and the cross-levelinteractions, which underscores the importance ofmodeling institutional trust hierarchically.

Empirical Findings

As a preliminary test of our hypotheses, we plot meanlevels of institutional trust among the highest andlowest educated in each of the 21 democracies in oursample. Figure 1 displays these mean levels, withcountries ranked according to their CPI score. First,congruent with Hypothesis 1, the highest educatedexhibit more political trust than the lowest educated,but only when corruption is low. Figure 1 shows thateducation boosts political trust in the countriestowards the right-hand side of the graph. The differ-ence in trust between the highest and lowest educateddisappears as corruption increases and even reversestoward the left-hand side of the graph. In Romania,Croatia, Greece, and Latvia, the least-educated dis-play higher levels of institutional trust than the most

educated. This graphical analysis lends tentativesupport to the hypothesis that the effect of educationon institutional trust is not uniform across countries,but rather depends on the context in which citizensare nested. Second, in line with Hypothesis 2, institu-tional trust decreases as corruption rises both amongthe highest and lowest educated groups, but thedecline is greatest among the highest educated. Ofcourse, as Figure 1 displays net effects of educationon political trust, it remains to be seen whether thesefindings will be reproduced in a fully specifiedhierarchical model. To this we turn next.

Table 1 presents the estimation results from ourrandom coefficients model. Before moving on to ourmain findings, however, we note that institutionaltrust is positively related to paternal education, income,religious attendance, and belonging to a religion, andnegatively related to age and being male. In addition,several indicators of civic and political engagementhave a positive effect on institutional trust. People whoare more trusting of other people are also more likelyto exhibit institutional trust compared to less sociallytrusting individuals (Zmerli and Newton 2008). More-over, those who evaluate the state of the economy morepositively will generally be more trusting of politicalinstitutions (Mishler and Rose 2001). Finally, non-voters and those having cast their votes for oppositionparties display lower levels of institutional trust thanelection winners (Anderson and Tverdova 2003). Asfor the country-level controls, most of these fail toreach conventional standards of statistical significance.We do note, however, that citizens of Central andEastern European countries report lower levels ofinstitutional trust than those of Western Europeancountries even after controlling for corruption andmacroeconomic performance.

FIGURE 1 Institutional Trust across Educational Levels

8Our dependent variable is normally distributed, so we useMaximum Likelihood to estimate a linear multilevel regression.All analyses were conducted using Stata 11.


Under which conditions can we expect a positiveor negative impact of education on institutionaltrust? The results of Table 1 show that education is

negatively related to institutional trust when theCorruption Perception Index is zero. Thus, in thehypothetical case of a perfectly corrupt society,the most educated are less politically trusting thanthe least educated (1.59 points less trusting to beexact). The interaction terms reveal that this gapdecreases as the CPI increases. In a perfectly cleansociety (CPI 5 10), the highest educated are esti-mated to be more trusting of political institutionsthan the lowest educated by 0.83 points. Of course,CPI values of 0 and 10 do not really exist. In the real-world settings of our sample of countries, the CPIruns from 3.8 in Romania to 9.3 in Denmark andSweden. Using the estimates from our model, Figure2 charts the marginal effect of education on institu-tional trust for different values of the CPI.9 Educationis negatively related to institutional trust for countrieswhose CPI is lower than 6.1 (which includes theCzech Republic, Croatia, Greece, Hungary, Latvia,Poland, Portugal, and Romania). The effect of edu-cation on institutional trust is nonsignificant if theCPI lies above 6.1 and below 7.0 (a range that coversEstonia, France, Slovenia, and Spain). Finally, edu-cation has a positive effect on institutional trust forcountries whose CPI exceeds 7.0 (encompassingBelgium, Denmark, Finland, Germany, Great Britain,the Netherlands, Norway, Sweden, and Switzerland).

In line with Hypothesis 1, in relatively corruptsocieties the most educated are more distrustful thanthe least educated, in moderately clean societies theyare equally trusting as the least educated, and inrelatively clean societies they are more trusting. Byexplicitly modeling the possibility that corruption andeducation interact, we are able to isolate the conditionsunder which education has a positive, nonsignificant,and negative effect on institutional trust and, in sodoing, we are able to resolve the inconsistency of theeffects of education as reported in previous research(see Table A1 in the online appendix).

Next, we deal with the conditioning effect ofeducation as formulated in Hypothesis 2. The coef-ficient for corruption in Table 1 captures the effect ofcorruption on institutional trust for the group with thelowest level of educational attainment. These arerespondents who have not completed primary educa-tion or who have obtained a qualification at theprimary or first stage of basic education. Corruptiondoes not seem to affect the institutional trust of theleast educated cohorts, controlling for a host of relevantindividual-level and country-level characteristics. While

TABLE 1 Effects of Education and Corruption onInstitutional Trust

Independent Variable Estimate (Std. Error)

Main variables of interestEducation

(Reference: notcompleted and basic)

Lower secondary – 0.100*** (0.172)Upper and postsecondary – 1.292*** (0.148)BA and higher – 1.587*** (0.201)

Corruption 0.093 (0.060)Corruption 3 Lower secondary 0.145*** (0.025)Corruption 3 Upper and

postsecondary0.189*** (0.022)

Corruption 3 BA and higher 0.242*** (0.029)

Individual-level controlsMale – 0.021 (0.019)Age – 0.029*** (0.003)Age2 / 1000 0.267*** (0.030)Father’s education 0.022* (0.008)Religious adherence 0.170*** (0.023)Attendance of religious services 0.063*** (0.008)Income 0.022*** (0.004)Election winner-loser status

(Reference: winner)Loser – 0.298*** (0.022)Nonvoter – 0.463*** (0.026)

Social trust 0.184*** (0.004)Satisfaction with the economy 0.290*** (0.005)

Country-level controlsGDP per capita / 1000 – 0.017 (0.014)Unemployment – 0.048 (0.037)Central and Eastern Europe – 0.533* (0.225)Disproportionality of the

electoral system0.009 (0.015)

Intercept 3.080*** (0.718)

Random effectsVariance (Country) 0.053**Variance (BA and higher) 0.023Variance (Upper and

postsecondary)0.005

Variance (Lower secondary) 0.009Variance (Individual) 2.353***N (Individuals) 27,785N (Countries) 21–2 3 Log Likelihood 102737

Source: European Social Survey, Round 4.Note: *** p , .001, ** p , .01, * p , .05. Two-tailed tests ofstatistical significance.

9We made use of the Stata syntax that accompanies Brambor,Clark, and Golder (2006).


the coefficient has the expected sign—the higher theCorruption Perception Index, the higher institutionaltrust—it fails to reach accepted levels of statisticalsignificance. However, corruption has a statisticallysignificant effect on the institutional trust of all othereducation groups. This leads us to conclude thatcorruption has a corrosive impact on trust in politicalinstitutions for all but the lowest-education cohorts andthat this corrosive effect increases as citizens attainhigher levels of education.

Figure 3 summarizes our statistical analyses bycalculating the predicted levels of institutional trustacross education groups and levels of corruption. Bykeeping all other variables in Model 1 at their meanor mode, we can determine the effect of education

alone on institutional trust at various values of theCorruption Perception Index. It is clear from thisfigure that the most educated trust or distrust politicalinstitutions based on the overall performance of theirpolitical system. Institutional trust among the mosteducated covaries strongly alongside the CorruptionPerception Index. The difference in institutional trustbetween two highly educated individuals, one fromSweden and the other from Romania, is approximately2.0 points on a 0–10 scale. By contrast, the samedifference is approximately less than half a point forthe least educated. Given that the results reported inTable 1 indicate that this difference is not statisticallysignificant at the .05 level, we cannot reliably infer thatcorruption affects institutional trust among citizenswith the lowest levels of education.

Conclusion

This study examines the context-specific effects ofeducation on institutional trust. The theory that wetest was developed around the following two prop-ositions. First, citizens grant or withhold trust as away of evaluating political institutions for theirperformance. Second, citizens with more educationare not only more likely to be better able to identifypractices that undermine the smooth functioning ofdemocratic institutions; they are also more likely tobe normatively troubled by such practices. In linewith our expectations, we find that the direction andmagnitude of the effect of education is conditionalupon the pervasiveness of public-sector corruption:in countries with low levels of corruption educationboosts institutional trust; in countries with compa-ratively high levels of corruption education dampensinstitutional trust. In addition, our analyses largelysupport our hypothesis that education also moderatesthe effect of corruption on institutional trust. Specif-ically, we find that the corrosive effect of corruptionon political trust worsens as education improves(except for the least educated whose trust in politicalinstitutions we find is unaffected by corruption).

The results of our analyses have implications notjust for future comparative research on the causes ofpublic opinion and the attitudinal effects of politicalcorruption, but also for better understanding thefunctioning of democracy. In the literature that hasemerged on corruption over the past 20 years one ofthe clearest and most consistent findings is thatcorruption undermines people’s trust and confi-dence in the actors and institutions of democratic

FIGURE 2 The Difference in Institutional Trustbetween the Highest and LowestEducated across Levels of Corruption

FIGURE 3 Predicted Values of Institutional Trustacross Levels of Corruption andEducation Groups


government. Thus far, almost all of this research hasfocused on the direct effects of corruption onpolitical support. In line however with a few otherexisting pieces (Anderson and Tverdova 2003;Herreros and Criado 2007), our study underscoresthe importance of developing and testing theoriesthat elucidate the conditional nature of the attitu-dinal (as well as the behavioral) consequences ofcorruption. Moreover, our finding that corruptionfails to impact the institutional trust of the leasteducated points to the need for additional scholar-ship on the micro- and/or macroconditions underwhich corruption and other forms of systemicdysfunctionality become, worryingly from a demo-cratic point of view, politically irrelevant for specificgroups of citizens.

Despite a long tradition of work emphasizing thelinks between educational achievement and democraticcommitment, scholars of political behavior exploringthe interactive effects of education have to date focusedtheir attention on the knowledge-enhancing, accuracy-inducing functions of education. A key contribution ofthe present study is that it incorporates the normativedimensions of education when theorizing about theconditional and conditioning effects of education onpublic opinion. In elaborating our hypotheses in a waythat is sensitive to both the cognitive and the moralproperties of education, the results of our analysesstrongly support a ‘‘rationalist’’ understanding of insti-tutional trust. Citizens with comparable levels ofeducation systematically express different levels ofpolitical trust depending on the overall performanceof their particular country’s political system. Thisclearly suggests that rather than being something fixedand deeply culturally embedded trust is generatedthrough a process that is fundamentally evaluative innature. As such, our findings warn against viewingsocieties that express high levels of institutional trust inthe aggregate as either politically docile or lacking theskepticism necessary for democratic maturity, or in-deed assuming that countries with low aggregate levelsof political trust are doomed in perpetuity.

Finally, the findings of our analyses point to someimportant potential political consequences arisingfrom the changing educational profile of many oftoday’s middle- and high-income democracies. Inrecent decades, and as a response to the demands andinsecurities arising from economic globalization,governments across the world have increasingly beenpursuing policies aimed at promoting a knowledge-based economic model. A key component of thesepolicies has been to invest heavily in raising educa-tional standards. For example, across OECD member

countries, completion rates of undergraduate degreeprograms rose by 21 percentage points between 1995and 2008 (OECD 2010, 61). Such rapidly increasingaggregate levels of education have particularly chal-lenging consequences where levels of corruption arehigh. As several of the countries in our sample attest,rooting out corruption can prove very difficult. Aftertwo decades of democratic rule corruption remainsrelatively high in a number of Central and EasternEuropean countries; the same is equally true ofGreece, for example, which has enjoyed democracyfor even longer. Given this contrast between, on theone hand, the speed with which educational stand-ards are rising across the democracies from which oursample of countries is drawn and, on the other hand,the slow pace at which political authorities are able orwilling to stamp out corrupt practices, our findingssuggest that the problem of low-level institutionaltrust in contexts of high corruption is likely only toworsen in the medium term. In and of itself thispotential development is normatively troubling. Inaddition however, as noted earlier existing compara-tive research shows that persistently low levels ofinstitutional trust can have serious ramifications forstate-society relations in these democracies. Possibleeffects include increasing levels of protest, politicalapathy, and disengagement from formal, electoralpolitics, an upturn in voting for antisystem parties,diminishing support for redistributive governmentaction aimed at tackling socioeconomic inequalities,and higher levels of noncompliance with governmentregulations.

Acknowledgments

Both authors contributed equally to this work. Wethank Imke Harbers, Sara Binzer Hobolt, AmaneyJamal, Jonas Pontusson, Jan Teorell, Tom Van DerMeer, the anonymous reviewers, and the editors fortheir valuable feedback on earlier versions. Any errorsremain our responsibility.

References

Aars, Jacob, and Kristin Strømsnes. 2007. ‘‘Contacting as aChannel of Political Involvement: Collectively MotivatedIndividually Enacted.’’ West European Politics 30 (1): 93–120.

Aarts, Kees, and Holli A. Semetko. 2003. ‘‘The Divided Elector-ate: Media Use and Political Involvement.’’ Journal of Politics65 (3): 759–84.

Anderson, Christopher J., and Matthew M. Singer. 2008. ‘‘TheSensitive Left and the Impervious Right: Multilevel Models


and the Politics of Inequality, Ideology, and Legitimacy inEurope.’’ Comparative Political Studies 41 (4/5): 564–99.

Anderson, Christopher J., and Yuliya V. Tverdova. 2003. ‘‘Cor-ruption, Political Allegiances, and Attitudes toward Govern-ment in Contemporary Democracies.’’ American Journal ofPolitical Science 47 (1): 91–109.

Anderson, Christopher J., Andre Blais, Shaun Bowler, ToddDonovan, et al. 2005. Losers’ Consent: Elections and DemocraticLegitimacy. Oxford and New York: Oxford University Press.

Berry, William D., Matt Golder, and Daniel Milton. 2012.‘‘Improving Tests of Theories Positing Interaction. ’’ Journalof Politics 74 (3).

Bobo, Lawrence, and Frederick C. Licari. 1989. ‘‘Education andPolitical Tolerance: Testing the Effects of Cognitive Sophisti-cation and Target Group Affect.’’ Public Opinion Quarterly 53(3): 285–308.

Brambor, Thomas, William Roberts Clark, and Matt Golder.2006. ‘‘Understanding Interaction Models: Improving Em-pirical Analyses.’’ Political Analysis 14 (1): 63–82.

Canache, Damarys, and Michael E. Allison. 2005. ‘‘Perceptions ofPolitical Corruption in Latin American Democracies.’’ LatinAmerican Politics and Society 47 (3): 91–111.

Chang, Eric C. C., and Yun-han Chu. 2006. ‘‘Corruption andTrust: Exceptionalism in Asian Democracies?’’ Journal ofPolitics 68 (2): 259–71.

Chanley, Virginia A., Thomas J. Rudolph, and Wendy M. Rahn.2000. ‘‘The Origins and Consequences of Public Trust inGovernment.’’ Public Opinion Quarterly 64 (3): 239–56.

Cook, Timothy E., and Paul Gronke. 2005. ‘‘The SkepticalAmerican: Revisiting the Meanings of Trust in Governmentand Confidence in Institutions.’’ Journal of Politics 67 (3):784–803.

Dalton, Russell J. 1994. ‘‘Communists and Democrats: Demo-cratic Attitudes in the Two Germanies.’’ British Journal ofPolitical Science 24 (4): 469–93.

Dalton, Russell J. 2004. Democratic Challenges, DemocraticChoices: The Erosion of Political Support in Advanced IndustrialDemocracies. Oxford; New York: Oxford University Press.

Dee, Thomas S. 2004. ‘‘Are There Civic Returns to Education?’’Journal of Public Economics 88 (9–10): 1697–1720.

Delli Carpini, Michael X., and Scott Keeter. 1996. What Amer-icans Know about Politics and Why It Matters. New Haven, CT:Yale University Press.

Elvestad, Eiri, and Arild Blekesaune. 2008. ‘‘Newspaper Readersin Europe.’’ European Journal of Communication 23 (4): 425–47.

ESS Round 4: European Social Survey Round 4 Data. 2008. Datafile edition 4.0. Norwegian Social Science Data Services,Norway– Data Archive and distributor of ESS data.

Evans, Geoffrey, and Pauline Rose. 2007. ‘‘Support for Democ-racy in Malawi: Does Schooling Matter?’’ World Development35 (5): 904–19.

Gallagher, Michael. 1991. ‘‘Proportionality, Disproportionalityand Electoral Systems.’’ Electoral Studies 10 (1): 33–51.

Gallego, Aina. 2010. ‘‘Understanding Unequal Turnout: Educa-tion and Voting in Comparative Perspective.’’ ElectoralStudies 29 (2): 239–48.

Gerring, John, and Strom C. Thacker. 2004. ‘‘Political Institu-tions and Corruption: The Role of Unitarism and Parliamen-tarism.’’ British Journal of Political Science 34 (2): 295–330.

Gordon, Stacy B., and Gary M. Segura. 1997. ‘‘Cross-NationalVariation in the Political Sophistication of Individuals:Capability or Choice?’’ Journal of Politics 59 (1): 126–47.

Herreros, Francisco, and Henar Criado. 2007. ‘‘Political SupportTaking into Account the Institutional Context.’’ ComparativePolitical Studies 40 (12): 1511–32.

Hetherington, Marc J. 1998. ‘‘The Political Relevance ofPolitical Trust.’’ American Political Science Review 92 (4):791–808.

Hetherington, Marc J. 2005. Why Trust Matters: DecliningPolitical Trust and the Demise of American Liberalism. Princeton,NJ: Princeton University Press.

Hibbing, John R., and Elizabeth Theiss-Morse. 1995. Congress asPublic Enemy: Public Attitudes toward American PoliticalInstitutions. New York: Cambridge University Press.

Hibbing, John R., and Elizabeth Theiss-Morse. 2002. StealthDemocracy: Americans’ Beliefs about How Government ShouldWork. Cambridge and New York: Cambridge UniversityPress.

Huber, John D., Georgia Kernell, and Eduardo L. Leoni. 2005.‘‘Institutional Context, Cognitive Resources and Party Attach-ments across Democracies.’’ Political Analysis 13 (4): 365–86.

Hyman, Herbert Hiram, and Charles Robert Wright. 1979.Education’s Lasting Influence on Values. Chicago: ChicagoUniversity Press.

Hyman, Herbert Hiram, Charles Robert Wright, and JohnShelton Reed. 1975. The Enduring Effects of Education.Chicago: Chicago University Press.

Inglehart, Ronald. 1999. ‘‘Postmodernization Erodes Respect forAuthority, but Increases Support for Democracy.’’ In CriticalCitizens: Global Support for Democratic Government, ed. PippaNorris. Oxford and New York: Oxford University Press, 236–56.

Jacoby, Wade G. 1991. ‘‘Ideological Identification and IssueAttitudes.’’ American Journal of Political Science 35 (1): 178–205.

Jamal, Amaney A. 2006. ‘‘Reassessing Support for Islam andDemocracy in the Arab World?’’ World Affairs 169 (2): 51–63.

Kam, Cindy D., and Carl L. Palmer. 2008. ‘‘Reconsidering theEffects of Education on Political Participation.’’ Journal ofPolitics 70 (3): 612–31.

Klingemann, Hans-Dieter. 1999. ‘‘Mapping Political Support inthe 1990s: A Global Analysis.’’ In Critical Citizens: GlobalSupport for Democratic Government, ed. Pippa Norris. Oxfordand New York: Oxford University Press, 31–56.

Kotzian, Peter. 2011. ‘‘Public Support for Liberal Democracy.’’International Political Science Review 32 (1): 23–41.

Krause, George A. 1997. ‘‘Voters, Information Heterogeneity, andthe Dynamics of Aggregate Economic Expectations.’’ Ameri-can Journal of Political Science 41 (4): 1170–1200.

Letki, Natalia. 2006. ‘‘Investigating the Roots of Civic Morality:Trust, Social Capital, and Institutional Performance.’’ PoliticalBehavior 28 (4): 305–25.

Levi, Margaret, and Laura Stoker. 2000. ‘‘Political Trust andTrustworthiness.’’ Annual Review of Political Science 3 (1):475–507.

Lijphart, Arend. 1999. Patterns of Democracy: Government Formsand Performance in Thirty-Six Countries. New Haven, CT: YaleUniversity Press.

McClosky, Herbert, and John R. Zaller. 1984. The AmericanEthos: Public Attitudes Toward Capitalism and Democracy.Cambridge, MA: Harvard University Press.

van der Meer, Tom. 2010. ‘‘In What We Trust? A Multi-LevelStudy into Trust in Parliament as an Evaluation of StateCharacteristics.’’ International Review of Administrative Scien-ces 76 (3): 517–36.


Micheletti, Michele, Andreas Follesdal, and Dietlind Stolle. 2003.Politics, Products, and Markets: Exploring Political Consumer-ism Past and Present. New Brunswick, NJ and London:Transaction Publishers.

Milligan, Kevin, Enrico Moretti, and Philip Oreopoulos. 2004.‘‘Does Education Improve Citizenship? Evidence from theUnited States and the United Kingdom.’’ Journal of PublicEconomics 88 (9–10): 1667–95.

Milner, Henry. 2002. Civic Literacy: How Informed Citizens MakeDemocracy Work. Hanover, NH: University Press of NewEngland.

Mishler, William, and Richard Rose. 1997. ‘‘Trust, Distrust andSkepticism: Popular Evaluations of Civil and Political Insti-tutions in Post-Communist Societies.’’ Journal of Politics 59(2): 418–51.

Mishler, William, and Richard Rose. 2001. ‘‘What Are theOrigins of Political Trust? Testing Institutional and CulturalTheories in Post-Communist Societies.’’ Comparative PoliticalStudies 34 (1): 30–62.

Mishler, William, and Richard Rose. 2005. ‘‘What Are thePolitical Consequences of Trust?’’ Comparative Political Stud-ies 38 (9): 1050–78.

Nagler, Jonathan. 1991. ‘‘The Effect of Registration Laws andEducation on U.S. Voter Turnout.’’ American Political ScienceReview 85 (4): 1393–1405.

Newton, Kenneth. 1999. ‘‘Social and Political Trust in EstablishedDemocracies.’’ In Critical Citizens: Global Support for Demo-cratic Government, ed. Pippa Norris. Oxford; New York:Oxford University Press, 169–87.

Nie, Norman H., Jane Junn, and Kenneth Stehlik-Barry. 1996.Education and Democratic Citizenship in America. Chicago:University of Chicago Press.

Norris, Pippa, ed. 1999a. Critical Citizens: Global Support forDemocratic Government. Oxford and New York: OxfordUniversity Press.

Norris, Pippa, ed. 1999b. ‘‘Introduction: The Growth of CriticalCitizens?’’ In Critical Citizens: Global Support for DemocraticGovernment, ed. Pippa Norris. Oxford and New York: OxfordUniversity Press, 1–27.

OECD. 2010. Education at a Glance 2010: OECD Indicators. Paris:OECD.

Pattie, Charles J., Patrick Seyd, and Paul Whiteley. 2004. Citizen-ship in Britain: Values, Participation, and Democracy. Cam-bridge and New York: Cambridge University Press.

Pharr, Susan J., and Robert D. Putnam, eds. 2000. DisaffectedDemocracies: What’s Troubling the Trilateral Countries?Princeton, NJ: Princeton University Press.

Powell, G. Bingham. 1986. ‘‘American Voter Turnout in Com-parative Perspective.’’ American Political Science Review 80:17–44.

Rohrschneider, Robert, and Rudiger Schmitt-Beck. 2002. ‘‘Trustin Democratic Institutions in Germany: Theory and EvidenceTen Years after Unification.’’ German Politics 11 (3): 35–58.

Rothstein, Bo, and Eric M. Uslaner. 2005. ‘‘All for One:Equality, Corruption, and Social Trust.’’ World Politics 58(1): 41–72.

Scharpf, Fritz. 1999. Governing in Europe: Effective and Demo-cratic? New York: Oxford University Press.

Shehata, Adam, and Jesper Stromback. 2011. ‘‘A Matter ofContext: A Comparative Study of Media Environments andNews Consumption Gaps in Europe.’’ Political Communica-tion 28 (1): 110–34.

Seligson, Mitchell A. 2002. ‘‘The Impact of Corruption onRegime Legitimacy: A Comparative Study of Four LatinAmerican Countries.’’ Journal of Politics 64 (2): 408–33.

Seligson, Mitchell A., and Julio F. Carrion. 2002. ‘‘PoliticalSupport, Political Skepticism, and Political Stability in NewDemocracies.’’ Comparative Political Studies 35 (1): 58–82.

Snijders, T.A.B., and Roel J. Bosker. 1999. Multilevel Analysis: AnIntroduction to Basic and Advanced Multilevel Modeling.London, New Delhi, and Thousand Oaks, CA: SagePublications.

Steenbergen, Marco R., and Bradford S. Jones. 2002. ‘‘ModelingMultilevel Data Structures.’’ American Journal of PoliticalScience 46 (1): 218–37.

Treisman, Daniel. 2000. ‘‘The Causes of Corruption: A Cross-National Study.’’ Journal of Public Economics 76 (3): 399–457.

Tyler, Tom R. 1990. Why People Obey the Law. New Haven, CT:Yale University Press.

Verba, Norman H. Nie, and Jae-On Kim. 1978. Participation andPolitical Equality: a Seven-Nation Comparison. Cambridge andNew York: Cambridge University Press.

Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady.1995. Voice and Equality: Civic Voluntarism in AmericanPolitics. Cambridge, MA: Harvard University Press.

Vogt, W. P. 1997. Tolerance and Education: Learning to Live withDiversity and Difference. London, New Delhi, and ThousandOaks, CA: Sage Publications.

Warren, Mark E. 2004. ‘‘What Does Corruption Mean in aDemocracy?’’ American Journal of Political Science 48 (2):328–43.

Zmerli, Sonja, and Kenneth Newton. 2008. ‘‘Social Trust andAttitudes toward Democracy.’’ Public Opinion Quarterly 72(4): 706–24.

Armen Hakhverdian is Assistant Professor in theDepartment of Political Science at the University ofAmsterdam, Netherlands.

Quinton Mayne is Assistant Professor of PublicPolicy in the Kennedy School of Government,Harvard University, Cambridge, MA 02138.


institutional trust, education, and corruption: a micro-macro interactive approach

Documents