Rolling in the deep: How values and personality …...!1! ECPR General Conference Montréal, August 2015 Rolling in the deep: How values and personality traits affect individual perceptions

! 1!

ECPR General Conference Montréal, August 2015

Rolling in the deep: How values and personality traits affect individual perceptions of electoral integrity

Camille Reynolds† and Alessandro Nai††

† University of Lausanne ([email protected]) †† University of Sydney ([email protected])

****** preliminary draft, do not quote

version: 2015.08.23 ******

Abstract An established literature assesses the substantial conditions for the presence of electoral integrity (e.g., Norris 2014); within this framework, recent research focuses on how individual perceive the integrity of elections, that is, if and how citizens perceive that elections in their country are "free and fair" and globally lacking manipulations and malpractices (e.g., Aarts and Thomassen 2008; Birch 2008, 2010). This contribution investigates the deep individual underpinnings of those perceptions. More specifically, our aim is to uncover how values (Schwartz 1992) and the Big Five personality traits (John et al. 1991, 2008) affect how citizens perceive the conduct of elections, and their integrity, in their country. To the best of our knowledge no existing research has yet provided empirical evidence in this sense; compensating this gap is paramount for a full understanding of electoral dynamics, given that both values and personality traits have been shown to matter greatly for attitudes, opinions and perceptions. Empirical analyses rely on the sixth wave of the World Value Survey (2014), the only individual dataset providing information simultaneously on values, personality traits and perceptions of electoral integrity. The dataset allows us, furthermore, to compare results across 13 countries, thus controlling for differences across party and electoral systems, and foremost for varying levels of measured electoral integrity (PEI index; Norris et al. 2013, 2014): rather high (Germany, Netherlands, Rwanda, South Africa), average (Colombia, Ecuador, India, Kuwait, Pakistan, Thailand), rather low (Algeria, Iraq, Jordan). Keywords Electoral integrity, perceptions, values, personality traits, comparative study

! 2!

Introduction

Elections are probably the most important event in any given country’s political life,

as they establish the roots for the future governance of the country through an

interactive process between political elites and citizens. Within this normative

framework, many authors stress the fundamental role of citizens’ electoral confidence

for the electoral process and the political system on the whole (Elklit 1999; Elklit and

Reynolds 2002, 2005; Hartlyn and McCoy 2006; Lehoucq 2003; Lyons 2004;

Mozaffar 2002; Mozaffar and Schedler 2002; Schedler 2002a; 2002b; 2006;

Anderson and Tverdova 2003; Birch 2007, 2008, 2010). This literature has however

less to say about how citizens perceive the conduct of elections (but see, e.g., Aarts

and Thomassen 2008; Birch 2008, 2010). In fact, how do citizens assess the overall

quality of the elections that take place in their country? And even more profoundly,

why, facing the same electoral contest, some citizens tend to evaluate the elections

rather positively while others are more skeptical about their overall quality?

The growing body of literature on electoral integrity mostly focuses on contextual

determinants and so far we know little about the roots of individual perceptions of

electoral integrity. Existing research on those micro-level determinants has mainly

examined the role of basic socio-demographic background, such as age, gender,

education, religion or ethnicity (e.g. Norris 2002a, 2004). Although important, this

literature has not yet provided empirical evidence assessing how, and under which

conditions, differences in individual predispositions define assessments of electoral

integrity. Even more, existing literature is silent on the effect of deep psychological

constructs – values and personality traits – on such assessments. However, given the

profound implications of such constructs for social and political behaviors, chances

are that perceptions of electoral integrity strongly depend on them. Thus, as noticed

by Newton and Norris (2000: 59), “because of their psychological history and make-

up, some individuals have an optimistic view of life and are willing to help others,

cooperate, and trust. Because of their own early life experiences, others are more

pessimistic and misanthropic. They are thus inclined to be guarded or alienated, more

distrustful and cautious of others, and pessimistic about social and political affairs and

about people and politicians in general”.

! 3!

The purpose of this paper is to explore the impact of two major psychological

constructs on citizens’ perceptions of electoral integrity: core values and personality

traits. Indeed, both values and personality traits have been shown to matter greatly for

attitudes, opinions and perceptions in the political field (Gerber et al. 2011). However,

to the best of our knowledge, there is no previous research examining the link

between values and/or personality traits and individual perceptions of electoral

integrity. In this paper, our aim is thus to fill this gap and investigate how values and

personality traits affect how citizens perceive the (mis)conduct of elections in their

country. Furthermore, setting a comparative research framework, we also show that

those individual-level relationships vary according to the political context.

Our paper unfolds as follows. First, we introduce the theoretical foundations of

individual values and personality traits and discuss some existing research linking

those psychological constructs to political trust, which allows us develop some

preliminary theoretical expectations. Second, we present the two datasets we rely on,

that is, the sixth wave of the World Value Survey (2010-2014) for individual data and

the Perceptions of Electoral Integrity Index (PEI data) for contextual-level

determinants. We also present the variables and methods used to test our hypotheses.

Third, we discuss our main results regarding the direct and mediated effects of values

and personality traits on individual perceptions of electoral integrity. The last section

concludes.

Values and personality traits as deep underpinnings of attitudes, perceptions and

behaviors

Basic personal values and personality traits are major psychological constructs. The

recent renewed interest in those individual characteristics has led to a plethora of

studies regarding their influence on many different types of psychological middle-

level constructs (e.g. attitudes) and behaviors, especially in the field of political

science1. However, to the best of our knowledge, there is no previous research on the

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!1 For an excellent review, see for instance Gerber et al. (2011).

! 4!

role of personality traits or basic values regarding individuals’ perception of electoral

integrity. Thus, systematic hypotheses have not yet been formulated nor empirically

investigated. Nonetheless, a few empirical researches have examined the link between

values or traits and political trust (e.g. Devos et al. 2002; Mondak and Halperin 2008;

Ackermann and Freitag, forthcoming; Morselli et al. 2012). As political trust “reflects

evaluations of whether or not political authorities and institutions are performing in

accordance with normative expectations held by the public” (Miller and Listhaug

1990: 358) and thus “stems from how people perceive political institutions to work”

(Luhiste 2006: 478, cited in Kong 2014: 2), it seems reasonable to think that political

trust and perception of electoral integrity are narrowly linked, as already shown by

some recent research (e.g. Nunnally 2011). Therefore, based on the few prior research

examining the link between personality and political trust, we can build up some

expectations about the effects of personality traits and basic values on individuals’

perception of electoral integrity.

Basic personal values and the perception of electoral integrity

The value concept is a central one in social sciences and its spread has led to many

different conceptualizations. In this paper, we adopt the theoretical approach

developed by Schwartz (1992, 1994) who defines human values (or “basic personal

values”) as “desirable, transsituational goals that vary in their importance as guiding

principles in people’s lives” (Roccas et al. 2002: 790). In this sense, values differ

from attitudes – which refer to object-specific evaluations – because of their

abstractness and generality (Schwartz 1992: 4). Accordingly, values are assumed to

be causally prior to attitudes, the latter being the concrete expression of the former.

Furthermore, values constitute “the criteria people use to select and justify actions and

to evaluate people (including the self) and events” (Schwartz 1992: 1). They guide

people’s choices and behavior by identifying desirable goals. As values are

hierarchically ordered by their relative importance, individuals use them to select (a

priori) or justify (a posteriori) choices and actions (Roccas et al. 2002: 790). As such,

they could influence the way people perceive, interpret, and respond to environmental

stimuli.

Schwartz (1992, 1994) distinguishes among 10 universal types of values according to

the type of motivational concern they express: universalism, benevolence, conformity,

! 5!

tradition, security, power, achievement, hedonism, stimulation and self-direction. As

those 10 different types can be compatible or may conflict to different degrees, the

dynamic relations are conceived as organized along a circular motivational continuum.

Moreover, the 10 types of values can be summed up in two broad dimensions: (1)

self-enhancement vs. self-transcendence, and (2) openness to change vs. conservation.

To begin with the first dimension, self-enhancement includes achievement and power

values. It refers to “the extent to which [people are motivated] to enhance their own

personal interests (even at the expense of others)” (Schwartz 1992: 44). Individuals

who foster those values tend to focus on their own social esteem. They are motivated

by personal success, prestige and control or dominance (Schwartz 1992: 8-9). Thus,

they should not express deep concern about electoral rules or political fairness.

Moreover, they tend to “take a negative view of human nature and have a disregard

for ethical means” (Ghosh 2009: 287). Indeed, empirical studies have reported a

strong relationship between self-enhancement values and (social) cynicism (e.g. Bond

et al. 2004; Chen et al. 2006; Leung et al. 2007). Because of this cynical world view,

we expect self-enhancement values to be negatively related to perception of electoral

integrity. In other words, individuals who pursue self-enhancement goals should

generally perceive politics and especially election procedures in a negative manner,

regardless of the political situation.

At the contrary, self-transcendence arrays universalism and benevolence values.

Those values motivate people “to transcend selfish concerns and promote the welfare

of others, close and distant, and of nature” (Schwartz 1992: 44). Individuals who

cherish those values are known to adopt more pro-social behaviors and to show a

great concern for common goods (e.g. Caprara et al. 2012; Schwartz 2010: 236).

Moreover, universalism values are related to a concern for equality and fairness in the

whole society (Schwartz, 1992: 7). As political corruption and electoral fraud

seriously undermine democracy, seeking those values should lead to a particular

attention to the electoral rules and thus an accurate perception of electoral integrity;

that is a negative one when facing unfair electoral contests and a positive one when

things go smoothly.

On the second dimension, self-direction and stimulation constitute the openness to

change type of values. It refers to the motivation for people to “follow their own

intellectual and emotional interests in unpredictable and uncertain directions”

(Schwartz 1992: 43). Individuals who pursue those goals value independence of

! 6!

thought and action, and seek novelty and challenges (Schwartz 1992: 6-8). Moreover,

they seem to be more likely than others to adopt critical attitudes, especially towards

authorities. This could be explained by the fact that when seeking individual

autonomy, “institutions are more likely to be appraised as constraining and less

confidence is prompted” (Morselli et al. 2012: 50). For people who hold openness

values, we thus expect a negative perception of political institutions in general, which

could lead to a pessimistic evaluation of electoral integrity even when elections are

carried out democratically.

Finally, in contrast to openness to change, conservation encompasses conformity,

tradition and security values. Individuals who endorse those values wish to “preserve

the status quo and the certainty it provides in relationships with close others,

institutions, and traditions” (Schwartz 2002: 43). Accordingly, those individuals

attach a great importance to rules, social norms and expectations, in order to preserve

social and political order. At the individual level, they value self-restriction, respect

and commitment: group membership is particularly important and thus individuals

have to conform themselves to the community’s rules in order to ensure stability and

predictability (Schwartz 2002: 9-10). Prior research has shown that people oriented

towards those values are more likely to trust institutions because they provide stability

and preserve social and political order (e.g. Devos et al. 2002; Morselli et al. 2012).

Accordingly, we expect conservation values to be related to a more positive

perception of electoral integrity.

Personality traits and the perception of electoral integrity

The concept of personality refers to “a multifaceted and enduring internal, or

psychological, structure” (Mondak 2010: 6) and traits might be described as the

“foundational personality dispositions” (Gerber et al. 2010: 111). They constitute a

highly consistent internal structure which is believed to “shape how [individuals]

respond to the vast array of stimuli they encounter in the world” (Gerber et al. 2011:

266). Because of their partially biological foundation, those core psychological

characteristics are believed to be quite stable throughout an individual’s life (Plomin

et al. 1990; Bouchard and Loehlin 2001; Van Gestel and Van Broeckhoven 2003;

McCrae and Costa 2003; Hatemi and McDermott 2011). They appear to exert an

influence on the development of other psychological constructs (i.e. “characteristic

! 7!

adaptations”, such as attitudes, interests, and so on) and also to predict behavior

(McAdams and Pals 2006: 207; Mondak et al. 2010: 2). In other words, personality

traits represent “dimensions of individual differences in tendencies to show consistent

patterns of thought, feelings, and actions” (McCrae and Costa 2003: 25).

Although there are several ways to describe – and measure – personality traits (see

John et al. 2008 for a good review), the most prominent one, which seems to reach

consensus among psychologists, is the so-called “Big-Five Model”. This holistic trait

model allows “to represent the central elements of trait structure with parsimonious

measures that are focused on five trait dimensions” (Mondak 2010: 9). According to

this view, personality traits can be combined into five broad factors: Openness to

experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism (an

inverse function of emotional stability). As they appear to be universal human

characteristics and thus to transcend cultural disparities (McCrae and Costa 1997),

they provide a powerful framework to investigate personality dispositions and its

influence on individuals’ behavior.

In the first place, openness to experience characterizes people who enjoy new,

stimulating ideas and experiences. Open-minded individuals tend to be creative,

curious and to think critically. This latter characteristic should logically lead to lower

levels of political trust. Though Ackermann and Freitag (forthcoming) found no direct

effects of this factor on institutional trust, Mondak and Halperin (2008) did found a

significant negative relationship. In other words, open citizens are more prone to

adopt a critical view about the way the political institutions work and then a general

distrustful stance towards them. They also should be particularly sensitive to any sign

that something is probably wrong. Consequently, it seems logical to expect those

people to have a negative perception of electoral integrity, especially when facing

unfair political situations.

Second, conscientiousness describes people who consistently follow rules in order to

preserve order, both at the societal and individual levels. They are perceived as

organized, responsible and reliable individuals, with a great sense of duty and a high

level of conformity. Furthermore, they appreciate structure and control. Recent

research found a negative relationship between conscientiousness and political trust

(Gabriel and Völkl 2005; Ackermann and Freitag, forthcoming). According to us, this

could be explained by the fact that conscientious citizens are less inclined to believe

that “the system […] will do what is right even in the absence of constant scrutiny"

! 8!

(Miller and Listhaug 1990: 358, cited in Kong 2014: 2, our emphasis). In other words,

because of their need for control, conscientious people are less prone than others to

express blind and absolute confidence in something. Accordingly, people who score

high on conscientiousness dimension should pay special attention to the scrupulous

observance of electoral rules. Consequently, we expect higher levels of

conscientiousness to be associated with an accurate perception of electoral integrity;

that is a negative evaluation in bad situations and a positive one in a fair electoral

context.

Third, people with a high level of extraversion are assumed to be particularly sociable

and dynamic. Moreover, extroverts tend to be optimistic and this could logically lead

to a higher level of both social and political trust. However, previous research finds

no significant effect of this factor on political trust (Gabriel and Völkl 2005; Mondak

and Halperin 2008; Ackermann and Freitag, forthcoming). We thus have no

theoretical expectations about the influence of extraversion on the perception of

electoral integrity.

Fourth, agreeableness refers to a tendency to please other people and to avoid conflict.

Thus, agreeable individuals are notably kind and sympathetic, and most often exhibit

cooperative and pro-social behavior. As noted by Mondak and Halperin (2008: 346),

“the most obvious political correlates of agreeableness are interpersonal and political

trust”. Indeed, past research has repeatedly found a robust positive relationship

between this personality dimension and both social and political trust (McCrae and

Costa 1996; Mondak and Halperin 2008; Hirashi et al. 2008; Dinesen et al. 2014;

Ackermann and Freitag, forthcoming). In other words, individuals highly agreeable

are generally trusting – to such an extent that general trust is often considered as a

“sub-trait” of agreeableness. We thus expect individuals who score high on

agreeableness dimension to have a markedly positive perception of electoral integrity,

even when they face unfair or corrupt electoral contests.

Finally, the last factor identified by the Big Five approach is neuroticism – an inverse

function of emotional stability. It indicates a tendency to feel anxious, nervous, and to

experience psychological distress or negative emotions. Moreover, individuals with

low levels of emotional stability are characterized by negativity and “view many

developments as unfair and often unsatisfactory” (Mondak and Halperin 2008: 345).

Accordingly, and even if prior work has found no relationship between this factor and

political trust (Gabriel and Völkl 2005; Mondak and Halperin 2008; Ackermann and

! 9!

Freitag, forthcoming), it seems logical to expect neuroticism to be associated with

pessimistic views of political functioning. Thus, for individuals higher on this last

dimension, the likelihood for a negative perception and evaluation of electoral

integrity should be high, whether things actually “go right” or not. In other words,

positive perceptions increase with increasing levels of emotional stability.

To sum up, regarding the influence of basic personal values, we expect conservation

to have a positive effect on the perception of electoral integrity, while self-

enhancement and openness to change values should lead to lower levels of perceived

integrity. With regard to personality traits, people who are open to experience or

characterized by a high level of neuroticism should hold a negative view of electoral

integrity, whereas a higher level of agreeableness should be associated to a positive

perception of political fairness2.

On the relationship between personal values and personality traits

As reflected by the previous section, basic values and personality traits are distinct

psychological constructs. In particular, traits are descriptive variables that depict an

individual’s patterns of feelings, thoughts, and behaviors; whereas values express

motivations or goals – and then may not be directly reflected in one’s behavior.

Thereby, while values refer to “what people consider important”, personality traits

refer to dispositions describing “what people are like” (Roccas et al. 2002: 790).

But at the same time, values and traits share certain similarities: (1) in their nature,

and (2) in their content (Parks-Leduc et al. 2014). First, in their nature because values

and traits are (at least partially) cognitively based. While this is true for all basic

values, the strength of traits’ cognitive component may vary because some traits (e.g.

neuroticism) are more emotionally based. At the contrary, openness to experience, for

example, has a strong cognitive component (Pytlik et al. 2002) and thus should be

highly correlated with some values – namely, openness to change values. Second,

regarding their content, some traits and values are close to each other. This is the case,

for example, for agreeableness which shares similarities in orientation with self-!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!2 As can be seen in the formulation of our hypotheses, notably regarding the influence of self-transcendence values and conscientiousness personality factor, we do not reject the possibility of a moderating effect of contextual-level determinants on individual-level relationships. However, due to the exploratory nature of our research, we choose not to go ahead in the theoretical discussion and formalization of such interaction effects. We will return to this question when discussing empirical results.

! 10!

transcendence values, especially with benevolence values. Indeed, both of them are

other-oriented and thus refer to a concern for others’ well-being, especially people in

one’s immediate circle. Empirical analyses support this view by establishing a strong

correlation between those traits and values (see Parks-Leduc et al. 2014; Fischer and

Boer 2014).

Thus, how are traits and values causally related? Some researchers have argued that

they constitute two different components of one’s personality (McAdams 1995;

Caprara et al. 2012; McCrae and Costa 2008; Saroglou and Munoz-Garcia 2008).

These kinds of integrative models generally view traits as antecedent (i.e. causally

prior) to values. According to this view, traits may influence values because people

tend to value the goals served by their own traits in order to justify their behaviors and

then reduce cognitive dissonance (Roccas et al. 2002: 791). Moreover, many studies

have already demonstrated the mediational role of values in linking personality traits

to different attitudes or behaviors (@@@). But as stated by Roccas and her

colleagues (2002) it could also be that values influence traits, because people

generally tend to act in a manner consistent with the values they cherish (Schwartz

1996). In sum, traits and values may likely be correlated and mutually influence one

other. Even if this is beyond the scope of this study, we must be aware of those

possible interplays and keep them in mind when interpreting our empirical results.

Data and methods

Our study combines data coming from two different sources: the World Value Survey

data (sixth wave) to measure variables at the individual level, and the Perceptions of

Electoral Integrity Index (PEI data) at the contextual level. Measures for individual

perceptions of electoral integrity (our dependent variable), values and personality

traits are available only for 22 countries out of the 57 in the WVS sixth wave. Among

those countries 9 (Brazil, Palestine, Hong Kong, Lebanon, Libya, Singapore, Tunisia,

Egypt, Yemen) do not have an overall score of electoral integrity in the 3.0 release of

the PEI index3. Our analyses are thus run on the remaining 13 countries, i.e., Algeria,

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!3 Note that some of those countries, for which an election occurred recently, have a electoral integrity score in the newest release of the PEI index (PEI 3.5, July 2015); we decided not to take those countries into account, however, given that the lapse between the election date and the survey dates is

! 11!

Colombia, Ecuador, Germany, India, Iraq, Jordan, Kuwait, Netherlands, Pakistan,

Rwanda, South Africa, and Thailand.

At the contextual level, we measure the quality of recent elections, for each country,

by relying on data gathered within the Electoral Integrity Project (Harvard University

and University of Sydney)4 . Developed within the project, the Perceptions of

Electoral Integrity Index (PEI; Norris 2014; Norris et al. 2014; Martinez i Coma and

van Ham 2015) relies on experts’ assessments of recent national elections. For each

national election (legislative or presidential) since June 2012 a survey has been sent to

a sample of local and international scholars working or having published on elections.

The survey asks to evaluate the “quality” of elections using 49 indicators, grouped

into eleven categories reflecting the whole electoral cycle (election laws, electoral

procedures, boundaries, voter registration, party and candidate registration, campaign

media, campaign finance, voting process, vote count, results and EMBs). The overall

PEI index is computed by adding the 49 separate indicators for each election and for

each country5.

Information about the 13 countries covered in our study (WVS survey dates, election

date, PEI scores) is detailed in Table 1 below.

[Table 1 about here]

To measure individual variables we rely on the sixth wave of the World Value Survey

data, which introduces for the first time several questions that allow us to provide an

estimation of individual perceptions of electoral integrity. Furthermore, to the best of

our knowledge, the sixth wave of the WVS provides virtually the only systematic

cross-sectional data that contain measures for both values and personality traits.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!too high to draw any consistent causal inference between individual perceptions and overall quality of those elections. 4 www.electoralintegrityproject.com 5 For more details see the « Year in Elections 2014 » report : https://sites.google.com/site/electoralintegrityproject4/projects/expert-survey-2/the-year-in-elections-2014.

! 12!

We measure individual perceptions of electoral integrity, our dependent variable, by

combining answers to nine questions asking respondents to provide evaluations of the

state of electoral processes in their countries. More specifically, respondents were

asked to evaluate, on a 4-points scale ranging from ‘very often’ to ‘not at all often’

how often in their country’s elections votes are counted fairly, opposition candidates

are prevented from running, TV news favor the governing party, voters are bribed,

journalists provide fair coverage of elections, election officials are fair, rich people

buy elections, voters are threatened with violence at the polls, and voters are offered

with a genuine choice in the elections. We reversed the coding of some variables (e.g.,

‘voters are bribed’) so that all items are coded positively, and computed a simple

additive scale that cumulates answers for all nine questions6. The resulting variable

ranges from 0 ‘very negative individual perception of electoral integrity’ to 27 ‘very

positive individual perception of electoral integrity’. A factor analysis (PCA)

confirmed the reliability of our additive measure, as our variable is strongly correlated

with the principal underlying factor extracted (Pearson’s R = .96***).

Unsurprisingly, average individual perceptions of electoral integrity strongly and

significantly differ between countries (Eta2 = .27***), as shown in Figure 1 below.

[Figure 1 about here]

Basic personal values were measured through the Portrait Values Questionnaire

(PVQ), a well-established measure of personal values developed by Schwartz

(Schwartz et al. 2001). The short version included in the World Values Survey (WVS)

consists of 11 short portraits of different people each describing “a person’s goals,

aspirations or wishes that point implicitly to the importance of a value” (Schwartz

2003: 273). Each item refers to one of the 10 motivational types (except for

universalism which is measured by two different statements). For each of them,

respondents had to indicate how similar to them is the person described, and thus

what is important to them. The original variables were first recoded to range from ‘not

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!6 Cronbach’s alpha = .75

! 13!

at all like me’ to ‘very much like me’. Then, 4 additive scales measuring the four

higher-order value orientations were computed by aggregating the relevant items:

openness to change (hedonism, stimulation, self-direction), self-transcendence

(universalism, benevolence), conservation (conformity, tradition, security) and self-

enhancement (power, achievement). Missing data systematically excluded.

We measured the five personality traits through a series 10 questions asking

respondents how much they agree with a series of qualifying sentences (e.g., ‘I'm

someone who is reserved’, ‘who has few artistic interests’); possible answers range on

a 5-points scale from ‘disagree strongly’ to ‘agree strongly’. Those questions bear a

strong similitude to the Ten Item Personality Measure (TIPI; Gosling et al. 2003),

which is considered the shortest survey inventory currently available to efficiently

measure personality traits (Gerber et al 2011: 267). To be sure, more complex

alternative measures of the five personality traits exist, form the very lengthy NEO-

Personality Inventory-Revised (NEO-PI-R; Costa and McCrae 1992) based on dozens

of items to the BFI inventory based on 44 questions (John et al. 1991; John et al.

2008). Those measures have the advantage of higher internal reliability, but are very

time-consuming to administer (Gerber et al 2011: 267).

We reversed the coding of some of the variables to provide comparable measures, and

computed five additive standardized scales, one for each personality trait: openness to

experience (many artistic interest, active imagination), conscientiousness (not lazy,

does a thorough job), agreeableness (never finds faults in others, trusting),

extraversion (not reserved, outgoing/sociable), and emotional stability (never gets

nervous, relaxed). Missing data were systematically excluded.

To increase robustness, our models are controlled by age (year of birth), gender,

education (9-points scale ranging from 1 ‘no formal education’ to 9 ‘university’),

interest in politics (from 0 ‘not at all interested’ to 3 ‘very interested’), satisfaction

with institutions (10-point scale). Respondents in the WVS were asked to assess the

importance of honest elections for them personally and for the economic development

of the country (1-4 scale); we use those two variables as additional controls, as they

might affect how individuals perceive the quality of elections. Descriptive statistics

for all individual variables are presented in Table 2 below.

! 14!

[Table 2 about here]

Empirics

Is the way individuals perceive the conduct of elections related to their innermost

traits and values? Empirical evidence discussed here shows that, indeed, this seems to

be the case. This section presents several models that provide information on the

direct and joint effect of values and personality traits on individual perceptions of

electoral integrity, in the 13 countries in our database.

First, the direct effect on individual perceptions of electoral integrity (IPEI) of each

individual value (self-transcendence, self-enhancement, openness to change, and

conservatism) and personality trait (openness to experience, conscientiousness,

extraversion, agreeableness, and emotional stability) appears as statistically

significant in Table 3 (last row).

[Table 3 about here]

These results are however not controlled by any intervening factors related to the

respondents’ sociodemographic profile or the other components of their innermost

values and personality traits. Furthermore, those results do not take into account the

fact that individual perceptions of electoral integrity are highly likely to be affected by

the quality of elections itself (that is, how elections in any given country scored in the

overall electoral integrity index). In order to control for the confounding effect of

those intervening factors we ran a series of multilevel OLS models able to take into

account that individual observations (respondents in the WVS data) are nested within

different countries, which in turn are characterized by rather different levels of

electoral integrity.

! 15!

First, Table 4 presents a “baseline” model (M1) in which individual perceptions of

electoral integrity are estimated as a function of sociodemographic factors (gender,

age, education) and other relevant controls (interest in politics, overall satisfaction

with institutions, importance of honest elections for the respondent personally and for

the economy). Model M1 also tests for the direct effect of country-level electoral

integrity (PEI classification). Results for this baseline model show that individual

perceptions of electoral integrity are significantly more positive when “honest”

elections are deemed important for the respondent him/herself personally and for the

economy, for male and more educated respondents, and when interest in politics and

overall satisfaction with institutions are higher. More importantly, our results show a

clear and positive association, ceteris paribus, between the experts’ assessment of

electoral integrity (PEI scores) and the way respondents in any given country assess

the quality of elections in their country: when experts assess elections as having a

high electoral integrity, respondents are strongly and significantly more likely to have

very positive perceptions of electoral integrity in their country, and vice-versa.

Although this baseline model does not provide any substantial information regarding

our main research question, it allows us to establish that (i) individual perceptions of

electoral integrity, ceteris paribus, strongly vary between individuals and that (ii)

contextual variations have to be taken into account.

[Table 4 about here]

Model M2 adds to this baseline model the direct effect of the four basic values, for

whose only self-enhancement has a significant effect. The direction of such effect –

negative – is consistent with our expectations, as individuals who foster self-

enhancement values tend to focus on their own social esteem, are motivated by

personal success, prestige and control or dominance (Schwartz 1992: 8-9), and tend to

“take a negative view of human nature and have a disregard for ethical means”

(Ghosh 2009: 287). Our results highlight that strong self-enhancement values

significantly decrease a positive evaluation of electoral integrity – that is, people high

in self-enhancement tend to assess the conduct of elections in their country as globally

poor and far from the ideal of “free and fair” elections, which corroborates the

! 16!

expectations that such individuals tend to have more cynical views of the social and

political world (e.g. Bond et al. 2004; Chen et al. 2006; Leung et al. 2007).

Model M3 replicates the same analysis, but instead of introducing the direct effect of

values as in M2 we now test for the direct effect of the five components of personality

according to the BFI model. Results highlight that all five personality traits have,

ceteris paribus, a rather strong and significant effect on individual perceptions of

electoral integrity. Furthermore, the direction of those effects is positive, that is,

individual perceptions of electoral integrity are higher with increasing levels of

openness to experience, conscientiousness, extraversion, agreeableness, and

emotional stability.

For some traits, the result is in line with our expectations. Agreeable people tend to

avoid conflict, usually display cooperative and pro-social behaviours, and score high

in interpersonal and political trust (McCrae and Costa 1996; Mondak and Halperin

2008; Hirashi et al. 2008; Dinesen et al. 2014), which should naturally lead to more

positive perceptions of electoral integrity. Similarly, individuals with low levels of

emotional stability are characterized by negativity and “view many developments as

unfair and often unsatisfactory” (Mondak and Halperin 2008: 345), which implies that

higher levels of emotional stability are more likely to result in positive evaluations,

even if no prior work has established a clear link between emotional stability and

political trust (Gabriel and Völkl 2005; Mondak and Halperin 2008; Ackermann and

Freitag, forthcoming). This is also the case for extraversion, for which no clear effect

on trust has been established but which usually implies optimistic views that could

lead to more positive assessments on the conduct of elections.

Results for openness to experience and conscientiousness are less straightforward.

First, individuals scoring high in openness to experience tend to think more critically,

and have lower political trust (Mondak and Halperin 2008), which should normally

lead to more dire perceptions of electoral integrity; our results suggest that the

opposite is at work, the relationship being clearly positive. The data we dispose of is

not detailed enough to pinpoint to specific causal underpinnings of the positive

association between openness to experience and perceptions of electoral integrity. We

might however imagine that, beyond fostering critical thinking, high openness to

experience creates favourable conditions for a less argumentative and antagonistic

approach to politics, as previous research showed positive effects for this trait on

! 17!

internal efficacy and the feeling that politics is not too complicated (Mondak 2010),

political efficiency (Vecchione and Caprara 2009), knowledge and interest in politics

(Mondak and Halperin 2008), and decreasing self-reported anger and hostility

(Tremblay and Ewart 2005). As we will discuss below, this effects becomes clearer

when we interact this specific personality trait with the actual level of electoral

integrity at the country level (PEI scores).

Second, we had no clear expectations for the direct effect of conscientiousness,

beyond the fact that respondents scoring high in this trait should provide an

assessment in line with the one made by the experts (that is, a positive assessment

when electoral integrity is high, and vice versa). Our results highlight that individuals

with high conscientiousness are significantly more likely to assess the quality of

elections positively.

Model M4 in Table 4 introduces the direct effect of both values and personality traits

simultaneously. Under mutual control, some interesting effects appear. First, the

direct effect of openness to experience is no longer statistically significant, whereas

the effects for the other four personality traits remain significant and of about the

same strength (and direction) as in the previous model (M3). More interestingly,

concerning values, self-enhancement no longer has a significant effect, and openness

to change appears as the only value having a significant direct effect on individual

perceptions of electoral integrity, ceteris paribus. This is in line with what we

expected: individuals high in openness to change value independence of thought and

action (Schwartz 1992). For them, “institutions are more likely to be appraised as

constraining and less confidence is prompted” (Morselli et al. 2012: 50), which leads

to a greater tendency towards critical attitudes, especially towards authorities.

The multilevel nature of the data we dispose of – that is, individual observations

nested within countries – allow us to assess the contextual conditions under which

respondents’ innermost traits (values and personality dimensions) affect perceptions

of electoral integrity. More specifically, our data allow us to test if different structural

situations regarding the conduct of elections in a given country affect the relationships

described before.

! 18!

[Table 5 about here]

Table 5 presents a final model in this sense (M5), in which individual perceptions of

electoral integrity are estimated as a function of sociodemographic and political

controls (our baseline model), the direct effect of values and personality traits, and

crosslevel interactions between country-level PEI scores and values and personality

traits.

Model M5 shows, first, that self-transcendence values interact significantly with high

levels of electoral integrity (PEI scores), and the direction of the interaction is as

expected: perceptions about the quality of elections in their country are significantly

more positive, when respondents score high in self-transcendence values and in

countries that know high levels of integrity according to PEI experts. Self-

transcendence values promote universalism, benevolence, and motivate people “to

transcend selfish concerns and promote the welfare of others, close and distant”

(Schwartz 1992: 44). Pro-social behaviors, pleas for common goods, equality and

universal fairness are common concerns for individuals with strong self-enhancement

values (Caprara et al. 2012; Schwartz 2010: 236; Schwartz, 1992: 7). We thus

expected those individuals to be particularly responsive of the actual quality of

elections in their country, and to provide a negative evaluation when elections

perform badly, and a positive evaluation when things go smoothly. Our results

provide empirical evidence that supports this expectation, as perceptions of integrity

for individuals who score high in self-transcendence are significantly more positive

when elections are evaluated positively by PEI experts, and negative when elections

perform poorly. This provides a first, substantial proof that how voters perceive

elections depends on a delicate balance between contextual and individual

determinants.

Turning to the joint effect of country-level PEI scores and personality traits, model

M5 highlights that openness to experience has statistically significant and positive

effect on individual perceptions of electoral integrity (as showed before in model M3),

but only when elections have an average or high integrity at the country level (PEI

scores); when elections are poorly rated, as for instance in Algeria, Iraq or Jordan,

increasing levels of openness to experience decrease the individual perception of

! 19!

electoral integrity (i.e., respondents rate elections more poorly). Figure 2 depicts

marginal changes in individual perceptions of electoral integrity as a function of the

interaction between openness to experience (x-axis) and the three values for country-

level PEI scores of electoral integrity (low, average and high). Respondents high in

openness to experience fosters critical thinking (Mondak and Halperin 2008), which

lead us to anticipate a direct negative effect on individual perceptions of electoral

integrity (that is, respondents high in openness to experience should be more critical

towards the actual quality of elections). Results for the direct effect of this personality

trait showed that, in reality, the opposite seems likely to happen as respondents high

in openness to experience are significantly more likely to have positive perceptions of

electoral integrity. This, we concluded, might be related to the fact that beyond

critical thinking openness to experience might foster a less antagonistic and

argumentative approach to politics, which could lead to more positive evaluation of

the political system itself (and the elections that are held within it).

[Figures 2 and 3 about here]

Results presented in Model M5 bring a new element to the picture, and nuance this

assessment. Indeed, respondents high in openness to experience have more positive

perceptions of electoral integrity, but this happens only when elections are evaluated

as positive (high or average integrity) by the independent experts in the PEI scores;

when this is not the case – that is, elections are of poor quality – openness to

experience decreases positive individual perceptions. A more complex than expected

logic seems to be at work here. Openness to experience, as shown in existing

literature, enhances critical thinking. The effect of such cognitive tool on individual

perceptions, in turn, depends on the evaluative situation people face: when the

situation is favourable, critical thinkers have the tools to recognize that and provide

positive assessments, and vice-versa they are able to pinpoint problematic situations

when facing negative realities. In other terms, openness to experience provides the

cognitive tools to properly assess a given situation. This moderation effect is far from

trivial, as it shows that personality traits strongly affect perceptions, but they do so in

accordance with the specific context in which the evaluation takes place. An

! 20!

additional interesting interaction effect is shown in model M5 between agreeableness

and PEI scores. In this case, as depicted in Figure 3, agreeableness globally enhances

positive perceptions of electoral integrity, and especially when the context is positive

(high PEI scores). In this sense, the context acts as a reinforcing factor on the

relationship between personality traits and individual perceptions.

Discussion and conclusion

@@@

! 21!

References Anderson, C. J. and Tverdova, Y. V. (2003) “Corruption, political allegiances, and attitudes

toward government in contemporary democracies”, American Journal of Political Science, 47(1): 91-109.Birch, S. (2007) “Electoral systems and electoral misconduct”, Comparative Political Studies, 40(12) : 1533-56.

Birch, S. (2008) “Electoral institutions and popular confidence in electoral processes: A crossnational analysis”, Electoral Studies, 27(2): 305-320.

Birch, S. (2010) “Perceptions of electoral fairness and voter turnout”, Comparative Political Studies December, 43(12): 1601-1622.

Bond, M. H., Leung, K., Au, A., Tong, K.-K. and Chemonges-Nielson, Z. (2004) “Combining social axioms with values in predicting social behaviours”, European Journal of Personality, 18(3): 177-191.

Bouchard, T. J. Jr. and Loehlin, J. C. (2001) “Genes, evolution and personality”, Behavior Genetics, 31(3): 243-273.

Caprara, G.-V., Alessandri, G. and Eisenberg, N. (2012) “Prosociality: The contribution of traits, values, and self-efficacy beliefs”, Journal of Personality and Social Psychology, 102(6): 1289-1303.

Chen, S. X., Fok, H.-K., Bond, M. H. and Matsumoto, D. (2006) “Personality and beliefs about the world revisited: Expanding the nomological network of social axioms”, Personality and Individual Differences, 41: 201-211.

Costa, P. T. and McCrae, R. R. (1992) NEO PI-R Professional Manual, Odessa: Psychological Assessment Resources, Inc.

Devos, T., Spini, D. and S. H. Schwartz (2002) “Conflicts among human values and trust in institutions”, British Journal of Social Psychology, 41: 481-494.

Dinesen, P. T., Nørgaard, A. S. and Klemmensen, R. (2014) “The civic personality: Personality and democratic citizenship”, Political Studies, 62(1): 134-152.

Elklit, J. (1999) “Electoral institutional change and democratization: You can lead a horse to water, but you can’t make it drink”, Democratization, 6(4): 28-51.

Elklit, J. and Reynolds, A. (2002) “The impact of election administration on the legitimacy of emerging democracies”, Commonwealth and Comparative Politics, 40(2): 86-119.

Elklit, J. and Reynolds, A. (2005) “A framework for the systematic study of election quality”, Democratization, 12(2): 147-62.Fischer, R. and Boer, D. (2014) “Motivational basis of personality traits: A meta-analysis of value-personality correlations”, Journal of Personality, DOI: 10.1111/jopy.12125.

Freitag, M. and Ackermann, K. (2015) “Direct democracy and institutional trust: Relationships and differences across personality traits”, Political Psychology, forthcoming.

Gabriel, O. W. and Völkl, K. (2005) “Persönlichkeitseigenschaften und institutionenvertrauen”, in Schumann, S. (ed) Persönlichkeit. Eine vergessene Grösse der Sozialforschung, Wiesbaden: Verlag für Sozialwissenschaften, pp. 175-192.

Gerber, A. S., Huber, G. A., Doherty, D. and Dowling, C. M. (2011) “The big five personality traits in the political arena”, Annual Review of Political Science, 14: 265-287.

! 22!

Gerber, A. S., Huber, G.A., Doherty, D. and Dowling, C. M. (2010) “Personality and political attitudes: Relationships across issue domains and political contexts”, American Political Science Review, 104(1): 111-133.

Ghosh, A. (2009) “Social axioms and individualistic-collectivist orientations in Indian college students”, in Leung, K. and Bond, M. H. (eds) Psychological Aspects of Social Axioms. Understanding Global Belief Systems, New-York: Springer, pp. 283-292.

Gosling, S., Rentfrow, P. and Swann, W. (2003) ‘A very brief measure of the Big-Five personality domains’, Journal of Research in Personality 37(6): 504.528.

Hartlyn, J. and McCoy, J. (2006) “Observer paradoxes: How to assess electoral manipulation”, in Schedler, A. (ed) Electoral authoritarianism: The dynamics of unfree competition, Boulder and London: Lynne Rienner, pp. 41-54.

Hatemi, P. K. and McDermott, R. (eds) (2011) Man is by nature a political animal. Evolution, biology, and politics, Chicago: The University of Chicago Press.

Hirashi, K., Yamagata, S., Shikishima, C. and Ando, J. (2008) “Maintenance of genetic variation in personality through control of mental mechanisms: A test of trust, extraversion, and agreeableness”, Evolution and Human Behavior, 29(2): 79-85.

John, O. P., Donhaue, E. M., and kentrle, R. L. (1991) ‘The Big Five Inventory--Versions 4a and 54’, Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research.

John, O. P., Naumann, L. P. and Soto, C. J. (2008) “Paradigm shift to the integrative big five trait taxonomy: history, measurement, and conceptual issues”, in John, O. P., Robins, R. W. and Pervin, L. A. (eds) Handbook of Personality Psychology: Theory and Research (3rd ed.), New-York: The Guilford Press, pp. 114-158.

Kong, D. T. (2014) “Perceived competence and benevolence of political institutions as culturally universal facilitators of political trust: Evidence from Arab countries”, Cross-Cultural Research, 48(4): 385-399.

Lehoucq, F. E. (2003) “Can parties police themselves? Electoral governance and democratization”, International Political Science Review, 23(1): 29-46.

Leung, K., Au, A., Huang, X., Kurman, J., Niit, T. and Niit, K. K. (2007) “Social axioms and values: A cross-cultural examination”, European Journal of Personality, 21: 91-111.

Luhiste, K. (2006) “Explaining trust in political institutions: Some illustrations from the Baltic states”, Communist and Post-Communist Studies, 39: 475-496.

Lyons, T. (2004) “Post-conflict elections and the process of demilitarizing politics: The role of electoral administration”, Democratization, 11(3): 36-62.

Martínez i Coma, F., & van Ham, C. (2015) ‘Can experts judge elections? Testing the validity of expert judgments for measuring election integrity’, European Journal of Political Research, 54(2): 305-325.

McAdams, D. P. (1995) “What do we know when we know a person?”, Journal of Personality, 63: 365-396.

McAdams, D. P. and Pals, J. L. (2006) “A new big five: Fundamental principles for an integrative science of personality”, American Psychologist, 61: 204-217.

McCrae, R. R. and Costa, P.T. (1996) “Toward a new generation of personality theories: theoretical contexts for the five-factor model”, in Wiggins, J. S. (ed) The five-factor model of personality: theoretical perspectives, New-York: The Guilford Press, pp. 51-87.

! 23!

McCrae, R. R. and Costa, P.T. (1997) “Personality trait structure as a human universal”, American Psychologist, 52: 509-516.

McCrae, R. R. and Costa, P.T. (2003) Personality in Adulthood. A five-factor theory perspective, New-York: The Guilford Press.

McCrae, R. R. and Costa, P.T. (2008) “The five-factor theory of personality”, in John, O. P., Robins, R. W. and Pervin, L. A. (eds) Handbook of personality: Theory and Research (3rd ed.), New-York: The Guilford Press.

Miller, A. H. and Listhaug, O. (1990) “Political parties and confidence in government: A comparison of Norway, Sweden and the United States”, British Journal of Political Science, 20: 357-386.

Mishler, W. and Rose, R. (2001a) “What are the origins of political trust? Testing institutional and cultural theories in post-communist societies”, Comparative Political Studies, 34(1): 30–62.

Mondak, J. J. (2010) Personality and the Foundations of Political Behavior, Cambridge: Cambridge University Press.

Mondak, J. J. and Halperin, K. D. (2008) “A framework for the study of personality and political behaviour”, British Journal of Political Science, 38(2): 335-362.

Mondak, J. J., Hibbing, M. V., Canache, D., Seligson, M. A. and Anderson, M. R. (2010) “Personality and civic engagement: an integrative framework for the study of trait effects on political behavior”, American Political Science Review, 104(1): 85-110.

Morselli, D., Spini, D. and Devos, T. (2012) “Human values and trust in institutions across countries: A multilevel test of Schwartz’s hypothesis of structural equivalence”, Survey Research Methods, 6(1): 49-60.

Mozaffar, S. (2002) “Patterns of electoral governance in africa’s emerging democracies”, International Political Science Review, 23(1).

Mozaffar, S. and Schedler, A. (2002) “The comparative study of electoral governance – introduction”, International Political Science Review, 23(1): 5-27.

Newton, K. and Norris, P. (2000) “Confidence in public institutions: Faith, culture, or performance”, in Pharr, S. J. and Putnam, R. D. (eds) Disaffected democracies: What’s troubling the trilateral countries, Princeton: Princeton University Press, pp. 52-73.

Norris, P. (2002) “Ballots not bullets: Testing consociational theories of ethnic conflict, electoral systems, and democratization”, in Reynolds, A. (ed) The architecture of democracy: Constitutional design, conflict management, and democracy, Oxford: Oxford University Press, pp. 206-247.

Norris, P. (2004) Electoral engineering: Voting rules and political behavior, Cambridge: Cambridge University Press.

Norris, P. (2014). Why electoral integrity matters. Cambridge University Press.

Norris, P., Frank, R. W., & Martínez i Coma, F. (2014) ‘Measuring Electoral Integrity around the World: A New Dataset’ PS: Political Science & Politics, 47(4): 789-798.

Nunnally, S. C. (2011) “(Dis)counting on democracy to work: Perceptions of electoral fairness in the 2008 Presidential Election”, Journal of Black Studies, 42(6): 923-942.

Parks-Leduc, L., Feldman, G. and Bardi, A. (2014). “Personality traits and personal values: A meta-analysis”, Personality and Psychology Review, 19: 3-29.

! 24!

Plomin, R., DeFries, J. C. and McClearn, G. E. (1990) Behavioral genetics: A primer, New-York: Freeman.

Pytlik Zillig, L. M., Hemenover, S. H., and Dienstbier, R. A. (2002) “What do we assess when we assess a Big 5 trait? A content analysis of the affective, behavioral, and cognitive processes represented in Big 5 personality inventories”, Personality and Social Psychology Bulletin, 28: 847-858.

Roccas, S., Sagiv, L., Schwartz, S. H. and Knafo, A. (2002) “The big five personality factors and personal values”, Personality and Social Psychology Bulletin, 28(6): 789-801.

Saraglou, V. and Munoz-Garcia, A. (2008) “Individual differences in religion and spirituality: An issue of personality traits and/or values”, Journal for the Scientific Study of Religion, 47: 83-101.

Schedler, A. (2002a) “Elections without democracy: The menu of manipulation”, Journal of Democracy, 13(2).

Schelder, A. (2002b) “The nested game of democratization by elections”, International Political Science Review, 23(1): 103-22.

Schedler, A. (2006) “The logic of electoral authoritarianism”, in Schedler, A. (ed) Electoral authoritarianism: The dynamics of unfree competition, Boulder and London: Lynne Rienner, pp. 1-23.

Schwartz, S. H. (1992) “Universals in the content and structure of values: theoretical advances and empirical tests in 20 countries”, Advances in Experimental Social Psychology, 25: 1-65.

Schwartz, S. H. (1994) “Are there universal aspects in the content and structure of values?”, Journal of Social Issues, 50: 19-45.

Schwartz, S. H. (1996) “Value priorities and behavior: Applying a theory of integrated value systems”, in Seligman, C., Olson, J. M. and Zanna, M. P. (eds) The Ontario Symposium: Vol. 8. The psychology of values, Hillsdale: Erlbaum, pp. 1-24.

Schwartz, S. H. (2003) “A Proposal for Measuring Value Orientations across Nations”, Chapter 7 in the Questionnaire Development Package of the European Social Survey. Website: http://www.europeansocialsurvey.org/index.php?option=com_docman&task=doc_view&gid=126&Itemid=80.

Schwartz, S. H. (2010) “Basic values: How they motivate and inhibit prosocial behavior”, in Mikulincer, M. and Shaver, P. R. (eds) Prosocial motives, emotions, and behavior: The better angels of our nature, Washington: American Psychological Association, pp. 221-241.

Schwartz, S., Melech, G., Lehmann, A., Burgess, S., Harris, M., and Owens, V. (2001) “Extending the cross-cultural validity of the theory of basic human values with a different method of measurement”, Journal of Cross-Cultural Psychology, 32: 519-542.

Tremblay, P. and Ewart, L. (2005) ‘The Buss and Perry Aggression Questionnaire and its relations to values, the Big Five, provoking hypothetical situations, alcohol consumption patterns, and alcohol expectancies’, Personality and Individual Differences 38: 337-346.

Van Gestel, S. and Van Broekhoven, C. (2003) “Genetics of personality: are we making progress?”, Molecular Psychiatry, 8(10): 840-852.

Vecchione, M. and Caprara, G. V. (2009) ‘Personality determinants of political participation: The contribution of traits and self-efficacy beliefs’, Personality and Individual Differences 46: 487-492.

! 25!

Tables and figures

Table 1: Countries and elections details Country WVS wave 6 survey dates Election date (type) PEI score (level) Algeria 17.12.2012 – 12.01.2013 17.04.2013 (Presidential) 54.2 (low) Colombia 17.11.2012 – 06.12.2012 09.03.2014 (Legislative) 69.1 (average) Ecuador 10.06.2013 – 29.08.2013 17.02.2013 (Presidential) 63.9 (average) Germany 22.07.2013 – 13.11.2013 22.09.2013 (Legislative) 84.2 (high) India 24.10.2013 – 17.04.2014 12.05.2014 (Legislative) 67.0 (average) Iraq 18.11.2013 – 29.11.2013 30.04.2014 (Legislative) 55.2 (low) Jordan 19.02.2014 – 01.03.2014 23.01.2013 (Legislative) 56.6 (low) Kuwait 01.01.2014 – 20.03.2014 27.06.2013 (Legislative) 66.9 (average) Netherlands 01.12.2012 – 31.12.2014 12.09.2012 (Legislative) 82.6 (high) Pakistan 04.10.2012 – 19.10.2012 11.05.2013 (Legislative) 60.0 (average) Rwanda 28.11.2012 – 10.12.2012 06.09.2013 (Legislative) 71.2 (high) South Africa 18.08.2013 – 06.10.2013 07.05.2014 (Legislative) 70.3 (high) Thailand 01.09.2013 – 20.09.2013 02.02.2014 (Legislative) 60.6 (average)

Table 2: Descriptive statistics for individual variables Variable Valid obs. Mean Std. dev. Min Max Individual perception of electoral integrity 18928 14.30 5.98 0 27 Honest elections: important personally 20335 2.24 1.02 0 3 Honest elections: important for economy 20314 2.33 0.95 0 3 Female 20525 0.50 0.50 0 1 Year of birth 20519 1972.63 16.23 1914 1999 Education 20394 5.19 2.38 1 9 Interest in politics 20443 1.47 0.98 0 3 Satisfaction with institutions 20437 6.25 2.33 1 10 Value: Self-transcendence 18023 0.75 0.18 0.17 1 Value: Self-enhancement 20142 0.65 0.20 0.17 1 Value: Openness to change 19971 0.67 0.18 0.17 1 Value: Conservatism 20087 0.74 0.18 0.17 1 Personality: Openness to experience 11785 0.64 0.17 0.25 1 Personality: Conscientiousness 13619 0.72 0.21 0.25 1 Personality: Extraversion 13652 0.66 0.16 0.25 1 Personality: Agreeableness 13270 0.67 0.19 0.25 1 Personality: Emotional stability 12529 0.66 0.18 0.25 1

! 26!

Table 3: Bivariate correlations of main variables Self

trans. Self enhanc.

Openn. change

Conserv Openn. exp.

Consc. Extrav. Agreeab. Emo. stab.

Self-transcendence 1 Self-enhancement .34*** 1 Openness to change .40*** .53*** 1 Conservatism .63*** .40*** .38*** 1 Openness to exp. -.02 .05*** -.01 -.06*** 1 Conscientiousness -.01 .04 -.07*** .02 .36*** 1 Extraversion -.01 .02* .04*** -.04*** .03** -.01 1 Agreeableness -.07*** -.06*** -.05*** -.07*** .15*** .38*** -.10*** 1 Emotional stability .01 -.04*** -.04*** -.07*** .05*** .28*** .06*** .27*** 1 IPEI -.05*** -.14*** -.12*** -.14*** .11*** .16*** .05*** .21*** .17*** *p<0.05, **p<0.01, ***p<0.001

Table 4: Multilevel determinants of individual perceptions of electoral integrity (IPEI) – direct effects M1 M2 M3 M4 Coeff. Se Sig. Coeff. Se Sig. Coeff. Se Sig. Coeff. Se Sig. Constant 45.12 (5.40) *** 96.53 (5.85) *** 78.00 (9.84) *** 55.86 (10.75) *** Honest elections: - important personally 0.37 (0.05) *** 0.30 (0.05) *** 0.12 (0.10) 0.24 (0.11) * - important economy 0.25 (0.05) *** 0.17 (0.06) ** 0.20 (0.12) 0.16 (0.12) Female -0.29 (0.07) *** -0.26 (0.08) ** -0.05 (0.15) -0.10 (0.16) Year of birth -0.02 (0.00) *** -0.04 (0.00) *** -0.04 (0.00) *** -0.03 (0.01) *** Education 0.13 (0.02) *** 0.18 (0.02) *** 0.15 (0.03) *** 0.17 (0.03) *** Interest in politics 0.47 (0.04) *** 0.59 (0.04) *** 0.56 (0.08) *** 0.49 (0.08) *** Satisf with institutions 0.12 (0.02) *** 0.16 (0.02) *** 0.12 (0.03) *** 0.10 (0.03) *** Country PEI: - Average integrity 1.26 (1.64) 0.81 (0.12) *** 1.83 (0.24) *** 1.85 (0.24) *** - High integrity 3.97 (1.77) * 1.69 (0.13) *** 4.73 (0.24) *** 3.40 (0.26) *** Values: - Self-transcendence . -0.23 (0.32) . -0.31 (0.61) - Self-enhancement . -2.21 (0.26) *** . 0.45 (0.46) - Openness to change . -0.07 (0.30) . -1.08 (0.50) * - Conservatism . -0.35 (0.32) . -0.70 (0.60) Personality traits: - Openness to exp. . . 1.87 (0.46) *** 0.91 (0.49) - Conscientiousness . . 2.09 (0.40) *** 2.80 (0.42) *** - Extraversion . . 1.82 (0.47) *** 2.15 (0.49) *** - Agreeableness . . 3.59 (0.43) *** 2.47 (0.45) *** - Emotional stability . . 1.94 (0.43) *** 1.86 (0.44) *** N (individuals) 18,492 16,012 5,041 4,491 N (countries) 13 12 13 12 R2 .12 .09 .23 .16 Note: Dependent variable is individual perceptions of electoral integrity, continuous. Models are random-effects GLS regressions, estimated with Stata 12.1. *p<0.05, **p<0.01, ***p<0.001

! 27!

Table 5: Multilevel determinants of individual perceptions of electoral integrity (IPEI) – crosslevel interactive effects (country PEI * values/personality traits) M5 Coeff. Se Sig. Constant 45.07 (11.10) *** Honest elections: - important personally 0.25 (0.11) * - important economy 0.14 (0.12) Female -0.04 (0.15) Year of birth -0.02 (0.01) *** Education 0.14 (0.03) *** Interest in politics 0.49 (0.08) *** Satisfaction with institutions 0.09 (0.03) ** Country PEI: - Average integrity 0.47 (2.37) - High integrity -2.90 (2.60) Values: - Self-transcendence -1.55 (1.26) - Self-enhancement 1.08 (1.04) - Openness to change -1.49 (1.03) - Conservatism 1.44 (1.49) Personality traits: - Openness to exp. -5.29 (1.18) *** - Conscientiousness 4.44 (1.09) *** - Extraversion 2.92 (1.29) * - Agreeableness 0.87 (0.93) - Emotional stability 2.63 (0.91) ** Country PEI * values: - Average integrity * Self-transcendence 0.32 (1.50) - High integrity * Self-transcendence 4.12 (1.82) * - Average integrity * Self-enhancement 0.54 (1.21) - High integrity * Self-enhancement -2.38 (1.44) - Average integrity * Openness to change 1.11 (1.22) - High integrity * Openness to change -0.19 (1.55) - Average integrity * Conservatism -0.92 (1.68) - High integrity * Conservatism -3.78 (1.94) Country PEI * personality traits: - Average integrity * Openness to exp. 7.64 (1.33) *** - High integrity * Openness to exp. 7.19 (1.56) *** - Average integrity * Conscientiousness -1.92 (1.24) - High integrity * Conscientiousness -3.12 (1.43) * - Average integrity * Extraversion -2.29 (1.45) - High integrity * Extraversion 0.78 (1.58) - Average integrity * Agreeableness 0.70 (1.13) - High integrity * Agreeableness 4.67 (1.38) *** - Average integrity * Emotional stability -2.68 (1.08) * - High integrity * Emotional stability 2.29 (1.34) N (individuals) 4,491 N (countries) 12 R2 .18 Note: Dependent variable is individual perceptions of electoral integrity, continuous. Models are random-effects GLS regressions, estimated with Stata 12.1. *p<0.05, **p<0.01, ***p<0.001

! 28!

Figure 1: Mean and standard deviations of individual perceptions of electoral integrity, by country (WVS data, wave 6)

Figure 2: Individual perceptions of electoral integrity by openness to experience and levels of electoral integrity

0

5

10

15

20

25

30

05

1015

2025

IPEI

0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1Openess to experience

Low electoral integrity Average electoral integrityHigh electoral integrity

! 29!

Figure 3: Individual perceptions of electoral integrity by agreeableness and levels of electoral integrity

!

05

1015

2025

IPEI

0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1Agreeableness

Low electoral integrity Average electoral integrityHigh electoral integrity