Personality’s effect on European identification

Abstract

A superordinate identity improves intergroup relations and bolsters support for the political system. Yet, why do only some identify superordinately? I argue that personality is an important determinant. I test this using an original survey in the United Kingdom, where European Union integration has increased the salience and feasibility of the “European” identity option in addition to a national one. Several Big Five traits matter: openness and extraversion increase identification with Europe while agreeableness decreases it. Mediation analysis subsequently shows that personality’s effects also travel through the mechanisms of risk aversion, knowledge, and ideology. Results imply that certain predispositions prompt some to be more receptive than others to seeing themselves in superordinate terms and that European identification may be at least partly more primordial than previously thought.

Keywords

Big Five causal mediation analysis European Union personality superordinate identification

Introduction

A superordinate identity that encompasses multiple subgroups under an overarching umbrella is often viewed as desirable—if not outright necessary—for peaceful intergroup relations, political legitimacy, and democratic success. This kind of identity helps individuals of different ethnic or racial backgrounds get along by reducing bias and increasing tolerance (Gaertner et al., 1993; Gaertner and Dovidio, 2000). Its effects transfer to the policy sphere by making members of the majority subgroup more supportive of redistributive policies for minorities (Transue, 2007). A superordinate identity also greases the wheels of democracy by instilling a sense of shared fate among citizens, making them more willing to participate and sacrifice for the benefit of all (Hooghe and Marks, 2004; Kritzinger, 2003). Unsurprisingly, people are more supportive of their political system and more compliant with institutional rules when they identify with the polity (Gibson and Caldeira, 1995; McDonough, 1995). In sum, collective identity is vital for political cohesion since achieving the common good necessitates overcoming parochial interests (Huddy, 2003). So why are some more likely than others to identify superordinately?

I argue that personality provides an answer: certain predispositions make people more amenable to seeing themselves in inclusive terms. After reviewing several reasons why individuals would not identify at the superordinate level, I justify expectations for how the Big Five personality traits of openness to experience, conscientiousness, extraversion, agreeableness, and emotional stability should help citizens overcome these psychological barriers. I test these projections on political-territorial identities in Europe, where European Union (EU) officials ponder how to construct a supranational identity that will bolster support for integration. Using original data from a nationally representative survey in the United Kingdom (UK), I find several traits indeed play a role. I then conduct causal mediation analysis to illuminate three mechanisms through which personality’s effects travel: risk aversion, political knowledge, and ideology. These results provide a novel explanation for whether and why individuals see themselves in superordinate terms and re-raise important questions about the nature of identity formation.

Psychological barriers to superordinate identification

Despite its appeal, I submit that superordinate identification is actually quite counterintuitive. Humans have an innate need for order and understanding, which they achieve through categorization and comparison. To make sense of a disorderly world, people engage in “prejudicial thinking” and attempt to fit everything new into their existing mental framework (Allport, 1954; Sears et al., 2003). This translates into a strong preference for homophily; love of one’s own reigns. The old adage, “birds of a feather flock together,” is based on the enduring observation that similarity connects people more quickly and easily than almost anything else (McPherson et al., 2001). This tendency is supported by myriad evidence¹ and appears quite justifiable. So how does it come about?

Cognitive development starts early in life as children emulate what they observe in their family setting. They construct a bubble of what they know and interpret the rest of the world in relation to it (Piaget, 1972). Once this schema is in place, people selectively perceive new information and rationalize away anything that contradicts their existing beliefs (Aronson, 1969). Dissonant information is physically and psychologically uncomfortable, so individuals go to great lengths to minimize— or altogether ignore—inconsistencies (Festinger, 1957). This produces two complementary inclinations. The first, disconfirmation bias, acknowledges that people actively disregard arguments that challenge their priors (Edwards and Smith, 1996). The second, confirmation bias, notes that individuals purposely seek out information that conforms to their expectations (Nickerson, 1998).

These two cognitive proclivities are reinforced by affective, emotional memories tied to past experience (Taber and Lodge, 2006). This means individuals are naturally motivated to be skeptical of anything different, without even realizing the strength of their convictions. Change is difficult because these powerful subconscious biases make us reject anything that does not easily align with our preconceptions. Altogether, these established tenets of homophily, cognitive dissonance, and (dis)confirmation bias suggest people are inherently wired to love who and what is most like them while resisting or rejecting new, contradictory, or unfamiliar things. In the modern world, political-territorial identification is dominated by attachment to the nation-state (Gellner, 1983). This implies that where national identity is deeply ingrained, citizens should be hesitant to identify with a superordinate level beyond the nation-state. And yet, as the case of Europe illustrates, a substantial portion do.

Superordinate identity in context: The case of the EU

With over 500 million citizens across 28 member states, the EU stands as the world’s foremost example of supranational governance and is but the latest in a long history of attempts to form a superordinate identity. Integration has made “Europe” a salient category for self-identification. The EU’s motto, “United in Diversity,” proclaims its desire to build a sense of community among those within it. EU officials have echoed this sentiment for decades (Tugendhat, 1977). Similarly, scholars prescribe a common European identity to help reduce the EU’s democratic deficit (Rohrschneider, 2002), overcome public opposition to integration (Carey, 2002; Kritzinger, 2003), and combat anti-immigrant sentiment (Curtis, 2014; Sides and Citrin, 2007). Thus, many insist that helping individuals identify superordinately will boost EU support and curtail prejudice. Yet, given the psychological aspects mentioned above, it is no wonder citizens vary in the propensity and strength of their European identification. Cross-national Eurobarometer data consistently show that many more identify as national than European (Figure 1). On a scale of 0–3 (not at all to a great extent), European identification is much lower and more variable (mean = 2.083; SD = 0.858) than national identification (mean = 2.783, SD = 0.545).

Figure 1.

Strength of national versus European identification in 27 EU member states.

This is unsurprising given how well-entrenched national identities are in Europe (Díez Medrano, 2003). But what is less instinctive are those who do identify with Europe. What characteristics might help explain the 31% of citizens who claim feeling European “to a great extent”? Why have some apparently overcome the cognitive barriers to superordinate identification while others have not?

The Big Five personality traits

I contend that personality helps explain how likely people are to extend their sense of self to include the superordinate category. Personality traits are stable and predictable internal systems that determine affective, cognitive, and motivational responses to the world around us (John and Srivastava, 1999). Particularly relevant is the fact that these traits develop early in life and then persist relatively unchanged over time (Caprara et al., 2006; Hibbing et al., 2011; Mondak et al., 2010). This stability makes personality exogenous to almost any subsequent attitude or behavior.²

The primary personality typology comes from the “Big Five” dimensions of openness to experience, conscientiousness, extraversion, agreeableness, and emotional stability (John and Srivastava, 1999; McCrae and Costa, 1992). These traits are widely recognized as the most appropriate test of personality’s effects (Mondak et al., 2010; Schoen and Schumann, 2007), and numerous studies underscore their cross-cultural validity (e.g. Schmitt et al., 2007). Openness to experience is demonstrated by intellectual curiosity, creativity, wide interests, and aesthetic appreciation. Conscientiousness reflects self-discipline, organization, reliability, and adherence to social norms. Extraversion measures how energetic, assertive, self-confident, talkative, and socially oriented one is. Agreeableness captures those who are compassionate, sympathetic, and kind. Finally, emotional stability is defined as being calm, confident, and relaxed.

Numerous studies explore personality’s political effects.³ Most relevant here, personality often plays a profound role in partisan identification (Cooper et al., 2013; Gerber et al., 2011b, 2012; Vecchione et al., 2011), so it should theoretically affect other types of identification as well.

Direct effects: Personality's role in superordinate identification

Openness has already been suggested as “the most important personality trait in terms of impact on identity development” (Ozer and Benet-Martínez, 2006: 408). Associated with cognitive flexibility (Carney et al., 2008), citizens who score highly on the openness dimension should, quite literally, be more open to seeing themselves as part of a larger group (such as a supranational community like Europe). Openness is also linked with acceptance of unconventional things (Lewis and Bates, 2011). Since subgroup attachments like national identity are by far the norm (Citrin and Sides, 2004), superordinate identification requires that individuals be willing to see themselves in a whole new way. European integration is anything but traditional, both in terms of organizational structure and geographic scope, so citizens must be open to expanding their horizons beyond the national level. Hence, openness to experience should be positively associated with greater superordinate identification.

H1: The more open one is to new experiences, the more she will identify with the superordinate group.

Conscientiousness is associated with the psychological need for structure and order (Costa and McCrae, 1992), so conscientious citizens may have difficulty envisioning themselves within a large, seemingly distant superordinate group. In Europe, the disorganized, abstract nature of the EU may be especially off-putting to these individuals. Furthermore, conscientious people feel a strong obligation to conform with societal expectations (Gallego and Oberski, 2011). Given that patriotism and national allegiance are the typical qualities of good citizenship (Theiss-Morse, 2009), conscientiousness should be negatively related to superordinate identification.

H2: The more conscientious one is, the less she will identify with the superordinate group.

Extraverts tend to be highly active and have larger, more diverse social networks (Gallego and Oberski, 2011; Wolak and Marcus, 2007). They may thus be more likely to interact with people from other subgroups (here, citizens of other EU member states), which could—per the “Contact Hypothesis” (Allport, 1954; Pettigrew, 1998)—help them internalize commonalities and be more amenable to superordinate self-categorization. Additionally, extraversion is associated with greater self-esteem (Mondak et al., 2010), so extraverted individuals, being more confident in their preexisting identities, may feel less threat from extending their sense of self even further.

H3: The more extroverted one is, the more she will identify with the superordinate group.

While friendly and compassionate, agreeable people go to great lengths to maintain social cohesion and are very conflict-avoidant (Gerber et al., 2011a). Thus, regarding identification, agreeableness should reinforce proclivities for the familiar, subordinate group. Given that higher agreeableness predicts national identification (Duckitt and Sibley, 2016), it may then hamper supranational identification if the superordinate option is perceived as incompatible. Additionally, highly agreeable individuals tend to possess more traditional values (Roccas et al., 2002) which, especially in the European case, should inhibit superordinate identification.

H4: The more agreeable one is, the less she will identify with the superordinate group.

Finally, emotionally stable individuals exhibit relaxed demeanors and low levels of anxiety (Gerber et al., 2011a), while neurotic individuals tend to dislike politics (Freitag and Ackermann, 2015) and are particularly sensitive to threats of social exclusion (Hibbing et al., 2011). In this way, emotional stability may induce greater superordinate identification because these individuals should be less susceptible to social pressure to maintain subgroup primacy and may not perceive as much threat from being part of a superordinate group.

H5: The more emotionally stable one is, the more she will identify with the superordinate group.

Indirect effects: Personality’s potential pathways

An understanding of personality’s effect on identification is incomplete without an assessment of the causal mechanisms, and recent research suggests personality may actually have a greater indirect than direct effect on political outcomes (e.g. Schoen and Schumann, 2007; Wolak and Marcus, 2007). Several factors may mediate personality’s influence in addition to the basic relationships just postulated.

The role of risk aversion

Behavioral economics consistently finds that people dislike uncertainty and are fearful of choices that could negatively impact their future welfare. Prospect theory further asserts that individuals judge a decision by the gains and losses it incurs relative to some inherent reference point. Instead of always choosing the option that maximizes gains, individuals assess the cost/benefit tradeoff asymmetrically depending on what is at stake (Kahneman and Tversky, 1979). Because people value what they have more than what they do not, they engage in significantly more risky behavior to avert loss than to achieve comparable gain (Camerer, 2005; Levy, 2003). This implies that citizens will be most risk-averse in the domain of gains. Whether issues are framed as gains or losses, then, affects one’s willingness to take risks (Mercer, 2005).⁴

These considerations should apply to identification, as most superordinate identities are promoted as additional options for self-categorization rather than something to completely replace preexisting attachments. Risk-averse individuals may feel more threatened by the superordinate alternative and thus be less likely to self-identify this way. Especially when national identity is the strongly engrained reference point, identifying with Europe could be perceived as risky, making risk aversion negatively related to superordinate identification:

H6: The more risk-averse an individual is, the less she will identify with the superordinate group.

Importantly, however, individuals vary quite substantially in their proclivity for risk (Arceneaux, 2012; McDermott, 2004). So what determines risk-aversion in the first place? Personality is a starting factor. Despite its conceptual similarity to the Big Five traits, risk aversion is well established as an independent dimension; in fact, personality is already a known determinant of it. Existing findings show risk aversion is often rooted in low openness, extraversion, and emotional stability combined with high agreeableness and conscientiousness (Ehrlich and Maestas, 2010; Nicholson et al., 2005). The ultimate contribution here is to note risk aversion’s as-yet unexamined effect on identification, both as a direct cause and as a mediating pathway.

H7: Personality determines how risk-averse someone is, which then affects her likelihood of identifying with the superordinate group.

The benefits of political knowledge

Politically sophisticated individuals should better grasp the complex, abstract nature of a superordinate group, particularly when it pertains to a multilevel polity like the EU. Cognitive mobilization was an original explanation for identification with Europe (Inglehart, 1970), and recent work confirms its enduring influence (Bellucci et al., 2012). Extending upon this, I expect that knowledge improves superordinate identification by making the broader identity more palatable to those who are more informed about its existence.

H8: The more objectively knowledgeable an individual is, the more she will identify with the superordinate group.

Again, there is reason to expect that personality predates how much people know about politics. Openness to experience causes individuals to be more intellectually curious and more politically savvy (Gerber et al., 2011a; Mondak et al., 2010). This translates into a desire to seek out additional, alternative information; one study found that openness increased an individual’s desire to learn more about political issues by 30% (Wolak and Marcus, 2007). Contrarily, extraverted individuals tend to be less informed about politics (Mondak and Halperin, 2008). Thus, a second mediational hypothesis is as follows:

H9: Personality determines how politically knowledgeable an individual is, which then affects her likelihood of identifying with the superordinate group.

The importance of ideology

Exclusive nationalism is highly correlated with far-right ideology (Druckman, 1994; Mummendey et al., 2001) and a greater propensity to support the radical right (Givens, 2005), whereas more “postmaterialist” values (emphasizing self-expression and group solidarity vs. physical and economic security) are typical of those on the left (Inglehart and Abramson, 1999). Therefore, those with a right-leaning political ideology should be more hesitant to identify beyond the nation. Fligstein (2009: 141) confirms this tendency; others show right-leaning ideology dampens EU support more generally (e.g. Aspinwall, 2002).

H10: The more conservative an individual is, the less she will identify with the superordinate group.

Several studies highlight personality’s effect on ideological orientation. Conscientiousness is positively related to conservatism (Carney et al., 2008; Gerber et al., 2010) and negatively associated with a liberal viewpoint (Caprara et al., 2006; Cooper et al., 2013). Agreeableness leads to more progressive economic views (Gerber et al., 2010). And liberals appear to be significantly more open-minded (Caprara et al., 2006; Carney et al., 2008; Gerber et al., 2010).

H11: Personality determines an individual’s ideological orientation, which then affects her likelihood of identifying with the superordinate group.

Data, measurement, and method

I test the above hypotheses using an original Political Attitudes and Identities Survey conducted in the UK. The London-based polling firm Opinium Research, LLP administered the online survey to a nationally representative sample of UK citizens between 28 June and 3 July 2012. Opinium invited 6913 members of their standing panel to participate; 2123 respondents completed the survey for a total response rate of 30.71%.⁵

The UK offers a prime opportunity to assess the above expectations. Although personality’s effects should be identical regardless of where it is studied empirically (Sibley and Duckitt, 2008), the UK actually provides a difficult test case because its aggregate levels of identification with Europe are lower than any other EU member state (Bruter and Harrison, 2012). UK citizens remain much more Eurosceptic than citizens of other EU member states (Spiering, 2004) and most in the UK do not see themselves as very European, thus making it harder for variation in personality to explain the few instances of those who do.

Dependent variable

Superordinate identification is operationalized as an individual’s self-professed degree of attachment to Europe using the question,

“We are all part of different groups. Some groups are more important to us than others when we think of ourselves. How important are each of the following in describing how you personally see yourself? Please consider the following scale, where 7 means that you identify very strongly with it and 1 means that you do not identify with it at all.”

This wording builds off of Bruter (2005) and is preferred because it measures an individual’s political-territorial loyalties independently of one another rather than forcing respondents to select their primary attachment (Sinnott, 2005). Respondents were asked separately about their identification with Europe versus the UK. Figure 2 displays the distribution of responses to these alternatives. The overall mean of Identification with Europe is only 3.064 (SD = 1.600), compared with 4.728 (SD = 1.592) for national identification. Clearly, few in the UK see themselves as very European to begin with.

Figure 2.

Distribution of political-territorial identification in the UK.

Measuring the Big Five

I measure personality using an adapted version of Goldberg’s (1992) format as utilized in the 2005 National Jury Survey that several prominent analyses employ.⁶ Respondents received the following prompt, followed by 25 pairs of bipolar adjectives (five items per trait):

“The following section contains pairs of words. On a scale of zero to ten, please indicate which of the words best describes you. For example, the number zero means ‘relaxed’, the number ten means ‘tense’, and the number five is exactly in the middle—neither relaxed nor tense. On this scale, what number best describes you? You can use any number from 0 to 10.”

After confirming a five-factor solution through principal components factor analysis, I generated mean indices for each Big Five dimension. Openness is based on the extent to which one is “imaginative,” “creative,” “curious,” and “intellectual”; excluding “analytical” increased its reliability, so I ultimately use only four items for this trait (α = 0.731). Conscientiousness reflects being “systematic,” “hardworking,” “neat,” “careful,” and “responsible” (α = 0.697). Extraversion consists of those who are “extraverted,” “talkative,” “bold,” “spontaneous,” and “outgoing” (α = 0.855). Agreeableness captures people who are most “warm,” “gentle,” “kind,” “polite,” and “sympathetic” (α = 0.843). Lastly, Emotional Stability indicates that a person is “calm,” “relaxed,” “at ease,” “steady,” and “content” (α = 0.816).

Mediating variables

Risk Aversion ranges from one to seven based on strength of agreement with the statement, “I like taking risks.” Answers were recoded so that higher responses indicate less willingness to take risks. Objective Knowledge reflects a mean index of 10 items based on a series of true/false statements testing respondents’ factual correctness regarding EU and UK affairs (α = 0.748). The final measure ranges from zero to one with 11 possible values, indicating low to high comprehension. Lastly, L-R Ideology is based on respondents’ self-placement on an 11-point scale of zero (far left) to 10 (far right).

Controls

My choice of covariates is informed by the standard sociodemographic characteristics prevalent in extant findings. First, younger citizens often identify with Europe more than their older counterparts (Citrin and Sides, 2004; Green, 2007; Pichler, 2008). Age ranges from 18 to over 80. Second, education is associated with greater superordinate identification because it increases “exposure” to Europe (Bellucci et al., 2012; Spannring et al., 2008). Education ranges from zero to five depending on one’s highest level of schooling: none, high school, some higher education but no degree, a university undergraduate degree, or an advanced degree. Third, those employed in high socioeconomic status jobs report feeling more European (Fligstein, 2009; Pichler, 2008). Occupational Prestige utilizes a standard measure of social grade based on field of employment. Responses for five categories were recoded so that higher values correspond to more skilled and managerial jobs. Fourth, males may report feeling more European than females (Fligstein, 2009; Green, 2007), and living in a cosmopolitan area is linked to stronger identification with Europe (Pichler, 2008). Gender is a dichotomous variable coded “one” for Male; Urban ranges from one to four based on the size of community in which one lives. Finally, those who possess the ethnolinguistic or religious attributes deemed desirable for group membership typically identify strongly with that group (Theiss-Morse, 2009). Though cultural considerations are more common predictors of national identification (Ruiz Jiménez et al., 2004), I tap ethnicity and religion with dummies for whether one is White (a physical marker associated with UK national identification that should thus be negatively signed) and Christian (which should be positively related to European identification as many retain a cultural view of the EU as a predominantly Christian “club”).

Given the range of the dependent variable, the analyses below employ ordinary least squares (OLS) regression. Results are robust to ordered logit as well (see online Appendix). I report OLS coefficients for ease of interpretation and because mediation analysis cannot yet accommodate an ordinal mediator or dependent variable (Imai et al., 2013). Standard errors are clustered by the 12 UK regions to control for sampling procedures and any underlying contextual differences between individuals nested within the same geographic unit. Lastly, all explanatory variables were standardized to range from zero to one so that the substantive effect of moving from their minimum to maximum is immediately comparable.

Baseline results

Table 1, Model 1 first presents the sociodemographic predictors by themselves. Like past studies, education is positive and significant. And, as predicted, being White has a significant negative effect while being Christian has a significant positive one. Contrary to prior findings, age, gender, and occupational prestige seem not to matter. Other recent work also finds sociodemographics unimportant (e.g. Bakker and de Vreese, 2016), perhaps because the cleavages previously associated with these elements have become less politically divisive.

Table 1.

Personality’s effect on identification with Europe.

	(1)	(2)
	Baseline model	Adding personality
Personality variables
Openness		0.622* (0.266)
Conscientiousness		−0.457 (0.419)
Extraversion		0.489** (0.141)
Agreeableness		−0.772** (0.229)
Emotional Stability		0.289 (0.287)
Sociodemographic controls
Age	0.081 (0.237)	0.096 (0.226)
Male	0.0910 (0.089)	0.0340 (0.091)
Education	0.400** (0.112)	0.332* (0.117)
Occupational Prestige	0.279 (0.165)	0.234 (0.150)
Urban	0.097 (0.147)	0.085 (0.143)
White	−0.467* (0.172)	−0.436* (0.162)
Christian	0.222* (0.093)	0.231* (0.088)
Constant	2.920** (0.267)	2.987** (0.201)
N	1704	1704
R ²	0.021	0.036

OLS: ordinary least squares.

OLS regression estimates with robust standard errors in parentheses.

p < 0.01, *p < 0.05

Next, Model 2 adds in the Big Five personality traits. Hypotheses 1, 3, and 4 are immediately confirmed. Openness and Extraversion are both positive and significant, showing that more intellectually curious and outgoing individuals are indeed more likely to see themselves in a superordinate light. Agreeableness’s effect is negative and significant, confirming that these types of people may hesitate to identify beyond the norm of strong national identity. Since these individuals have a high need for social acceptance and strongly dislike disagreement, they likely fear not comporting with the stereotypical UK norm of low European identification. In other words, agreeable citizens may be catering to social pressure from co-nationals not to identify superordinately if they feel it will jeopardize fellow citizens’ view of them. Hypotheses 2 and 5 are also supported, as the coefficients on Conscientiousness and Emotional Stability are in the expected directions; they fail to reach statistical significance, so their ultimate effects remain speculative. In terms of substantive importance, personality plays a stronger role than any sociodemographic aspect. Agreeableness has the largest impact of any variable: moving from the least to most agreeable reduces identification with Europe by −0.772 on the seven-point scale. Openness and extraversion take second and third place, increasing superordinate identification by 0.622 and 0.489, respectively. Ethnicity comes next at −0.436, followed by education at 0.332 and religion at 0.231.

Mediation analysis

To better understand the relationship between personality and identity, I turn to isolating the causal mechanisms at work. Mediation analysis facilitates just this; however, its inherent challenges are increasingly documented (Imai et al., 2011; Keele, 2015). Traditional mediation analysis consists of regressing the dependent variable on the explanatory variable, regressing the mediator on the explanatory variable, and regressing the dependent variable on both the explanatory variable and the mediator together (Baron and Kenny, 1986). Mediation exists when the relationship between the explanatory and dependent variables is weaker or altogether insignificant in the presence of the mediating variable.

Importantly, recent advances now advocate nonparametric identification using structural equation modeling within a potential outcomes framework to decompose a treatment’s effect into its statistically distinguishable direct and indirect effects.⁷ This more rigorous approach also assesses results’ sensitivity to the sequential ignorability assumption, which presumes that both the explanatory variable (treatment) and mediator are exogenous and that no confounding omitted variable bias exists (Imai et al., 2010). Since this assumption cannot be tested directly, one instead estimates the correlation between the errors (rho) of the latter two models described above at which the observed average causal mediation effect (ACME) would lose significance (Hicks and Tingley, 2011a). To do this, I employ Hicks and Tingley’s (2011b) mediation package for Stata. This cannot accommodate multiple mediators at once, so the proposed mechanisms are tested separately for each Big Five trait, controlling for the other four traits and sociodemographic characteristics.⁸

Mechanism 1: Risk aversion

All five traits significantly determine one’s aversion to risk (Table 2, Model 3). Consistent with others’ findings, openness, extraversion, and emotional stability are negatively related, while conscientiousness and agreeableness are positively related. Since all variables have the same range (0–1), the coefficient size indicates the magnitude of each trait’s effect. Extraversion has the strongest impact at −0.342; emotional stability contributes least at −0.159. Apart from personality, risk aversion increases slightly with age (which has the third-strongest impact), and males are less likely to be risk-averse than females (an almost-zero substantive effect).

Table 2.

The mediating role of Risk Aversion.

	(3)	(4)	(5)
	Identification w/ Europe	Risk Aversion	Identification w/ Europe
Personality variables
Openness	0.622** (0.269)	−0.312*** (0.040)	0.372 (0.272)
Conscientiousness	−0.457 (0.302)	0.200*** (0.045)	−0.297 (0.302)
Extraversion	0.489* (0.265)	−0.342*** (0.040)	0.216 (0.269)
Agreeableness	−0.772*** (0.295)	0.222*** (0.044)	−0.594** (0.295)
Emotional Stability	0.289 (0.253)	−0.159*** (0.038)	0.161 (0.252)
Mediating variable
Risk Aversion			−0.800*** (0.161)
Sociodemographic controls
Age	0.096 (0.176)	0.223*** (0.026)	0.274 (0.178)
Male	0.034 (0.080)	−0.082*** (0.012)	−0.0313 (0.081)
Education	0.332** (0.161)	0.015 (0.024)	0.344** (0.160)
Occupational Prestige	0.234 (0.163)	−0.036 (0.024)	0.205 (0.162)
Urban	0.085 (0.116)	0.002 (0.017)	0.086 (0.115)
White	−0.436*** (0.159)	0.044* (0.024)	−0.401** (0.158)
Christian	0.231*** (0.079)	−0.005 (0.012)	0.227*** (0.078)
Constant	2.987*** (0.304)	0.678*** (0.046)	3.530*** (0.321)
N	1704	1704	1704
R ²	0.036	0.197	0.050

OLS: ordinary least squares.

OLS regression estimates with standard errors in parentheses.

Column heading indicates dependent variable in each model.

***

p < 0.01, **p < 0.05, *p < 0.1

Model 5 supports Hypotheses 6 and 7: not only is risk aversion associated with less identification with Europe, but most of the Big Five’s direct effects disappear when risk aversion is included; only that of agreeableness remains. Though personality’s ultimate impact is largely indirect, it is still important via this mediating pathway as confirmed by significant ACMEs: 0.251 for openness, −0.159 for conscientiousness, 0.275 for extraversion, −0.177 for agreeableness, and 0.129 for emotional stability.⁹ In other words, risk aversion captures much of personality’s total effect, something particularly common when variables’ direct and indirect effects are oppositely signed (Imai et al., 2010). Table 5 clarifies the nature of this attenuation and compares, among other things, the percent of total effect mediated for each trait and mediator. More than half (52.97%) of extraversion’s effect on identification with Europe is mediated by risk aversion, as is 39.05% of openness and 31.89% of conscientiousness.

Table 5.

Summary of mediation effects.

Causal pathway	Average causal mediation effect	Sig.	Direct effect	Sig.	Total effect	Sig.	% of Total effect mediated	Rho at which ACME = 0
Openness → Risk Aversion	0.251	**	0.377		0.627	**	39.05%	−0.120
Conscientiousness → Risk Aversion	−0.159	**	−0.292		−0.451		31.89%	“ ”
Extraversion → Risk Aversion	0.275	**	0.220		0.495		52.97%	“ ”
Agreeableness → Risk Aversion	−0.177	**	−0.589	**	−0.766	**	23.22%	“ ”
Emotional Stability → Risk Aversion	0.129	**	0.165		0.294		33.10%	“ ”
Openness → Objective Knowledge	0.091	**	0.534	**	0.625	**	14.15%	0.079
Extraversion → Objective Knowledge	−0.069	**	0.561	**	0.492		−13.14%	“ ”
Conscientiousness → L-R Ideology	−0.163	**	−0.288		−0.451		32.80%	−0.111
Agreeableness → L-R Ideology	0.230	**	−0.995	**	−0.766	**	−30.01%	“ ”
Emotional Stability → L-R Ideology	−0.065	**	0.359		0.294		−16.89%	“ ”

ACME: average causal mediation effect.

Note: Obtained using Hicks and Tingley’s (2011b) ‘mediation’ package for Stata.

p < 0.05

Mechanism 2: Knowledge

Table 3 tests whether objective political knowledge is another causal pathway through which the Big Five’s effects travel. As Model 7 demonstrates, openness is indeed positively and significantly associated with more accurate awareness of national and EU affairs; after age, it has the strongest effect on knowledge followed by education and extraversion (which is negatively related to factual correctness). In the full model (Model 8), extraversion and agreeableness are the only personality traits to maintain a direct effect on identification with Europe, but again, the ACME results in Table 5 corroborate Hypothesis 9 and indicate that significant mediation of both openness (0.091) and extraversion (−0.069) is taking place. Hypothesis 8 is also substantiated since objective knowledge is positive and significant on its own. It has the second-strongest substantive effect on identification after agreeableness (which is not mediated by knowledge), followed by openness and extraversion (which are).

Table 3.

The mediating role of Objective Knowledge.

	(6)	(7)	(8)
	Identification w/ Europe	Objective Knowledge	Identification w/ Europe
Personality variables
Openness	0.622** (0.269)	0.163*** (0.038)	0.530* (0.270)
Conscientiousness	−0.457 (0.302)	−0.069 (0.042)	−0.418 (0.301)
Extraversion	0.489* (0.265)	−0.120*** (0.037)	0.557** (0.265)
Agreeableness	−0.772*** (0.295)	−0.003 (0.041)	−0.770*** (0.294)
Emotional Stability	0.289 (0.253)	0.011 (0.035)	0.283 (0.252)
Mediating variable
Objective Knowledge			0.565*** (0.173)
Sociodemographic controls
Age	0.096 (0.176)	0.257*** (0.025)	−0.0493 (0.181)
Male	0.034 (0.080)	0.115*** (0.011)	−0.0309 (0.0822)
Education	0.332** (0.161)	0.146*** (0.023)	0.249 (0.163)
Occupational Prestige	0.234 (0.163)	0.077*** (0.023)	0.190 (0.163)
Urban	0.085 (0.116)	0.030* (0.016)	0.0678 (0.116)
White	−0.436*** (0.159)	−0.035 (0.022)	−0.416*** (0.159)
Christian	0.231*** (0.079)	0.014 (0.011)	0.223*** (0.0787)
Constant	2.987*** (0.304)	0.050 (0.043)	2.959*** (0.303)
N	1704	1704	1704
R ²	0.036	0.196	0.042

OLS: ordinary least squares.

OLS regression estimates with standard errors in parentheses.

Column heading indicates dependent variable in each model.

***

p < 0.01, **p < 0.05, *p < 0.1

The percentages of total effects mediated by objective knowledge are 14.15% for openness and −13.14% for extraversion, helping detect an additional mechanism through which personality affects identification.

Mechanism 3: Ideology

A final route through which personality’s effects may run is ideological orientation. In keeping with others’ findings, Table 4, Model 10 shows that conscientiousness and agreeableness significantly affect ideology. Emotional stability also matters here, while, in contrast to extant studies, openness does not. Conscientiousness and emotional stability are linked to more conservative views, while agreeableness is linked to less conservative ones. Agreeableness has the strongest pull (−0.272 compared to 0.195 for conscientiousness and 0.079 for emotional stability).

Table 4.

The mediating role of Ideology.

	(9)	(10)	(11)
	Identification w/ Europe	L-R Ideology	Identification w/ Europe
Personality variables
Openness	0.622** (0.269)	0.017 (0.036)	0.636** (0.268)
Conscientiousness	−0.457 (0.302)	0.195*** (0.040)	−0.293 (0.302)
Extraversion	0.489* (0.265)	0.054 (0.035)	0.534** (0.264)
Agreeableness	−0.772*** (0.295)	−0.272*** (0.039)	−1.000*** (0.298)
Emotional Stability	0.289 (0.253)	0.079** (0.034)	0.355 (0.252)
Mediating variable
L-R Ideology			−0.840*** (0.182)
Sociodemographic controls
Age	0.096 (0.176)	0.021 (0.023)	0.113 (0.175)
Male	0.034 (0.080)	0.015 (0.011)	0.0468 (0.0795)
Education	0.332** (0.161)	−0.064*** (0.021)	0.278* (0.161)
Occupational Prestige	0.234 (0.163)	0.036* (0.022)	0.264 (0.162)
Urban	0.085 (0.116)	−0.040*** (0.015)	0.0511 (0.116)
White	−0.436*** (0.159)	−0.028 (0.021)	−0.460*** (0.158)
Christian	0.231*** (0.079)	0.052*** (0.010)	0.274*** (0.0790)
Constant	2.987*** (0.304)	0.506*** (0.040)	3.413*** (0.316)
N	1704	1704	1704
R ²	0.036	0.065	0.048

OLS: ordinary least squares.

OLS regression estimates with standard errors in parentheses.

Column heading indicates dependent variable in each model.

***

p < 0.01, **p < 0.05, *p < 0.1

Model 11 supports Hypothesis 10 since L-R Ideology is negative, significant, and substantively strong; more conservative individuals are less likely to identify superordinately. Openness and extraversion retain independent positive effects on identification with Europe; agreeableness also remains strong and negative, per usual. Even though the direct effects of conscientiousness and emotional stability are insignificant in Model 11, they still have a significant indirect impact on identification via their contribution to one’s ideological position. Table 5 confirms Hypothesis 11 with significant ACME results of −0.163 for conscientiousness, 0.230 for agreeableness, and −0.065 for emotional stability. The percentage of total effect mediated by ideological orientation is 32.8% for conscientiousness, −30.01% for agreeableness, and −16.89% for emotional stability. Once more, the central conclusion is that ideology mediates several traits’ impact.

The above analyses outline numerous ways that personality affects superordinate identification. In addition to direct effects, a fair amount of its power flows through the intermediary variables of risk aversion, objective political knowledge, and ideology. Importantly, no mediating relationship was found in tests that reversed the treatment and mediator. Neither risk aversion nor objective knowledge nor ideology’s effects were significantly mediated by any Big Five traits, thereby boosting confidence in the causal order posited here. When all three mechanisms are included simultaneously with the Big Five traits in one overarching analysis (Table 6, Model 12), agreeableness and extraversion still maintain independent effects. In this complete model, ideology emerges as the strongest determinant of superordinate identification (−0.842), closely followed by agreeableness (−0.826), then risk aversion (−0.780), objective knowledge (0.471), ethnicity (−0.409), extraversion (0.324), and religion (0.264). In sum, seeing oneself as part of the superordinate group is partially determined—both directly and indirectly—by aspects of one’s personality.

Table 6.

Determinants of Identification with Europe.

	(12) Complete model
Personality variables
Openness	0.316 (0.273)
Conscientiousness	−0.104 (0.421)
Extraversion	0.324** (0.132)
Agreeableness	−0.826** (0.293)
Emotional Stability	0.226 (0.266)
Mediating variables
Risk Aversion	−0.780*** (0.156)
Objective Knowledge	0.471*** (0.115)
L-R Ideology	−0.842*** (0.185)
Sociodemographic controls
Age	0.166 (0.203)
Male	−0.0710 (0.0883)
Education	0.221* (0.123)
Occupational Prestige	0.200 (0.137)
Urban	0.0381 (0.136)
White	−0.409** (0.149)
Christian	0.264** (0.0917)
Constant	3.919*** (0.291)
N	1704
R ²	0.067

OLS: ordinary least squares.

OLS regression estimates with robust standard errors in parentheses.

***

p < 0.01, **p < 0.05, *p < 0.1

Conclusion

Identification is a complex psychological process, and identification with a superordinate political-territorial group seems improbable since people naturally seek to surround themselves with similar peers. Why, then, identify in superordinate terms? In the case of the EU, what allows some citizens to overcome the cognitive biases that prevent others from seeing themselves as European?

Linking personality to political-territorial identification adds an important explanation for why individuals differ in the extent of their inclusivity. I show that certain predispositions make citizens more receptive to superordinate identification than others. Highly open and extraverted individuals are more likely to adopt a superordinate identity. In contrast, those with an agreeable personality appear much less open to superordinate self-categorization; this trait retains one of the strongest effects across all models. A major implication of these findings is that some individuals inherently experience difficulty extending their sense of self because particular traits affect the extent to which a superordinate identity is perceived to conflict with preexisting, subordinate identity attachments.

Three causal mechanisms (risk aversion, knowledge, and ideology) transfer personality’s effect forward, as well as affect identification on their own. First, risk aversion significantly mediates all five traits’ effects (to the point that little direct impact is left after accounting for this indirect conduit). This hints that superordinate European identification may be more of an instrumental calculation than a deep emotional attachment like its national counterpart and helps explain why many citizens might be hesitant to adopt a supranational identity in addition to their existing ties. Second, more open and introverted individuals have greater factual accuracy about politics; objective political knowledge then mediates these traits’ effects on superordinate identification. Third, much of the influence of conscientiousness, agreeableness, and emotional stability runs through ideological orientation, insinuating that personality leads individuals to sort themselves into party affiliations and media preferences that only reinforce their prior identity inclinations. Altogether, these channels help explain why it is much more common for citizens to identify with a subordinate (e.g. national) than a superordinate (e.g. European) group.

While identities may be politically constructed and malleable (Fearon and Laitin, 2000), these results re-raise the question of whether identification (the extent to which one feels personally attached to a particular identity) is, instead, something people have less control over—something at least partly more primordial than previously thought. This issue deserves further attention as it has implications for intergroup relations and democratic success, especially in multicultural polities attempting to establish a cohesive overarching identity. Though at face value these findings suggest identification is significantly less manipulable than policymakers may hope, predispositions rarely operate in isolation; they are often conditioned by situational and cultural context (Freitag and Ackermann, 2015; Greenstein, 1967). And while I show that certain personality types are more likely to identify with Europe, personality is not the only prerequisite of superordinate identification. Other elements affect it as well, including the three mechanisms studied here. Thus, policymakers wishing to increase collective identification could stress the additional gains to be had by seeing oneself superordinately rather than allowing integration to be framed as a risk that challenges preexisting allegiances. They could also coordinate greater information campaigns to increase citizens’ objective knowledge of the effort. For now, recognizing that personality plays at least somewhat of a role opens the door for EU officials to adjust their communication accordingly (Bakker and de Vreese, 2016).¹⁰

Future research should extend this analysis in several ways. Most obviously, these findings stem from a single country. While the relationship at hand should theoretically extend to any superordinate political-territorial identity, and though I have attempted to justify why the UK is a helpful test case, results should be replicated elsewhere to confirm their generalizability. Since mean Big Five levels appear quite similar across country contexts—especially among neighboring states (e.g. Schmitt et al., 2007)—we would expect the same traits to matter in the same way for superordinate identification, at least in other EU member states. Likewise, the data here come from a cross-sectional snapshot in time. Though the early formation and subsequent stability of personality offers strong causal leverage to refute criticisms of endogeneity (Block and Block, 2006; Caprara et al., 2006; Costa and McCrae, 1992), static analyses cannot speak to dynamic trends in identification over time (Ruiz Jiménez et al., 2004). Thus, additional methodological approaches should be utilized to further solidify causality. Next, future analyses might parse out which causal mechanism matters most and what conditional circumstances temper or amplify personality’s tendencies. For instance, education often moderates certain traits’ effect (Canache et al., 2013; Osborne and Sibley, 2015), and women are significantly more risk-averse than men (Borghans et al., 2009). The interactive relationships between these elements and the findings above could generate further tangible policy recommendations. Finally, knowing that European identity indeed operates as a superordinate identity in terms of reducing outgroup hostility and increasing inclusivity (Curtis, 2014), it would be valuable to investigate personality’s link to other instances of superordinate attachment as well.

Altogether, my findings shed light on the psychological processes individuals experience when confronted with a superordinate identity alternative. Continuing to determine the causes of superordinate identification will elucidate ways in which scholars and policymakers can attain its beneficial consequences: increased political legitimacy and stability, enhanced citizen willingness to sacrifice for the common good, and greater intergroup harmony in diverse societies. Each of these facets affects political collectivities’ ability to endure, making the discernment of factors that promote or hinder superordinate identification key to achieving peace and prosperity.

Footnotes

Acknowledgments

Earlier versions of this paper were presented at the 71st Annual Meeting of the Midwest Political Science Association, Chicago, 11–14 April 2013 and the “Mass Attitudes and the Political Economy of European and Global Integration” workshop hosted by the Colorado European Union Center of Excellence, 30–31 January 2014. I gratefully thank Jennifer Fitzgerald, Joseph Jupille, Matthew Hibbing, Sara Hobolt, David Leblang, Jennifer Wolak, and all conference participants for their helpful comments on earlier drafts, as well as Opinium Research, LLP for its help administering this survey.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This survey was made possible by grants from the Colorado European Union Center of Excellence and the Center for British and Irish Studies at the University of Colorado Boulder.

Notes

References

Allport

(1954) The Nature of Prejudice, Reading, MA: Addison-Wesley.

Arceneaux

(2012) Cognitive biases and the strength of political arguments. American Journal of Political Science 56(2): 271–285.

Aronson

(1969) The theory of cognitive dissonance: A current perspective. In: Berkowitz

(ed.) Advances in Experimental Social Psychology Volume 4, New York: Academic Press, Inc, pp. 2–34.

Aspinwall

(2002) Preferring Europe: Ideology and national preferences on European integration. European Union Politics 3(1): 81–111.

Bakker

de Vreese

(2016) Personality and EU attitudes: Relationships across EU attitude dimensions. European Union Politics. 17(1): 25-45.

Baron RM and Kenny DA (1986) The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51(6): 1173–1182.

Bellucci

Sanders

Serricchio

(2012) Explaining European identity. In: Sanders

Bellucci

Tóka

Torcal

(eds) The Europeanization of National Polities? Citizenship and Support in a Post-Enlargement Union, New York: Oxford University Press, pp. 61–90.

Block

(2006) Nursery school personality and political orientation two decades later. Journal of Research in Personality 40(5): 734–749.

Bloeser

Canache

Mitchell

D-G

(2015) The temporal consistency of personality effects: Evidence from the British Household Panel Survey. Political Psychology 36(3): 331–340.

10.

Borghans

Heckman

Golsteyn

BHH

(2009) Gender differences in risk aversion and ambiguity aversion. Journal of the European Economic Association 7(2–3): 649–658.

11.

Bruter

(2005) Citizens of Europe? The Emergence of a Mass European Identity, New York: Palgrave Macmillan.

12.

Bruter M and Harrison S (2012) How European do you feel? The psychology of European Identity. Available at: http://www.lansons.com/pdfs/eu-identity-brochure.pdf (accessed 19 February 2013).

13.

Camerer C (2005) Three cheers–psychological, theoretical, empirical–for loss aversion. Journal of Marketing Research 42(2): 129–133.

14.

Canache

Hayes

Mondak

(2013) Openness, extraversion, and the intention to emigrate. Journal of Research in Personality 47(4): 351–355.

15.

Caprara

Schwartz

Capanna

(2006) Personality and politics: Values, traits, and political choice. Political Psychology 27(1): 1–28.

16.

Carey

(2002) Undivided loyalties: Is national identity an obstacle to European integration? European Union Politics 3(4): 387–413.

17.

Carney

Jost

Gosling

(2008) The secret lives of liberals and conservatives: Personality profiles, interaction styles, and the things they leave behind. Political Psychology 29(6): 807–840.

18.

Citrin

Sides

(2004) More than nationals: How identity choice matters in the new Europe. In: Herrmann

Risse

Brewer

(eds) Transnational Identities: Becoming European in the EU, New York: Rowman and Littlefield Publishers, Inc, pp. 161–185.

19.

Cooper

Golden

Socha

(2013) The Big Five personality factors and mass politics. Journal of Applied Social Psychology 43(1): 68–82.

20.

Costa, Jr. PT and McCrae RR (1992) Normal personality assessment in clinical practice: The NEO personality inventory. Psychological Assessment 4(1): 5–13.

21.

Curtis

(2014) Inclusive versus exclusive: A cross-national comparison of the effects of subnational, national, and supranational identity. European Union Politics 15(4): 521–546.

22.

Díez Medrano

(2003) Framing Europe: Attitudes to European Integration in Germany, Spain, and the United Kingdom, New Jersey: Princeton University Press.

23.

Dinesen

Klemmensen

Nørgaard

(2016) Attitudes toward immigration: The role of personal predispositions. Political Psychology 37(1): 55–72.

24.

Druckman

(1994) Nationalism, patriotism, and group loyalty: A social psychological perspective. Mershon International Studies Review 38(1): 43–68.

25.

Duckitt

Sibley

(2016) Personality, ideological attitudes, and group identity as predictors of political behavior in majority and minority ethnic groups. Political Psychology 37(1): 109–124.

26.

Edwards

Smith

(1996) A disconfirmation bias in the evaluation of arguments. Journal of Personality and Social Psychology 71(1): 5–24.

27.

Ehrlich

Maestas

(2010) Risk orientation, risk exposure, and policy opinions: The case of free trade. Political Psychology 31(5): 657–684.

28.

Fearon

Laitin

(2000) Violence and the social construction of ethnic identity. International Organization 54(4): 845–877.

29.

Festinger

(1957) A Theory of Cognitive Dissonance, Stanford, CA: Stanford University Press.

30.

Fligstein

(2009) Who are the Europeans and how does this matter for politics? In: Checkel

Katzenstein

(eds) European Identity, New York: Cambridge University Press, pp. 132–166.

31.

Freitag

Ackermann

(2015) Direct democracy and institutional trust: Relationships and differences across personality traits. Political Psychology. Epub ahead of print 1 September 2015. DOI: 10.1111/pops.12293.

32.

Gaertner

Dovidio

(2000) Reducing Intergroup Bias: The Common Ingroup Identity Model, Philadelphia: Psychology Press.

33.

Gaertner

Dovidio

Anastasio

(1993) The common ingroup identity model: Recategorization and the reduction of intergroup bias. European Review of Social Psychology 4(1): 1–26.

34.

Gallego

Oberski

(2011) Personality and political participation: The mediation hypothesis. Political Behavior 34(3): 425–451.

35.

Gellner

(1983) Nations and Nationalism, Ithaca, NY: Cornell University Press.

36.

Gerber

Huber

Doherty

(2010) Personality and political attitudes: Relationships across issue domains and political contexts. American Political Science Review 104(1): 111–133.

37.

Gerber

Huber

Doherty

(2011a) Personality traits and the consumption of political information. American Politics Research 39(1): 32–84.

38.

Gerber

Huber

Doherty

(2011b) The Big Five personality traits in the political arena. Annual Review of Political Science 14: 265–287.

39.

Gerber

Huber

Doherty

(2012) Personality and the strength and direction of partisan identification. Political Behavior 38(4): 653–668.

40.

Gibson

Caldeira

(1995) The legitimacy of transnational legal institutions: Compliance, support, and the European Court of Justice. American Journal of Political Science 39(2): 459–489.

41.

Givens

(2005) Voting Radical Right in Western Europe, New York: Cambridge University Press.

42.

Goldberg

(1992) The development of markers for the Big-Five factor structure. Psychological Assessment 4(1): 26–42.

43.

Green

(2007) The Europeans: Political Identity in an Emerging Polity, Boulder, CO: Lynne Rienner.

44.

Greenstein

(1967) The impact of personality on politics: An attempt to clear away the underbrush. American Political Science Review 61(3): 629–641.

45.

Hibbing

Ritchie

Anderson

(2011) Personality and political discussion. Political Behavior 33: 601–624.

46.

Hicks

Tingley

(2011a) Causal mediation analysis. Stata Journal 11(4): 1–15.

47.

Hicks R and Tingley D (2011b) Mediation (for Stata). Statistical package for calculating mediation effects. Available at: http://scholar.harvard.edu/dtingley/software/mediation-stata (accessed 15 December 2013).

48.

Hooghe

Marks

(2004) Does identity or economic reality drive public opinion on European integration? Political Science and Politics 37: 415–420.

49.

Huddy

(2003) Group identity and political cohesion. In: Sears

Huddy

Jervis

(eds) The Oxford Handbook of Political Psychology, New York: Oxford University Press, pp. 511–558.

50.

Imai

Keele

Tingley

(2010) A general approach to causal mediation analysis. Psychological Methods 15(4): 309–334.

51.

Imai

Keele

Tingley

(2011) Unpacking the black box of causality: Learning about causal mechanisms from experimental and observational studies. American Political Science Review 105(4): 765–789.

52.

Imai

Keele

Tingley

(2013) Unpacking the black box of causality: Learning about causal mechanisms from experimental and observational studies. APSA Short Course. Available at: http://imai.princeton.edu/talk/files/APSA13.pdf (accessed 14 September 2015).

53.

Imai

Yamamoto

(2013) Identification and sensitivity analysis for multiple causal mechanisms: Revisiting evidence from framing experiments. Political Analysis 21(2): 141–171.

54.

Inglehart

(1970) Cognitive mobilization and European identity. Comparative Politics 3(1): 45–70.

55.

Inglehart

Abramson

(1999) Measuring postmaterialism. American Political Science Review 93(3): 665–677.

56.

John

Srivastava

(1999) The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In: Pervin

John

(eds) Handbook of Personality: Theory and Research, New York: Guilford Press, pp. 102–138.

57.

Kahneman

Tversky

(1979) Prospect theory: An analysis of decision under risk. Econometrica 47(2): 263–292.

58.

Keele

(2015) Causal mediation analysis: Warning! Assumptions ahead. Journal of Evaluation. Epub ahead of print 16 July 2015. DOI: 10.1177/1098214015594689.

59.

Kritzinger

(2003) The influence of the nation-state on individual support for the European Union. European Union Politics 4(2): 219–241.

60.

Levy JS (2003) Applications of prospect theory to political science. Synthese 135(2): 215–241.

61.

Lewis

Bates

(2011) From left to right: How the personality system allows basic traits to influence politics via characteristic moral adaptations. British Journal of Psychology 102(3): 546–558.

62.

McCrae

Costa

PT Jr

(1992) Normal personality assessment in clinical practice: The NEO personality inventory. Psychological Assessment 4(1): 5–13.

63.

McDermott

(2004) Prospect theory in political science: Gains and losses from the first decade. Political Psychology 25(2): 289–312.

64.

McDonough

(1995) Identities, ideologies, and interests: Democratization and the culture of mass politics in Spain and Eastern Europe. Journal of Politics 57(3): 649–676.

65.

McPherson

Smith-Lovin

Cook

(2001) Birds of a feather: Homophily in social networks. Annual Review of Sociology 27: 415–444.

66.

Mercer

(2005) Prospect theory and political science. Annual Review of Political Science 8: 1–21.

67.

Mollica

Gray

Treviño

(2003) Racial homophily and its persistence in newcomers’ social networks. Organization Science 14(2): 123–136.

68.

Mondak

Halperin

(2008) A framework for the study of personality and political behavior. British Journal of Political Science 38(2): 335–362.

69.

Mondak

Hibbing

Canache

(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.

70.

Mummendey

Klink

Brown

(2001) Nationalism and patriotism: National identification and out-group rejection. British Journal of Social Psychology 40(2): 159–172.

71.

Nicholson

Soane

Fenton-O’Creevy

(2005) Personality and domain-specific risk taking. Journal of Risk Research 8(2): 157–176.

72.

Nickerson

(1998) Confirmation bias: A ubiquitous phenomenon. Review of General Psychology 2(2): 175–220.

73.

Osborne

Sibley

(2015) Within the limits of civic training: Education moderates the relationship between openness and political attitudes. Political Psychology 36(3): 295–313.

74.

Ozer

Benet-Martínez

(2006) Personality and the prediction of consequential outcomes. Annual Review of Psychology 57: 301–321.

75.

Papacosta

(2009) Eurobarometer 71.3: Globalization, Personal Values and Priorities, European Identity, Future of the European Union, Social Problems and Welfare, and European Elections, June–July, 2009. [Computer file]. ICPSR28184-v1. Cologne, Germany: GESIS/Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributors], 2010-08-13. DOI: 10.3886/ICPSR28184.v1.

76.

Pettigrew

(1998) Intergroup contact theory. Annual Review of Psychology 49: 65–85.

77.

Piaget

(1972) The Principles of Genetic Epistemology, New York: Routledge & Kegan Paul.

78.

Pichler

(2008) Socio-structural differences in identification with Europe. Perspectives on European Politics and Society 9(4): 381–396.

79.

Preacher

Hayes

(2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavioral Research Methods 40(3): 879–891.

80.

Roccas

Sagiv

Schwartz

(2002) The Big Five personality factors and personality values. Personal and Social Psychology Bulletin 28: 789–801.

81.

Rohrschneider

(2002) The democracy deficit and mass support for an EU-wide government. American Journal of Political Science 46(2): 463–475.

82.

Ruiz Jiménez

Górniak

Kandulla

(2004) European and national identities in the EU’s old and new member states: Ethnic, civic, instrumental and symbolic components. European Integration online Papers (EIoP). Vol. 8, No. 11. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=570601 (accessed 9 March 2013).

83.

Schmitt

Allik

McCrae

(2007) The geographic distribution of Big Five personality traits. Journal of Cross-Cultural Psychology 38(2): 173–212.

84.

Schoen

Schumann

(2007) Personality traits, partisan attitudes, and voting behavior: Evidence from Germany. Political Psychology 28(4): 471–498.

85.

Schuck

ART

de Vreese

(2006) Between risk and opportunity: News framing and its effects on public support for EU enlargement. European Journal of Communication 21(1): 5–32.

86.

Sears

Huddy

Jervis

(2003) The psychologies underlying political psychology. In: Sears

Huddy

Jervis

(eds) The Oxford Handbook of Political Psychology, New York: Oxford University Press, pp. 3–16.

87.

Sibley

Duckitt

(2008) Personality and prejudice: A meta-analysis and theoretical review. Personality and Social Psychology Review 12(3): 248–280.

88.

Sides

Citrin

(2007) European opinion about immigration: The role of identities, interests and information. British Journal of Political Science 37(3): 477–504.

89.

Sinnott

(2005) An evaluation of the measurement of national, subnational, and supranational identity in crossnational surveys. International Journal of Public Opinion Research 18(2): 211–223.

90.

Spannring

Wallace

Datler

(2008) What leads young people to identify with Europe? An exploration of the impact of exposure to Europe and political engagement on European identity among young Europeans. Perspectives on European Politics and Society 9(4): 480–498.

91.

Spiering

(2004) British Euroscepticism. European Studies 20: 127–149.

92.

Taber

Lodge

(2006) Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science 50(3): 755–769.

93.

Theiss-Morse

(2009) Who Counts as an American? The Boundaries of National Identity, New York: Cambridge University Press.

94.

Transue

(2007) Identity salience, identity acceptance, and racial policy attitudes: American national identity as a uniting force. American Journal of Political Science 51(1): 78–91.

95.

Tugendhat C (1977) The European community: Its moral identity and mission. Speech by Christopher Tugendhat, European Commissioner, to Cambridge University Conservative Association. November 6. Available at: http://aei.pitt.edu/11110/ (accessed 1 November 2010).

96.

Vecchione

Schoen

González Castro

(2011) Personality correlates of party preference: The Big Five in five big European countries. Personality and Individual Differences 51(6): 737–742.

97.

Wolak

Marcus

(2007) Personality and emotional response: Strategic and tactical responses to changing political circumstances. The Annals of the American Academy of Political and Social Science 614(1): 172–195.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.11 MB

0.00 MB

0.43 MB