The association between agreeableness ...

Personality and Individual Differences 107 (2017) 102–107

Contents lists available at ScienceDirect

Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid

The association between agreeableness, extraversion, and support for secessionist movements: Evidence from a large survey of more than 33,000 respondents in Catalonia☆ Joan Barceló Washington University in St. Louis, Department of Political Science, Campus Box 1063, One Brookings Drive, St. Louis, MO 63130-4899, United States.

a r t i c l e

i n f o

Article history: Received 5 August 2016 Received in revised form 3 November 2016 Accepted 11 November 2016 Available online xxxx Keywords: Big Five Personality Agreeableness Extraversion Referendum Independence Secessionism Political behavior

a b s t r a c t Does personality influence attitudes toward secessionism? Personality traits are important determinants of a wide range of political attitudes and behaviors. Prior work has mainly focused on the study of personality during regular processes of democratic political systems (e.g., vote choice, turnout, information acquisition, etc.). By contrast, this paper assesses that personality traits have an influence over the decision of individuals to support a secessionist movement. To test this, I rely on a survey distributed to more than 33.000 respondents in Catalonia, a region with a salient secessionist movement. Beyond the effects of demographic characteristics, political factors, descent-based attributes, economic considerations, and even identity attachments, the results confirm that the personality trait of Agreeableness is associated with supporting the Catalan secessionist movement. Although less consistently, there is some preliminary evidence that extraverted individuals are more likely to support the movement. I conclude that combining insights from political science, sociology and psychology in the study of dispositional and situational determinants of attitudes toward secessionism is fertile research ground. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction A secession is defined as a formal separation from a larger organized body to which the secessionist region has been tied to. In a world dominated by states, it is not surprising that the study of secessionism is of vital importance. The number of countries in the world has tripled during the second half of the twentieth century due to a process of decolonization and separation from larger entities. Today, some recent successful secessionist cases such as East Timor, Montenegro, South Sudan or Kosovo, have been paralleled by increases in the demands for secession in advanced democracies. Even though political elites tend to play a substantial role in these processes, mass support for independence usually becomes a crucial driver for the successful culmination of a secessionist project. A wide range of factors have been found to influence an individual's attitudes, preferences and behaviors concerning secessionism. Demographic, political variables, descent-based, as well as identity variables are common predictors in most empirical models of secessionism in advanced democracies (Barceló, 2014, Burg, 2015, Orriols & Rodón, 2016, Serrano, 2013). Moreover, some scholars suggest that economic considerations such as future expectation in the current compared to the new ☆ I thank James L. Gibson, Chris Dawes, Dawn Brancati, Nicole Simonelli, Vera Heuer, and Luke Cooper for helpful comments on earlier versions of this article. All remaining errors are my own. Replication data and code are available on the author's website.

http://dx.doi.org/10.1016/j.paid.2016.11.029 0191-8869/© 2016 Elsevier Ltd. All rights reserved.

regime may also underlie the surge in secessionist movements (Burg, 2015, Muñoz & Tormos, 2015). Alternatively, other scholars also suggest that elite-led cuing is the major driving force behind people's support for a secession (Fernández-Albertos & Lago, 2015, Miley, 2014, Torcal & Mota, 2014). In addition, there is a growing body of scholarship which debates the impact of psychological factors such as moral sentiments (Lewis & Brown, 2015) or risk attitudes (Nadeau, Martin, & Blai, 1999; Verge, Guinjoan, & Rodón, 2015) on secessionist aspirations. Nevertheless, scholars have not considered the potential impact of individuals' personality in their support for secessionist movements. To fill this gap, I provide a systematic analysis of the role of personality traits in supporting secessionist movements. 1.1. Personality as a disposition for attitudes and behavior Personality is the psychological organization responsible for the way individuals react to external stimuli. Everyone has a unique personality, which influences how we think and behave (Allport & Odbert, 1936). Personality psychologists have accumulated much evidence suggesting that individuals are characterized by dispositional personality traits that can be categorized under the umbrella of the Big Five Factor Model, or simply, the Big Five: Extraversion, Openness to experience, Agreeableness, Conscientiousness, and Neuroticism (John, Naumann, & Soto, 2008; McCrae & Costa, 1999).

J. Barceló / Personality and Individual Differences 107 (2017) 102–107

Extraversion is linked to facets such as gregariousness, activity and emotion-seeking, and people who score high in this trait have an energetic, optimistic and active way of looking at the world. Openness to experience characterizes individuals by their curiosity, attraction to the unknown, and willingness to explore new experiences, which includes facets like imagination and artistic interests. Agreeableness assesses facets related to sincerity, altruism and modesty. Conscientiousness includes dispositions that tend toward impulse control and facets such as orderliness, self-discipline and achievement striving. Lastly, Neuroticism, or emotional stability, defines individuals by their tendencies toward anxiety, self-consciousness and angry hostility. The salience of the 5-factor structure lies in its capability to reduce personality to this small number of universal core traits that are robust across time and space, highly stable over time (Costa & McCrae, 1988) and comparable across different languages and cultures around the globe (McCrae & Terracciano, 2005). Social scientists have increasingly paid major attention to the role of personality traits because of two major reasons. First, they are endogenous basic tendencies that remain rather stable across individuals' lifespan (Anusic, 2016). Preferences were traditionally thought to appear as a combination of both more immediate considerations and long-standing acquisitions as a result of socialization processes (Zaller, 1992), especially those that occurred early in childhood (Jennings, Stoker, & Bowers, 2009). Recent research on personality came to complement traditional insights which shed light on the interactive nature of the relationship between dispositional traits and the situational environment to determine attitudes and behavior. Second, personality traits are powerfully associated with a wide range of attitudes and behaviors, including political participation and ideology (Gerber, Huber, Doherty, & Dowling, 2011a). 1.2. The present study: rationale and predictions Thus far, there has been no attempt to integrate personality characteristics to explain the fundamental question of why some people want to secede from their host country. While personality traits provide us with some leverage to understand this question, we should not expect all traits to exert an effect on secessionism. In particular, I expect Agreeableness, Extraversion, and Conscientiousness to exert effects on the likelihood of supporting a secessionist movement. The personality trait of Agreeableness comprises my strongest expectation. Agreeableness reflects inter-individual differences in people's tendencies to cooperate, their generalized interpersonal trust, and concern about others' well-being. For this reason, agreeable people tend to strongly exhibit positive ties with others, which lead them to have a profound dislike for disagreement and an avoidance to put themselves in a position of conflict. When a region considers secession, it becomes a highly controversial political issue. In the political arena, individuals are pressed to respond to this environment by standing favor of or against a hypothetical referendum. Thus, individuals are asked to decide whether they would rather uphold the status quo, seek a policy change within the system, or support the anti-system position of the independence. In this context, we should expect that agreeable individuals would rather move away from anti-system positions, that is, political positions that “aim to overhaul a country's existing political system” (Aksoy & Carter, 2014: 187–188) such as supporting irredentist and separatist groups. Secondly, Extraversion is a personality trait that manifests itself in sociable and excitement-seeker individuals. Extraverted people are more likely to have an active social life, leading them to join more civic groups (Weinschenk, 2014), participate more in politics, especially in those activities that involve social interaction (Gerber et al., 2011b). In this regard, pro-secessionist movements could be viewed as belonging to a larger family of collective action movements and, as a consequence, extraverted people would be more likely to join proindependence movements.

103

Finally, I also expect a high score of Conscientiousness to decrease an individual's propensity to support secessionist movements. People high in Conscientiousness are characterized by norm compliance, deliberateness, sense of responsibility and status-quo-preserving attitudes, which should be negatively associated with a political enterprise that involves radically changing a country's constitutional order. In this vein, Lewis and Brown (2015) have recently investigated the relationship between moral sentiments and support for Scottish independence. The authors found that the binding dimension of moral sentiment, which has to do with people's concerns of norm compliance, is negatively related to support for secession in the Scottish case. Therefore, I expect that an individual's willingness to comply with the current social and political norms would lead to a negative association between Conscientiousness and secessionism. While my expectations for Agreeableness, Conscientiousness and Extraversion are directional, my expectations for the other personality traits are mixed in their directionality. To begin with, emotionally unstable – or neurotic – personalities may more easily feel hostility and anxiety regarding economic, political or socially disadvantageous policies. Furthermore, they may react more emotionally toward a threat and hold more prejudice against the social out-group when they feel threatened by them. However, the effect can go both ways since individuals may feel uneasy and anxious in the situation within their current country, leading them to pursue a different territorial solution such as secession. At the same time, they may also be more likely to support staying in their current country due to their fear of an uncertain transitional period. Therefore, I do not have strong expectations of the effect of Emotional stability or Neuroticism. Support for secession could also be influenced by Openness to experience. On the one hand, individuals who score high in Openness tend to be more accepting of social diversity and be more involved in heterogeneous social networks, which increases their exposure to different groups and depresses their willingness to break bridges across groups. Therefore, the self-selection of individuals who score high in Openness to environments that are more heterogeneous leads me to expect a negative relationship between this trait and support for secession. However, open individuals are also more likely to explore new things, and obviously moving their country toward a secessionist path should excite them. These two hypotheses, thus, appear to lead to contradictory expectations with respect to the role of Openness; thus, no hypothesis was made regarding its effect on pro-secessionist attitudes.

2. Materials and method 2.1. The survey context Catalonia has historically been an economically advantaged region within Spain, with a linguistic singularity, and a strong competition between state wide parties and non-state wide parties in both national and regional elections since the advent of Spain's democracy. Therefore, Catalonia provides a prototypical case study of secessionism that captures most of the variables that are common in secessionist movements, at least, among developed countries.1

1 Evidence for the cross-sectional determinants of secessionism from Sorens (2005) shows that the list of variables that are more strongly correlated with a region having a secessionist movement include identity-based attributes such as history of independence or a regional language, but also a “lack of irredentist potential, relative affluence, geographical noncontiguity, and multiparty political system” (304). Catalonia is a prototypical case of secessionism since it contains most of the variables that make a region more likely to have a secessionist movement. Obviously, this statement is probabilistic and, consequently, there may be important secessionist movement such as in Quebec or Scotland that are not relatively affluent.

104


2.2. Participants Participants consisted of 37,535 adults recruited through an opt-in Internet survey distributed in Catalonia and publicized via social and mass media.2 All participants identified themselves as residents of Catalonia. The Internet survey has been weighted to be representative of the national population in order to minimize self-selection bias.3 For this purpose, a number of demographic and attitudinal questions were matched with a probabilistic survey, the Baròmetre d'Opinió Política (BOP), carried out by the Centre d'Estudis d'Opinió (CEO). The link of the probabilistic to the non-probabilistic sample is made with the 2nd wave of the BOP of 2013, which is the wave temporarily closest to the non-probabilistic sample. To generate the sampling weights, I follow a general technique for the treatment of samples with unequal selection probabilities (Horvitz and Thompson 1952). These sampling weights allow me to minimize bias between my convenience sample and the population at large.4

comprised of three values: intention to vote in favor of the secession, abstain, or vote against the secession. The wording of the question is the following: “If a referendum for independence were to be held tomorrow, what would you do? I would vote in favor of the independence/I would vote against the independence/I would abstain. 2.4. Data analysis A set of multinomial logistic regression models were performed to explore the role of personality traits on attitudes toward the secession across model specifications with different sets of controls. The output reported here presents the log-odds coefficients of abstaining compared to voting against the secession (odd columns of Table 2); and, the logodds coefficients of voting in favor of the secessions compared to voting against the secession (even columns of Table 2). 3. Results and discussion

2.3. Measures

3.1. Main findings

2.3.1. Demographics, social, political, and identity controls A self-reported questionnaire was used to obtain the participants' demographic characteristics, including age, sex, education, subjective social class, household income, and province of residence. In addition, the questionnaire included questions on political, descent-based, and identity factors that are predictive of support for secession: 1) ideological self-placement (left-right self-placement, from 0 to 7); 2) a variable capturing whether the participant participated in the previous regional election; 3) territorial origin (whether born in Catalonia, first or second generation regional immigrant from other parts of Spain, or abroad); 4) mother language (Catalan, Spanish, both or neither); and 5) the LinzMoreno scale of subjective national identity, whereby participants had to choose the statement that was closest to their feelings among the following: “I feel: only Catalan, more Catalan than Spanish, as Catalan as Spanish, more Spanish than Catalan or only Spanish”.

Table 2 reports the results from three multinomial logistic regressions where I regress an individual's intention to abstain or vote in favor of the secession in a hypothetical self-determination referendum on the Big Five personality traits, as well as controls. The results reveal a significantly negative association between an individual's Agreeableness score and the likelihood of intending to vote in favor of the secession in a self-determination referendum. This coefficient remains negative and statistically significant at the 95% confidence level across the three model specification, i.e. after adjusting for the demographic variables (age, sex, province, education, subjective social class, household income, place of birth, and mother tongue), political variables (ideology and electoral turnout in the last election), and subjective national identity. Moreover, we can see that the negative effect of agreeableness becomes even stronger once it is adjusted for individual's national identity. Just as for Agreeableness, the estimate for Extraversion shows the expected pattern. Individuals who score high in Extraversion are significantly more likely to report the intention to vote in favor of independence. However, the effect is not consistent throughout the different model specifications. While it is not significant at a 95% confidence level when the model is adjusted for demographics, political controls, and decent-based attributes, it becomes significant only in a fully-controlled model. Even though the effect in the final model seems relatively strong in statistical terms, we should be cautious in its interpretation given the lack of stability of the coefficient. The finding that personality traits become larger once we control for national identity deserves further discussion. In the models, national identity seems to be a negative confounder; that is, the lack of adjustment for it biases the estimate toward the null hypothesis. There are three results that cause this: a) The conditional correlation between Agreeableness and secessionism is positive; b) the conditional correlation between national identity and secessionism is positive; and, c) the conditional correlation between national identity and Agreeableness is negative. Therefore, because traits are associated with identity in a different direction than with secessionism, then the national identity variable negatively confounds the relationship between traits and secessionism, increasing their impact once we adjust for it. As for the rest of personality traits, individuals' high in neuroticism are persistently less likely to either abstain or vote in favor of secession, although none of these effects are statistically significant. In addition, respondents who score high in Openness are statistically more likely to abstain and to vote in favor of secession, yet the relationship falls short of statistical significance once political and identity controls are included in the model. Finally, I do not find support for the hypothesized negative association between Conscientiousness and support for independence. The results reveal that the association is very close to having

2.3.2. Personality measures The Big Five personality traits were evaluated through a self-administered Ten-Item Personality Inventory (TIPI) available for the participants in Spanish and Catalan. This instrument has been shown to reliably estimate personality in multiple languages and be highly correlated with longer instruments (Gosling, Rentfrow, & Swann, 2003). The TIPI items were rated on a 5-point Likert scale, ranging from “strongly disagree” to “strongly agree.” Table 1 reports the mean (scale range from 0 to 8), standard deviation, and Pearson's correlation of the items from the TIPI questionnaire. Correlations range from 0.17 for the two items of Agreeableness to 0.50 for the two items of Neuroticism. Even though the low correlations for Agreeableness and Openness are far from ideal, similar works have relied upon factors with similar low correlations (Gosling et al., 2003; Mondak, Hibbing, Canache, Seligson, & Anderson, 2010). 2.3.3. Support for the secession To measure support for the secession, I rely on the self-reported intention to support the secessionist movements in a hypothetical selfdetermination referendum. This variable is a categorical factor 2 Social media used for the recruitment of the respondents: Facebook and Twitter. The survey also appeared in digital newspaper such as Nació Digital, Racó Català, or Vilaweb, among others, which explains the large sample size of the survey. 3 For the purpose of linking the probabilistic to the non-probabilistic sample, I use the 2nd wave of the BOP. This has 2000 respondents from the four Catalan provinces (Barcelona, Girona, Lleida and Tarragona) and was fielded between June 3 and June 12, 2013. 4 See the Appendix A for more information about the technical procedures of the survey, the weighting procedures, and the balance of covariates between the convenience and the probabilistic survey. See Appendix B for the descriptive statistics of the variables used in the analysis.


105

Table 1 Survey-based indicators of the Big Five. Personality traits

Components

Mean (s.d.)

Pearson's r

Number of cases

Openness

Open to new experiences, complex. Conventional, uncreative (reversed). Dependable, self-disciplined. Disorganized, careless (reversed). Extraverted, enthusiastic. Reserved, quiet (reversed). Critical, quarrelsome (reversed). Sympathetic, warm. Anxious, easily upset. Calm, emotionally stable (reversed).

2.50 (1.62)

.23

35,809

2.20 (1.54)

.33

35,692

3.36 (1.74) 2.84 (1.38)

.46 .17

35,609 36,463

4.39 (1.74)

.50

36,273

Conscientiousness Extraversion Agreeableness Neuroticism/emotional stability

no-effect. However, we cannot rule out the possibility that the lack of the effect of Conscientiousness is due to the lack of coverage of the TIPI of important facets that may well be related to the support for the status quo such as a sense of responsibility or deliberateness, for example. Although Table 2 shows that the findings are consistent with the expectations, it could be argued that the large sample size is what allows us to detect small effects (e.g. personality traits), even when their effect size may be argued to be meaningless. To see if this is the case, I proceed to show the size of the effects of Agreeableness and Extraversion. Fig. 1 illustrates the changes in the predicted probability in the likelihood to vote for secession as a function of the standardized values of Agreeableness (panel A) and Extraversion (panel B). In the left panel, we can observe that a shift in an individual's Agreeableness score from two standard deviations below the mean to the mean or from the mean to two standard deviations above the mean leads to a reduction of 6.6 and 10.6 percentage points – 7.1% and 12.3% decrease – in the probability to vote in favor of independence, respectively. This implies a cumulative reduction in the probability of support for independence in a hypothetical self-determination referendum of a 18.6% decrease, or a decrease of 17.2 percentage points. As for the effect of Extraversion, the positive effect reported in Table 2 is illustrated in Fig. 1. The size of the effect is smaller and remarkably less reliable than the effect of Agreeableness. A shift in two standard deviations below the mean to the mean and from the mean to two standard deviations above the mean would lead to an increase in the chance of voting in favor of secession of 4.6 and 3.5 percentage points – 5.3% and 3.9% rise – in the probability to vote in favor of the independence, respectively. Altogether, this implies a cumulative increase of a change of four standard deviations in the distribution of Extraversion,

nearly a shift along the full distribution, of 8.1 percentage points, or a 10.3% increase in the likelihood to support the independence. 3.2. Item-level analyses This section presents item-level analyses as suggested by Mõttus et al. (2016) and Vainik et al. (2015). If the trait-outcome association ought to be interpreted as causal, the authors suggest that we should observe the relationship between each item and the outcome, since the relationship should hold up regardless of how we operationalize the trait. Table 3 reports the trait-outcome relationships at the level of the item. There are three findings from these analyses. First, the effect of Agreeableness is not driven by individual items, but both items present a similarly significant effect on pro-secessionism. Second, the effect of Extraversion maintains its positive sign in both items, but its significance vanishes. The combined lack of consistency of the estimate of Extraversion in Table 2, and the lack of significance for each individual item, suggests that we need to be cautious when interpreting the correlation of Extraversion as causal. Finally, the overall Conscientiousness-outcome relationship has been consistently not significant in the models from Table 2. Yet, when we explore the specific items, we can see that this is due to the mixed effects of the two items included in the questionnaire. The item that taps into the facet of self-discipline presents a strongly significant effect, with those individuals who report being more self-disciplined being less likely to support the secession, but those who are more organized are generally more supportive of the secession. The mixed results in each item suggest that better instruments to test the facet-level correlates of secessionism are required to yield stronger conclusions about the (lack of) the existence of a relationship.

Table 2 The Big Five personality traits and the support for secession. Multinomial logistic regressions. Model 1

Agreeableness Extraversion Neuroticism Openness Conscientiousness Controls Demographics/descent Political National identity Observations Log likelihood Pseudo R2

Model 2

Model 3

Abstention

Secessionist

Abstention

Secessionist

Abstention

Secessionist

−.02 (.89) .12 (.35) −.06 (.57) .24** (.02) .02 (.89)

−.19** (.01) .09 (.22) −.00 (.97) .15 (.03) −.07 (.30)

.00 (.99) .07 (.51) −.07 (.48) .22** (.04) −.00 (.94)

−.15** (.04) .10 (.17) −.02 (.74) .12 (.10) −.03 (.71)

−.02 (.88) .08 (.45) −.05 (.61) .22* (.05) .02 (.84)

−.32** (.00) .19** (.03) −.02 (.78) .07 (.47) .01 (.91)

✓ ✘ ✘ 33,990 −22,397 .23

✓ ✓ ✘ 33,990 −20,557 .29

✓ ✓ ✓ 33,990 −13,719 .53

Note: ⁎p b 0.1; ⁎⁎p b 0.05. Weighted analysis. The number of observations in the dependent variable is distributed as follows: 9420 respondents would vote against the secession (25.4% of the weighted sample); 4748 respondents would abstain (12.8% of the weighted sample); and, 22,886 would vote in favor of the secession (61.8% of the weighted sample). Control variables are described in the text. See Appendix C for the full models with the coefficients of all controls.

106


Fig. 1. Predicted probabilities of voting in favor of the secession at different values of agreeableness and extraversion.

3.3. A discussion of alternative theories An argument could be made that the previous models do not address the potential role of economic considerations/expectations and the elite-led inducement of secessionist attitudes. However, there is little reason to expect a spurious correlation between our main independent variables, personality traits, and attitudes toward the secession. A noncausal correlation (X ➔ Y) can be spurious if there is an antecedent variable (Z) which causes both (Z ➔ X and Z ➔ Y). Therefore, the inclusion of variable tapping into these alternative theories would only be necessary for the identification of a causal effect if either economic considerations/expectations or elite-led mobilizations are antecedent variables that cause both personality and pro-secessionist attitudes. However, besides maturation and major life events, personality seems to change little over the entire life span (Specht et al. 2011). Therefore, it is reasonable to believe that personality traits are exogenous from economic considerations/expectations of the future of Catalonia and the information of political acquisition from the elites.5 While the questionnaire does not include ideal questions to tap into each of these theories, it includes some questions that can provide us with some leverage in assessing whether the role of personality traits is affected by either of the two theories. On the one hand, the survey asks respondents about the reasons they support their decision in the self-determination referendum. For those who reported to be in favor of the referendum, the survey asks them to evaluate from 0 to 100 the extent to which eleven items were important in the decision. Four of them have to do with increasing the economic welfare of Catalan society. Unfortunately, those who reported to either vote against the secession or to abstain were presented with nine items, yet none of these items overlapped with those presented to those who would vote in favor of the secession. Even though none of the items refers directly to an economic reason to stay, one of them states that the secession would not be positive for Catalonia. I take the mean of the four economic items for those who would vote in favor of the secession and the value of the item stating that it would not be positive for Catalonia to construct a proxy variable for the subjective importance of economic considerations. On the other hand, the questionnaire includes a question of whether the individual follows the news related to the process of independence of Catalonia with the following options: a lot, quite, not much, or none at all of attention. It is generally believed that mass opinion change takes the form of a two-stage process: 1) the reception of political communication; 5 This does not mean that personality traits are unrelated to either economic considerations/expectations of the future of Catalonia or the information of political acquisition, but that personality traits are more likely to antecede any of these other two variables.

Table 3 The Big Five personality traits and the support for secession. Multinomial logistic regressions of item-level analyses. Traits

Items

DV: referendum vote Abstention Secessionist

Agreeableness Critical, quarrelsome (reversed). Sympathetic, warm.

.12 (.12)

−.25* (.10)

−.10 (.11)

−.21* (.09)

Extraverted, enthusiastic. Reserved, quiet (reversed).

−.03 (.11) .07 (.14)

.11 (.08) .08 (.10)

Anxious, easily upset. Calm, emotionally stable (reversed).

.12 (.12) −19 (.13)

.02 (.09) −.06 (.11)

Open to new experiences, complex. Conventional, uncreative (reversed).

.22* (.09)

.05 (.08)

.07 (.13)

.07 (.09)

Dependable, self-disciplined. Disorganized, careless (reversed).

−.11 (.11) .21 (.14)

−.20* (.09) .29** (.10)

Extraversion

Neuroticism

Openness

Conscientiousness

Controls Demographics/Descent Political National Identity Observations Log Likelihood Pseudo R2

✓ ✓ ✓ 33,990 −14,750 .51

Note: ⁎p b 0.1; ⁎⁎p b 0.05. Weighted analysis. Control variables are described in the text. See Appendix D for the full models with the coefficients of all controls. Items have been reversed so that the expected direction at the trait-level is the same between the two traits.

and, 2) acceptance/rejection of that communication (Zaller, 1992). Following this basic model, we should expect that attention to the news on the secessionist process would be a relatively good proxy for people's susceptibility for opinion change in the secessionist issue.6 To check the sensitivity of the estimates with regards to the Big Five, Table 4 reports the results after controlling for the proxy variable tapping into these different theoretical frameworks. Reassuringly, the variable “economic considerations” increases the Pseudo-R2 of the model,

6

See the wording of these questions in the Appendix E.


107

Table 4 The Big Five personality traits and the support for secession. Multinomial logistic regressions with controls for alternative theories. Model 1

Agreeableness Extraversion Neuroticism Openness Conscientiousness Interested in news on secessionism (1–4)

Model 2 Secessionist

Abstention

Secessionist

Abstention

Secessionist

−.04 (.10) .06 (.11) −.09 (.11) .24* (.11) .01 (.11) −.46** (.13)

−.28** (.09) .20* (.09) −.00 (.09) .02 (.09) −.03 (.09) .78** (.12)

.03 (.11) .16 (.12) −.04 (.11) .24* (.12) −.05 (.12)

−.34** (.10) .23* (.10) −.08 (.10) .02 (.10) −.01 (.10)

−.03** (0.00)

.04** (0.00)

.03 (.11) .17 (.12) −.07 (.12) .26* (.13) .04 (.12) −.44** (.14) −.03** (0.00)

−.30** (.10) .26* (.10) −.07 (.10) .00 (.11) −.05 (.10) 0.62** (.13) .03** (.00)

Economic Considerations (0−100)

Controls Demographics/Descent Political National Identity Observations Log Likelihood Pseudo R2

Model 3

Abstention

✓ ✓ ✓ 33,903 −14,092 .52

✓ ✓ ✓ 33,282 −10,571 .60

✓ ✓ ✓ 33,216 −10,214 .62

Note: ⁎p b 0.1; ⁎⁎p b 0.05. Weighted analysis. See Appendix F for the full models with the coefficients of all controls. The drop in the number of observations reflects the pattern of missingness in the variables, which makes the log-likelihood from Table 2 not fully comparable.

which should lead us to believe that it is actually capturing some important variance that is related to support for the secession. After including each variable in turn (models 1–2), or both in the same model (3), the main findings from the fully-specified model 3 in Table 2 remain unaltered, with the significant effects of both Agreeableness and Extraversion in decreasing and encouraging support for independence, respectively. 4. Conclusion This research presents one of the first attempts to empirically gauge the effects of psychological attributes in the attachment to secessionist movements. While recent work on the support for secessionist movements have generally highlighted the role of economic interests, namely, an individual's utilitarian argument on the political and economic consequences of independence, as well as individuals' national identity, this present provides a step forward in our understanding of the correlates of personality traits, as well as pro-secessionist attitudes. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.paid.2016.11.029. References Aksoy, D., & Carter, D. B. (2014). Electoral institutions and the emergence of terrorist groups. British Journal of Political Science, 44(1), 181–204. Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study. Psychological monographs, 47(1) (pp. i). Anusic, I., & Schimmack, U. (2016). Stability and change of personality traits, self-esteem, and well-being: Introducing the meta-analytic stability and change model of retest correlations. Journal of Personality and Social Psychology, 110(5), 766–781. Barceló, J. (2014). Contextual effects on subjective national identity. Nations and Nationalism, 20(4), 701–720. Burg, S. L. (2015). Identity, grievances, and popular mobilization for independence in Catalonia. Nationalism and Ethnic Politics, 21(3), 289–312. Costa, P. T., & McCrae, R. R. (1988). Personality in adulthood: A six-year longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology, 54(5), 853. Fernández-Albertos, J., & Lago, I. (2015). Gobiernos autonómicos e identidades regionales en España, 1980–2012. Política y Gobierno, 22(2), 283–315. Gerber, A. S., Huber, G. A., Doherty, D., & Dowling, C. M. (2011a). The big five personality traits in the political arena. Annual Review of Political Science, 14, 265–287.

Gerber, A. S., Huber, G. A., Doherty, D., Dowling, C. M., Raso, C., & Ha, S. E. (2011b). Personality traits and participation in political processes. The Journal of Politics, 73(3), 692–706. Gosling, S. D., Rentfrow, P. J., & Swann, W. B., Jr. (2003). A very brief measure of the bigfive personality domains. Journal of Research in Personality, 37(6), 504–528. Horvitz, D. G., & Thompson, D. J. (1952). A generalization of sampling without replacement from a finite universe. Journal of the American statistical Association, 47(260), 663–685. Jennings, M. K., Stoker, L., & Bowers, J. (2009). Politics across generations: Family transmission reexamined. The Journal of Politics, 71(3), 782–799. John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative big five trait taxonomy. Handbook of Personality: Theory and Research, 3, 114–158. Lewis, G. J., & Brown, G. (2015). Moral sentiment predicts desire for Scottish independence. Personality and Individual Differences, 74, 254–258. McCrae, R. R., & Costa, P. T., Jr. (1999). A five-factor theory of personality. Handbook of personality: Theory and research, 2, 139–153. McCrae, R. R., & Terracciano, A. (2005). Universal features of personality traits from the observer's perspective: Data from 50 cultures. Journal of Personality and Social Psychology, 88(3), 547–561. Miley (2014). Democratic representation and the national dimension in Catalan and Basque politics. International Journal of Politics, Culture, and Society, 27(3), 291–322. Mõttus, R., Kandler, C., Bleidorn, W., Riemann, R., & McCrae, R. R. (2016). Personality Traits Below Facets: The Consensual Validity, Longitudinal Stability, Heritability, and Utility of Personality Nuances. Journal of Personality and Social Psychology. Mondak, J. J., Hibbing, M. V., Canache, D., Seligson, M. A., & 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. Muñoz, J., & Tormos, R. (2015). Economic expectations and support for secession in Catalonia: Between causality and rationalization. European Political Science Review, 7(2), 315–341. Nadeau, R., Martin, P., & Blai, A. (1999). Attitude toward risk-taking and individual choice in the Quebec referendum on sovereignty. British Journal of Political Science, 29(3), 523–539. Orriols, L., & Rodón, T. (2016). The 2015 Catalan election: The independence bid at the polls. Southern European Society and Politics, 21(3), 359–381. Serrano, I. (2013). Just a matter of identity? Support for independence in Catalonia. Regional & Federal Studies, 23(5), 523–545. Sorens, J. (2005). The cross-sectional determinants of secessionism in advanced democracies. Comparative Political Studies, 38(3), 304–326. Specht, J., Boris, E., & Schmukle, S. C. (2011). Stability and change of personality across the life course: the impact of age and major life events on mean-level and rank-order stability of the Big Five. Journal of personality and social psychology, 101(4), 862–882. Torcal, M., & Mota, F. (2014). The role of political parties in shaping citizens' political preferences for the territorial organization of the state: The Spanish case. European Political Science Review, 6(3), 477–502. Vainik, U., Mõttus, R., Allik, J., Esko, T., & Realo, A. (2015). Are trait–outcome associations caused by scales or particular items? Example analysis of personality facets and BMI. European Journal of Personality, 29(6), 622–634. Verge, T., Guinjoan, M., & Rodón, T. (2015). Risk aversion, gender and constitutional change. Politics & Gender, 11(3), 499–521. Weinschenk, A. C. (2014). Personality traits and the sense of civic duty. American Politics Research, 42(1), 90–113. Zaller, J. (1992). The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press.