Q Routledge Taylor &. Francis Group West European Politics ISSN: 0140-2382 (Print) 1743-9655 (Online) Journal homepage: https://www.tandfonline.com/loi/fwep20 The personality of populists: provocateurs, charismatic leaders, or drunken dinner guests? Alessandro Nai & Ferran Martinez i Coma To cite this article: Alessandro Nai & Ferran Martinez i Coma (2019) The personality of populists: provocateurs, charismatic leaders, or drunken dinner guests?, West European Politics, 42:7, 1337-1367, DOI: 10.1080/01402382.2019.1599570 To link to this article: https://doi.Org/10.1080/01402382.2019.1599570 Q © 2019 The Author(s). Published by Informa Pa view supplementary material B" UK Limited, trading as Taylor & Francis Group fffl Published online: 03 May 2019. or Submit your article to this journal D? UiiL Article views: 9708 View related articles □? O CrossMark View Crossmark data D? Citing articles: 17 View citing articles D? Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journal lnformation?journalCode=fwep20 WEST EUROPEAN POLITICS 2019, VOL 42, NO. 7, 1337-1367 https://doi.Org/10.1080/01402382.2019.1599570 Taylor &. Francis Group Routledge 3 OPEN ACCESS A Check for updates The personality of populists: provocateurs, charismatic leaders, or drunken dinner guests? Alessandro Naia O and Ferran Martinez i Comab (j Amsterdam School of Communication Research (ASCoR), University of Amsterdam, Amsterdam, The Netherlands; bCentre for Governance and Public Policy, Griffith University, Brisbane, Australia The 'populist phenomenon' has received a lot of attention in recent years. Yet little is known about the populists themselves: who are they? They are often described as bad-mannered provocateurs disrupting the political game, but also as charismatic leaders able to persuade and motivate. Can a populist 'style' or 'personality' be identified? This article assesses to what extent populists score differently from 'mainstream' politicians on established personality inventories. Using a new dataset based on expert ratings for 152 candidates (including 33 populists) having competed in 73 elections worldwide, it is found that populists score lower on agreeableness, emotional stability and conscientiousness. At the same time, populists score higher on extraversion, narcissism, psychopathy and Machiavellianism. These results have important implications for the study of the success of populists in contemporary democracies and beyond. KEYWORDS Populism; political leaders; personality; Big Five; Dark Triad It is not uncommon to describe populists as adopting a 'transgressive political style' (Oliver and Rahn 2016: 191) that 'emphasises agitation, spectacular acts, exaggeration, calculated provocations, and the intended breech of political and socio-cultural taboos' (Heinisch 2003: 94). Acting at odds with social norms, taking pleasure in displaying 'bad manners' (Moffitt 2016) and behaving as 'drunken dinner guest [s]' (Arditi 2007: 78), populists rely on provocation and a more aggressive rhetoric that sets them apart from other 'mainstream' candidates. Examples abound that support this image of populist candidates as 'bad-mannered provocateurs', from The Netherlands' Geert Wilders being accused of having 'a CONTACT Alessandro Nai @ a.nai@uva.nl © Supplemental data for this article can be accessed online at: http://doi.org/10.1080/01402382.2019.1599570. © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.Org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. ABSTRACT 1338 (g) A. NAI AND F. MARTINEZ I COMA controversial attitude and aberrant political style' (De Landsheer and Kalkhoven 2014: 27) and 'not trying at all to be agreeable' (McBride 2017) to Russia's Vladimir Zhirinovsky described in turn as 'the insane clown prince of Russian politics' (Bruk 2013), 'Russia's Trump' (Nemtsova 2016) and one of 'the usual nut-jobs' (Simpson 2012). At the same time, it is also not uncommon to associate populists with qualities of leadership and charisma. 'Charismatic leadership', a political style that 'helps instill confidence in the leader's capacity to perform' (Barr 2009: 45), seems to be a common characteristic of Latin American left-wing populists, from Chavez in Venezuela to Haya de la Torre in Peru (Roberts 2007), though it is not limited to that region or ideology (Mudde 2004). According to this narrative, populists are able to establish a direct and effective connection with their followers, allowing them to mobilise and persuade them through their energetic, emotional and bold political style (Canovan 1999; Weyland 2001). Charisma is particularly useful in demagogic communication, as it helps politicians 'overcome gaps between their messages and reality' (Barr 2009: 32). Not all scholars agree that charisma is an intrinsic quality of populist leaders - rather, going back to Weberian foundations, some argue that charisma exists in the relationship between the leader and their followers and the way the former is perceived by the latter (e.g. McDonnell 2016). In this case as well, and perhaps even more strongly, the reputation and image of the leader matters. Much attention has been given to the 'populist phenomenon' in recent years, and yet little is known about the populists themselves: who are they? Is there anything like a populist 'style' or 'personality' that can be identified? Two reasons justify such an exercise. First, candidates' personality and public personas are likely to matter with respect to their chances of electoral success and realisation when holding an office (Bartels 2002; Bittner 2011). Evidence exists for instance that narcissism is associated in US presidents with public persuasiveness and 'presidential greatness', but also with unethical behaviour (Watts et al. 2013), or that candidates scoring low in agreeableness tend to be more electorally successful (Joly et al. 2018). At the individual level, several studies suggest that voters assess the personality of candidates and take it into account in their voting decisions (e.g. Caprara and Zimbardo 2004). Thus, showing that populists with a specific personality might contribute to explaining their momentum (or absence thereof) in elections worldwide. In addition, there is virtually no description of a populist today which does not refer directly or indirectly to their 'unusual' character or peculiar political style and, perhaps even more fundamentally, the term 'populism' is often used in news WEST EUROPEAN POLITICS @ 1339 media in a pejorative yet imprecise and inconsistent way (Bale et at. 2011); hence, providing systematic evidence, able to substantiate the perceived character or personality of populists, could help make clear which are warranted characterisations and which rhetoric embellishments. This article provides a systematic comparison of the personality of populists and non-populists worldwide, as assessed by selected samples of experts (Lilienfeld et at. 2012; Parry et at. 2014; Rubenzer et at. 2000; Visser et at 2017). We describe the candidates' personality in terms of both the 'Big Five' (extraversion, agreeableness, conscientiousness, emotional stability and openness: Gerber et at. 2011; Mondak 2010; Vecchione et at. 2011) and the 'Dark Triad' (narcissism, psychopathy and Machiavellianism: Jonason and Webster 2010; Paulhus and Williams 2002). Our analyses rely on a novel dataset (Nai 2018a) that includes information about the personality of 152 candidates (including 33 populists) who have competed in 73 national elections between June 2016 and December 2018. The dataset covers recent national elections in countries across the globe, from Albania to Zimbabwe. It includes recent elections in 'Western' democracies from the USA to France, the UK, Germany, Italy, Sweden, The Netherlands, Iceland, Spain, Austria, Australia and beyond. Furthermore, it includes data about regions of the globe that are less often studied in the literature, such as the African continent (e.g. Ghana, Cote d'lvoire, Morocco, Kenya, Rwanda, Madagascar, Cameroon, Zimbabwe), Eastern Europe (e.g. Bulgaria, Moldova, Romania, Czech Republic, Hungary, Turkey), the Balkans (e.g. Serbia, Montenegro, Croatia, Macedonia, Kosovo), Eastern Asia (Japan, Hong Kong, Malaysia), and Latin America (e.g. Argentina, Mexico, Brazil, Chile, Ecuador, Colombia, Costa Rica). Most importantly, the dataset contains information about the campaigning style of a wide palette of candidates - including many populist candidates, such as Donald Trump, Viktor Orban, Jair Bolsonaro, Recep Tayyip Erdogan, Geert Wilders, Marine Le Pen, Norbert Hofer, Andrej Babis, Matteo Salvini. The full list of elections and candidates is in Online Appendix A. Largely confirming our expectations, our analyses show that populists score significantly lower on agreeableness, emotional stability and conscientiousness; right-wing populists score particularly low on this last trait. At the same time, populists score significantly and substantially higher on the 'dark' traits of narcissism, psychopathy and Machiavellianism, but also on extraversion. The personality of populists The personality of political figures The systematic study of psychological personality traits follows a rich tradition. Within the wealth of approaches to classify the human 1340 (g) A. NAI AND F. MARTINEZ I COMA personality according to a fixed set of traits, the Big Five inventory (Goldberg 1990) is undoubtedly one of the most important. The model identifies five personality traits: extraversion (sociability, energy, charisma), agreeableness (cooperative and pro-social behaviours, conflict avoidance and tolerance), conscientiousness (discipline, responsibility and a sense that life should be organised), emotional stability (calm, detachment, low emotional distress and anxiety) and openness (curiosity, a tendency to create new experiences). Research in political science has fully embraced the Big Five model, which has been shown to affect a wide range of phenomena, such as political attentiveness (e.g. Mondak 2010), attitudes towards political issues (e.g. Gerber et al. 2010), party preferences (e.g. Vecchione et al. 2011) or voting behaviour (e.g. Caprara et al. 1999). This approach is, however, incomplete as 'it does not easily discriminate between various antisocial tendencies, such as a propensity for lying or for being vengeful' (Visser et al. 2017: 282). Indeed, the literature points to an alternative set of personality traits, either as independent constructs (Dark Triad: Jones and Paulhus 2014; Paulhus and Williams 2002) or in conjunction with the other traits (e.g. the HEXACO model: Jonason and McCain 2012). Three main traits usually comprise the 'dark' side of personality: narcissism (ego-reinforcement behaviours, tendency to seek attention and admiration), psychopathy (lack of affect, lack of remorse, insensitivity), and Machiavellianism (tendency to use manipulation and strategic behaviours). These three traits can be qualified as 'malevolent' or 'aversive' but are still 'within the normal range of functioning' (Furnham et al. 2013), in that they do not represent clinical manifestations of disorders (but can be associated with them). Most importantly, these three 'darker' tendencies are separate constructs from the Big Five, and do not simply represent their absence - for instance, narcissism cannot be captured conceptually as the absence of conscientiousness or emotional stability. In recent years, several studies have assessed the personality traits of political figures - a relatively complex endeavour, as we discuss in the methodological section below. Some work assessed the personality of specific candidates (e.g. Donald Trump: Nai and Maier 2018; Visser et al. 2017), whereas other studies presented an assessment of the personality of political figures in general. Thus, for instance, Caprara et al. (2003) assessed the personality of 103 Italian politicians and revealed significantly higher levels of extraversion and agreeableness in politicians when compared with the general public; a similar study of women MPs in Italy found that they scored higher than women voters with respect to extra-version, emotional stability and openness (Caprara et al. 2010). Joly et al. WEST EUROPEAN POLITICS @ 1341 (2018) found that Belgian MPs score particularly high in conscientiousness and emotional stability, and Norgaard and Klemmensen (2018) found that Danish MPs were more extravert, conscientious and open than the average Danish voter. The 'populist personality': three narratives Does anything like a 'populist personality' exist? The considerable (and rapidly expanding) literature on populism does not, to the best of our knowledge, pay particular attention to this question. The maelstrom of research on populism sees it in turn as 'a pathology, a style, a syndrome and a doctrine' (Stanley 2008: 95). Facing the risk of providing an oversimplified picture, we might classify the existing studies on populism in three categories; they see populism, in turn, as an ideology (Albertazzi and McDonnell 2008; Mudde 2004), a communication frame (Aalberg et at. 2017; Jagers and Walgrave 2007) or a performative style (Moffitt 2016; Moffitt and Tormey 2014). The first strand sees populism as an ideological feature of parties and candidates. Within this approach, most research adopts Cas Mudde's definition of populism as a 'thin-centered' ideology 'that considers society to be ultimately separated into two homogeneous and antagonistic groups, 'the pure people' versus 'the corrupt elite', and which argues that politics should be an expression of the volonte generate (general will) of the people' (Mudde 2004: 543). The second strand of research moves beyond the ideological nature of parties and candidates and focuses on the features of their discourse (Aalberg et at. 2017; Jagers and Walgrave 2007; Rooduijn and Pauwels 2011). In this case, populism becomes 'a communication frame that appeals to and identifies with the people, and pretends to speak in their name ...[,] a conspicuous exhibition of closeness to (ordinary) citizens' (Jagers and Walgrave 2007: 322). Research in this tradition focuses on different communication features of the 'populist' language, from the use of appeals to the people to anti-establishment rhetoric, or the use of a simpler language defining communication as a colloquial and informal exercise (Ernst et at. 2019). A third strand of research looks at the performative act of populism and assumes that populists are characterised by a particular political style. Populists refer to 'the people', because they are both their 'central audience ... as well as the subject that populists attempt to render present' (Moffitt and Tormey 2014: 391). Acting at odds with social norms and taking pleasure in displaying 'bad manners' (Moffitt 2016), populists tend to adopt a political style that demarcates them from other 'mainstream' candidates, often exhibiting more bombastic, exaggerated, 1342 (g) A. NAI AND F. MARTINEZ I COMA spectacular and ostensibly provocative behaviour intended to breach all 'political and socio-cultural taboos' (Heinisch 2003: 94). Our article echoes this third line of work. In assessing the existence of a 'populist personality' - or, more precisely, to what extent some personality traits are more likely to appear in populists than in other 'mainstream' political figures - we focus not on populism as a feature of the candidates' ideology, or on their discursive production or campaigning prowess, but rather on who the populists intrinsically are. We are not aware of studies systematically assessing the personality of populist candidates, and we thus enter uncharted territory. We believe that three intuitive narratives help define the boundaries of what a populist personality might entail. First, according to what we might call the 'drunken dinner guesf narrative (Arditi 2007), populists take pleasure in displaying 'bad manners' (Moffitt 2016) by introducing 'a more negative, hardened tone to the debate' (Immerzeel and Pickup 2015: 350) and displaying overall a 'low' style of politics (Ostiguy 2009). This narrative highlights a certain aggressiveness in populist candidates, for instance using 'offensive' discourse 'filled with invectives, ironies, sarcasm, and even personal attacks' (Corbu et at. 2017: 328). In terms of personality traits, this narrative reflects low scores of agreeableness, conscientiousness and emotional stability. The principal dimension of agreeableness is a desire to promote pro-social and communal interactions through conflict avoidance (John et at. 2008), and it thus comes as no surprise that agreeable individuals are usually associated with lower levels of physical and verbal aggression (Tremblay and Ewart 2005). The fact, according to recent comparative evidence (Bakker et at. 2016), that voters low in agreeableness seem to prefer populist parties, provides additional support for this rationale. Simultaneously, conscientious individuals usually show constraint in social interactions, characterised by high self-control (Roberts et at. 2005) and low anger (Jensen-Campbell et at. 2007). The 'drunken' and 'aggressive' display of bad manners thus seems at odds with both agreeableness and conscientiousness. Similarly, individuals low in emotional stability are often described as on edge, anxious and nervous (Mondak 2010), which could explain why they tend to score higher on anger and hostility scales (Tremblay and Ewart 2005). According to this narrative, populists should be perceived as having low scores for those three traits. A second, somewhat similar narrative, which we label the 'agent provocateur narrative, portrays populists as having a political style that 'emphasises agitation, spectacular acts, exaggeration, calculated provocations, and the intended breech of political and socio-cultural taboos' (Heinisch 2003: 94). The 'carnivalesque' style (MacMillan 2017) of the WEST EUROPEAN POLITICS @ 1343 left-wing populists of Podemos in Spain or the Movimento Cinque Stelle (M5S) in Italy illustrates this narrative well. Trump's 'outrageous threats' (Ahmadian et at. 2017) in his 2016 presidential bid - to close off the southern US border, or to throw his opponents in jail - as well as his frequent twitter tantrums are also a good example. The personality components in this narrative are high extraversion, naturally associated with colourful and outrageous rhetoric intended to capture the attention of the audience (Ashton et at. 2007), and low emotional stability. Also central in this narrative is the fact that populists are often accused of playing their outrageous role knowingly, in a conscious attempt to disrupt politics as usual (Oliver and Rahn 2016). For instance, Schmuck et at. (2017: 88) discuss how Austrian Freedom Party (FPO) candidates are known for 'intentionally provoking scandals' for political advantage. Such strategic behaviour is reflected in the 'dark' trait of Machiavellianism: individuals high in this trait are 'cynical, unprincipled, believe in interpersonal manipulation as the key for life success, and behave accordingly' (Furnham et at. 2013: 201). Individuals high in this trait have no qualms in adopting emotionally manipulative behaviours (Austin et at. 2007), and consistent evidence also exists outside the realm of politics that those high in Machiavellianism are willing to provoke scandals and adopt unethical behaviour for the greater good of their organisation or association (Castille et at. 2016). Finally, research on aggressive behaviour shows that individuals scoring high on psychopathy and emotional detachment tend to engage in unprovoked aggression (Reidy et at. 2011) - that is, individuals high in psychopathy are more likely to be the instigators of physical and verbal aggression. Third, the 'charismatic leaders narrative portrays populists as being particularly skilled in establishing a direct and effective connection with their followers, mobilising and persuading them through their energetic, emotional and bold political style (Barr 2009). Some have argued that this charismatic component is a characteristic of the populist leaders themselves (Canovan 1999; Weyland 2001), whereas others suggest that it rather determines the relationship between the leaders and their followers (McDonnell 2016). In this article, in agreement with Mudde (2004), we do not argue that charisma is a defining component of populism; rather, our point is that populists are more likely than other candidates to exhibit traits that are associated with charismatic leadership. In this sense, two personality traits seem relevant: extraversion and narcissism. Extraversion has been shown as a strong and consistent factor determining charismatic leadership (Bono and Judge 2004), especially during turbulent times or in highly competitive situations (De Hoogh et at. 2005), due to the importance of social dominance and bold social interactions for mobilisation 1344 (g) A. NAI AND F. MARTINEZ I COMA Table 1. Narratives and expectations. Big Five Dark Triad Narrative Logic E A C Es 0 N P M 'Drunken dinner guests' Taking pleasure in displaying bad manners, 'low' style of politics, low self-control 'Agents provocateurs' Agitation, spectacular acts, exaggeration, provocations, and breaching of social taboos + + + 'Charismatic leaders' Energetic and bold style, charisma, self-assurance, self-promotion, fearless dominance + + + Expected profile of populists + — — — x + + + +'lncreased prevalence of trait'; - 'Decreased prevalence of trait'; x 'No different prevalence of trait'. E 'Extraversion'; A 'Agreeableness'; C 'Conscientiousness'; Es 'Emotional Stability'; 0 'Openness'; N 'Narcissism'; P 'Psychopathy'; M 'Machiavellianism'. and persuasion of followers. Similarly, political charisma is often associated with narcissistic tendencies (Post 1993; Rosenthal and Pittinsky 2006), due to the dimensions of self-assurance and self-promotion common to both traits. Watts et at. (2013), for instance, show that grandiose narcissism is associated, in US presidents, with public persuasiveness and 'presidential greatness'. Beyond charisma, the 'leadership' component of this narrative might be associated with a reputation of boldness and (subclinical) psychopathy. At the individual level, high scores in subclinical psychopathy have been shown to lead to more successful trajectories in socially competitive 'niches' like business (Babiak and Hare 2006) and politics (Lilienfeld et at. 2012). Those niches reward people who show high levels of individualism and adaptive behaviour, as well as social boldness (or 'fearless dominance'; Lilienfeld et at. 2012) - defined as the 'capacity to remain calm and focused in situations involving pressure or threat' (Patrick et at. 2009: 926), a key component of psychopathic personality (Lilienfeld et at. 2015). In this sense, a charismatic and bold personality shows signs of psychopathic traits. It is not our aim to compare and evaluate the heuristic power of these different narratives, and or to suggest that they are individually (or equally) meaningful in making sense of the populist personality and profile. Rather, we use these narratives broadly to support our intuition that populists behave differently to 'mainstream' candidates, and as guiding lines to sketch the frontiers of a new research agenda on 'populist personality'. Taken together, these three narratives set up a framework for the expected differences in personality traits between populists and non-populists, as illustrated in Table 1. WEST EUROPEAN POLITICS @ 1345 As seen in the bottom row, populists are expected to score higher than mainstream candidates on extroversion and on the Dark Triad, and to score lower on agreeableness, conscientiousness and emotional stability. Only with respect to openness do we not expect to see significant differences between populists and mainstream candidates. Data and methods Measuring candidates' personality via expert ratings Most studies on human personality rely on self-assessments or clinical examinations by psychologists. In the case of political elites, the lack of direct contact between the researchers and the subjects makes these two approaches unpractical. Some studies were able to collect survey data directly from the elites themselves (Dietrich et at. 2012; Joly et at. 2018; Norgaard and Klemmensen 2018), but these studies are isolated and concern very specific populations. A second approach for the study of the personality of political elites consists of relying on psychohistoric analyses of secondary data, such as content analysis of political speeches, for instance via machine learning techniques (Ramey et at. 2016, 2017). This approach has recently shown promising results, and advances in computational power and sophistication of algorithms will enhance even further the possibilities offered. In this article, as others before us (Lilienfeld et at 2012; Nai and Maier 2018; Rubenzer et at. 2000; Visser et at. 2017; Watts et at. 2013), we rely on expert assessments to measure the perceived personality of political figures. Although often limited to selected traits such as narcissism (Glad 2002; Watts et at. 2013), psychopathy (Lilienfeld et at. 2012) or intellectual brilliance (Simonton 2006), all these studies share the idea that external observers ('experts', 'judges') can provide systematic opinions that can be transformed into quantified and comparable measures of perceived character and personality (Parry et at. 2014). An outstanding debate in the discipline is whether external observers are able adequately to assess the psychological profile of individuals without directly interacting with them in diagnostic practices - a debate that reignited in recent times due to the attention given to the current US president's state of mind.1 Historically, the so-called 'Goldwater rule' proscribed the establishment of psychological assessments without direct examination, but many voices have emerged in recent months for its relaxation (e.g. Lilienfeld et at. 2018), based on a misconceived idea that only direct examination can provide systematic information about a psychological profile. Instead, many argue now that observing the behaviour of public figures can provide unbiased and systematic information about 1346 (g) A. NAI AND F. MARTINEZ I COMA their psychological profile (Visser et al. 2017). We agree, and this article follows that line of enquiry. As far as external observers go, we rely on the ratings provided by experts in politics and elections, instead of ratings provided by the public at large; beyond the fact that mass data for as many candidates as in our dataset is unlikely to exist, there is evidence that voters tend to evaluate the personality of public figures in a relatively simplified way (Caprara et al. 2007: 394); for instance, research on 'thin-sliced' decisions suggests that the 'image' of candidates is perceived by voters along the two main heuristics of threat/dominance and competence (Spezio et al. 2012). We expect experts, because they are more informed and 'neutral', to be less likely to provide excessively simplified assessments. Dataset We use a new comparative dataset about the campaigning strategies of candidates competing in elections worldwide (NEGex; Nai 2018a; Nai and Maier 2018).2 It covers all national elections held across the world between June 2016 and December 2018. Data is gathered through a systematic survey distributed to election-specific samples of national and international election experts in the weeks following each election. The experts evaluated several aspects of the election, including the personality of a subsample of the 2-3 leading candidates through separate batteries for the Big Five and the Dark Triad (see below). Evaluations were then aggregated at the candidate level. After excluding missing values on all relevant variables and considering only candidates for which at least two3 experts provided evaluations, our models are run on 152 candidates having competed in 73 elections worldwide. Information is based on answers provided by 1280 experts, aggregated at the candidate level. Online Appendix A lists all elections and candidates in our dataset and specifies the number of expert opinions gathered for each candidate. The geographical coverage of the data-set is illustrated in Figure 1. In the absence of a large comparative survey, expert judgements are the most efficient and reliable approach to provide systematic information about perceptions of candidates' reputations. Data gathering is cost-effective, and questioning scholars with proven expertise increases the chances that the main concepts tapped are understood in a similar fashion - thus reducing the risk of cross-cultural biases in comparison. Furthermore, relying on scholars allows expansion of the coverage of the data, as issues such as linguistic expertise or knowledge of cultural dynamics are 'outsourced' to the experts themselves - virtually all contexts can be studied, providing that relevant experts are identified. Finally, expert 1348 (g) A. NAI AND F. MARTINEZ I COMA ratings can be controlled by each expert's familiarity with the topic studied, which increases the reliability of the aggregated scores. Experts As discussed in Nai (2018a), in our case an 'expert' is a national or international scholar with expertise in electoral politics, political communication and/or electoral behaviour for the country where the election was held. We established expertise via the presence of at least one of the following criteria: (1) relevant publications; (2) position held (e.g., professor of electoral behavior) and classes taught; (3) membership of a relevant research group, professional network or organised section of such a group; (4) explicit self-assessed expertise on a professional webpage (e.g. bio on university web-page). Experts were contacted during the week following the election and invited to complete an online questionnaire, in English; they were sent two reminders, respectively one and two weeks afterwards. On average the 1280 experts that rated the candidates presented in this article lean slightly to the left on a 0-10 left-right scale (M = 4.33, SD = 1.78), 76% are domestic (that is, work in the country for which they were asked to evaluate the election) and 33% are female. Experts declared themselves very familiar with the elections (M = 8.06, SD = 1.75), and estimated that the questions in the survey were relatively easy to answer (M = 6.56, SD = 2.36); both variables range between 0 'very low' and 10 'very high'. Table Dl (Online Appendix D) presents the average profile of experts, according to these five dimensions, for each election surveyed. The profile of experts can, potentially, alter their assessments (Martinez i Coma and Van Ham 2015; Steenbergen and Marks 2007; Wright and Tomlinson 2018). To assess the extent of these profile effects, we ran a series of models where the experts' evaluations (that is, how they evaluated the personality profile of the candidates) were regressed on their profile. The results, in Online Appendix D (Tables D2 and D3), show that the experts' profile affects their evaluations only very marginally. We also tested whether the composition of the experts' sample for each candidate (that is, the average profile of experts evaluating each candidate) affected how candidates were perceived. Results, discussed in the robustness checks section (Tables B6 and B7 in the online appendix), show again that this is not the case. Personality measures The Big Five are measured through the Ten Items Personality Inventory (TIPI: Gosling et at. 2003). For each trait experts had to evaluate two statements (e.g. the candidate might be someone that is 'critical, WEST EUROPEAN POLITICS @ 1349 quarrelsome'), and the underlying personality trait exists as the average value for those statements. Compared to other measures the TIPI is relatively short, and thus less nuanced; however, it has been shown to provide satisfactory results in terms of convergent validity (Ehrhart et al. 2009). The battery of questions yields five variables ranging between 0 'very low' and 4 'very high'. The measure of the three 'dark' personality traits is usually done through lengthy batteries of question such as the 40-item NPI for narcissism, the 20-item Mach-IV for Machiavellianism or the 31-item SRP III for psychopathy (Paulhus and Williams 2002). We designed a shorter version of the 'Dirty Dozen' battery by Jonason and Webster (2010), based on the principal component analyses described in their study (Jonason and Webster 2010: 422); we selected the two items that correlated the highest with each trait and used them as a battery (see supplementary material and Table 1). As for the Big Five, the Dark Triad variables range from 0 'very low' to 4 'very high'. Online Appendix E provides more details about the measures of personality used here. Reliability of the eight traits is relatively high overall,4 even if, in the case of very short inventories using two statements per trait, like the TIPI, priority is often given to theoretical validity over reliability (Gosling et al. 2003) because statements are chosen to reflect different facets of each trait, which improves validity but potentially reduces reliability. Construct consistency is also high (Tables El and E2). The relationship between the Big Five and Dark Triad traits has been shown to be sometimes erratic (e.g. Lee and Ashton 2005), but several similar patterns are often reported - for instance, agreeableness has been shown to correlate negatively with all the three 'dark' traits, conscientiousness is negatively associated with psychopathy and Machiavellianism, and narcissism is positively associated with extraversion (Paulhus and Williams 2002). We find all those patterns in our data as well (Table E3). The external validity of our measures is harder to assess, due to the absence of comparable data about the personality, character or public personas of candidates competing in elections worldwide. We were, however, able to collect information about the public personas, character and personality for 68 candidates as described independently in news media, reports and scientific publications (see Table E5 in the online appendix). As appears quite clearly from the online appendix, very often the image of those candidates converges closely with the measure of personality in our dataset. Identifying populist candidates No comprehensive 'repertoire of populism' exists that covers all candidates in our database. To create such a repertoire, we assessed whether or 1350 (g) A. NAI AND F. MARTINEZ I COMA not each candidate in our dataset was referred to as 'populist' in relevant published research. We relied on the few existing comparative works (Inglehart and Norris 2016; Mudde 2007), systematic collections of case studies (Aalberg et at. 2017; Albertazzi and McDonnell 2008) and additional single case studies for selected countries (e.g. Bos and Brants 2014; Gurov and Zankina 2013), all based on similar definitions of 'populism' as an ideology that advocates people-centrism and anti-elitism (Mudde 2004) or more generally an opposition between the 'common people' and the (corrupt, wicked) elites. Some of the work collected refers to populism in general (Aalberg et at. 2017; Rooduijn and Pauwels 2011), whereas some focuses on particular types such as right-wing populism (Ennser 2012; Mudde 2007). We thus identified 33 candidates (out of 152)5 who can be qualified as 'populists'. In the majority of cases multiple independent scientific references per candidate were identified,6 which we use as an indicator of converging consensus over the populist nature of any given candidate in the list. The list of all populist candidates, including the references used to establish the classification, is presented in Online Appendix C. Covariates Our models control for a series of covariates. At the candidate level, we control for the incumbency status and ideology of candidates through a scale ranging from 1 'far left' to 7 'far right'; this control is important, as some traits are sometimes associated with political ideology (e.g. conscientiousness with conservative/right-wing ideology and openness with liberal/ left-wing ideology; Gerber et at. 2011). We then control for the gender of candidates, their age and their electoral success (percentage of votes received during the election). At the contextual level, we use a binary variable that separates countries with a PR electoral system (including mixed member proportional) from countries with a plurality/majority system (including mixed member majoritarian; Gallagher 2014); we adapt the formula proposed by Laakso and Taagepera (1979) for the effective number of parties (ENPP) to measure the total (effective) number of candidates. Finally, a binary variable sorts presidential (2) from legislative (1) elections. Descriptive statistics are presented in Table 2. Results Populists Table 3 illustrates the personality scores of the 33 populist candidates in our dataset, obtained by aggregating the expert ratings on the two batteries for the Big Five and Dark Triad. At a glance, it appears that Table 2. Descriptive statistics (missing values excluded). Level Variable Measure N Mean Std. dev. Min Max Candidate Extraversion 0 'Very low' to 4 'Very high' 152 2.30 0.77 0.13 3.75 Agreeableness 0 'Very low' to 4 'Very high' 152 1.80 0.72 0.18 4.00 Conscientiousness 0 'Very low' to 4 'Very high' 152 2.67 0.67 0.68 3.80 Emotional stability 0 'Very low' to 4 'Very high' 152 2.27 0.81 0.43 4.00 Openness 0 'Very low' to 4 'Very high' 152 1.98 0.72 0.13 3.83 Narcissism 0 'Very low' to 4 'Very high' 152 2.77 0.70 0.63 4.00 Psychopathy 0 'Very low' to 4 'Very high' 152 2.19 0.84 0.17 3.85 Machiavellianism 0 'Very low' to 4 'Very high' 152 2.19 0.77 0.33 3.75 Populist 0 'Not populist', 1 'Populist' 152 0.22 0.41 0.00 1.00 Incumbent 0 'Challenger', 1 'Incumbent' 152 0.32 0.47 0.00 1.00 Success Percentage of votes received (0-100%) 152 30.25 16.47 0.48 88.61 Left-right 1 'Far left' to 7 'Far right' 152 4.24 1.57 1.00 7.00 Female 0 'Male', 1 'Female' 152 0.14 0.35 0.00 1.00 Year born Year of birth 152 1961.13 12.00 1925 1990 Election Electoral system: PR 0 'Plurality/majority', 1 'PR' 73 0.66 0.48 0.00 1.00 Effective N of candidates Continuous score (Laakso and Taagepera 1979) 73 3.96 1.80 1.02 8.55 Presidential election 1 'Legislative', 2 'Presidential' 73 1.42 0.50 1.00 2.00 1352 (g) A. NAI AND F. MARTINEZ I COMA Table 3. Personality of populists in our sample; average scores. Big Five Dark Triad E A C Es 0 N P M Jimmie Äkesson Sweden 2.50 0.71 2.50 1.64 1.14 2.07 2.64 1.93 Andrej Babis Czech Republic 2.86 1.14 2.86 1.82 2.50 3.44 2.83 2.91 Jair Bolsonaro Brazil 2.75 1.15 1.83 1.03 1.60 3.50 3.17 3.33 Boyko Borisov Bulgaria 3.00 1.33 1.83 1.17 2.02 3.50 2.60 3.10 Luigi Di Maio Italy 2.48 1.21 2.60 1.83 1.64 2.83 1.83 2.28 Milo Dukanovic Montenegro 2.40 1.60 3.40 2.20 1.60 2.98 2.83 2.75 Ivan Duque Märquez Colombia 2.22 1.95 3.19 3.02 1.26 2.58 1.83 1.75 Recep Tayyip Erdogan Turkey 3.16 0.63 2.69 0.75 2.06 3.63 3.50 2.92 Arlene Foster Northern Ireland 1.73 0.85 2.17 1.71 1.29 2.54 3.69 2.66 Alexander Gauland Germany 2.21 0.34 2.39 1.99 1.27 3.10 3.58 2.62 Nikola Gruevski Macedonia 2.13 1.07 2.01 1.23 1.22 3.83 3.85 3.29 Norbert Hofer Austria 2.33 1.45 3.16 2.20 1.31 3.06 2.84 3.28 Siv Jensen Norway 2.67 1.00 3.00 2.00 2.08 2.39 2.57 1.68 Imran Khan Pakistan 2.88 1.75 2.38 1.88 2.75 2.95 2.19 1.96 Cr. Fernandez de Kirchner Argentina 3.00 0.33 2.97 0.50 2.30 3.13 2.28 2.50 Albin Kurti Kosovo 3.08 2.13 3.58 2.92 2.88 2.83 1.33 0.67 Marine Le Pen France 2.93 1.07 2.43 2.14 2.01 3.00 3.26 3.28 Andres M. Lopez Obrador Mexico 2.54 1.43 2.14 1.75 1.96 3.29 2.16 2.08 Jean-Luc Melenchon France 3.42 1.50 2.33 0.73 3.05 3.25 1.96 2.67 Paul Nuttall UK 2.33 1.08 1.00 1.50 1.17 3.25 3.00 2.13 Michelle O'Neill Northern Ireland 2.92 2.35 3.06 3.10 2.17 2.61 1.90 2.17 Tomio Okamura Czech Republic 2.90 0.60 2.28 0.98 1.63 3.73 3.20 2.92 Viktor Orbän Hungary 3.10 0.80 3.33 2.00 2.10 3.50 3.25 2.88 Daniel Ortega Nicaragua 0.83 0.33 2.17 0.67 0.67 3.33 3.50 2.50 Vladimir Putin Russia 2.24 2.00 2.92 2.92 1.93 3.67 3.25 2.67 Matteo Salvini Italy 3.03 1.40 1.98 1.10 2.43 2.83 2.42 2.33 Ivan Vilibor Sincic Croatia 2.40 0.80 2.10 1.30 2.80 2.83 1.50 2.17 Heinz-Christian Strache Austria 3.33 1.17 3.33 2.33 1.00 3.31 2.38 2.11 Donald Trump USA 3.61 0.18 0.68 0.43 1.88 3.91 3.66 3.44 Aleksandar Vucic Serbia 2.30 1.20 2.10 0.90 1.60 3.67 2.67 2.50 Geert Wilders The Netherlands 2.50 0.41 2.73 1.63 1.50 3.11 3.58 1.86 Vladimir Zhirinovsky Russia 3.55 0.75 0.89 0.72 1.50 3.31 2.86 3.38 Gennady Zyuganov Russia 1.63 1.00 2.57 2.53 0.94 2.25 2.63 2.50 Note: all variables range between 0 'very low' and 4 'very high'. E 'Extraversion'; A 'Agreeableness'; C 'Conscientiousness'; Es 'Emotional Stability'; 0 'Openness'; N 'Narcissism'; P 'Psychopathy'; M 'Machiavellianism'. populists tend to score high on perceived extraversion (an exception being Nicaragua's Daniel Ortega, who a former colleague described as having 'a prison personality: Lonely, solitary, mistrustful, hard'7) but lower on all four remaining socially 'desirable' traits, and especially on agreeableness. Their scores on the three 'dark' traits are also relatively high overall. As perhaps the best symptomatic example, our data reveal that Trump's personality is characterised by extreme extraversion, off-the-charts narcissism, high psychopathy and Machiavellianism, very low agreeableness and low emotional stability - all of which is in line with several accounts published elsewhere (e.g. Hamblin 2016; McAdams 2016, 2016; Nai and Maier 2018; Nai et al. 2019; Olbermann 2016; Visser et al. 2017). A series of t-tests reveal that the personality of these 33 populists is relatively homogeneous in terms of left-right ideology and gender. Even WEST EUROPEAN POLITICS @ 1353 though the sample is not big enough to detect small but meaningful differences due to low statistical power, t-tests highlight only a handful of significant differences for these two characteristics. Right-wing populists8 score significantly higher on psychopathy (M = 2.99, SD = 0.11) than left-wing populists (M = 2.32, SD = 0.23); the difference between the two is quite substantial,9 £(29) = -2.88, p<0.007, d=1.07, and in line with studies finding a positive association between right-wing ideology and social dominance orientation (e.g. Perry and Sibley 2012). Female populists (N—5) score significantly lower on narcissism (M = 2.73, SD = 0.14) than male populists (M = 3.20, SD = 0.08), £(31) = 2.22, p < 0.034, d = 0.80, which is in line with results found for leaders in general (e.g. Jorstad 1996). Incumbency status does play a bigger role, with four traits out of eight showing significantly different scores across challengers (N—26) and incumbents (N—7); incumbent populists score higher than challengers on narcissism (incumbents: M = 3.43, SD = 0.16; challengers: M = 3.04, SD = 0.09, £(31) = -2.03, p < 0.050, d = 0.73), psychopathy (incumbents: M = 3.38, SD = 0.15; challengers: M = 2.58, SD = 0.12, £(31) = -3.14, p< 0.004, d=1.13), and Machiavellianism - this last only atp<0.1 (incumbents: M = 2.86, SD = 0.10; challengers: M = 2.43, SD = 0.13, £(31) = —1.70, p < 0.099, d — 0.61). However, incumbents score lower than challengers on extraversion at£<0.1 (incumbents: M = 2.31, SD = 0.32; challengers: M = 2.72, SD = 0.09, £(31) = 1.71, p < 0.097, d = 0.61). These results suggest that some personality traits are more likely to drive electoral success for populist candidates, much in the same way as they are important predictors of electoral success for all candidates in general (Joly et al. 2018; Nai 2018a, 2018b). Populists vs. mainstream candidates Moving towards our main research question, Figure 2 illustrates the average differences in personality traits between populists and 'mainstream' candidates. A series of t-tests suggest that all these differences are statistically significant, even if only at p < 0.1 in one case (openness). Thus, populists score significantly higher than mainstream candidates on extraversion (£(150) = -2.85, p < 0.005, d = 0.47), narcissism (£(150) = -3.47, p < 0.001, d = 0.57), psychopathy (£(150) =-4.62, p< 0.000, d = 0.75) and Machiavellianism (£(150) =-2.85, p < 0.005, d = 0.47). At the same time, populists score significantly lower than mainstream candidates on agree-ableness (£(150) = 7.15, p < 0.000, d—1.17), conscientiousness (£(150) = 2.18, p< 0.031, d = 0.36), emotional stability (£(150) = 5.30, p< 0.000, d = 0.87), and openness - but only at p<0.1 (£(150) = 1.67, p< 0.097, d =0.27). 1354 (g) A. NAI AND F. MARTINEZ I COMA Extraversion -1-1- Mainstream Populist Agreeableness -1-1- Mainstream Populist Conscientiousness 8 3 CO I 2- -1-1- Mainstream Populist Emotional stability Openness Narcissism Mainstream Populist Mainstream Populist Mainstream Populist Psychopathy Machiavellianism Mainstream Populist 8 2- l-H Mainstream Populist Figure 2. Average personality scores of populists and mainstream candidates. N (mainstream candidates) = 119 N (populists) = 33 These trends are all in line with the expectations presented earlier (Table 1). To reduce the risk of spurious results, we next test for the effect of populism on the personality of candidates, controlling for several covariates at the candidate and context levels. Table 4 reports models for the five 'socially desirable' traits (Big Five), whereas Table 5 reports the same models but for the 'socially malevolent' traits (Dark Triad). Table 4. Perceived Big Five traits by candidate profile and context. Extraversion Agreeableness Conscientiousness Emotional stability Openness Coef Se Sig Coef Se Sig Coef Se Sig Coef Se Sig Coef Se Sig Populist 0.41 (0.17) * -0.83 (0.13) -0.41 (0.14) ** -0.86 (0.16) -0.17 (0.15) Incumbent -0.21 (0.16) -0.22 (0.13) t -0.28 (0.13) * -0.12 (0.15) -0.53 (0.15) Success -0.00 (0.00) 0.00 (0.00) 0.01 (0.00) ** 0.01 (0.00) 0.01 (0.00) Left-right 0.04 (0.04) -0.09 (0.03) ** 0.01 (0.03) -0.04 (0.04) -0.08 (0.04) Female 0.21 (0.19) -0.05 (0.15) 0.03 (0.15) 0.01 (0.18) 0.25 (0.17) Year born 0.00 (0.00) -0.00 (0.00) -0.00 (0.00) -0.00 (0.00) * 0.00 (0.00) Proportional rep. 0.21 (0.16) 0.03 (0.12) 0.31 (0.12) * 0.08 (0.14) 0.05 (0.14) EN candidates -0.01 (0.05) -0.01 (0.04) 0.04 (0.04) -0.00 (0.04) 0.01 (0.04) Presidential elect -0.12 (0.18) -0.21 (0.14) -0.13 (0.14) -0.07 (0.16) -0.20 (0.16) MENAa 0.08 (0.27) -0.12 (0.21) -0.21 (0.22) -0.44 (0.25) t 0.19 (0.24) Sub-Sah Africa 0.68 (0.32) * -0.03 (0.25) -0.53 (0.25) * -0.89 (0.29) ** 0.18 (0.28) Lat Am & Car 0.08 (0.25) -0.20 (0.19) -0.54 (0.19) ** -0.72 (0.22) ** -0.21 (0.21) Ctr & S Asia 0.61 (0.37) t -0.03 (0.28) -0.04 (0.29) -0.75 (0.33) * 0.12 (0.32) E & SE Asia 0.17 (0.28) -0.58 (0.21) ** -0.52 (0.21) * -0.63 (0.25) * 0.00 (0.24) Eastern Europe 0.18 (0.22) -0.38 (0.16) * -0.41 (0.17) * -0.59 (0.20) ** -0.19 (0.19) Southern Europe -0.03 (0.23) -0.29 (0.17) t -0.51 (0.18) ** -0.54 (0.20) ** 0.00 (0.20) Intercept 0.91 (1.69) 4.43 (1.32) 4.04 (1.36) ** 6.03 (1.57) 1.38 (1.52) N (candidates) N (elections) R2 Model Chi2 152 73 0.136 21.07 152 73 0.379 82.40 152 73 0.249 44.72 152 73 0.316 62.42 152 73 0.185 30.71 Note: All models are random-effect hierarchical linear regressions (HLM) where candidates are nested within elections. All dependent variables vary between 0 'very low' and 4 'very high'. aFor all regions, the reference category is 'Western and Northern Europe' (includes the USA, Australia and New Zealand). East & SE Asia includes Melanesia, Micronesia and Polynesia. p < 0.001; **p < 0.01; *p < 0.05; +p < 0.1. 1356 (g) A. NAI AND F. MARTINEZ I COMA Table 5. Perceived Dark Triad traits by candidate profile and context. Narcissism Psychopathy Machiavellianism Coef Se Sig Coef Se Sig Coef Se Sig Populist 0.37 (0.14) ** 0.60 (0.16) 0.35 (0.15) * Incumbent 0.21 (0.13) t 0.41 (0.16) ** 0.46 (0.15) ** Success -0.00 (0.00) 0.00 (0.00) 0.00 (0.00) Left-right 0.12 (0.03) 0.16 (0.04) 0.13 (0.04) ** Female -0.42 (0.16) ** 0.01 (0.18) -0.09 (0.18) Year born 0.00 (0.00) t 0.00 (0.00) 0.00 (0.00) Proportional rep. -0.06 (0.16) -0.12 (0.15) -0.30 (0.15) * EN candidates 0.02 (0.05) 0.06 (0.05) 0.08 (0.05) Presidential elect 0.25 (0.17) 0.21 (0.17) 0.33 (0.17) t MENAa -0.28 (0.26) -0.12 (0.26) -0.17 (0.26) Sub-Sah Africa 0.24 (0.30) 0.11 (0.30) 0.22 (0.30) Lat Am & Car 0.27 (0.24) 0.22 (0.23) 0.44 (0.23) t Ctr & S Asia 0.01 (0.35) 0.26 (0.34) -0.10 (0.35) E & SE Asia 0.10 (0.27) 0.19 (0.26) -0.04 (0.26) Eastern Europe 0.19 (0.22) 0.36 (0.20) t 0.40 (0.21) t Southern Europe 0.13 (0.22) 0.20 (0.21) 0.31 (0.22) Intercept -0.61 (1.44) -1.43 (1.62) -1.67 (1.56) N (candidates) N (elections) R2 Model Chi2 152 73 0.255 47.47 152 73 0.319 63.13 152 73 0.264 48.90 Note: All models are random-effect hierarchical linear regressions (HLM) where candidates are nested within elections. All dependent variables vary between 0 'very low' and 4 'very high'. aFor all regions, the reference category is 'Western and Northern Europe' (includes the USA, Australia and New Zealand). East & SE Asia includes Melanesia, Micronesia and Polynesia. ***p < 0.001; **p < 0.01; *p < 0.05; +p < 0.1. Looking first at the Big Five (Table 4), the image of populists as 'drunken dinner guests' with bad manners seems to find confirmation across the candidates in our database. Keeping constant characteristics of candidates such as their incumbency status, electoral success, ideological position and socio-demographic profile, as well as characteristics of the context in which the election took place, our models show that populists score significantly lower on agreeableness, conscientiousness and emotional stability. The difference between populists and 'mainstream' candidates is particularly strong for agreeableness and emotional stability, suggesting that populists suffer from a disadvantage in terms of image -at least in terms of socially desirable reputation traits. Table 4 shows, furthermore, the higher score on agreeableness for populists, shown in a bivariate way, resisting the inclusion of controls at the candidate and context levels. We did not expect any particular effect for openness - and, indeed, this is the only trait that does not differ significantly between populists and mainstream candidates. All these trends confirm our expectations of the 'socially desirable' Big Five. Our expectations also find support when it comes to the three 'dark' traits (Table 5). Populists score significantly higher on narcissism, Machiavellianism and especially WEST EUROPEAN POLITICS @ 1357 psychopathy. Coupled with the higher scores shown for extraversion, high scores on the three 'dark' traits go in the direction of both the 'agent provocateur' and 'charismatic leader' narratives described earlier. Robustness checks Online Appendix B presents several robustness tests. First, we reversed the logic and tested to what extent personality traits determine whether candidates are populists or not. Table Bl presents four logistic regressions for the Big Five (Ml and M2) and Dark Triad (M3 and M4), respectively without and with controls at the candidate and context levels. These models test for the joint effect of all personality traits within each inventory, to ensure for mutual control. Results show that candidates scoring high in extraversion and low on agreeableness are more likely to be classified as populists, and so are candidates scoring high in narcissism and psychopathy. Indeed, with the data at hand, we are of course not able to assess to what extent people with (latent) populist traits do not self-select into populist movements - that is, we are not able to test for the fully reversed causal story that personality drives the populist status. In this sense, results in Table Bl simply suggest that reversing the independent and dependent variables in our models does not lead, overall, to a radically different conclusion on the association between personality traits and populism. Further research should disentangle the direction of causality in a more detailed fashion, for instance via longitudinal data. Second, we showed elsewhere that populist candidates are more likely to campaign in a harsh and negative way (Nai 2018a, 2018b); to ensure that it is not the populists' campaign that drives their personality perception, we added to the general models an interaction effect between the tone of the candidates' campaign (positive vs. negative) and the variable that sorts populists from mainstream candidates; results (Tables B2 and B3) show that this should not be the case, as no significant interaction is found across the board. Third, we assessed to what extent the dynamics at play are influenced by differences between a left-wing versus a right-wing ideology of candidates. We replicated the main models adding an interaction effect between the ideology of the candidate and the variable that sorts populists from mainstream candidates (Tables B4 and B5) and found no effects except for conscientiousness: right-wing populists score lower on conscientiousness than right-wing non-populists, whereas the difference between the two is not significant for left-wing candidates; the effect is substantiated in Figure 3, via marginal effects. Fourth, we replicated the models controlling for the average experts' profile - that is, the composition of each country sample of experts along 1358 (g) A. NAI AND F. MARTÍNEZ I COMA 25 20 15 10 Q- Left-right ideology of the candidate Figure 3. Conscientiousness by populism * left-right. Note: marginal effects with 95% CI, based on coefficients in Table B4. Left-right varies between 1 'Far left' and 7 'Far right'. ---Mainstream - Populist some major characteristics (average familiarity with the election, average difficulties in answering the questionnaire, percentage of female experts, average left-right position and percentage of domestic experts); results are presented in Tables B6 and B7. We find some scattered effects (e.g. candidates are evaluated as more extrovert when on average more female experts are in the sample), but most importantly all results discussed above resist and are thus not affected by the composition of the expert samples. Finally, Tables B8 and B9 replicate the models but only for candidates for which at least five independent experts provided measures of candidates' reputation. Results, although based on a smaller subsample of candidates, are robust. Discussion and conclusion As we write these lines, populism is increasing in momentum. Although the electoral success of populist candidates and parties in some countries might be weaker than some observers predicted, it is undisputable that the topic is a regular 'guest' in the news media headlines and at scientific conferences. Beyond its ubiquity, this attention seems to be accompanied by a specific set of narratives that describe populists as particular political animals with distinct character traits - unpleasantness, proclivity for provocation, charisma, aggressiveness and a political style ostensibly at odds with social norms (Arditi 2007; Heinisch 2003; Moffitt 2016). Yet, to WEST EUROPEAN POLITICS (g) 1359 the best of our knowledge, very little comparative evidence exists that the public persona of populist candidates indeed differs from the one of their 'establishment' counterparts. In this article, we have provided systematic evidence of this, by comparing the personality of populist and non-populist candidates competing in elections worldwide over the course of a year, via ratings provided by independent scholars. Based on the reputation of 152 candidates (including 33 populists) having competed in 73 elections across the globe, our results suggest that populists score significantly lower than non-populists in perceived agreeableness, conscientiousness and emotional stability but score significantly higher in perceived extraversion, narcissism, psychopathy and Machiavellianism. All in all, from our analyses emerges a pattern of 'populist reputation' which portrays them as disagreeable, narcissistic and potentially unhinged, yet extrovert and socially bold - in short, bad-tempered and provocative, but charismatic. These results matter for a deeper understanding of the populist phenomenon in four ways. First, the large-scale scope of the data illustrates dynamics that are not (or less) bound to specific geographical and political contexts. In this sense, we contribute to research on populism beyond some well-known cases, towards the development of a comparative systematic understanding of the phenomenon, in line with recent collections of systematic multi-country studies (Aalberg et al. 2017; Albertazzi and McDonnell 2008). Second, our results are informative about the resonance of entrenched narratives that are commonly used when describing populists worldwide. If, for a given candidate, a consensus seems to exist between scholars and commentators about their character and public image, then the chances are that such traits are also perceived by the public at large - and, potentially, resonate in its electoral choices. Third, there is evidence that the public personas of political figures are not unrelated to their performance (Bartels 2002; Joly et al. 2018). For instance, Watts et al. (2013) show that US presidents higher in reported narcissism score more favourably in public persuasiveness and 'presidential greatness' but are also more likely to be associated with unethical behaviour (see also Glad 2002). Fourth, and more importantly, our result could help shed light on the populist phenomenon: that is, the (real or alleged) increase of populist movements across established democracies. Indeed, two alternative rationales link candidates' personality with their electoral success. On the one hand, voters might be more likely to support candidates with personalities that 'match' their own, as individuals with congruent personality profiles tend to 'like' each other (e.g. Selfhout et al. 2010). In politics, evidence of congruence exists between party leaders and their supporters (Caprara et al. 2003); more generally, voters tend to select candidates 1360 (g) A. NAI AND F. MARTINEZ I COMA whose traits match their own (Caprara and Zimbardo 2004; but see Klingler et at. 2018). Fortunato et at. (2018) suggest that similar mechanisms were also at play during the 2016 US election. On the other hand, certain candidate personality profiles could be more appealing for some voters but not for others. Although the evidence in this sense is more limited, some studies suggest, for example, that voters scoring low in agreeableness are more likely to support populist candidates (Bakker et at. 2016), who have been shown to exhibit a specific set of personality traits - as we discussed in this article. Similarly, we could expect candidates high in conscientiousness to be preferred by voters on the right, and candidates high in openness and agreeableness by voters on the left, reflecting correlations between ideology and personality found at the individual level (Gerber et at. 2011; Vecchione and Caprara 2009). In this sense, then, providing systematic evidence of the personality of populists might contribute to explaining their momentum (or absence thereof) in elections worldwide, now and in the future. Notes 1. See, for instance: Dan P. McAdams, 'The mind of Donald Trump', The Atlantic, June 2016; Chris Cillizza, 'The case for Donald Trump's mental fitness', CNN online, 4 January 2018. 2. https://www.alessandro-nai.com/negative-campaigning-comparative-data 3. Below, we discuss a series of robustness checks that replicate the analyses but with a more restrictive condition (minimum five independent experts per candidate); results are overall very similar, although based on a smaller sample of candidates. 4. The reliability scores are, respectively, a = 0.74 (extraversion), a = 0.59 (agreeableness), a = 0.78 (conscientiousness), a = 0.84 (emotional stability), a = 0.65 (openness), a = 0.85 (narcissism), a = 0.89 (psychopathy), and a = 0.78 (Machiavellianism). 5. Two additional candidates could have made the list: Spain's Pablo Iglesias (Podemos) and Hungary's Gabor Vona {Jobbik). For these two candidates not enough experts provided their ratings on the personality batteries, which is the reason why we excluded them from our analysis altogether. 6. We also ran tests based on a more restrictive classification that codes candidates for which only one scientific reference was found as non-populists (this is the case of three candidates; Kosovo's Albin Kurti, Montenegro's Milo Dukanovic and Russia's Gennady Zyuganov). All effects are virtually identical, regardless of the classification used. Results for these additional tests are available from the authors upon request. 7. Ed Vulliamy, 'In the lions' den again', The Guardian, 2 September 2001. 8. We used a simplified variable that sorts populists on the left (including centre-left; N=8) from populists on the right (including centre-right; N=23); two candidates are excluded from this t-test because they are considered as ideologically centrists (Croatia's Ivan Vilibor Sincic and Italy's Luigi di Maio). WEST EUROPEAN POLITICS @ 1361 9. To compute the effect size (Cohen's d) for independent samples t-tests, we used the following approximation: (2 x t)/^{df). Acknowledgements We are very grateful to the anonymous reviewers and the journal editors for their constructive comments and suggestions; any remaining mistakes are our responsibility alone. A. Nai acknowledges financial support from the Swiss National Science Foundation (Grant ref P300P1_161163) and the material support provided by the Electoral Integrity Project (Harvard University and University of Sydney). Special thanks to Pippa Norris and Duncan McDonnell for their inputs. The dataset described in this article can be found at https://www.alessandro-nai. com/negative - campaigning- comparative - data. Disclosure statement No potential conflict of interest was reported by the authors. Notes on contributors Alessandro Nai is Assistant Professor in Political Communication and Journalism at the University of Amsterdam. Recent publications have appeared in European Journal of Political Research, Political Psychology, Electoral Studies, Personality and Individual Differences, Government & Opposition, Journal of Political Marketing, Presidential Studies Quarterly, and more. He is Associate Editor of the Journal of Social and Political Psychology, [a.nai@uva.nl] Ferran Martinez i Coma is lecturer at Griffith University. His research interests include: political behaviour, public opinion and electoral integrity. He has published in European Journal of Political Research, Electoral Studies, Democratization and Party Politics, amongst others, [f.martinezcoma@griffith.edu.au] ORCID Alessandro Nai © http://orcid.org/0000-0001-7303-2693 Ferran Martinez i Coma http://orcid.org/0000-0002-3486-6596 References Aalberg, T., F. Esser, C. Reinemann, J. Stromback, and C. H. de Vreese (2017). Populist Political Communication in Europe. New York: Routledge. Ahmadian, S., S. Azarshahi, and D. L. Paulhus (2017). 'Explaining Donald Trump via Communication Style: Grandiosity, Informality, and Dynamism', Personality and Individual Differences, 107, 49-53. Albertazzi, D., and D. McDonnell, eds. (2008). Twenty-First Century Populism. The Spectre of Western European Democracy. Houndmills: Palgrave. Arditi, B. (2007). Politics on the Edges of Liberalism: Difference, Populism, Revolution, Agitation. Edinburgh: Edinburgh University Press. 1362 (g) A. NAI AND F. MARTINEZ I COMA Ashton, M., J. Kushner, and D. Siegel (2007). 'Personality Traits of Municipal Politicians and Staff, Canadian Public Administration, 50:2, 273-89. Austin, E. J., D. Farrelly, C. Black, and H. Moore (2007). 'Emotional Intelligence, Machiavellianism and Emotional Manipulation: Does EI Have a Dark Side?', Personality and Individual Differences, 43:1, 179-89. Babiak, P., and R. D. Hare (2006). Snakes in Suits: When Psychopaths Go to Work. New York: Regan Books. Bakker, B. N., M. Rooduijn, and G. Schumacher (2016). 'The Psychological Roots of Populist Voting: Evidence from the United States, the Netherlands and Germany', European Journal of Political Research, 55:2, 302-20. Bale, T., S. Van Kessel, and P. Taggart (2011). Thrown Around with Abandon? Popular Understandings of Populism as Conveyed by the Print Media: A UK Case Study', Acta Politica, 46:2, 111-31. Barr, R. R. (2009). 'Populists, Outsiders and Anti-Establishment Polities', Party Politics, 15:1, 29-48. Bartels, L. M. (2002). 'The Impact of Candidate Traits in American Presidential Elections', in A. King (ed.), Leaders' Personalities and the Outcomes of Democratic Elections. New York: Oxford University Press, 44-70. Bittner, A. (2011). Platform or Personality? The Role of Party Leaders in Elections. Oxford: Oxford University Press. Bono, J. E., and T. A. Judge (2004). 'Personality and Transformational and Transactional Leadership: A Meta-Analysis', Journal of Applied Psychology, 89: 5, 901-10. Bos, L., and K. Brants (2014). 'Populist Rhetoric in Politics and Media: A Longitudinal Study of the Netherlands', European Journal of Communication, 29:6, 703-19. Bruk, D. (2013). 'The Best of Vladimir Zhirinovsky, the Clown Prince of Russian Polities', Vice, 11 August 2013. Canovan, M. (1999). 'Trust the People! Populism and the Two Faces of Democracy', Political Studies, 47:1, 2-16. Caprara, G., D. Francescato, M. Mebane, R. Sorace, and M. Vecchione (2010). 'Personality Foundations of Ideological Divide: A Comparison of Women Members of Parliament and Women Voters in Italy', Political Psychology, 31:5, 739-62. Caprara, G. V., and P. G. Zimbardo (2004). 'Personalizing Politics: A Congruency Model of Political Preference', American Psychologist, 59:7, 581-94. Caprara, G. V., C. Barbaranelli, C. Consiglio, L. Picconi, and P. G. Zimbardo (2003). 'Personalities of Politicians and Voters: Unique and Synergistic Relationships', Journal of Personality and Social Psychology, 84:4, 849-56. Caprara, G. V., C. Barbaranelli, and P. G. Zimbardo (1999). 'Personality Profiles and Political Parties', Political Psychology, 20:1, 175-97. Caprara, G. V., C. Barbaranelli, R. C. Fraley, and M. Vecchione (2007). 'The Simplicity of Politicians' Personalities across Political Context: An Anomalous Replication', International Journal of Psychology, 42:6, 393-405. Castille, C. M., J. E. Buckner, and C. N. Thoroughgood (2016). 'Prosocial Citizens without a Moral Compass? Examining the Relationship between Machiavellianism and Unethical Pro-Organizational Behavior', Journal of Business Ethics, 149:4, 919-30. https://doi.org/10.1007/sl0551-016-3079-9 Corbu, N., D. Balaban-Balas, and E. Negrea-Busuioc (2017). 'Romania: Populist Ideology without Teeth', in T. Aalberg, F. Esser, C. Reinemann, J. Stromback, WEST EUROPEAN POLITICS @ 1363 and C. H. de Vreese (eds.), Populist Political Communication in Europe. New York: Routledge, 326-38. De Hoogh, A. H., D. N. Den Hartog, and P. L. Koopman (2005). 'Linking the Big Five-Factors of Personality to Charismatic and Transactional Leadership; Perceived Dynamic Work Environment as a Moderator', Journal of Organizational Behavior, 26:7, 839-65. De Landsheer, C. and L. Kalkhoven (2014). 'The Imagery of Geert Wilders, Leader of the Dutch Freedom Party (PVV)', Paper presented at the IPSA World Congress, Canada, July 2014. Dietrich, B. J., S. Lasley, J. J. Mondak, M. L. Remmel, and J. Turner (2012). 'Personality and Legislative Politics: The Big Five Trait Dimensions among US State Legislators', Political Psychology, 33:2, 195-210. Ehrhart, M. G., K. H. Ehrhart, S. C. Roesch, B. G. Chung-Herrera, K. Nadler, and K. Bradshaw (2009). 'Testing the Latent Factor Structure and Construct Validity of the Ten-Item Personality Inventory', Personality and Individual Differences, 47:8, 900-05. Ennser, L. (2012). 'The Homogeneity of West European Party Families: The Radical Right in Comparative Perspective', Party Politics, 18:2, 151-71. Ernst, N., S. Blassing, S. Engesser, F. Biicher, and F. Esser (2019). 'Populists Prefer Social Media over Talk Shows: An Analysis of Populist Messages and Stylistic Elements across Six Countries', Social Media & Society, 5:1, 205630511882335. Fortunato, D., M. V. Hibbing, and J. J. Mondak (2018). 'The Trump Draw: Voter Personality and Support for Donald Trump in the 2016 Republican Nomination Campaign', American Politics Research, 46:5, 785-810. Furnham, A., S. C. Richards, and D. L. Paulhus (2013). 'The Dark Triad of Personality: A 10 Year Review', Social and Personality Psychology Compass, 7:3, 199-216. Gallagher, M. (2014). 'Electoral Institutions and Representation', in L. LeDuc, R Niemi, and P. Norris (eds.), Comparing Democracies 4. London: Sage, 11-31. Gerber, A. S., G. A. Hiber, D. Doherty, C. M. Dowling, and S. E. Ha (2010). 'Personality and Political Attitudes: Relationshis across Issue Domains and Political Context', American Political Science Review, 104:1, 111-33. Gerber, A. S., G A. Huber, D. Doherty, and C. M. Dowling (2011). 'The Big Five Personality Traits in the Political Arena', Annual Review of Political Science, 14, 265-87. Glad, B. (2002). 'Why Tyrants Go too Far: Malignant Narcissism and Absolute Power', Political Psychology, 23:1, 1-37. Goldberg, L. R. (1990). 'An Alternative "Description of Personality": The Big-Five Factor Structure', Journal of Personality and Social Psychology, 59:6, 1216-29. Gosling, S. D., P. J. Rentfrow, and W. B. Swann (2003). 'A Very Brief Measure of the Big-Five Personality Domains', Journal of Research in Personality, 37:6, 504-28. Gurov, B., and E. Zankina (2013). 'Populism and the Construction of Political Charisma: Post-Transition Politics in Bulgaria', Problems of Post-Communism, 60:1, 3-17. Hamblin, J. (2016, July). 'Donald Trump: Sociopath?' The Atlantic. https://www. theatlantic.com/health/archive/2016/07/trump-and-sociopathy/491966/ 1364 (g) A. NAI AND F. MARTINEZ I COMA Heinisch, R. (2003). 'Success in Opposition-Failure in Government: Explaining the Performance of Right-Wing Populist Parties in Public Office', West European Politics, 26:3, 91-130. Immerzeel, T., and M. Pickup (2015). 'Populist Radical Right Parties Mobilizing 'the People'? The Role of Populist Radical Right Success in Voter Turnout', Electoral Studies, 40, 347-60. Inglehart, R., and P. Norris (2016). Trump, Brexit, and the Rise of Populism: Economic Have-Nots and Cultural Backlash. HKS Faculty Research Working Paper Series RWP16-026, August 2016. Jagers, J., and S. Walgrave (2007). 'Populism as Political Communication Style: An Empirical Study of Political Parties' Discourse in Belgium', European Journal of Political Research, 46:3, 319-45. Jensen-Campbell, L. A., and K. T. Malcolm (2007). 'The Importance of Conscientiousness in Adolescent Interpersonal Relationships', Personality and Social Psychology Bulletin, 33:3, 368-83. John, O. P., L. P. Naumann, and C. J. Soto (2008). 'Paradigm Shift to the Integrative Big Five Trait Taxonomy: History, Measurement, and Conceptual Issues', in O. P. John, R W. Robins, and L. A. Pervin (eds.), Handbook of Personality: Theory and Research. New York: Guilford Press, 114-58. Joly, J., S. Soroka, and P. Loewen (2018) 'Nice Guys Finish Last: Personality and Political Success', Acta Politica. https://doi.org/10.1057/s41269-018-0095-z Jonason, P. K., and J. McCain (2012). 'Using the HEXACO Model to Test the Validity of the Dirty Dozen Measure of the Dark Triad', Personality and Individual Differences, 53:7, 935-8. Jonason, P. K., and G. D. Webster (2010). 'The Dirty Dozen: A Concise Measure of the Dark Triad', Psychological Assessment, 22:2, 420-32. Jones, D. N., and D. L. Paulhus (2014). 'Introducing the Short Dark Triad (SD3): A Brief Measure of Dark Personality Traits', Assessment, 21:1, 28-41. Jorstad, J. (1996). 'Narcissism and Leadership: Some Differences in Male and Female Leaders', Leadership & Organization Development Journal, 17:6, 17-23. Klingler, J. D., G. E. Hollibaugh, and A. J. Ramey (2018). 'What I Like about You: Legislator Personality and Legislator Approval', Political Behavior, Laakso, M. and R Taagepera (1979) "Effective' Number of Parties: A Measure with Application to West Europe', Comparative Political Studies, 12, 3-27. Lee, K., and M. C. Ashton (2005). 'Psychopathy, Machiavellianism, and Narcissism in the Five-Factor Model and the HEXACO Model of Personality Structure', Personality and Individual Differences, 38:7, 1571-82. Lilienfeld, S. O., A. L. Watts, and S. F. Smith (2015). 'Successful Psychopathy: A Scientific Status Report. Current Directions in Psychological Science, 24:4, 298-303. Lilienfeld, S. O., I. D. Waldman, K. Landfield, A. L. Watts, S. Rubenzer, and T. R Faschingbauer (2012). 'Fearless Dominance and the US Presidency: Implications of Psychopathic Personality Traits for Successful and Unsuccessful Political Leadership', Journal of Personality and Social Psychology, 103:3, 489-505. Lilienfeld, S. O., J. D. Miller, and D. R Lynam (2018). 'The Goldwater Rule: Perspectives from, and Implications for, Psychological Science', Perspectives on Psychological Science, 13:1, 3-27. WEST EUROPEAN POLITICS @ 1365 MacMillan, C. (2017). 'Welcome to the Carnival? Podemos, Populism and Bakhtin's Carnivalesque', Journal of Contemporary European Studies, 25:2, 258-73. Martinez i Coma, F., and C. Van Ham (2015). 'Can Experts Judge Elections? Testing the Validity of Expert Judgments for Measuring Election Integrity', European Journal of Political Research, 54:2, 305-25. McAdams, D. P. (2016, May). 'The Mind of Donald Trump'. The Atlantic. https://www.theatlantic.com/magazine/archive/2016/06/the-mind-of-donald- trump/480771/ McBride, J. (2017). 'Dutch Elections and the Future of the EU'. Council on Foreign Relations, 10 March 2017. McDonnell, D. (2016). 'Populist Leaders and Coterie Charisma', Political Studies, 64:3, 719-33. Moffitt, B. (2016). The Global Rise of Populism: Performance, Political Style, and Representation. Stanford: Stanford University Press. Moffitt, B., and S. Tormey (2014). 'Rethinking Populism: Politics, Mediatisation and Political Style', Political Studies, 62:2, 381-97. Mondak, J. J. (2010). Personality and the Foundations of Political Behavior. New York: Cambridge University Press. Mudde, C. (2004). 'The Populist Zeitgeist', Government and Opposition, 39:4, 542-63. Mudde, C. (2007). Populist Radical Right Parties in Europe. New York: Cambridge University press. Nai, A. (2018a). 'Disagreeable Narcissists, Extroverted Psychopaths, and Elections. A New Dataset to Measure the Personality of Candidates Worldwide', European Political Science. Nai, A. (2018b). 'Fear and Loathing in Populist Campaigns? Comparing the Communication Style of Populists and Non-Populists in Elections Worldwide', Journal of Political Marketing. Nai, A., and J. Maier (2018). 'Perceived Personality and Campaign Style of Hillary Clinton and Donald Trump', Personality and Individual Differences, 121, 80-3. Nai, A., F. Martinez i Coma, and J. Maier (2019). 'Donald Trump, Populism, and the Age of Extremes: Comparing the Personality Traits and Campaigning Style of Trump and Other Leaders Worldwide', Presidential Studies Quarterly. https://doi.org/10J 11 l/psq.12511 Nemtsova, A. (2016). 'Russia's Trump, Vladimir Zhirinovsky, Wants to Build a Wall, Ban Muslims, and Nuke the White House'. Daily Beast, 7 September 2016. Norgaard, A. S., and R. Klemmensen (2018). 'The Personalities of Danish MPs: Trait-and Aspect-Level Differences', Journal of Personality, 87, 267-275. Olbermann, K. (2016, July). 'Could Donald Trump Pass a Sanity Test?' Vanity Fair, http://www.vanityfair.com/news/2016/07/donald-trump-keith-olbermann-sanity-test Oliver, J. E., and W. M. Rahn (2016). 'Rise of the Trumpenvolk: Populism in the 2016 Election', The ANNALS of the American Academy of Political and Social Science, 667:1, 189-206. Ostiguy, P. (2009). 'The High and Low in Politics: A Two-Dimensional Political Space for Comparative Analysis and Electoral Studies'. Kellogg Institute 1366 (g) A. NAI AND F. MARTINEZ I COMA Working Paper 360, July 2009. https://kellogg.nd.edu/publications/workingpa-pers/WPS/360.pdf Parry, K., M. D. Mumford, I. Bower, and L. L. Watts (2014). 'Qualitative and Historiometric Methods in Leadership Research: A Review of the First 25 Years of the Leadership Quarterly', The Leadership Quarterly, 25:1, 132-51. Patrick, C. J., D. C. Fowles, and R. F. Krueger (2009). 'Triarchic Conceptualization of Psychopathy: Developmental Origins of Disinhibition, Boldness, and Meanness', Development and psychopathology, 21(03): 913-38. Paulhus, D. L., and K. M. Williams (2002). 'The Dark Triad of Personality: Narcissism, Machiavellianism, and Psychopathy', Journal of Research in Personality, 36:6, 556-63. Perry, R, and C. G. Sibley (2012). 'Big-Five Personality Prospectively Predicts Social Dominance Orientation and Right-Wing Authoritarianism', Personality and Individual Differences, 52:1, 3-8. Post, J. M. (1993). 'Current Concepts of the Narcissistic Personality: Implications for Political Psychology', Political Psychology, 14:1, 99-121. Ramey, A. J., J. D. Klingler, and G. E. Hollibaugh (2016). 'Measuring Elite Personality Using Speech', Political Science Research and Methods, 7:1, 163-84. Ramey, A. J., J. D. Klingler, and G. E. Hollibaugh (2017). More than a Feeling: Personality, Polarization, and the Transformation of the US Congress. Chicago: University of Chicago Press. Reidy, D. E., A. Zeichner, and L. A. Seibert (2011). 'Unprovoked Aggression: Effects of Psychopathic Traits and Sadism', Journal of Personality, 79:1, 75-100. Roberts, B. W., O. S. Chernyshenko, S. Stark, and L. R. Goldberg (2005). 'The Structure of Conscientiousness: An Empirical Investigation Based on Seven Major Personality Questionnaires', Personnel Psychology, 58:1, 103-39. Roberts, K. M. (2007). 'Latin America's Populist Revival', SAIS Review of International Affairs, 27:1, 3-15. Rooduijn, M., and T. Pauwels (2011). 'Measuring Populism: Comparing Two Methods of Content Analysis', West European Politics, 34:6, 1272-83. Rosenthal, S. A., and T. L. Pittinsky (2006). 'Narcissistic Leadership', The Leadership Quarterly, 17:6, 617-33. Rubenzer, S. J., T. R. Faschingbauer, and D. S. Ones (2000). 'Assessing the US Presidents Using the Revised NEO Personality Inventory', Assessment, 7:4, 403-19. Schmuck, D., J. Matthes, and H. Boomgaarden (2017). 'Austria: Candidate-Centered and Anti-Immigrant Right-Wing Populism', in T. Aalberg, F. Esser, C. Reinemann, J. Stromback, and C. H. de Vreese (eds.), Populist Political Communication in Europe. New York: Routledge, 85-98. Selfhout, M., W. Burk, S. Branje, J. Denissen, M. Van Aken, and W. Meeus (2010). 'Emerging Late Adolescent Friendship Networks and Big Five Personality Traits: A Social Network Approach', Journal of Personality, 78:2, 509-38. Simonton, D. K. (2006). 'Presidential IQ, Openness, Intellectual Brilliance, and Leadership: Estimates and Correlations for 42 US Chief Executives', Political Psychology, 27:4, 511-26. Simpson, J (2012). 'Putin Power'. The Spectator, 2 March 2012. Spezio, M. L., L. Loesch, F. Gosselin, K. Mattes, and R. M. Alvarez (2012). 'Thin-Slice Decisions Do Not Need Faces to be Predictive of Election Outcomes', Political Psychology, 33:3, 331-41. WEST EUROPEAN POLITICS @ 1367 Stanley, B. (2008). 'The Thin Ideology of Populism', Journal of Political Ideologies, 13:1, 95-110. Steenbergen, M. R, and G. Marks (2007). 'Evaluating Expert Judgments', European Journal of Political Research, 46:3, 347-66. Tremblay, P. F., and L. A. Ewart (2005). 'The Buss and Perry Aggression Questionnaire and Its Relations to Values, the Big Five, Provoking Hypothetical Situations, Alcohol Consumption Patterns, and Alcohol Expectancies', Personality and Individual Differences, 38:2, 337-46. Vecchione, M., and G. V. Caprara (2009). 'Personality Determinants of Political Participation: The Contribution of Traits and Self-Efficacy Beliefs', Personality and Individual Differences, 46:4, 487-92. Vecchione, M., H. Schoen, J. L. G. Castro, J. Cieciuch, V. Pavlopoulos, and G. V. Caprara (2011). 'Personality Correlates of Party Preference: The Big Five in Five Big European Countries', Personality and Individual Differences, 51:6, 737-42. Visser, B. A., A. S. Book, and A. A. Volk (2017). 'Is Hillary Dishonest and Donald Narcissistic? A HEXACO Analysis of the Presidential Candidates' Public Personas', Personality and Individual Differences, 106, 281-6. Watts, A. L., et al. (2013). 'The Double-Edged Sword of Grandiose Narcissism: Implications for Successful and Unsuccessful Leadership among US Presidents', Psychological Science, 24:12, 2379-89. Weyland, K. (2001). 'Clarifying a Contested Concept: Populism in the Study of Latin American Polities', Comparative Politics, 34:1, 1-22. Wright, J. D., and M. F. Tomlinson (2018). 'Personality Profiles of Hillary Clinton and Donald Trump: Fooled by Your Own Polities', Personality and Individual Differences, 128, 21-4.