PL14CH13-Gerber ARI 11 April 2011 13:52 The Big Five Personality Traits in the Political Arena Alan S. Gerber, Gregory A. Huber, David Doherty, and Conor M. Dowling Yale University, Department of Political Science, Institution for Social and Policy Studies, New Haven, Connecticut 06520-8209 email: alan.gerber@yale.edu, gregory.huber@yale.edu, david.doherty@yale.edu, conor.dowling@yale.edu Annu. Rev. Polit. Sci. 2011. 14:265���87 The Annual Review of Political Science is online at polisci.annualreviews.org This article���s doi: 10.1146/annurev-polisci-051010-111659 Copyright c 2011 by Annual Reviews. All rights reserved 1094-2939/11/0615-0265$20.00 Keywords dispositional traits, political behavior, participation, political psychology, Five-Factor Model Abstract Recent political science research on the effects of core personality traits���the Big Five���contributes to our understanding of how peo- ple interact with their political environments. This research examines how individual-level variations in broad, stable psychological charac- teristics affect individual-level political outcomes. In this article, we review recent work that uses the Big Five to predict political attitudes and behavior. We also replicate some of these analyses using new data to examine the possibility that prior findings stem from sampling error or unique political contexts. Finally, we discuss several of the challenges faced by scholars who are currently pursuing or are interested in pur- suing this line of inquiry. These challenges include refining theoretical explanations of how the Big Five shape political outcomes, addressing important measurement concerns, and resolving inconsistencies across studies. 265 Annu. Rev. Polit. Sci. 2011.14:265-287. Downloaded from www.annualreviews.org by Syddansk University on 08/14/11. For personal use only.

PL14CH13-Gerber ARI 11 April 2011 13:52 INTRODUCTION Understanding how people interact with and evaluate their political environment is a cen- tral focus of research on political behavior. Research has examined how a broad range of factors affect an individual���s behaviors and attitudes. These include: sociological factors (such as socioeconomic class, group affiliation, and social networks), political context (includ- ing campaign effects and geographic variation in political culture), and psychological factors (such as partisan identity, altruism, authoritari- anism, and patriotism). Recent research on the effects of core personality traits���the Big Five��� contributes to this tradition by examining how individual-level variations in broad, stable psy- chological characteristics affect individual-level political outcomes. Personality psychologists are broadly con- cerned with ���provid[ing] an integrative frame- work for understanding the whole person��� (McAdams & Pals 2006, p. 204). Over the past 20 years, the Big Five traits have emerged in psychology as the dominant framework for measuring personality traits. Psychologists re- fer to these traits as dispositional or core traits. This label differentiates the Big Five from other aspects of individuals��� personalities such as their characteristic adaptations (values, attitudes, in- terests), self-concepts (self-esteem, identity), and objective biography (careers, background) (McAdams & Pals 2006, McCrae & Costa 1996). In contrast to these other components of personality that develop and change through- out the life cycle, dispositional traits are be- lieved to be stable aspects of individuals that shape how they respond to the vast array of stimuli they encounter in the world. As such, they affect behaviors and attitudes across a wide array of situations.1 1Our focus on the Big Five traits in this review does not in any way imply that research on other aspects of personality (in- cluding other trait-based approaches) is either inferior or ob- solete. In our discussion of directions for future research, we note the importance of integrating Big Five���based accounts of political attitudes and behavior with findings regarding the THE ORIGINS AND PROPERTIES OF THE BIG FIVE According to McCrae & Costa (2008), trait- based research on personality is premised on four assumptions about human nature: (a) personality traits exist and are measurable, (b) these traits vary across individuals, (c) the causes of human behavior are rooted within the individual (e.g., personality traits affect individual behavior), and (d ) people ���can un- derstand themselves and others��� (p. 161). The Big Five trait domains were identified through extensive lexical analysis. This approach rests primarily on the fourth assumption���that people understand themselves and others and that an important historical function of language has been to provide a way for people to describe enduring differences between individuals (Allport & Odbert 1936 John et al. 2008a). Thus, over time, languages have come to include words that facilitate identifying the most salient and enduring individual-level differences in what people are like. Lexical analysis involves gathering extensive lists of adjectives or phrases that can be used to describe enduring individual-level characteris- tics. Subjects are then asked to rate how well each word or phrase describes themselves or another individual. Researchers then use fac- tor analysis to identify the broad superfactors or trait domains that underlie these responses. Although some scholars find evidence of more (Ashton & Lee 2005, Paunonen & Jackson 2000) or fewer (Blackburn et al. 2004, Musek 2007) factors, most analyses identify five: Extraversion, Agreeableness, Conscientious- ness, Emotional Stability (sometimes referred to by its inverse���Neuroticism), and Openness to Experience. This five-factor structure has been replicated in a variety of languages and contexts (see John et al. 2008a for a review), as well as in unusual subpopulations (Yang et al. relationships between other components of personality and these outcomes. 266 Gerber et al. Annu. Rev. Polit. Sci. 2011.14:265-287. Downloaded from www.annualreviews.org by Syddansk University on 08/14/11. For personal use only.

PL14CH13-Gerber ARI 11 April 2011 13:52 Table 1 Description of the Big Five traits (adapted from John et al. 2008a) Trait Definition Extraversion . . .energetic approach toward the social and material world. Agreeableness Contrasts a prosocial and communal orientation toward others with antagonism. . . Conscientiousness . . .socially prescribed impulse control that faciliates task- and goal-directed behavior. . . Emotional stability Contrasts. . .even-temperedness with negative emotionality. . . Openness to experience . . .the breadth, depth, originality, and complexity of an individuals��� mental and experiential life. 1999).2 Table 1 presents a brief description of each of these traits. Evidence suggests that the Big Five traits are highly stable through the life cycle (Caspi et al. 2005, but see Srivastava et al. 2003) and are heritable (Bouchard 1997, Plomin et al. 1990, Van Gestel & Van Broeckhoven 2003). There is also a growing body of evidence that variation in dispositional traits is associated with specific, measurable biological factors. Some research has identified relationships between specific ge- netic markers and Big Five traits (Lesch et al. 1996 see Canli 2008 for a review). DeYoung et al. (2010) find associations between four of the Big Five traits and the size (volume) of theo- retically associated regions of the brain (see also DeYoung et al. 2009). For example, they find that Conscientiousness is associated with the volume of the lateral prefrontal cortex, which plays an important role in planning and impulse control. Measuring the Big Five Numerous batteries have been developed to measure the Big Five trait domains. Each consists of a list of adjectives [e.g., ���temper- amental��� (Goldberg 1992)] or phrases [e.g., ���Sometimes I do things on impulse that I later regret��� (Costa & McCrae 1992)] and asks the respondent to rate how well each adjective or 2Alternative approaches to the measurement of personality traits include trait typologies based on a priori, theoretical expectations regarding essential personality traits (e.g., the Myers-Briggs Type Indicator) and idiographic approaches that attempt to identify a unique set of traits that are par- ticularly relevant for a specific individual. See Barenbaum & Winter (2008) for a discussion of these approaches. phrase describes the individual whose person- ality is being rated���typically the respondent. Researchers use these ratings to calculate scores for each of the Big Five traits. There are a variety of instruments that can be used to measure the Big Five, ranging from brief batteries of ten items [Ten Item Personality Measure (TIPI) (Gosling et al. 2003 see also Langford 2003)] to batteries that use dozens of items [Big Five Questionnaire (BFQ) (Caprara et al. 1993), Big Five Inventory (BFI) ( John et al. 1991), Mini-Markers (Saucier 1994), NEO-Five-Factor Inventory (NEO-FFI) (Costa & McCrae 1992)] or even hun- dreds [NEO���Personality Inventory���Revised (NEO-PI-R) (Costa & McCrae 1992), Interna- tional Personality Item Pool (IPIP) (Goldberg et al. 2006)].3 The most important trade-off researchers face when deciding which personality battery to use is between internal reliability and brevity. For example, the NEO-PI-R measures each Big Five trait with dozens of items. It has the ben- efit of producing trait measures with high in- ternal reliability [alpha coefficients greater than 0.85 (Costa & McCrae 1992)]. However, the NEO-PI-R takes more than 30 minutes to com- plete, making its inclusion extremely difficult on surveys where other content (e.g., political content) needs to be measured. In contrast, the TIPI measures each of the Big Five dimensions with scores from only two items. For example, the Extraversion scale is calculated on the ba- sis of a respondent���s level of agreement with 3Additionally, some researchers have constructed customized batteries of questions using items based on these batteries (e.g., Mondak & Halperin 2008, Mondak et al. 2010). www.annualreviews.org ��� Big Five Personality Traits 267 Annu. Rev. Polit. Sci. 2011.14:265-287. Downloaded from www.annualreviews.org by Syddansk University on 08/14/11. For personal use only.

PL14CH13-Gerber ARI 11 April 2011 13:52 two statements: ���I see myself as extraverted, enthusiastic��� and ���I see myself as reserved, quiet��� (reverse coded). Thus, this battery has the benefit of being short enough to be included in large political surveys. However, with only two items per trait domain, TIPI scales can- not achieve the same level of internal reliability as longer batteries. We discuss the theoretical and empirical implications of battery selection below, under Challenges: Refining Theory and Measurement. THE BIG FIVE AS PREDICTORS OF BEHAVIORS AND OTHER OUTCOMES The stability and replicability of the Big Five framework have spurred a great deal of schol- arly interest in these traits. Although a large share of this research has been conducted by social psychologists, scholars in fields includ- ing public health, economics, education, so- ciology, and clinical psychology have recog- nized that dispositional traits have the potential to improve our understanding of fundamental, individual-level differences in how people eval- uate and respond to the world around them. This rapidly expanding body of research finds that the Big Five personality traits predict a vast array of behaviors and other outcomes. Public health researchers have identified a variety of associations between Big Five traits and alcohol and tobacco consumption (McAdams & Donnellan 2009, Mezquita et al. 2010, Paunonen & Ashton 2001), frequency of physical exercise (Rhodes & Smith 2006), cholesterol and triglyceride levels (Sutin et al. 2010), longevity (Friedman et al. 2010, Roberts et al. 2007), and overall mental and physical health (Goodwin & Friedman 2006, Ozer & Benet-Mart�� ��nez 2006). Economists find evidence that dispositional personality traits predict behavior in economic games (Ben-Ner et al. 2008, Koole et al. 2001) as well as wages (Nyhus & Pons 2005) and occupational status (De Fruyt & Mervielde 1999). Other social research finds evidence that Big Five traits predict parenting style (Huver et al. 2010), satisfaction with intimate relationships (Malouff et al. 2010), and occupational choice and satisfaction (Borghans et al. 2008, Hogan & Holland 2003, Ozer & Benet-Mart��nez�� 2006, Roberts et al. 2007, Salgado 1997). This is not to suggest that all research finds relationships between Big Five traits and outcomes of interest. For example, whereas Paunonen & Ashton (2001) find relationships between Big Five traits and a variety of out- comes, they do not find evidence that these traits predict buying lottery tickets, obesity, or peer ratings of an individual���s intelligence and popularity. More broadly, we note that in most cases only some of the Big Five traits signif- icantly predict outcomes of interest. In sum- mary, these traits have predictive power in an impressive variety of domains but are not uni- versal predictors of all outcomes. THE BIG FIVE IN THE POLITICAL ARENA The Big Five taxonomy has opened a promising new frontier in research on political attitudes and behavior. Previous research on the relation- ship between personality and political attitudes and behaviors focused on characteristic adapta- tions and self concepts that are likely to be par- ticularly relevant to politics, such as racial re- sentment [a measure based on agreement with statements such as ���It���s really a matter of some people not trying hard enough if blacks would only try harder they could be just as well off as whites��� (e.g., Feldman & Huddy 2005, Kinder & Sears 1981, Sniderman & Carmines 1997)], right-wing authoritarianism [RWA, a measure based on agreement with statements such as ���What our country really needs is a strong, de- termined leader who will crush evil, and take us back to our true path��� (e.g., Adorno et al. 1950, Altemeyer 1996)], and partisanship (e.g., Campbell et al. 1960). This research has yielded a vast array of findings that offer important in- sight into the relationship between political at- titudes and other characteristic adaptations. One of the key differences between this pre- vious research and research on the Big Five and 268 Gerber et al. Annu. Rev. Polit. Sci. 2011.14:265-287. Downloaded from www.annualreviews.org by Syddansk University on 08/14/11. For personal use only.

PL14CH13-Gerber ARI 11 April 2011 13:52 political outcomes is that, on their face, the Big Five traits are not obviously associated with po- litical attitudes and behaviors. Instead, they are broad dispositions that are theorized to shape responses to the full range of stimuli people en- counter in the world. Thus, just as socioeco- nomic status is associated with a broad range of forms of political and social engagement, po- litical research on Big Five traits may provide a way to situate political judgments and behav- iors within the context of a broader theoretical account of how individuals engage with their environments. In this section, we summarize current find- ings regarding the relationships between Big Five traits and political outcomes. The magni- tudes of the effects we discuss are summarized in Table 2. Specifically, the table presents the es- timated effect of a two���standard deviation (SD) increase in each personality trait on the out- come of interest. We also report the effects of similar changes in education and income. As the table illustrates, in many cases the magnitudes of the effects of Big Five traits are comparable to those associated with these canonical predic- tors of political attitudes and behaviors (e.g., Rosenstone & Hansen 1993). The relation- ships between characteristic adaptations such as RWA and political outcomes such as ideology are often much stronger than the relationships between Big Five traits and political outcomes (see Gerber et al. 2010c, p. 123). This is to be expected. As noted above, RWA and racial at- titudes have clear political corollaries, whereas the Big Five personality traits do not. Ideology and Political Attitudes Much of the early political science research on the Big Five has focused on the relationships between these traits and political ideology.4 The most consistent findings from this line of research are an association between Openness 4See Gerber et al. (2010c), table 1, for a summary of find- ings regarding the relationships between Big Five traits and ideology. to Experience and liberalism and between Conscientiousness and conservatism (Alford & Hibbing 2007 Carney et al. 2008 Gerber et al. 2010c Gosling et al. 2003 Jost et al. 2003, 2007 McCrae 1996 Mondak 2010 Mondak & Halperin 2008 Riemann et al. 1993 Van Hiel et al. 2000 Van Hiel & Mervielde 2004). Open- ness to Experience is associated with positive responses to novel stimuli. Thus, researchers posit that individuals high on this trait are more likely to respond favorably to liberal social policies, which often involve acceptance of un- conventional behaviors, and liberal economic policies, which may involve a willingness to support proposals that entail new government involvement in the economy. By contrast, individuals high on Conscientiousness tend to be attracted to social norms and achievement striving. These response tendencies likely ex- plain why those high on this trait are more likely to reject the challenges to social norms that often accompany liberal social policies, as well as liberal economic policies, which may be seen as undermining incentives for individual effort. According to Carney et al. (2008), the relationships between both Openness to Experience and Conscientiousness and po- litical ideology comport well with earlier theoretical accounts of the relationships be- tween personality and ideology. For example, previous researchers posited that individuals who are ���creative, imaginative, [and] curious��� (characteristics associated with Openness to Experience) are more likely to be attracted to a liberal or left-wing ideology, whereas those who are ���orderly [and] organized��� (character- istics associated with Conscientiousness) are more likely to be attracted to a conservative or right-wing ideology.5 Recently scholars have also begun to exam- ine the relationships between core personality traits and more specific political attitudes. This new line of research offers sharper insight into 5Carney et al. (2008, p. 816, table 1) summarize recent find- ings regarding the relationships between Big Five traits and ideology and compare these findings with previous work on the relationship between personality and ideology. www.annualreviews.org ��� Big Five Personality Traits 269 Annu. Rev. Polit. Sci. 2011.14:265-287. Downloaded from www.annualreviews.org by Syddansk University on 08/14/11. For personal use only.