Show Summary Details

Discussion of several new citations.

Updated on 11 January 2018. The previous version of this content can be found here.
Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 23 February 2019

Candidate Traits and Political Choice

Abstract and Keywords

What role do presidential candidate character traits play in vote decisions? To some, the answer is obvious as campaigns, journalists, pundits, and voters frequently differentiate presidential candidates in terms of their personal qualities—traits are deemed important. On the other hand, past research suggests that, while candidate character traits are short term forces, they hold relatively limited in influence on vote preference. However, theoretical and methodological limitations may have hindered past research ability to detect the true influence of character traits in voter decisions. This author reviews past literature, offers a clear conceptualization of candidate character traits, presents ways in which trait may influence vote choice, and suggests areas for future research.

Keywords: traits, character, personality, personal qualities, leadership, presidential candidates, elections, vote decision

Introduction

What role do presidential candidate character traits play in vote decisions? To some, the answer is obvious as journalists and pundits frequently differentiate candidates in terms of their personal qualities. Campaign strategists are among those who ardently believe this (e.g., Sosnik, Dowd, and Fournier, 2006). In advertisements, on the stump, and in debates, candidates routinely attack the integrity, honesty, and leadership credentials of their rivals and as such inherently praise their own. Casual political conversations among citizens often pivot on candidates’ personalities. When pressed to explain why someone supports a candidate, personal attributes are often offered as a central reason. For example, after Republican Senator John McCain won the Florida primary on January 29, 2008, CNN.com’s1 readers responded to a request to explain his success by attributing it to his character: Debbie Pike of St. Louis, Missouri, wrote “It is very simple, actually. He comes across as an honest, trustworthy, real person.” Mike Bodina of Edina, Minnesota, commented, “It’s sad to say that my vote has come down to this … I just want to see an honest candidate. I don’t necessarily agree with him on all of the issues.” Geno Galindo of Santa Barbara explained, “Like myself, I think many feel character counts and McCain gets big points for character.” The conventional wisdom holds that candidate character influences vote decisions. Yet during the general election, the Arizonan lost to Democratic senator Barack Obama, perhaps in part because of the Democratic campaign’s success in presenting its candidate as inspiring and as an agent of change while framing the Vietnam War hero as “erratic,” “too old to be president,” and “out of touch” (Kenski, Hardy, and Jamieson 2010).

As the suggested by a US News & World Report article,2 in 2012 President Obama perhaps could attribute a portion of his reelection to “Mitt Romney’s Personality Problem.” The article’s subtitle continued: “Mitt Romney will lose the election on Tuesday because of his inability to connect to voters.” Enhancing the article is a cartoon that presents a caricature of the Republican nominee along with five other objects—1) pudding, 2) a lump of dough, 3) an amoeba, 4) mayonnaise, and 5) oatmeal—and then asks: “One of these is not like the other…. Can you spot the difference? Answer: While all are bland and amorphous, the pudding, dough, amoeba, mayonnaise, and oatmeal don’t change their core beliefs based on the latest polling data.” The conventional wisdom holds that candidate character influences vote decisions.

Nevertheless, this conventional wisdom and the belief of those orchestrating campaigns appear to be at odds with some older empirical research that suggests that candidate character traits are important short-term forces but relatively limited in their influence on vote preference (Bartels, 2002; Miller and Shanks, 1982; 1996; Shanks and Miller, 1990, 1991). Additionally, many of the classic vote-choice and forecasting models practically ignore the role of candidate traits but still hold predictive power. In this essay, I attempt to resolve this conflict by presenting a clear conceptualization of the concept of traits, explaining how and why they may play a role in vote preference, reviewing past empirical research on the (p. 438) topic, and suggesting some new avenues for research on candidate character traits that may shed light on their importance in vote decisions.

Conceptualizing Candidate Character Traits

Much of the research on this topic fails to clearly define the concept of traits—a failure that sabotages a nuanced understanding of their function and importance in vote decisions. To define candidate character traits, it is helpful to turn to a conceptualization of general personality traits offered in 1938 by Carr and Kingsbury:

A trait is a conceptual attribute or definition of the reactive nature of an individual. The nature of the individual is defined on the basis of certain observable behavioral characteristics. Not all observable characteristics are used for this purpose. The definition is based only upon those characteristics (1) which society regards as of sufficient importance to identify and name, and (2) which are regarded as expressions or manifestations of the constitutional nature of the individual. The term ‘constitutional nature’ refers to all of those relatively permanent and enduring organic conditions that characterize a given individual and differentiate him from his fellows, and these organic conditions may both be innate and acquired in respect to origin

(Carr and Kingsbury, 1938, 497).

These University of Chicago psychologists argued that traits are lexical categorizations of how an individual responds to his or her environment—a person’s reactive nature. Yet, traits are much more than just descriptors. When a person reacts to his or her environment in a persistent pattern, traits are used by others to characterize the constitutional nature of the individual—the core of a person’s character or personality. There is a sequential order in understanding a person’s character. First, we view one’s reaction to the environment and then name it with a trait that “society regards as of sufficient importance to identify and name” (Carr and Kingsbury, 1938, 497). After viewing this reaction repeatedly, we use this trait to characterize the person’s constitutional nature (Carr and Kingsbury, 1938).

The Predictive Value of Traits

Most human transactions require trust and a level of confidence regarding how others will act. Specific behaviors are bounded to specific situations and over time individuals manifest consistent behavioral patterns that represent their “true self” or “constitutional nature.” Personality traits differ from transient mood states in that the former are relatively enduring. The identification of personality traits in others fosters interpersonal (p. 439) relationships because, by forecasting future behavior, trait ascription minimizes uncertainty, risk, and doubt. Relationships would be extremely difficult to manage without a means of categorizing behavior that informs expectations of future interactions. Once an individual is characterized in terms of traits, a readily accessible heuristic is available to predict his or her future behavior.

In sum:

  1. 1. Traits are attributes reflecting the reactive nature of an individual that define his or her constitutional nature.

  2. 2. The selection of important traits is a social and lexical (or communication) process.

  3. 3. Trait inferences are drawn from observed behavior.

  4. 4. Traits are useful because they provide a predictive value for future behavior.

The Predicative Value of Traits in Assessing Presidential Candidates

The predictive value of traits can be extended to voters’ assessments of presidential candidates to foretell behavior in office (see Barber, 1972). Consider the 2000 contest between George W. Bush and Vice President Al Gore. This election focused mainly on domestic issues such as Social Security, Medicare, health insurance, and taxes. Nothing about the policy issues that commanded the center stage during this election would forecast how the candidates would respond to the terrorist attacks a year later. Character traits may play an important role in vote decision because of their predictive value:

  1. 1. Traits are used to capsulate trends in behavior of the presidential candidates into descriptive attributes.

  2. 2. Traits are then used in the prediction of future behavior of the candidate if elected.

  3. 3. These predictions of future behavior are then calculated into voters’ decision making processes.

Lexical Approaches to Identifying “Prototype” Presidential Candidate Traits and Higher Order Constructs

In past research, presidential candidate traits have been operationalized as “prototypes” consisting of only those most relevant to voters. For example, in an influential study by Kinder et al. (1980), participants responded to open-ended questions asking (p. 440) them to describe an “ideal” and an “anti-ideal” president. From the responses, two lists of traits, each containing 16 items, were compiled. Respondents then selected the six most important ones from each list to construct a “profile of an ideal president” (Kinder et al., 1980, 319).3 From these results, Kinder (1986) identified four second-order content dimensions of presidential traits that he labeled competence, leadership, integrity, and empathy. Using a confirmatory factor analysis, Funk (1996) grouped these components into two higher-order factors―competence and integrity―and argued that these are universally relevant in the evaluation of presidential candidates. Work by Geer shows that most attack advertising on candidates from opposing camps focus on these two dimensions (Geer, 2006, Chapter 4).

Consistent with the Kinder et al. (1980) approach, Miller, Wattenberg, and Malanchuk (1986) wrote that voters’ general “schema” of a presidential candidate “will be evoked during the actual campaign period when people receive the appropriate stimuli to trigger these pre-existing cognitions” (523). This view is consistent with research that has shown that people organize their past experiences into cognitive structures known as schemas that are structured sets of expectations and rules that help make sense out of seemingly pattern-less life experiences (Fiske and Taylor, 1991). “Candidate schemas thus reduce the complexity of our impressions by enabling us to categorize and label an individual politician according to certain abstract or representative feature” (Miller, Wattenberg, and Malanchuk, 1986, 524). Examining responses to open-ended questions on the American National Election Studies (ANES) from 1952 to 1984 these researchers found that perceptions of candidates were mostly focused on personality characteristics. Following Kinder and his colleagues, they constructed general categories that voters use in the evaluation of a candidate: competence, integrity, reliability, charisma, and personal.4

Candidate Traits as Decision-making Shortcuts

Voters rely on informational shortcuts and heuristics in making political decisions (Lodge and Stroh, 1993; Lupia and McCubbins, 1998; Popkin, 1994). Candidate traits are particularly useful heuristics because they are relatively easy to assess compared to intricate policy positions (Kinder, 1986). Candidate traits “offer an appealing shortcut for citizens to evaluate candidates on their performances without having to invest considerable time and energy into following public affairs or uncovering candidate issues” (Funk, 1996, 97). Voters use candidate traits as a relatively inexpensive way to gain information about the candidates and simplify vote decisions (Funk, 1999; Kinder et al., 1980; Miller, Wattenberg, and Melanchuk, 1986; Popkin, 1994; Rahn, Aldrich, Borgida, and Sullivan, 1990). This type of evaluation is easy and people do it all the time. Political scientist Wendy Rahn and her colleagues (1990) suggested that voters’ assessments of candidates’ traits mirror their assessment of people they meet in their everyday lives.

(p. 441) Past Research on the Influence of Presidential Character Traits in Vote Decisions

Some of the more thorough tests of the influence of traits were conducted by Miller and Shanks in a series of articles and their book, The New American Voter (Miller and Shanks, 1982, 1996; Shanks and Miller, 1990, 1991) using data from the ANES. Focusing on the 1992 election, these researchers first examined the bivariate relationships between comparative trait evaluations and vote choice and found strong relationships. As they incrementally entered antecedent variables into their statistical model, however, the influence of traits dropped substantially suggesting a large impact from exogenous antecedent variables—i.e., party identification and policy predisposing. Yet the influence of candidate character traits never completely disappeared.

These researchers do find unique variance in vote choice explained by candidate character traits. They conclude that “these results suggest a visible—but limited—extent to which voters’ choices between [Democratic candidate Bill] Clinton and [Republican incumbent President George W.] Bush may have been influenced by their comparative evaluation of Clinton and Bush concerning … specific personal qualities” (Miller and Shanks, 1996, 429). Referring to the series of studies by Miller and Shanks, King (2002) concluded that presidential candidate characteristics had limited influence on vote choice and no impact on determining election outcomes.

When analyzing the ANES data from 1980 to 2000, Bartels (2002) too found detectable influence of candidate traits and concluded that, for the most part, traits do have a small detectable influence on vote choice; however, this influence is not generally in play in election outcomes:

[T]he net effects of candidate trait assessments are generally quite modest in magnitude. The average effect for the six election is 1.6 percentage points, and the largest effect (in 1992) is only 3.5 percentage points. By comparison, the average margin of victory in these six elections (that is the winning candidate’s plurality of the two-party popular vote, including George W. Bush’s small negative plurality in 2000) was about 8.7 percentage points. In three of the six elections—including the two with the largest net trait effects—the winning candidate would quite probably have won by a larger margin had personal qualities played no role in determining the election outcome. The only case in which it seems at all likely that perceptions of candidates’ personal qualities had a decisive impact on the election outcome is the 2000 election, where Bush’s half point advantage with respect to candidate traits was probably one of many “decisive” factors contributing to his razor-thin victory.

(Bartels, 2002, 65)

(p. 442) A strong case for the effects of trait ascription can be found in the 2000 presidential election where, according to the economic forecasting models, Democratic nominee Vice President Al Gore, the candidate for the incumbent party during a good economy, should have won the election with a comfortable margin (Johnston, Hagen, and Jamieson, 2004; see the special issue of PS: Political Science and Politics, 2001, 34, 1). Yet, Gore did not find himself in the White House. Johnston, Hagen, and Jamieson (2004) argued that toward the end of the election Republican candidate George W. Bush outspent Gore on advertising in key “battleground” states but Gore was more successful in getting his message across broadcast news. Therefore, Gore gained votes across the states but Bush gained more votes in the states that mattered and, thus, more electoral votes. Johnston et al. concluded:

The typical forecast placed Gore some eight points ahead of George W. Bush. Our own data indicate that in late September Gore’s margin was about where the forecasts said it should be, eight points. In most elections, this would have seemed insurmountable. Campbell’s (2000) review of postwar elections suggests that Gore would lose ground but, given the late date of this eight point lead, still win decisively. What no account predicted is that a lead of eight points would disappear overnight. The inference is obvious: the election was close because the campaign made it so. (186).

The downfall of Gore, according to Johnston et al. (2004), was the direct outcome of Republican attacks on his character underscored by news coverage that emphasized what reporters saw as a tendency to exaggerate his accomplishments and misuse data. These scholars argued that the interplay between advertisements and news stories that attacked Gore’s character diminished his share of the vote in the battleground and ultimately cost him the election.

A 2011 study found that during US Senate elections, campaign messages and news directly influence voters in the face of controls (Fridkin and Kenney, 2011). Combining a contextual analysis of campaign messages with the 2006 Cooperative Congressional Election Study (CCES), these researchers found that campaign messages focusing on character traits increase citizens’ willingness to evaluate those running for office on trait dimensions and that character traits weigh heavily in voters’ evaluations of competing candidates in US Senate campaigns. Fridkin and Kenney (2011) conclude, “With these data in hand, we find citizen’s trait assessments of senate candidates ‘matter’” (72). Relying on the same CCES data, Hayes (2010) showed that candidate character traits influenced voters during the 2006 US Senate elections, but this relationship is not conditional on campaign intensity and is strongest among those with low levels of political awareness.

The Dynamics of Presidential Candidate Traits

Much of the research in this area relies on the ANES, which, unlike the National Annenberg Election Survey (NAES), does not allow for the assessment of the dynamic nature of traits, During the 2004 election, the average NAES ratings of incumbent Republican President George W. Bush and Democratic challenger Senator John Kerry on the trait “strong leader” (Figure 31.1) was rather dynamic across the general election. The divergence of the rating of the two candidates in August and September coincides with the Swift Boat Veterans for Truth (SVBT) campaign against Democratic Senator John Kerry, which attacked the Democrat’s leadership credentials and trustworthiness. Similar dynamics are reported by Hardy and Jamieson (2005) in their examination of the impact of media coverage of a 2004 Los Angeles Times poll that reported the public thought incumbent president George W. Bush was more stubborn than his Democratic (p. 443) challenger Senator John Kerry. The findings from that study suggested that coverage of this poll magnified the perception that President Bush was indeed stubborn across time showing how trait ratings move across time. The dynamic nature of candidate character traits is further illustrated by Kenski, Hardy, and Jamieson (2010), who attributed changes in perception of candidate traits across the general election to media coverage and campaign communication.

Candidate Traits and Political ChoiceClick to view larger

Figure 31.1. Ratings of the trait “strong leader”

(0-to-10 point scale; 5-day prior moving average; from the 2004 NAES).

What this suggests is that capturing the impact of candidate traits on vote choice with a cross-sectional survey design is similar to shooting at a moving target. The rolling cross-sectional design of the NAES allows for a more nuanced examination of the impact of candidate character traits on vote decision.

Communication Environment, Campaign Context, and Character Traits

When thinking about the impact of candidate character, researchers need to be cognizant of the communication environment which is shaped by both the campaigns, (p. 444) the news media, “real world” concerns and conditions that are central to which traits are in voters’ minds, how they are used to frame the candidates, and ultimately which candidate voters support. Take the 2004 and 2008 presidential campaigns as examples. In 2004, the salience of the War on Terror and Iraq War potentially made leadership a focal trait in voters’ assessment of the candidates. In contrast the worst economic meltdown since the Great Depression marked the 2008 presidential election. Even though Republican Senator John McCain was rated higher than Democratic Senator Barack Obama (see Figures 31.2 through 31.4) on perceptions of which candidate was “ready to be Commander-in-Chief,” “strong leader,” and “has the experience needed to be president,” McCain did not find himself in the Oval Office on January 20, 2009.

Candidate Traits and Political ChoiceClick to view larger

Figure 31.2. Ratings of the candidates on “ready to be commander-in-chief”

(0-to-10 point scale; 5-day prior moving average; from the 2008 NAES).

Candidate Traits and Political ChoiceClick to view larger

Figure 31.3. Ratings of the candidates on “strong leader”

(0-to-10 point scale; 5-day prior moving average; from the 2008 NAES).

Candidate Traits and Political ChoiceClick to view larger

Figure 31.4. Ratings of the candidates on the trait “has the experience needed to be president”

(0-to-10 point scale; 5-day prior moving average; from the 2008 NAES).

To say that leadership qualities of the candidates did not matter in the 2008 presidential election would be misguided. If, instead of the economy being the most important problem, the War on Iraq was of most concern to Americans McCain might have been elected. More Americans believed that the Republican senior Senator from Arizona could better handle the Iraq War than his junior competitor from Illinois (Figure 31.5). When it came to handling the economy Obama consistently trumped McCain (Figure 31.6). As the Iraq War receded in importance in voters’ minds, McCain’s military leadership credentials became less central to voting decisions. Because of the sinking economy, an unpopular incumbent, and the fact that over three-fourths of American’s believed the country was on the wrong track, the 2008 election was one of “change” (see (p. 445) (p. 446) (p. 447) Kenski, Hardy, and Jamieson, 2010). Leadership was not framed by the media in terms of commander-in-chief, where McCain held the advantage, but in terms of a president’s ability to inspire change. Which traits matter in vote decisions is contextual and determined by the intersection of actual conditions, campaign strategies, and media.

Candidate Traits and Political ChoiceClick to view larger

Figure 31.5. Perceptions of which candidate would better handle Iraq war

(0-to-10 point scale; 5-day prior moving average; from the 2004 NAES).

Candidate Traits and Political ChoiceClick to view larger

Figure 31.6. Perceptions of which candidate would better handle the economy

(0-to-10 point scale; 5-day prior moving average; from the 2004 NAES).

Conclusion

In an era marked by an expanding media environment, astronomical campaign spending, and candidate-centered campaigns, it seems probable that candidate traits will continue to be emphasized in future presidential races. Even though the amount of money that presidential campaigns spend rivals the national budget of some small countries, they cannot control external factors such as the economy, war, and so forth, that heavily shape both the media’s and voters’ agendas. But within these boundaries, campaigns can have a significant influence on how salient issues and traits are framed and on their perceptions of which candidate possess the attributes best suited to handle the central issues.

At a minimum, researchers interested in the impact of the candidates’ traits need to hold their ears to the ground to make sure that they include questions on survey instruments that reflect campaign messages. Otherwise the researcher will be left with measures that do not capture the salient traits. A promising research design would track campaign messages as they are being implemented. Such a design may find that within the boundaries of salient issues, campaign messages prove effective in shifting public opinion and corresponding votes. Only a more sophisticated research agenda with a finger on the pulse of the campaigns will be able to test such hypotheses.

The immediate challenge for political communication scholars studying presidential elections is the refining theory and explicating the contextual boundaries that foster or hinder specific communication processes that may influence trait assessments. Theories that enjoyed empirical support during the 1970s, 1980s, and 1990s, when a majority of citizens got their news from print and broadcast television, require fine-tuning as the communication environment surrounding presidential campaigns changes. The informational tide has grown into a tsunami and the number of channels in which campaigns and other motivated groups and individuals can connect to citizens will continue to increase as more mobile devices are adopted and micro-targeting advertising techniques are perfected. Each of these channels presents research opportunities on the impact of candidate character traits.

If the 2008 election is any indication, an additional and significant change in campaign dynamics that requires new theorizing and conceptualization is the growing diversity of the candidates. Are there traits that can be more easily attributed to specific ethnicities or genders? Do these traits or their salience differ when different issues are central? Such questions will be difficult to answer empirically (for example, because they require three-way interactions) but will provide a more nuanced understanding of (p. 448) traits, the role played by the media in how they are framed, and how these interactions affect vote preference.

Another important but undertheorized and studied phenomenon is the increasing competiveness and visibility of presidential primaries. Primaries provide a potentially fruitful setting for trait research since arguably they are more focused on character than the general election because candidates from the same party share similar issue stands and voters are unable to rely on party cues to help form opinions of the candidates.

Perhaps the greatest threat to conclusions on the impact of traits is one that all research based on cross-sectional surveys faces—the issue of causal direction. It remains possible that vote preference leads to rationalized trait evaluation; if so a respondent who prefers one candidate will rate this candidate favorably on any trait.

Providing conclusive empirical support for causal relationships between presidential candidate trait rating and vote preference would be extremely difficult. One could use an experimental design but, given that the context of a presidential election and the flood of communication surrounding it, true replication could not be attained in a lab setting and, therefore, this approach would come at great expense to the validity of the findings. Alternatively researchers could examine the impact of traits in less-complex and media saturated elections (e.g., elections within organizations, local elections, off-year elections, etc.) but generalizations from such low information elections to high information elections like those for U.S. president would be tentative at best.

While the rolling cross-sectional design provides some advantages in addressing issue of causality, it is not a panacea. For example, the daily sample of the rolling cross-sectional design allows for trend analyses of cyclical patterns using statistical techniques such as Autoregressive Regressive Integrated Moving Average models. Such advanced techniques assume that causal shifts happen in longer time spans than the time in which each data point is collected. To find this type of causal relationship, a shift in vote preference would need to occur a day after, or more, a shift in a trait rating. It is quite possible that the shift in an assessment of a candidate trait and one’s vote preference could occur simultaneously, or at least faster than could be detected by the survey design.

More generally, future research needs to be more sensitive to the context-dependent nature of presidential (and perhaps other) elections. The possible impact of traits can only be understood if the communication environment surrounding a presidential election is taken into account. Certain traits matter when the communication environment makes them matter. Salient issues influence both the media’s selection of salient traits and their framing of them, which in turn influences vote preference.

References

Barber, J. D. 1972. The presidential character. Englewood Cliffs, NJ: Prentice-Hall.Find this resource:

Bartels, L. M. 2002. The impact of candidate traits in American presidential elections. In A. King (Ed.), Leader’s personalities and the outcome of democratic elections (pp. 44–69). Oxford, UK: Oxford University Press.Find this resource:

Campbell, J. E. 2000. The American campaign: U.S. Presidential campaigns and the national vote. College Station: Texas A&M University Press.Find this resource:

Carr, H. A., and F. A. Kingsbury. 1938. The concept of traits. Psychological Review, 45, 497–524.Find this resource:

Costa, P. T., Jr., and R. R. McCrae. 1992a. NEO PR-R professional manual. Odessa, FL: Psychological Assessment Resources.Find this resource:

Costa, P. T., Jr., and R. R. McCrae. 1992b. Four ways five factors are basic. Personality and Individual Differences, 135, 653–665.Find this resource:

De Raad, B. 2000. The big five personality factors: The psycholexical approach to personality. Seattle, WA: Hogrefe and Huber.Find this resource:

Fiske, S. T., and Taylor, S. E. 1991. Social cognition (2nd ed.). New York: McGraw-Hill.Find this resource:

Funk, C. L. 1996. Understanding trait inferences in candidate images. In M. X. Delli Carpini, L. Huddy, and R. Y. Shapiro (Eds.), Research in micropolitics: Rethinking rationality, Vol. 5 (pp. 97–123). Greenwich, CT: JAI.Find this resource:

Fridkin, K. L., and P. J. Kenney. 2011. The role of candidate traits in campaigns, Journal of Politics, 73, 61–73.Find this resource:

Funk, C. L. 1999. Bringing the candidate into models of candidate evaluation. Journal of Politics, 61, 700–720.Find this resource:

Geer, J. G. 2006. In defense of negativity: Attack ads in presidential campaigns. Chicago: Chicago University Press.Find this resource:

Hardy, B. W., and K. H. Jamieson. 2005. Can a poll affect the perception of candidate traits? Public Opinion Quarterly, 69, 725–743.Find this resource:

Hayes, D. 2010. Trait voting in US senate elections. American Politics Research, 38, 1102–1129.Find this resource:

Johnston, R., M. G. Hagen, and K. H. Jamieson. 2004. The 2000 presidential election and the foundation of party politics. Cambridge, UK: Cambridge University Press.Find this resource:

Kenski, K., B. W. Hardy, and K. H. Jamieson. 2010. The Obama victory: How media, money, and messages shaped the 2008 election. New York: Oxford University Press.Find this resource:

Kinder, D. R. 1986. Presidential character revisited. In R. R. Lau and D. O. Sears (Eds.), Political cognition. Hillsdale, NJ: Lawrence Erlbaum.Find this resource:

Kinder, D. R., M. D. Peters, R. P. Abelson, and S. T. Fiske. 1980. Presidential prototypes. Political Behavior, 2, 315–337.Find this resource:

King, A. 2002. Do leaders’ personalities really matter? In A. King (Ed.), Leader’s personalities and the outcome of democratic elections (pp. 1–43). Oxford: Oxford University Press.Find this resource:

Lodge, M., and P. Stroh. (p. 450) 1993. Inside the mental voting booth: An impression-driven process model of candidate evaluation. In S. Iyengar and W. McGuire (Eds.), Explorations in political psychology (pp. 225–263). Durham, NC: Duke University Press.Find this resource:

Lupia, A., and M. D. McCubbins. 1998. The Democratic dilemma: Can citizens learn what they need to know? Cambridge, UK: Cambridge University Press.Find this resource:

Miller, A. H., M. P. Wattenberg, and O. Malanchuk. 1986. Schematic assessment of presidential candidates. American Political Science Review, 80, 521–540.Find this resource:

Miller, W. E., and J. M. Shanks. 1982. Policy directions and presidential leadership: Alternative explanations of the 1980 election. British Journal of Political Science, 12, 299–356.Find this resource:

Miller, W. E., and J. M. Shanks. 1996. The new American voter. Cambridge MA: Harvard University Press.Find this resource:

Popkin, S. L. 1994. The reasoning voter: Communication and persuasion in presidential campaigns (2nd ed.). Chicago: University of Chicago Press.Find this resource:

Rahn, W., J. H. Aldrich, E. Borgida, and J. L. Sullivan. 1990. A social-cognitive model of candidate appraisal. In J. Ferejohn and J. Kuklinski (Eds.), Information and democratic processes (pp. 136–159). Champaign: University of Illinois Press.Find this resource:

Saucier, G., and L. R. Goldberg. 2001. Lexical studies of indigenous personality factors: Premises, products, and prospects. Journal of Personality, 69, 847–878.Find this resource:

Shanks, J. M., and W. E. Miller. 1990. Policy direction and performance evaluation: Complimentary explanations of the Reagan elections. British Journal of Political Science, 20, 143–235.Find this resource:

Shanks, J. M, and W. E. Miller. (1991). Partisanship, Policy and Performance: The Reagan Legacy in the 1988 Election. British Journal of Political Science, 21(2), 129–197. doi: 10.1017/S0007123400006098Find this resource:

Sosnik, M. J., M. J. Dowd, and R. Fournier. 2006. Applebee’s America. New York: Simon & Schuster.Find this resource:

Notes:

(3.) This approach in finding important candidate traits follows work in personality trait that focuses on “The Big Five” trait domains: Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness (Costa and McCrae, 1992a, 1992b). Each domain contains “trait facets” or individual traits and the selection of individual traits used to make up the five factors were initially selected by a lexical approach to find clusters of personality descriptors in language (De Raad, 2000). The rationale behind lexical studies of personality traits is based on the assumption that the most meaningful traits are encoded in language as single word descriptors (Carr and Kingsbury, 1938; Saucier and Goldberg, 2001).

(4.) The “personal” category refers to background factors such as military experience, religion, wealth, age, health, previous occupation and so on (Miller, Wattenberg, and Malanchuk, 1986, 528)