ArticlePDF Available

Abstract and Figures

The debate about the relative importance of the personality traits of presidents has a long history. Until the mid-1970s, scholars of the presidency extensively focused on the uniqueness of the individuals that held office. However, the difficulty in capturing presidential personalities and measuring their impact on executive politics led to a significant quantitative shift that focused more on the institutions within which presidents operate. This change produced a long-lasting divide between researchers interested in the "institutional" presidency and those focused on the "personal" presidency. I propose to integrate both approaches by incorporating insights from differential psychology to treat the personality traits of presidents as independent variables. In support of the argument, I use data from an expert survey that captured psychometric traits of presidents who governed the Western Hemisphere in 1945-2012 to reassess an influential study about Latin American presidents. The results show that adding openness to experience leads to a deeper understanding of presidential approval. I conclude by arguing that measuring the personality traits of all sorts of leaders is necessary to modernize the study of elites.
Content may be subject to copyright.
Political Psychology, Vol. 0, No. 0, 2020
doi: 10.1111/pops.12722
0162-895X © 2020 International Society of Political Psychology
Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, 9600 Garsington Road, Oxford, OX4 2DQ,
and PO Box 378 Carlton South, 3053 Victoria, Australia
The Personalities of Presidents as Independent Variables
Ignacio Arana Araya
Carnegie Mellon University
The debate about the relative importance of the personality traits of presidents has a long history. Until the
mid-1970s, scholars of the presidency extensively focused on the uniqueness of the individuals that held office.
However, the difficulty in capturing presidential personalities and measuring their impact on executive politics
led to a significant quantitative shift that focused more on the institutions within which presidents operate. This
change produced a long-lasting divide between researchers interested in the “institutional” presidency and
those focused on the “personal” presidency. I propose to integrate both approaches by incorporating insights
from differential psychology to treat the personality traits of presidents as independent variables. In support of
the argument, I use data from an expert survey that captured psychometric traits of presidents who governed
the Western Hemisphere in 1945–2012 to reassess an influential study about Latin American presidents. The
results show that adding openness to experience leads to a deeper understanding of presidential approval. I
conclude by arguing that measuring the personality traits of all sorts of leaders is necessary to modernize the
study of elites.
KEY WORDS: presidents, personality traits, leadership, interdisciplinary
Presidents receive universal attention in their countries. The press covers their agenda, most
citizens have an opinion about them, and numerous scholars and pundits routinely assess the leaders’
performance through media and academic outlets. Such attention is not without cause. These indi-
viduals are the most powerful politicians in presidential systems, and their decisions have enormous
consequences. Given the substantial resources dedicated to analyzing presidents, one would expect
that many questions about presidential behavior have been answered. However, there is still limited
understanding of how presidents’ personalities may matter.
The subject is of utmost importance. Using the leaders’ unique traits as independent variables
allows scholars to examine research questions related to the potential effect that presidents have
on domestic and international politics. Including the personality traits of heads of government also
allows us to revisit works whose findings could be biased because the personalities of presidents
were omitted, despite reasons to expect them to have an identifiable impact on executive politics.
Furthermore, if certain traits can be related to a better performance in the presidency, their identifi-
cation could allow voters to support the leaders best equipped to represent them and potentially raise
the standards of presidential accomplishment.
There seems to be a growing interest in understanding how the psychological characteristics
of leaders affect their performance in office. The election of Donald Trump to the US presidency
2Arana Araya
sparked controversy among medical practitioners and political scientists regarding his psychological
fitness for office (Ashcroft, 2016). In a recent book (Lee, 2017), more than two dozen psychiatrists
and psychologists agreed that Trump’s mental instability is dangerous to the United States—and the
The concern among scholars, practitioners, and the general public about the mental stability
of heads of government is not new. Researchers examining the personalities of world leaders have
claimed, for example, that Saddam Hussein was a “malignant narcissist” (Post, 2003), Adolf Hitler
a “psychopath” (Waite, 1977), and Joseph Stalin a “manic-depressive” (Hershman & Lieb, 1994). A
study of American presidents concluded that half of them suffered from mental illnesses, including
depression, substance abuse, and bipolar and anxiety disorders (Davidson, Connor, & Swartz, 2006).
In the last decades, Presidents Hugo Chávez of Venezuela and Abdalá Bucaram of Ecuador were
frequently in international headlines for their unconventional behavior (De la Torre, 1999; Koeneke,
2002). Bucaram, popularly nicknamed “El Loco,” was eventually ousted by Congress for—offi-
cially—being “mentally unfit” to govern.
While other subfields in political science are increasingly absorbing personality research (Kertzer
& Tingley, 2018), presidency scholars have not embraced it despite widespread concerns about the
effects of presidential behavior on democratic norms and institutions. For example, Levitsky and
Ziblatt (2018) draw direct lines between the behavior of presidents such as Juan Domingo Perón,
Alberto Fujimori, Hugo Chávez, Evo Morales, Rafael Correa, and Donald Trump to the erosion of
democracies. These associations, nonetheless, remain largely descriptive. If these leaders do repre-
sent an authoritarian threat, then examining how their personalities shape their behavior is a pressing
This article argues that presidential studies should embrace the theories and methods that differ-
ential psychology offers to test hypotheses using the personality traits of presidents as independent
variables. This branch of psychological research studies the individual differences of humans or how
the feelings, acts, thoughts, and behavior of people differ from each other. In support of the argument,
I introduce a survey in which experts completed psychometric questionnaires to assess the personal-
ities of 165 presidents who governed a Western Hemisphere country between 1945 and 2012.
The survey data is proposed as an instrument to measure the personality traits of presidents. To
prove its added value, a “proof of concept” exercise will reasess part of a study about Latin American
leaders. I selected Pérez-Liñán’s (2007) book because it is an influential piece of scholarship and is
among the few quantitative works that have tried to capture arguments related to the uniqueness of
leaders. However, the author did not incorporate individual characteristics of presidents as causal
factors. The exercise also allows for scrutiny of personality traits using a cross-national sample,
avoiding the potential biases of including leaders from a single country. I reassess two models in the
book that attempt to explain presidential approval including “openness to experience” as an indepen-
dent variable. The exercise shows that the revisited models gain analytical leverage.
The analysis centers on the Americas because the region comprises 20 of the 49 existent pres-
idential systems and presents large variation in economic and democratic development. Therefore,
its experience can serve as a reference for all presidential systems. In fact, the propositions advanced
here have implications for the study of all types of leaders.
The next section reviews the historical role of leaders’ personalities in presidential studies. The
focus is mainly on studies about American presidents because it is the richest literature, but a similar
evolution has occurred in Latin American research. The third section advances the main argument
by discussing how personality research can contribute to our understanding of presidents. The fourth
section introduces the expert survey conducted and shows the reassessment of Pérez-Liñán’s (2007)
work. I conclude by proposing that researchers must move from ignoring or theorizing about the
effect that the personalities of presidents and other elites have on politics to using techniques to
estimate their impact.
Personalities of Presidents
The Evolution in Presidential Studies
American and Latin American presidential studies can be divided in two dominant approaches.
President-centered researchers consider the attributes of leaders essential to understanding policy
outcomes and decision-making in the executive branch (Arana Araya, 2016a, 2016b; Arana Araya
& Guerrero Valencia, 2020; Barber, 1972; Corwin, 1940; Cronin & Greenberg, 1969; Greenstein,
2009; Hermann, 2003; Koenig, 1964; Neustadt, 1960; Renshon, 2008; Walker, 1990). In contrast,
presidency-oriented scholars minimize the importance of presidents as individuals and rather focus
on the institutional setting in which they work (Heclo, 1977; King, 1975, 1993; Lowi, 1986; Moe,
1993; Wayne, 1983). Both schools of thought have evolved through parallel corridors with limited
interconnection, leading to conflicting views on how the presidency works.
A brief review of the evolution in American scholarship can be illustrative. Mainstream research
on the American presidency considered the uniqueness of presidents an important factor in under-
standing executive politics until the mid-1970s (Edwards, Kessel, & Rockman, 1993). These studies
were primarily of a qualitative nature and can be grouped into three research streams that, at varying
degrees, combined the president-centered and the presidency-oriented approaches.
One group of studies centered on the legal structures and roles of the presidency. Corwin (1940)
analyzed various conceptions of the office and the role of the president as administrative chief. Other
works saw presidential roles as responding to external conditions and therefore expanding beyond
strictly legal definitions (e.g., Koenig, 1964). Despite the legalistic focus of these authors, presiden-
tial behavior was deemed central to executive politics, with Corwin stating: “What the presidency
is at any particular moment depends in important measure upon who is president” (1940, p. 338).
A second group focused on the exercise of presidential power and the operation of the White
House. Neustadt, for example, proposed that presidents who lead by persuasion are more successful
than those who rely on formal powers. He famously claimed that the presidency is a place for “ex-
perienced politicians of extraordinary temperament” (1960, p. 163). Cronin and Greenberg (1969)
analyzed the presidential advisory network, providing insights about how individuals and organiza-
tions influenced presidential decision-making and how the characteristics of whoever sat in the Oval
Office shaped the presidency.
A third research stream advanced a political psychology approach through the development of
typologies, operational code analyses, and psychobiographies. Barber (1972) categorized American
presidents according to their orientation toward their role as either passive or active and positive or
negative. More recently, Hermann used interview data to produce typologies about the personalities
of presidents and other world leaders (Hermann, 2003). Scholars have conducted operational code
analysis to infer how the values and beliefs of leaders explain how they process external events (e.g.,
Renshon, 2008; Walker, 1990). Although statistical analyses have been conducted for most American
presidents (e.g., Renshon, 2008), foreign policy researchers have used this procedure more than
presidency scholars. Finally, psychobiographers use psychological constructs to analyze the life of
prominent individuals to understand their behavior and decisions. That was the purpose behind the
aforementioned studies about Hitler, Hussein, and Stalin.
The (Partial) Depersonalization of the Presidency
In the mid-70s, influential researchers started to complain that presidential studies were lagging
behind other more sophisticated subfields (e.g., Heclo, 1977; King, 1975, p. 30). It was the begin-
ning of a strong intellectual movement towards the quantification of presidential studies that led to a
permanent split between president-centered and presidency-oriented research.
The criticism was blunt: “To read most general studies of the United States presidency… is to
feel that one is reading not a number of different books but essentially the same book over and over
4Arana Araya
again… The existing literature is mainly descriptive and atheoretical” (King, 1975, p. 173). The crit-
icism grew on the 1980s (e.g., Lowi, 1986; Wayne, 1983). Wayne (1983, p. 6) complained that “by
concentrating on personalities, on dramatic situations, and on controversial decisions and extraordi-
nary events, students of the presidency have reduced the applicability of social science techniques”
(1983, p. 6).
The criticism of qualitative president-centered research soon became mainstream, and the exam-
ination of the uniqueness of leaders became the main collateral damage. Lowi claimed that “an in-
stitutionalist approach does not deny the relevance of individual psychology but treats it as marginal
in the context of the tremendous historical forces lodged in the laws, traditions and commitments
of institutions” (1986, p. 20). Some scholars urged to abandon considering heads of state as units of
analysis for causal inferences because it was “extremely unlikely to yield reliable empirical conclu-
sions” (King, 1993, p. 403). Other calls were more extreme. Moe proposed “to stop thinking about
presidents as people and to start thinking of them generically: as faceless, nameless institutional
actors whose behavior is an institutional product” (1993, p. 379).
The “depersonalization” of the presidency led to a rise in the development of theories derived
from rational choice that were statistically tested. These presidency-oriented studies increasingly
deemed the personalities of presidents as analytically incomparable and unimportant. Meanwhile,
works centered on the “personal” presidency has continued to flourish in parallel, but in little con-
nection with students of the “institutional” presidency and personality psychology. A similar divide
has continued in Latin American scholarship despite works on political elites (Hart, 1977; Stevens,
Bishin, & Barr, 2006) and on mass behavior (Ribeiro & Borba, 2016) that have integrated insights
from personality psychology.
Toward an Interdisciplinary Agenda: Integrating Personality Research
Studying presidential personalities is pressing for at least three reasons. First, there are countless
situations in which the traits of the leaders can make a difference in executive politics. For example,
scholars could explore whether emotionally intelligent leaders are more likely to try to convince
legislators to support their legal initiatives, given that this trait has been associated with negotiating
skills (Der Foo, Anger Elfenbein, Hoon Tan, & Chuan Aik, 2004). Researchers could also examine
whether extraverted leaders and their tendency to be talkative attention seekers are more likely to
sidestep Congress and make public appeals on behalf of their legislative agendas.
Undoubtedly, estimating the effect of personalities has challenges. Researchers may misunder-
stand how context mediates the causal relationships explored (Zaccaro, Green, Dubrow, & Kolze,
2018). For example, the institutional context may strongly constrain the effect of the behavioral
dispositions of a leader. A president with scant legislative support may be unable to pass legisla-
tion—irrespective of her personality. However, even in these cases the leaders’ characteristics should
condition how institutions shape their attitudes and behaviors.
Second, treating the personality traits of presidents as independent variables enables us to revisit
theories and findings from research in which presidential personalities were unjustifiably not considered
as explanatory factors. Take, for instance, the role of presidents in their attempts to change the constitu-
tions of their countries to relax their term limits. Mckie (2019) revealed that 30% of the 221 presidents
who faced term limits in Latin America, the Middle East, Africa, and Asia from 1975 to 2018 attempted
to overstay in office. Existing quantitative explanations of the removal of term limits have focused on
the institutional context in which leaders govern (e.g., Corrales, 2016; Negretto, 2013). However, all
presidents would benefit from enjoying for more time the privileges of office. Arguably, whether the
unique traits of the leaders are behind the attempts to remove term limits needs to be empirically tested.
So far, most of the few works centered in the institutional presidency that have attempted to cap-
ture the uniqueness of leaders have used dichotomous variables to differentiate among incumbents
Personalities of Presidents
(e.g., Eshbaugh-Soha, 2003; Gilmour, 2002). This is problematic because dummy variables only
differentiate the years in which a leader governed, and therefore they do not capture anything specific
about the chief executives. As the psychologist Simonton (1987) showed, adding a personality trait
to reexamine presidency-oriented research can improve our understanding of executive politics. The
author found that the incumbents’ level of inflexibility moderates the impact that the magnitude of
the presidents’ electoral mandate and the support that leaders have in Congress have on the presi-
dents’ use of vetoes.
Third, voters and political parties can make a more informed decision when supporting a candi-
date for office if they have a better understanding of the differences that the traits of presidents can
make on executive politics. There is evidence that voters support candidates based on their traits. For
example, Cohen (2018) showed that voters are more likely to support the presidential candidates they
believe are more intelligent. Furthermore, Holian and Prysby (2014) found that perceptions of presi-
dential candidates’ traits such as empathy, competence, and integrity affected whom voters supported
in the 2012 American elections.
Moreover, voters informed about the potential consequences of their choice for office should
be better prepared to handle the misinformation spread by negative political advertising and “fake
news.” In sum, the electoral arena could resemble a widespread hiring practice in the corporate
world. By one account, two-thirds of medium-to-large American organizations use some type of
psychological testing to select their personnel in order to improve employee fit and reduce turnover
(Beagrie, 2005).
To integrate personality research into presidential studies, presidency-oriented and presi-
dent-centered researchers must first acknowledge three well-established facts. First, since the 1990s,
there is a broad agreement on the central components of human personality. The reigning psycho-
logical paradigm dictates that there are five core personality traits, known as the “Big Five”: neurot-
icism, extraversion, openness to experience, conscientiousness, and agreeableness (McCrae, 2009;
McCrae & Costa, 1997). This paradigm has proved to be robust and generalizable across virtually all
cultures (Benet-Martinez & John, 1998; McCrae, 2009; McCrae & Costa, 1997).
Second, personality research reached a general consensus that personality traits tend to remain
stable over time and therefore can be systematically examined (Corr & Matthews, 2009; Costa &
McCrae, 1992; McCrae, 2009). The study of traits has grown parallel to the consolidation of the Big
Five paradigm but also due to advances in genetics and neuroscience and rising integration among
various fields in psychology (Matthews, Deary, & Whiteman, 2003).
Third, psychologists have published thousands of personality traits scales that are constantly re-
fined (for a review, see Weiner & Greene, 2017). These scales have undergone construct validation, a
process necessary to measure latent or unobservable concepts, as is the case with all personality traits.
Therefore, political scientists do not need to create new scales of questionable scientific standard
or to measure presidential traits while overlooking personality research. For example, Greenstein
(2009) assessed the public communication, organizational capacity, political skills, policy vision,
cognitive style, and emotional intelligence of the first seven American presidents based mainly on
his own definition of these concepts.
The three facts described show that there is no impediment to using presidents as units of anal-
ysis and their personalities as independent variables. In fact, psychologists have measured nearly
all possible personality traits of American chiefs of state. According to Song and Simonton (2007),
psychologists have studied outstanding individuals, such as presidents, to examine whether findings
derived from samples of nonexceptional people are generalizable and to determine if certain attri-
butes of the leaders explain their exceptionality. Since remarkable individuals may be unwilling or
unable to participate in psychological studies, researchers have developed techniques to study them
at a distance. The four most used techniques are psychobiographies, content analysis, historiometry,
and expert surveys (Song & Simonton, 2007).
6Arana Araya
Psychobiographies focus on events in the life of individuals to have a deeper understanding of
the motivations behind some of the subjects’ actions. However, since psychobiographies are subjec-
tive analyses, the authors’ biases may affect them (see Flett, 2007). Content analysis has been defined
as “the systematic, objective, quantitative analysis of message characteristics” (Neuendorf, 2016,
p. 1). To elude reliability problems, researchers need to avoid documents that are inauthentic (e.g.,
a speech written by a ghost writer) or insincere (e.g., a strategically framed speech to Congress).
Historiometric analysis involves transforming qualitative information about people into quantitative
data to test theories about their individual differences (Ligon, Harris, & Hunter, 2012, pp. 1104–
1105). As in content analysis, historiometric works may inadvertently inspect information that is
unrepresentative of the leaders’ personalities, especially when the biographical data is limited or of
dubious trustworthiness (Simonton, 2018a, p. 331).
Finally, expert surveys allow scholars to gather data from multiple qualified individuals. This
technique allows researchers to determine estimates on new, rare, complex, or poorly understood
phenomena; forecast future events; interpret or integrate existing data; and capture the present
knowledge and potentially future pursuits in a field (Meyer & Booker, 2001). Expert surveys may
experience biases if the participants are not truly specialists or if their feelings affect their judgment.
However, these potential problems can be addressed by the identification of each experts’ credentials
and asking raters questions designed to capture their potential biases. Furthermore, research has
validated using observer ratings to assess personalities. One meta-analysis has even showed that ob-
server ratings have a predictive validity of academic achievement and job performance substantially
greater than self-ratings (Connelly & Ones, 2010).
Personality researchers have used the four techniques described to study the personality of pres-
idents in six broad topics. Some scholars have measured the Big Five (Rubenzer & Faschingbauer,
2004; Rubenzer, Faschingbauer, & Ones, 2000). Another group has tried to capture the presidents’
power, achievement, and affiliation motives (Holmes & Elder, 1989; Winter, 2002). A third group
has centered on leadership characteristics, such as the presidents’ charisma (Deluga, 1998; Emrich,
Brower, Feldman, & Garland, 2001; Simonton, 1988) and interpersonal, creative, deliberative, and
neurotic styles (Simonton, 1988). A fourth group has analyzed the decision-making style of presi-
dents based on their “integrative complexity” (e.g., Tetlock, 1981; Thoemmes & Conway, 2007),
a concept that alludes to the degree to which thinking integrates multiple perspectives. A fifth re-
search stream has explored relationships between the leaders’ intellect and their accomplishments
(Simonton, 1988, 2006). A final group has measured traits that do not fit a simple category, such
as moderation, friendliness, Machiavellianism, achievement drive, forcefulness, wit, inflexibility,
pacificism, activity inhibition, narcissism, and psychopathy (Davidson et al., 2006; Deluga, 1997;
Lilienfeld et al., 2012; Simonton, 1986).
Notably, assessments of presidential personalities that have used different techniques and sam-
ples are highly correlated, as it would be theoretically expected. For example, Winter and Stewart’s
(1977) content analysis of power motivation had a correlation of .80 with Simonton’s (1986, p. 152)
historiometric assessment of forcefulness. Furthermore, Rubenzer and Faschingbauer’s (2004) ex-
pert-based measurement of openness to experience has a correlation of .69 with Simonton’s (1986,
2006, p. 518) measurement of intellectual brilliance.
How Personalities Matter: A Proof of Concept
Expert Survey
Between 2012 and 2013, I led a research team that conducted a survey among experts about one
or more of the 315 presidents of the Western Hemisphere who governed between 1945 and 2012 for
at least six months. The end of World War II was used as the starting point because it marks the onset
Personalities of Presidents
of the most extensive democratic period in the Americas. Experts assessed personality traits and
other individual differences of leaders that may relate to executive governance.
The expert survey is the main component of the Presidential Database of the Americas (Arana
Araya, 2017), which also contains detailed biographical information about the heads of government.
An expert survey was conducted because it was the procedure best suited to overcome data-collec-
tion limitations that could affect the validity of the instrument used. After conducting field research
in seven countries and an extensive search for online documents, it became clear that even finding
reliable biographies for a third of the sample was difficult. The available material for leaders who
governed before 1970, did not finish their terms, or belonged to countries that were poor, small, or
had a small population tended to be extremely limited. Instead, finding experts was more feasible
because there were many potential participants that had not yet published biographies but still had
detailed knowledge about chief executives.
Five strategies were followed to identify the largest possible amount of raters. First, a search was
conducted in the WorldCat database using the names of each head of government and also using the
keywords “presidents” and presidentes (to cover publications in Spanish and Portuguese). This led
to a review of 14,202 book titles, of which 1184 authors were identified as potential raters. Second,
a search of scholarly publications about presidents was done in Google Scholar and Amazon. Third,
experts who participated in similar studies about American chief executives were identified; C-SPAN
Survey of Presidential Leadership (2000), Taranto and Leo (2004), and Rubenzer and Faschingbauer
(2004). Fourth, nearly 50 professional organizations of the Americas that group historians, political
scientists, and journalists were asked to provide names of potential raters. Finally, an invitation to
participate was extended to scholars whose names were suggested by other participants in the survey.
All this search led to a finding of 1879 experts. After subtracting those who had passed away or were
otherwise inaccessible, the contact information of 911 scholars was identified.
Participants in the survey answered three types of questions. First, they completed psychomet-
ric tests that captured the Big Five, as well as risk taking, assertiveness, and dominance. To mea-
sure these constructs, I used scales extensively used in personality research. Second, raters assessed
background characteristics of the leaders (e.g., ideology and socioeconomic origin). Third, experts
answered questions designed to measure their potential biases. They reported their gender, age, na-
tionality, city of residence, educational attainment, sympathy toward the president, approval of the
leader’s performance, number of times they met the chief of state (and if the contact was profes-
sional, friendly, or familiar), and their ideology.
Each online survey took about 20 minutes to be completed and was delivered via email in three
languages. In total, 361 experts from 29 nationalities filled out 531 surveys in which they assessed
165 leaders from 19 countries. The average age of the raters was 57; 73% of them were male and had
a high educational attainment; 96% completed college degrees, and 56% held a Ph.D. Most raters
were political scientists (27%), journalists (19%), and historians (18%). Furthermore, experts were
highly knowledgeable about the presidents; 216 questionnaires were answered by evaluators who
had met the leaders at least once.
Figure1 shows the geographical distribution of the raters. The dots in the figure identify the
location from where the survey was responded, and countries with more participants are colored
darker. The mean number of raters per head of state is 3, a number that allows the conduction of reli-
able analyses (Rubenzer & Faschingbauer, 2004). Even considering contacted experts who may have
not been reached, the response rate was 40%, which is above most online surveys (Hamilton, 2009).
Reassessment of an Influential Study
I use part of the data generated in the survey to examine whether a prominent work could have
gained analytical leverage had the author included personality traits of presidents as independent
8Arana Araya
variables. Pérez-Liñán (2007) studied six heads of government from five Latin American countries
who faced impeachment processes. He found that the decline in presidential approval contributes
to the occurrence of impeachments. I reexamine two models in which Pérez-Liñán (2007, p. 117)
tested whether presidential approval is contingent on government scandals, whether the president
was involved in the scandals, a honeymoon period, inflation, unemployment, and the implementation
of adjustment policies.
I expect that openness to experience is directly related to presidential approval—and therefore
indirectly associated with the impeachments. Openness reflects how broadly and deeply people think
(Costa & McCrae, 1992; McCrae & Costa, 1997). People high in openness tend to be creative, intel-
lectually curious, sensitive to aesthetics, attentive to inner feelings, self-directed, sensation seekers,
and prefer novel forms of stimulation (Dollinger, Leong, & Ulicni, 1996).
The relationship between openness and presidential popularity can be established through two
paths. First, openness is beneficial in a position that constantly demands making relevant decisions
about complex, multilayered issues. Being intellectually curious helps to be well informed, to process
large amounts of information, and to solve complicated problems. Creativity helps to find innovative
solutions to convoluted challenges. In support of this expectation, Rubenzer et al. (2000) found that
presidents who are more open to experience tend to perform better in office. The authors reached
this finding after measuring the Big Five across all American heads of government and then using
the scores to predict presidential success (as measured by Murray & Blessing, 1983). The authors at-
tributed their finding to the strong link between openness and cognitive ability. This relationship has
also been found on research on American presidents (Simonton, 2018b), and intelligence has been
positively associated with presidential greatness (Simonton, 1986, 1988, 2006). Furthermore, since
voters are more likely to support a candidate they perceive as intelligent (Cohen, 2018), this type of
leaders could be expected to be more popular.
Second, personality research has associated openness with charismatic leadership because
some characteristics of this trait—especially creativity and unconventional thinking and behavior—
have been related to the ability of leaders to motivate and inspire followers (Oreg & Berson, 2015).
Furthermore, research has found a positive correlation between charismatic presidents and success
in office (Deluga, 1998; Emrich et al., 2001; Simonton, 1988).
Figure 1. Geographic distribution of survey experts. Source: Author’s elaboration based on the IP addresses of the raters.
Personalities of Presidents
In sum, since previous works suggest that leaders who are high in openness tend to excel in
office and attract followers, by extension they should also tend to be popular leaders. Therefore, I
reassess the work of Pérez-Liñán in which he explored the causes of presidential approval adding
presidents’ openness to experience to his models.
The expert survey measured the Big Five using the Big Five Inventory (BFI), a scale composed
of 44 short phrases (John, Donahue, & Kentle, 1991). The BFI has internal consistency, retest reli-
ability, and strong convergence and discriminant validity with longer Big Five measurements. It also
has a high agreement between self- and peer-reports and works well across cultures (Benet-Martinez
& John, 1998). Cronbach’s alpha was computed to measure the interrater reliability of the BFI scores
obtained in the expert survey. The alpha was .82 for the general sample and .85 for the six presidents
in Pérez-Liñán’s work. These numbers are above the .70–.80 conventional threshold in psychomet-
rics (Song & Simonton, 2007, p. 315).
Table1 shows the BFI statements that measured openness. Like the other components of the
BFI, the variable ranges between 1 (no openness) and 5 (total openness). As is conventional, the
score for each president is the average value provided by raters (Steenbergen & Marks, 2007).
Pérez-Liñán analyzed the presidencies of Fernando Collor de Mello (Brazil, 1990–92), Carlos
Andrés Pérez (Venezuela, 1989–93), Ernesto Samper (Colombia, 1994–98), Abdalá Bucaram
(Ecuador, 1996–97), Raúl Cubas Grau (Paraguay, 1998–99), and Luis Ángel González Macchi
(Paraguay, 1999–2003). In the survey, these leaders were assessed by between three (González
Macchi and Cubas Grau) and seven experts (Samper).
Figure2 shows the distribution of openness across the six leaders, which varied from 2.6 (Collor
de Mello) to 3.78 (Samper). The horizontal line contrasts the values of these leaders to the 3.21 aver-
age score received by the 152 presidents whose openness was assessed in the survey. As a reference, I
also included the minimum (1.6 for Ecuadorean Lucio Gutiérrez) and maximum (4.7 for Costa Rican
José Figueres Ferrer) scores assigned to any leader. Noticeably, the presidents in Pérez-Liñán’s study
did not approximate extreme values.
Table 1. Openness to Experience
Here are a number of personality traits that may or may not apply to the president. Please indicate the extent to which
you agree or disagree with the following statements:
Strongly Disagree Disagree
Neither Agree
nor Disagree Agree Strongly Agree
1. Was original, came up
with new ideas
2. Was curious about
many different things
3. Was ingenious, a deep
4. Had an active
5. Was inventive
6. Valued artistic, aesthetic
7. Preferred work that is
routine (reversed)
8. Liked to reflect, play
with ideas
9. Had few artistic inter-
ests (reversed)
10. Was sophisticated in
art, music, or literature
Source: John et al. (1991).
10 Arana Araya
Revealingly, there is a negative association between openness and conservatism, a relation also
widely documented among citizens (e.g., Caprara, Schwartz, Capanna, Vecchione, & Barbaranelli,
2006) and politicians (e.g., Best, 2011) and present among the presidents assessed in the expert sur-
vey (there is a .45 negative correlation between both variables). Relatedly, studies on American presi-
dents found that conservatism has a negative correlation with intellectual brilliance (Simonton, 1986)
and that intellectually brilliant leaders are more open to experience (Simonton, 2018b). Political
ideology was assessed in a 7-point scale, where 1 was far left and 7 far right. In the sample, Samper
scored 3, Pérez 3.5, Macchi 4, Bucaram 4.75, and Cubas Grau and Collor de Mello 5.5.
Table2 reexamines two models in which Pérez-Liñán (2007, p. 117) explores the causes of
presidential approval. Model 2.1 presents Pérez-Liñán’s base fixed-effects model, while 2.4 shows
the robustness-check model in which he linearized the dependent variable (bounded between 0 and
100) by taking the natural logarithm of the odds of approval. In models 2.2 and 2.5, I reestimate 2.1
and 2.4, respectively, adding Openness. Model 2.2 presents a random-effects estimation because it
is consistent with the argument that the individual differences of presidents should be captured in-
stead of simply controlling for “unobserved heterogeneity” as fixed-effects models do (in any case,
the statistical significance of the variables in model 2.2 do not change when a fixed-effects “hybrid
method” is conducted). Following Allison (2009), in model 2.4 I conduct a “hybrid method” suitable
for time-series data that includes variables that do not change over time, such as Openness.
Noticeably, the results in models 2.2 and 2.5 do not change the core findings in models 2.1 and
2.4. Model 2.2 shows that each one-unit increase in the presidents’ Openness increases by 11% the
popularity of a president. Estimating predicted probabilities shows that a leader who scores 1 in
Openness (the lowest value) will have an average popularity of 13%, while a president who scores
5 (the highest value) will enjoy a support of 57%. Four other variables become statistically signifi-
cant. Presidential approval decreases by 18% when leaders adopted neoliberal reforms (Adjustment
Policy), by 8% when presidents were directly involved in a corruption or abuse of power scandal
(President Involved), and 6% when a scandal affected the government every month (an average of
100 points in Scandal Index). Finally, presidents enjoyed an advantage of 20% of approval during the
first three months of their terms (Honeymoon).
Model 2.3 reruns model 2.2 but excludes Openness to uncover the difference that it makes. In
line with the argument, the explanatory power of the model decreases (the “within” R2 changes from
Figure 2. Openness to experience.
Personalities of Presidents
.45 to .43). Additionally, the effect size of the statistically significant variables changes, supporting
the proposition that adding Openness mitigates an omitted variable bias problem.
Arguably, the scores of Openness can be traced in the impeachment processes that Collor
de Mello (2.76 in openness) and Samper (3.78) faced. Samper was the only leader—along with
González Macchi—who avoided being ousted from power. According to Pérez-Liñán (2007, pp.
22–24), Samper’s popularity and majority support in Congress helped him to navigate through the
impeachment process. The story of Collor de Mello is different: “Once in power, he had built an
isolated command in order to preserve his autonomy vis-à-vis traditional interest groups and to cen-
tralize power in his personal office instead of building a stable, but costly, legislative coalition” (p.
17). Collor de Mello was stubborn, unimaginative, and unable to face the impeachment process with
new ideas. In contrast, Samper made creative decisions, such as asking Congress to investigate the
accusations against him, and played the role of a victim of political elites and the US government.
While the personality of Samper helped him to survive in office, the traits of Collor de Mello con-
tributed to his downfall.
For illustrative purposes, Figure3 shows the predicted popularity of Collor de Mello (blue line)
versus a counterfactual Collor de Mello (red line) that has the Openness of Samper. Noticeably, the
popularity of the counterfactual leader would have been so high that he might have avoided the con-
stitutional crisis altogether.
Openness, nonetheless, does not shield the popularity of a president. For example, Pérez scored
only second to Samper in Openness (3.43). When he took office in 1989, he was considered “charis-
matic” and a “living myth” among politicians (Pérez-Liñán, 2007, pp. 99–103). However, his initial
approval of 70% sharply declined after he implemented severe adjustment policies, which triggered
violent riots and two military coups attempts. Corruption scandals also undermined Pérez’s govern-
ment, and his fate was sealed once he was connected to a financial fraud. Congress swiftly removed
him from office in 1993.
Table 2. Models of Presidential Approval
(2.1) (2.2) (2.3) (2.4) (2.5)
Fixed-Effects Random Effects 2.2 Revisited Log-Odds Hybrid Log-Odds
Scandal Index −.11** −.06*** −.07*** −.01** −.01**
(.05) (.01) (.01) (.00) (.00)
President Involved −9.70*** −7.80** −7.40* −.49*** −.49***
(3.08) (2.91) (3.01) (.16) (.15)
Honeymoon 17.37*** 19.67*** 17.90*** .82*** .82***
(3.36) (2.98) (3.04) (.17) (.16)
Inflation .05 .03 −.06 −.00 −.00
(.16) (.13) (.13) (.01) (.01)
Unemployment −1.82** −.92 −.43 −.07* −.07*
(.69) (.53) (.53) (.04) (.03)
Adjustment Policy −15.76*** −17.69*** −21.77*** −.90*** −.90***
(3.09) (2.84) (2.69) (.16) (.15)
Openness 11.21*** 14.44**
(3.18) (5.94)
#R2.46 .45 .43 .46 .46
Wald chi2 225.94 199.09 254.69
Prob>chi2 .000 .000 .000 .000 .000
Observations 165 165 165 165 165
Note: Unit of analysis is president-month. #Corresponds to “within” R2.
Standard errors in parentheses.
*p<.1; **p<.05; ***p<.01.
12 Arana Araya
Robustness Checks
The results presented on Table2 may be subject to three main concerns. First, the experts’ char-
acteristics could potentially bias their survey answers. To investigate this possibility, I correlated the
experts’ 22 answers to their personal characteristics. Their judgment of the presidents’ performance
and their sympathy toward the leaders were the only features that correlated above .20 with one of
the scores the leaders received. These two variables have a correlation of .81, arguably because they
both measure the experts’ normative assessment of presidents. To account for a potential bias, model
3.1 on Table3 reruns model 2.2 excluding experts who strongly approved/disapproved of the leaders’
Second, an endogeneity problem could occur if the experts who participated in the survey as-
sessed Openness based on the leaders’ popularity in office. If such reverse-coding happened, then
Openness could be partially explained by the values of the dependent variable. Three reasons may
dissipate this concern. First, experts were asked to reflect on the presidents’ characteristics before
they reached office. Second, the content of the statements that capture Openness (presented on
Table1) is unrelated to the president’s popularity. Third, experts may be unable to associate presi-
dential approval to Openness because the former varied significantly (as shown in Appendix S1 in
the online supporting information) while the personality trait is stable. Still, model 3.2 reruns model
2.2, conducting an instrumental variable (IV) procedure. Since research has found that Openness is
associated with education (e.g., Costa et al., 1986), I use formal educational attainment as the IV.
The exclusion restriction requires that the IV is not directly related to presidential popularity. Thus,
I tested whether Education is statistically associated with approval after controlling for Openness,
which was not the case. Furthermore, the first-stage model F-statistic is .83, well above conventional
thresholds for instrument strength (Stock & Yogo, 2005).
Third, other Big Five traits could be related to presidential popularity. For example, in an extensive
meta-analysis, Judge, Bono, Ilies, and Gerhardt (2002) found that extraversion is consistently associ-
ated with leadership emergence and effectiveness. Therefore, I replaced Openness with Neuroticism,
Extraversion, Agreeableness, and Conscientiousness in models 3.3, 3.4, 3.5, and 3.6, respectively.
Finally, model 3.7 reestimates model 2.2 with a panel-corrected standard errors (PCSE) regression to
Figure 3. Predicted effect of openness to experience.
Personalities of Presidents
Table 3. Robustness Checks
(3.1) (3.2) (3.3) (3.4) (3.5) (3.6) (3.7)
2.2 Without
Extreme Raters Endogeneity Neuroticism Extraversion Agreeableness Concientiousness 2.2 With PCSE
Scandal Index −.06*** −.06*** −.05** −.09*** −.06*** −.05*** −.01
(.01) (.01) (.01) (.01) (.01) (.01) (.01)
President Involved −7.78** −7.88*** −7.98** −7.12* −7.98** −9.08** −1.94**
(2.91) (2.92) (2.91) (2.94) (2.92) (2.95) (.97)
Honeymoon 19.71*** 20.07*** 20.43*** 19.35*** 19.40*** 19.38*** 2.33
(2.98) (3.08) (3.01) (3.01) (2.97) (2.97) (1.96)
Inflation .03 .05 −.01 −.10 .06 −.05 .04
(.13) (.14) (.13) (.13) (.13) (.13) (.03)
Unemployment −.91 −1.03* −.93 −.45 −.96 −.88 .01
(.53) (.58) (.53) (.52) (.54) (.53) (.20)
Adjustment Policy −17.72*** −16.79*** −16.64*** −18.78*** −18.29*** −18.40*** −2.32
(2.84) (3.36) (2.96) (2.82) (2.79) (2.77) (1.81)
Approval (lag) .86***
Personality Trait 11.10*** 13.70** −6.54*** 4.59** 7.75*** 6.00*** 2.00*
(3.16) (5.84) (1.82) (1.57) (2.25) (1.72) (1.18)
First Stage: Instrumental Variable
Education .25***
#R2.45 .44 .45 .44 .45 .45 .93
Wald chi2 225.69 218.17 226.89 217.13 224.75 225.20 2751.76
Prob>chi2 .000 .000 .000 .000 .000 .000 .000
Observations 165 165 165 165 165 165 164
Note: Unit of analysis is president-month. #Corresponds to R2 for 3.7 and “within” R2 for the other models.
Standard errors in parentheses.
*p<.1; **p<.05; ***p<.01.
14 Arana Araya
account for potential heteroskedasticity, autocorrelation, or cross-sectional correlation due to the fact
that there are few presidents with many observations for each of them (Beck & Katz, 1995).
The results in models 3.1 and 3.2 are highly similar to the ones in model 2.2. This suggests that
the experts’ assessment of presidents performance did not systematically bias their evaluations and
that endogeneity is not a serious concern. Revealingly, all of the Big Five are associated with pres-
idential popularity, further supporting the argument that the presidents’ personality traits should be
included as independent variables. The results also show that Openness is the personality trait with
the strongest effect on the dependent variable: the magnitude of its effect (11.21% for each one-unit
increase, as shown in 2.2) is 37% higher than the effect of Agreeableness (7.75% for each one-unit
increase, as observable in model 3.5), the second trait with most impact on presidential popularity.
Finally, Openness remains statistically significant in the PCSE model (model 3.7).
Some caveats are in place. As of any reassessment, the exercise was restricted to the data used
in Pérez-Liñán (2007), and therefore the results do not allow us to make universal claims about the
relation between the personalities of presidents and their popularity. Furthermore, the exercise does
not rebut Pérez-Liñán’s theory and findings; only two models of his book were reassessed, and the
results do not contradict his findings. Nonetheless, the results do suggest that Pérez-Liñán’s study
would have provided a more thorough understanding of presidential popularity—and, by extension,
the impeachment processes—had the author considered treating the personalities of presidents as
independent variables.
This article proposed that presidential studies should embrace insights from differential psy-
chology to treat the personality traits of presidents as independent variables. I argued that doing so
would allow us to integrate president-centered and presidency-oriented hypotheses under a com-
mon research design, expanding our understanding of the true difference that presidents may make
across different situations. Moreover, it would improve the analytical leverage of studies that have
omitted the personalities of leaders as explanatory factors and, potentially, allow us to identify the
expected consequences of choosing certain individuals for office. That, in turn, would permit voters
and organizations to select as candidates individuals that more accurately represent them, facilitating
democratic representation and accountability. In support of the argument, the statistical analyses
conducted showed that reassessing a study about presidents was enriched when a measurement of
their openness to experience was included.
A research agenda treating the personalities of presidents as independent variables allows us
to answer pressing questions: in which situations are the personalities of presidents consequential
and in which ones do they have little explanatory power? Can the traits of powerful leaders explain
institutional transformations such as the relaxation of term limits that, in turn, change the constraints
that presidents face? Are the personal characteristic of presidents more relevant in nondemocratic
regimes, where leaders enjoy more leeway? Perhaps Thomas Carlyle exaggerated when he stated that
“the history of the world is but the biography of great men” (1841, p. 47). However, the influence that
presidents have in presidential systems certainly deserves a deeper examination of how their traits
may relate to relevant political outcomes.
The argument made in this article focused on presidents and personality psychology, but it is
associated with an older and broader question in human thought: what differences do leaders make?
As Zaccaro et al. claim: “The theme of individual differences that contribute to leadership is the
longest-standing research topic in the science of leadership” (2018, p. 1). All countries are led by a
small group of individuals. Therefore, students of all sorts of political elites—from majors to prime
ministers to revolutionary leaders and monarchs—would greatly benefit from the integration of in-
sights from personality psychology.
Personalities of Presidents
This project was funded by FONDECYT 3160357. The author thanks Alison Munden, Dov
Levin, Giorleny Altamirano, Daniel Silverman, John Chin, and Juan Pablo Luna for their thought-
ful comments to the first drafts of this manuscript. The author is also grateful to three anonymous
reviewers for helping him to substantially improve the quality of the manuscript. Correspondence
concerning this article should be addressed to Ignacio Arana Araya, Assistant Teaching Professor,
Institute for Politics and Strategy, Carnegie Mellon University, 117 Dunlap Street. Pittsburgh, PA
15214, USA. E-mail:
Allison, D. (2009). Fixed effects regression models. London, United Kingdom: SAGE.
Arana Araya, I. (2016a). ¿Cómo evaluar a los integrantes de la élite política? Una propuesta basada en los presidentes amer-
icanos. Política, 54(1), 219–254.
Arana Araya, I. (2016b). What drives evo’s attempts to remain in power? A psychological explanation. Bolivian Studies
Journal/Revista de Estudios Bolivianos, 22, 191–219.
Arana Araya, I. (2017). Presidential database of the Americas [v.2]. Unpublished Data Set, Carnegie Mellon University.
Arana Araya, I., & Guerrero Valencia, C. (2020). Executive-legislative relations: When do legislators trust the president? In M.
Alcántara Sáez, M. García Moreno, & C. Rivas Pérez (Eds.), Politics and political elites in Latin America, (pp.151–171).
Cham, Switzerland: Springer.
Ashcroft, A. (2016). Donald Trump: Narcissist, psychopath or representative of the people? Psychotherapy and Politics
International, 14(3), 217–222.
Barber, J. D. (1972). The presidential character: Predicting performance in the White House. Englewood Cliffs, NJ:
Beagrie, S. (2005). How to… excel at psychometric assessments. Personnel Today, 25(1). 25
Beck, N., & Katz, J. N. (1995). What to do (and not to do) with time-series cross-section data. American Political Science
Review, 89(3), 634–647.
Benet-Martinez, V., & John, O. (1998). Los Cinco Grandes across cultures and ethnic groups: Multitrait-multimethod analyses
of the Big Five in Spanish and English. Journal of Personality and Social Psychology, 75(3), 729–750.
Best, H. (2011). Does personality matter in politics? Personality factors as determinants of parliamentary recruitment and
policy preferences. Comparative Sociology, 10(6), 928–948.
Caprara, G. V., Schwartz, S., Capanna, C., Vecchione, M., & Barbaranelli, C. (2006). Personality and politics: Values, traits,
and political choice. Political Psychology, 27, 1–28.
Carlyle, T. (1841). On heroes, hero-worship, and the heroic in history. London, United Kingdom: James Fraser.
Cohen, J. E. (2018). Voters and presidential intelligence. Intelligence, 71, 54–65.
Connelly, B. S., & Ones, D. S. (2010). An other perspective on personality: Meta-analytic integration of observers’ accuracy
and predictive validity. Psychological Bulletin, 136(6), 1092–1122.
Corr, J., & Matthews, G. (Eds.). (2009). The Cambridge handbook of personality psychology. Cambridge, United Kingdom:
Cambridge University Press.
Corrales, J. (2016). Can anyone stop the president? Power asymmetries and term limits in Latin America, 1984–2016. Latin
American Politics and Society, 58(2), 3–25.
Corwin, E. S. (1940). The president, office and powers: History and analysis of practice and opinion. New York, NY: New
York University Press.
Costa, T., Jr., & McCrae, R. R. (1992). Normal personality assessment in clinical practice: The NEO Personality Inventory.
Psychological Assessment, 4(1), 5–13.
Costa, T., Jr., McCrae, R. R., Zonderman, A. B., Barbano, H. E., Lebowitz, B., & Larson, D. M. (1986). Cross-sectional stud-
ies of personality in a national sample: II. Stability in neuroticism, extraversion, and openness. Psychology and Aging,
1(2), 144.
Cronin, T. E., & Greenberg, S. D. (1969). The presidential advisory system. New York, NY: Harper and Row.
C-Span Survey of Presidential Leadership. (2000). Retrieved from http://www.ameri canpr eside y/histo rians/
overa ll.asp
16 Arana Araya
Davidson, J. R., Connor, K. M., & Swartz, M. (2006). Mental illness in US presidents between 1776 and 1974: A review of
biographical sources. The Journal of Nervous and Mental Disease, 194(1), 47–51.
De la Torre, C. (1999). Neopopulism in contemporary Ecuador: The case of Bucaram’s use of the mass media. International
Journal of Politics, Culture, and Society, 12(4), 555–571.
Deluga, R. J. (1997). Relationship among American presidential charismatic leadership, narcissism, and rated performance.
The Leadership Quarterly, 8(1), 49–65.
Deluga, R. J. (1998). American presidential proactivity, charismatic leadership, and rated performance. The Leadership
Quarterly, 9(3), 265–291.
Der Foo, M., Anger Elfenbein, H., Hoon Tan, H., & Chuan Aik, V. (2004). Emotional intelligence and negotiation: The tension
between creating and claiming value. International Journal of Conflict Management, 15(4), 411–429.
Dollinger, S. J., Leong, F. T., & Ulicni, S. K. (1996). On traits and values: With special reference to openness to experience.
Journal of Research in Personality, 30(1), 23–41.
Edwards, G. C., Kessel, J. H., & Rockman, B. A. (Eds.). (1993). Researching the presidency: Vital questions, new approaches.
Pittsburgh, PA: University of Pittsburgh.
Emrich, C. G., Brower, H. H., Feldman, J. M., & Garland, H. (2001). Images in words: Presidential rhetoric, charisma, and
greatness. Administrative Science Quarterly, 46(3), 527–557.
Eshbaugh-Soha, M. (2003). Presidential press conferences over time. American Journal of Political Science, 47(2), 348–353.
Flett, G. (2007). Personality theory and research: An international perspective. Toronto, ON: John Wiley & Sons.
Gilmour, J. B. (2002). Institutional and individual influences on the President’s veto. The Journal of Politics, 64(1), 198–218.
Greenstein, F. I. (2009). Inventing the job of president: Leadership style from George Washington to Andrew Jackson.
Princeton, NJ: Princeton University Press.
Hamilton, M. B. (2009). Online survey response rates and times: Background and guidance for industry. Cambridge, MA:
Tercent, Inc.
Hart, J. A. (1977). Cognitive maps of three Latin American policy makers. World Politics, 30(1), 115–140.
Heclo, H. (1977). Studying the presidency: A report to the Ford Foundation. New York, NY: Ford Foundation Press.
Hermann, M. G. (2003). Assessing leadership style: Trait analysis. In J. M. Post (Ed.), The psychological assessment of polit-
ical leaders: With profiles of Saddam Hussein and Bill Clinton, (pp.178–212). Ann Arbor, MI: University of Michigan
Hershman, D. J., & Lieb, J. (1994). A brotherhood of tyrants: Manic depression and absolute power. Amherst, NY: Prometheus
Holian, D. B., & Prysby, C. (2014). Candidate character traits in the 2012 presidential election. Presidential Studies Quarterly,
44(3), 484–505.
Holmes, J. E., & Elder jr, R.E. (1989). Our Best and Worst Presidents: Some Possible Reasons for Perceived Performance.
Presidential Studies Quarterly, 19, 529–557.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory: Versions 4a and 54. Technical Report, Institute
of Personality Assessment and Research.
Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and leadership: A qualitative and quantitative review.
Journal of Applied Psychology, 87(4), 765–780.
Kertzer, J. D., & Tingley, D. (2018). Political psychology in international relations: Beyond the paradigms. American Review
of Political Science, 21, 1–23.
King, A. (1975). Executives. In F. I. Greenstein & N. W. Polsby (Eds.), Handbook of political science (Vol. 5, (173–265).
Reading, MA: Addison-Wesley.
King, G. (1993). The methodology of presidential research. In G. Edwards, Ill, J. H. Kessel, & B. A. Rockman (Eds.),
Researching the presidency: Vital questions, new approaches (pp. 387–412). Pittsburgh, PA: University of Pittsburgh.
Koeneke, H. (2002). Rol, estilo presidencial y tolerancia cívica en la Venezuela de Hugo Chávez Frías. XI Jornadas
Venezolanas de Psicología Social. 1–16.
Koenig, L. W. (1964). The chief executive, New York: Harcourt, Brace & World.
Lee, B. X. (2017). The dangerous case of Donald Trump: 27 psychiatrists and mental health experts assess a president. New
York, NY: Thomas Dunne Books.
Levitsky, S., & Ziblatt, D. (2018). How democracies die. New York, NY: Crown.
Ligon, G. S., Harris, D. J., & Hunter, S. T. (2012). Quantifying leader lives: What historiometric approaches can tell us. The
Leadership Quarterly, 23(6), 1104–1133.
Lilienfeld, S. O., Waldman, I. D., Landfield, K., Watts, A. L., Rubenzer, S., & Faschingbauer, T. R. (2012). Fearless domi-
nance 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.
Personalities of Presidents
Lowi, T. J. (1986). The personal president: Power invested, promise unfulfilled. Ithaca, NY: Cornell University Press.
Matthews, G., Deary, I. J., & Whiteman, M. C. (2003). Personality traits. Cambridge, United Kingdom: Cambridge University
McCrae, R. R. (2009). The five-factor model of personality. In J. Corr & G. Matthews (Eds.), The Cambridge handbook of
personality psychology (pp. 148–161). Cambridge, United Kingdom: Cambridge University Press.
McCrae, R. R., & Costa, T., Jr. (1997). Personality trait structure as a human universal. American Psychologist, 52(5), 509–516.
McKie, K. (2019). Presidential term limit contravention: Abolish, extend, fail, or respect? Comparative Political Studies,
52(10), 1500–1534.
Meyer, M. A., & Booker, J. M. (2001). Eliciting and analyzing expert judgment: A practical guide 7, Philadelphia, PA: SIAM.
Moe, T. M. (1993). Presidents, Institutions, and Theory. In G. Edwards, Ill, J. H. Kessel, & B. A. Rockman (Eds.), Researching
the presidency: Vital questions, new approaches (pp. 337–385). Pittsburgh, PA: University of Pittsburgh.
Murray, R. K., & Blessing, T. H. (1983). The presidential performance study: A progress report. The Journal of American
History, 70(3), 535–555.
Negretto, G. L. (2013). Making constitutions: Presidents, parties, and institutional choice in Latin America, Cambridge:
Cambridge University Press.
Neuendorf, K. A. (2016). The content analysis guidebook (2nd ed.). Thousand Oaks, CA: Sage.
Neustadt, R. (1960). Presidential power: The politics of leadership. New York, NY: Wiley and Sons.
Oreg, S., & Berson, Y. (2015). Personality and charismatic leadership in context: The moderating role of situational stress.
Personnel Psychology, 68(1), 49–77.
Pérez-Liñán, A. (2007). Presidential impeachment and the new political instability in Latin America. Cambridge, United
Kingdom: Cambridge University Press.
Post, J. M. (Ed.). (2003). The psychological assessment of political leaders: With profiles of Saddam Hussein and Bill Clinton.
Ann Arbor, MI: University of Michigan Press.
Renshon, J. (2008). Stability and change in belief systems: The operational code of George W. Bush from governor to sec-
ond-term president. Journal of Conflict Resolution, 52(6), 820–849.
Ribeiro, E. A., & Borba, J. (2016). Personality, political attitudes and participation in protests: The direct and mediated effects
of psychological factors on political activism. Brazilian Political Science Review, 10(3).1–33.
Rubenzer, S. J., & Faschingbauer, T. R. (2004). Personality, character, and leadership in the White House: Psychologists
assess the presidents, Dulles, Virginia: Potomac Books Inc.
Rubenzer, S. J., Faschingbauer, T. R., & Ones, D. S. (2000). Assessing the US presidents using the revised NEO personality
inventory. Assessment, 7(4), 403–419.
Simonton, D. K. (1986). Presidential personality: Biographical use of the Gough Adjective Check List. Journal of Personality
and Social Psychology, 51(1), 149–160.
Simonton, D. K. (1987). Presidential inflexibility and veto behavior: Two individual-situational interactions. Journal of
Personality, 55(1), 1–18.
Simonton, D. K. (1988). Presidential style: Personality, biography, and performance. Journal of Personality and Social
Psychology, 55(6), 928–936.
Simonton, D. K. (2006). Presidential IQ, openness, intellectual brilliance, and leadership: Estimates and correlations for 42
US chief executives. Political Psychology, 27(4), 511–526.
Simonton, D. K. (2018a). Historiometric methods. In K. A. Ericsson, R. R. Hoffman, A. Kozbelt, & A. M. Williams (Eds.),
The Cambridge handbook of expertise and expert performance, (310–327). Cambridge, United Kingdom: Cambridge
University Press.
Simonton, D. K. (2018b). Intellectual brilliance and presidential performance: Why pure intelligence (or openness) doesn’t
suffice. Journal of Intelligence, 6(2), 18.
Song, A. V., & Simonton, D. K. (2007). Personality assessment at a distance. In R. W. Robins, R. C. Fraley, & R. F. Krueger
(Eds.), Handbook of research methods in personality psychology, (pp. 308–201). New York, NY: Guilford.
Steenbergen, M. R., & Marks, G. (2007). Evaluating expert judgments. European Journal of Political Research, 46(3),
Stevens, D., Bishin, B. G., & Barr, R. R. (2006). Authoritarian attitudes, democracy, and policy preferences among Latin
American elites. American Journal of Political Science, 50(3), 606–620.
Stock, J. H., & Yogo, M. (2005). Testing for weak instruments in linear IV regression. In D. W. K. Andrews & J. H. Stock
(Eds.), Identification and inference for econometric models: Essays in honor of Thomas Rothenberg (pp. 80–108). New
York, NY: Cambridge University Press.
Taranto, J., & Leo, L. (Eds.) (2004). Presidential leadership: Rating the best and the worst in the White House. New York,
NY: Wall Street Journal Books.
18 Arana Araya
Tetlock, E. (1981). Pre- to postelection shifts in presidential rhetoric: Impression management or cognitive adjustment.
Journal of Personality and Social Psychology, 41(2), 207–212.
Thoemmes, F. J., & Conway III, L. G. (2007). Integrative complexity of 41 US presidents. Political Psychology, 28(2),
Walker, S.G. (1990). The evolution of operational code analysis. Political Psychology, 11(2), 403–418.
Waite, R. (1977). The psychopathic god: Adolf Hitler. New York, NY: Basic Books.
Wayne, S. J. (1983). An introduction to research on the presidency. In G. C. Edwards, Ill & S. J. Wayne (Eds.), Studying the
presidency, (pp. 3–14). Knoxville, TN: University of Tennessee Press.
Weiner, I. B., & Greene, R. L. (2017). Handbook of personality assessment, (25–47). New York, NY: John Wiley & Sons.
Winter, D. G. (2002). Motivation and political leadership. In L. O. Valenty & O. Feldman (Eds.), Political leadership for the
new century: Personality and behavior among American leaders. (pp. 25–47). Westport, CT: Praeger.
Winter, D. G., & Stewart, A. J. (1977). Content analysis as a technique for assessing political leaders. In M. G. Hermann (Ed.),
The Psychological Examination of Political Leaders (pp. 27–61).New York: Free Press.
Zaccaro, S. J., Green, J. P., Dubrow, S., & Kolze, M. (2018). Leader individual differences, situational parameters, and lead-
ership outcomes: A comprehensive review and integration. The Leadership Quarterly, 29(1), 2–43.
Supporting Information
Additional supporting information may be found in the online version of this article at the publisher’s
web site:
Appendix S1. Complementary Figure and Tables
Figure S1. Presidential approval across presidents in the sample
Table S1.1. Variables in Pérez-Liñán’s (2007) Models of Presidential Approval
Table S1.2. Presidents of the Americas Assessed in Expert Survey
Full-text available
Thirty-one presidents from every Latin American country-excluding Mexico-who were governing from 1945-2012 tried forty times to change the constitution of their countries to overstay in office. These attempts often caused severe political instability. Current explanations of the variability of term limits have centered on the context in which presidents govern despite the protagonism of the leaders in the constitutional changes. I argue that the personality traits of presidents are an important driver of their overreaching behavior. Centered on the paradigm of the "Big Five," I propose hypotheses about a causal relationship between each of the five core personality factors-openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism-and the presidents' attempts to alter their term limits. To test the theory, I use data about presidents who governed from 1945-2012. The results of a discrete-time duration analysis show that three of the Big Five are associated to the likelihood of observing a president changing term limits. I conclude by discussing how this research agenda should be extended to uncover how the uniqueness of the leaders explains relevant outcomes in executive politics.
Full-text available
There is a growing scholarly consensus that overreaching heads of government are subverting democracies across the globe. However, the characteristics of these leaders remain unclear. This article examines a type of overreaching presidential behavior that has been commonplace in Latin America: between 1945 and 2012, 25 presidents from 14 countries tried to change their respective constitutions to increase their powers. Building on personality research and semi-structured interviews conducted with former presidents, this article proposes that risk-taking and assertive leaders are more likely to try to increase their powers. Using a novel database, I conduct discrete-time duration models to test the hypotheses on the presidents that governed from 1945-2012. The results demonstrate that the personalities of presidents are a strong force behind their attempts to consolidate their authority. These findings challenge current approaches in presidential studies and have implications for the study of all types of political elites.
Full-text available
The pandemic posed a dramatic challenge to practically all heads of government: the measures they took had an enormous impact on the lives and livelihoods of their citizens. Growing conventional wisdom proposes that populists failed to deal with the pandemic while women dignitaries thrived. However, these accounts are merely supported by anecdotal evidence. I conduct a systematic evaluation of the performance of world leaders during the pandemic, taking their speed of reaction as an indicator of achievement. I argue that populists should not have reacted slower than their counterparts because they faced contradictory motivations to handle the pandemic. I also propose that women and leaders with more formal and informal political experience reacted faster. The former because glass ceilings force women to endure a more difficult selection process to become heads of government, while the latter should have reacted faster given their superior expertise. Using a unique database with biographical information of 166 heads of government, a cross-sectional survival analysis reveals that populists, women, and leaders from political families reacted faster than their counterparts to the public health crisis.
Full-text available
Research on interbranch conflict has mostly focused on the effect that the institutional and political context has on executive-legislative relations. Little attention has been paid to the interpersonal dimension in interbranch relations, despite being characterized by intensive interactions among political elites. Arguably, the trust that legislators have in the incumbent signals their willingness to negotiate and reach agreements with the head of government. In this chapter, we begin to address the factors that explain the trust legislators have in presidents. We use data from two unique databases, the Presidential Database of the Americas and the Parliamentary Elites of Latin America Project, to examine legislative trust in presidents from 18 countries for the 1994-2014 period. We find that factors that capture the institutional and political environment as well as variables that measure psychological and non-psychological characteristics of the leaders are relevant to understanding the trust that legislators have in heads of government.
Full-text available
In recent years it has become popular on the internet to debate the IQ of the incumbent president of the United States. Yet, these controversies (and hoaxes) presume that IQ has some relevance to understanding the president's actual performance as the nation's leader. This assumption is examined by reviewing the empirical research on the intelligence-performance association in political leadership, with a special focus on U.S. presidents. The review starts by discussing at-a-distance assessment techniques, a method that has yielded reliable and valid measures of IQ, Intellectual Brilliance, and Openness to Experience; three correlated even if separable concepts. The discussion then turns to the reliable and valid measurement of presidential performance-or "greatness"-via successive surveys of hundreds of experts. These two lines of research then converged on the emergence of a six-predictor equation, in which Intellectual Brilliance plays a major role, to the exclusion of both IQ and Openness. The greatest presidents are those who feature wide interests, and who are artistic, inventive, curious, intelligent, sophisticated, complicated, insightful, wise, and idealistic (but who are far from being either dull or commonplace). These are the personal traits we should look for in the person who occupies the nation's highest office if we seek someone most likely to solve the urgent problems of today and tomorrow.
Full-text available
Este artículo repasa críticamente el estudio sobre la élite política, incluyendo la evolución histórica de su significado, rol, composición, independencia y las opciones para analizar a sus miembros. Argumenta que para avanzar en el estudio de los integrantes de la élite es necesario examinar sus diferencias individuales. A modo de ejercicio, se abordan algunas diferencias individuales de quienes están en la cúspide de la élite política en sistemas presidenciales: los presidentes. El análisis muestra que, tomados como grupo, los presidentes del continente tienen un origen socioeconómico medianamente acomodado, al menos un tercio ha sido abogado o participado en fuerzas de seguridad, y que tienden a puntuar bajo en amabilidad y neuroticismo, moderadamente alto en extraversión y apertura a la experiencia, y alto en responsabilidad. Este ejercicio sugiere una agenda de investigación que puede extenderse a otros integrantes de la élite.
Since presidential term limits were (re)adopted by many states during the third wave of democratization, 221 presidents across Latin America, Africa, the Middle East, and Asia have reached the end of their term(s) in office. Of these, 30% have attempted to contravene term limits, resulting in either full abolition, one-term extensions, or failure. What explains these divergent trajectories? I argue that trends in electoral competition over time best predict term limit outcomes, with noncompetitive elections permitting full abolition, less competitive elections allowing for one-term extensions, and competitive elections leading to failed bids. This is because electoral trends provide informational cues to the president’s co-partisan legislators and constitutional court judges (the actors who ultimately rule on constitutional term limit amendments) about the cost/benefit analysis that voting to uphold or repeal term limits would have on their own political survival. These findings suggest a linkage between political uncertainty and constitutional stability more generally.
Can voters accurately discern a president's intelligence and do the intelligence perceptions of voters affect their vote choice in presidential elections? Presidents scored as more intelligent by experts also score more highly on the presidential greatness rankings. Study 1 uses expert ratings of presidential intelligence to assess whether voter perceptions of presidential intelligence are accurate, using American National Election (ANES) survey data from 1980 through 2008 (n= 13276). Voters are moderately to weakly accurate in perceiving presidential intelligence (r = 0.24, p < 0.001). Study 2 asks whether perceptions of presidential candidate intelligence affects the vote choice, also using ANES survey data from 1980 through 2008 (n = 5965). There is approximately a 10% probability voting advantage for the maximum difference in perceptions of candidate intelligence, with controls for other variables that affect the vote choice. Results suggest that voters are moderately to weakly accurate in assessing presidential intelligence and prefer the more intelligent candidate in presidential elections.
Political psychology in international relations (IR) has undergone a dramatic transformation in the past two decades, mirroring the broader changes occurring in IR itself. This review examines the current state of the field.Webegin by offering a data-driven snapshot analyzing four years of manuscript classifications at a major IR journal to characterize the questions that IR scholars engaged in psychological research are and are not investigating. We then emphasize six developments in particular, both present-day growth areas (an increased interest in emotions and hot cognition, the rise of more psychologically informed work on public opinion, a nascent research tradition we call the first image reversed, and the rise of neurobiological and evolutionary approaches) and calls for additional scholarship (better integration of the study of mass and elite political behavior and more psychological work in international political economy). Together, these developments constitute some of the directions in which we see the next generation of scholarship heading. Expected final online publication date for the Annual Review of Political Science Volume 21 is May 11, 2018. Please see for revised estimates.
In this article, we provide a wide-ranging review of recent research on leader individual differences. The review focuses specifically on the explosion of such research in the last decade. The first purpose of this review is to summarize and integrate various conceptual frameworks describing how leader attributes influence leader emergence and leader effectiveness. The second purpose is to provide a comprehensive review of empirical research on this relationship. Also, most prior reviews primarily examined leader personality traits; this review includes a broader array of leader attributes, including cognitive capacities, personality, motives and values, social skills, and knowledge and expertise. The final broad purpose of this paper is to review and integrate situational and contextual parameters into our conceptual framing of leader individual differences. Few, if any, prior reviews have systematically accounted for the critical role of such parameters in cuing, activating, or delimiting the effects of particular leader attributes. We do so in this article.