Annotated Bibliography and Outline: Leadership

Annotated Bibliography and Outline: Leadership

Provide an annotated bibliography (750-1,000 words total) of the articles listed above. Including the following for each article: 1.The article citation and persistent link. These are provided above for you to paste into the assignment and are not included in the total word count. 2.A written summary of the key concept(s) of the article. Why was the study done? What was the population studied? What did the researcher(s) conclude? What other information about this study do you believe is unique or important to recall? Are there specific statements made by the author that you wish to retain?

 

PERSONALITY PROCESSES AND INDIVIDUAL DIFFERENCES Fearless Dominance and the U.S. Presidency: Implications of Psychopathic Personality Traits for Successful and Unsuccessful Political Leadership Scott O. Lilienfeld, Irwin D. Waldman, and Kristin Landfield Emory University Ashley L. Watts University of Georgia Steven Rubenzer Houston, Texas Thomas R. Faschingbauer Foundation for the Study of Personality in History, Houston, Texas Although psychopathic personality (psychopathy) is marked largely by maladaptive traits (e.g., poor impulse control, lack of guilt), some authors have conjectured that some features of this condition (e.g., fearlessness, interpersonal dominance) are adaptive in certain occupations, including leadership positions. We tested this hypothesis in the 42 U.S. presidents up to and including George W. Bush using (a) psychopathy trait estimates derived from personality data completed by historical experts on each president, (b) independent historical surveys of presidential leadership, and (c) largely or entirely objective indicators of presidential performance. Fearless Dominance, which reflects the boldness associated with psychopathy, was associated with better rated presidential performance, leadership, persuasiveness, crisis management, Congressional relations, and allied variables; it was also associated with several largely or entirely objective indicators of presidential performance, such as initiating new projects and being viewed as a world figure. Most of these associations survived statistical control for covariates, including intellectual brilliance, five factor model personality traits, and need for power. In contrast, Impulsive Antisociality and related traits of psychopathy were generally unassociated with rated presidential performance, although they were linked to some largely or entirely objective indicators of negative job performance, including Congressional impeachment resolutions, tolerating unethical behavior in subordinates, and negative character. These findings indicate that the boldness associated with psychopathy is an important but heretofore neglected predictor of presidential performance, and suggest that certain features of psychopathy are tied to successful interpersonal behavior. Keywords: psychopathy, antisocial behavior, leadership, politics, personality Psychopathic personality (psychopathy) is a constellation of personality traits encompassing superficial charm, egocentricity, dishonesty, guiltlessness, callousness, risk taking, poor impulse control (Cleckley, 1941/1988; Hare, 2003), and, according to many authors (Fowles & Dindo, 2009; Lykken, 1995; Patrick, 2006), fearlessness, social dominance, and immunity to anxiety. In contrast to the Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM–IV–TR; American Psychiatric Association, 2000), diagnosis of antisocial personality disorder (ASPD), which is primarily a behavioral condition that emphasizes a long-standing history of antisocial and criminal behavior, psychopathy is primarily a dispositional condition that emphasizes personality traits. Nevertheless, measures of these two conditions tend to be at least moderately correlated (Lilienfeld, 1994). Factor analyses of the most extensively validated measure of psychopathy, the Psychopathy Checklist–Revised (PCL-R; Hare, 2003), have often revealed two broad and moderately correlated dimensions. The first dimension (Factor 1) assesses the core interpersonal and affective features of psychopathy (e.g., guiltlessness, narcissism, glibness), whereas the second dimension (Factor 2) assesses an impulsive and antisocial lifestyle that is closely associated with ASPD (Harpur, Hare, & Hakstian, 1989; but see Cooke & Michie, 2001, and Hare, 2003, for alternative factor This article was published Online First July 23, 2012. Scott O. Lilienfeld, Irwin D. Waldman, and Kristin Landfield, Department of Psychology, Emory University; Ashley L. Watts, Department of Psychology, University of Georgia; Steven Rubenzer, Houston, Texas; Thomas R. Faschingbauer, Foundation for the Study of Personality in History, Houston, Texas. We thank Joanna Berg, Rachel Ammirati, David Molho, Gabriella Rich, Zack Babin, Marie King, and Barbara Greenspan for their helpful comments on previous drafts of this manuscript; Joshua Miller for his statistical assistance; Alan Abramowitz for his helpful advice; and Caroline Hennigar and Alyssa Redmon for their valuable assistance with data entry and library research. Correspondence concerning this article should be addressed to Scott O. Lilienfeld, Room 473, Psychology and Interdisciplinary Sciences Building, Emory University, 36 Eagle Row, Atlanta, GA 30322. E-mail: slilien@emory.edu Journal of Personality and Social Psychology, 2012, Vol. 103, No. 3, 489 –505 © 2012 American Psychological Association 0022-3514/12/$12.00 DOI: 10.1037/a0029392 489 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. solutions). Although the PCL-R is a semistructured interview that incorporates file information, its two major dimensions can be closely approximated by scores on normal range personality dimensions, such as those derived from the five-factor model (FFM) of personality. PCL-R Factor 1 is associated primarily with low scores on FFM Agreeableness, whereas PCL-R Factor 2 is associated primarily with low scores on both FFM Agreeableness and Conscientiousness (Miller, Lynam, Widiger, & Leukefeld, 2001). Most research demonstrates that psychopathy and its constituent traits are underpinned by dimensions rather than taxa (natural categories; see Edens, Marcus, Lilienfeld, & Poythress, 2006), offering empirical support for recent efforts to conceptualize and assess this condition within a general dimensional model of personality structure. Most research on the behavioral manifestations of psychopathy has focused on its relations with antisocial, criminal, and otherwise unsuccessful actions. Studies demonstrate that psychopathy is a risk factor for criminality and violent recidivism among prison inmates (Porter & Woodworth, 2006; Salekin, Rogers, & Sewell, 1996) as well as cheating among college students (Williams, Nathanson, & Paulhus, 2010). In addition, some authors have argued that psychopathy is associated with malignant workplace behavior. Babiak and Hare (2006) referred to psychopaths in business settings as “snakes in suits” and suggested that their propensity toward dishonesty and manipulativeness makes them destructive coworkers and bosses (see also Boddy, 2006; Heinze, Allen, Magai, & Ritzler, 2010). Despite the lengthy research tradition linking psychopathy to unsuccessful behavior, a consistent strand of clinical lore has tied psychopathy, or at least certain features of it, to socially successful behavior across a variety of domains, including the business world, politics, and everyday life (Lilienfeld, 1998). In his classic writings, Cleckley (1941/1988) referred to individuals with marked psychopathic traits whose “outward appearance may include business or professional careers that continue in a sense successful, and which are truly successful when measured by financial reward or even by the casual observer’s opinion of real accomplishment” (p. 191). Extending these observations, Lykken (1982) referred to psychopaths and heroes as “twigs from the same branch” (p. 22) and conjectured that the fearlessness associated with psychopathy can predispose to heroic behaviors. Other authors have raised the possibility of “subclinical” (Widom, 1977) or “successful” (Hall & Benning, 2006; Mullins-Sweatt, Glover, Miller, Derefinko, & Widiger, 2010) psychopaths, individuals with pronounced psychopathic traits who function effectively in circumscribed “adaptive niches” of society, such as politics, business, law enforcement, and high-risk sports. In one of the few studies to address this issue empirically, Babiak, Neumann, and Hare (2010) examined a sample of 203 corporate professionals and found that scores on the PCL-R and its component factors were associated not only with a more problematic management style and with being a poor team player but also with superior communication skills, creativity, and strategic thinking. These important results raise the possibility that psychopathy, or at least some features of it, are associated with certain aspects of adaptive functioning in workplace settings, although they may also be associated with certain aspects of maladaptive functioning. Nevertheless, because the PCL-R ratings in this study were conducted by a single individual who was not blind to other information about participants, including information potentially relevant to criterion ratings, these results should be viewed as preliminary. Still others have speculated that some psychopathic traits, such as interpersonal dominance, persuasiveness, and venturesomeness, may be conducive to acquiring positions of political power and to successful leadership (Hogan, Raskin, & Fazzini, 1990; Lobacweski, 2007). Indeed, Lykken (1995) speculated that British Prime Minister Winston Churchill and U.S. president Lyndon Baines Johnson possessed certain personality features of psychopathy: They started off life as “daring, adventurous, and unconventional youngsters who began playing by their own rules” (p. 116) but later managed to parlay these traits into political success. Nevertheless, the successful manifestations of psychopathy remain largely in the realm of clinical conjecture. Moreover, with the exception of the study by Babiak et al. (2010), the scattered research in this domain (e.g., Ishakawa, Raine, Lencz, Bihrle, & LaCasse, 2001; Widom, 1977) has focused almost exclusively on psychopathic individuals who have engaged in minimal antisocial behavior or managed to escape detection by the legal system, rather than those who are clearly successful from an interpersonal or societal standpoint (Hall & Benning, 2006). Recent work on a widely used and well-validated self-report psychopathy measure, the Psychopathic Personality Inventory (PPI; Lilienfeld & Andrews, 1996), may shed light on this issue. Exploratory factor analyses of the PPI (Benning, Patrick, Hicks, Blonigen, & Krueger, 2003) in community samples have identified two largely uncorrelated higher order dimensions, Fearless Dominance (FD) and Impulsive Antisociality1 (IA; but see Neumann, Malterer, & Newman, 2008, for an alternative factor structure of the PPI). FD, which assesses what Patrick, Fowles, and Krueger (2009) term “boldness,” comprises such traits as social dominance, charm, physical fearlessness, and immunity to anxiety; IA comprises such traits as egocentricity, manipulativeness, poor impulse control, rebelliousness, and tendency to externalize blame. Although these two factors bear some similarities to the two major PCL-R factors, they are not isomorphic with them empirically or conceptually. In particular, although IA and PCL-R Factor 2 are moderately to highly correlated, FD and PCL-R Factor 1 are only weakly correlated (Malterer, Lilienfeld, Newman, & Neumann, 2010), largely because FD assesses a more psychologically adaptive set of traits than does PCL-R Factor 1 (Patrick, 2006). Several studies have demonstrated that the boldness assessed by FD is associated with healthy psychological adjustment—and may reflect many of the traits commonly attributed to successful psychopathy—whereas IA is associated with psychological maladjustment. Offering provisional corroboration for Lykken’s (1982) conjecture regarding fearlessness and heroism, Patrick, Edens, Poythress, Lilienfeld, and Benning (2006) found that in a sample of 96 prisoners, FD scores derived from the PPI were significantly and positively associated with self-reported heroic behaviors (e.g., breaking up fights in public, helping stranded motorists), whereas IA scores were significantly and negatively associated with these behaviors. In addition, PPI-derived FD is negatively correlated 1 In the revised version of the PPI (Lilienfeld & Widows, 2005), this dimension is termed Self-Centered Impulsivity. Nevertheless, we use the term Impulsive Antisociality here to retain continuity with most of the extant literature (e.g., Benning et al., 2003). 490 LILIENFELD ET AL. This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. with measures of Axis I psychopathology, such as anxiety, depressive, and somatoform symptoms, as well as suicide attempts, whereas IA is positively associated with these indices (Benning et al., 2003; Douglas, Lilienfeld, Skeem, Edens, Poythress, & Patrick, 2008; Patrick et al., 2006). These findings are consistent with a “dual-process model” (Fowles & Dindo, 2009; see also Patrick et al., 2009, for an extended “triarchic model”) that conceptualizes psychopathy as the joint outcome of two separable etiological processes: (a) a bold temperament marked by largely adaptive functioning, assessed by FD and, to a substantially lesser extent, PCL-R Factor 1 and (2) a disposition toward disinhibition and externalizing behavior marked by largely maladaptive functioning, assessed by IA and PCL-R Factor 2. Nevertheless, the differential associations of these two components of psychopathy with both successful and unsuccessful interpersonal functioning, including job performance and leadership, have yet to be examined empirically. Patrick et al. (2009) conjectured that the boldness assessed by FD may be especially helpful in “the identification of individuals with psychopathic tendencies who ascend to positions of leadership and influence in society” (p. 925), but this intriguing hypothesis has yet to be put to an empirical test. In this study, we examined the implications of psychopathic personality traits for job performance and leadership in a remarkable sample of individuals whose successful and unsuccessful behaviors are a matter of well-documented public record: the 42 U.S. presidents up to and including George W. Bush. Inspired by the pioneering research of Simonton (1986, 1987) on presidential personality, Rubenzer, Faschingbauer, and Ones (2000) found that some personality traits, most notably high levels of openness to experience (see also Simonton, 2006), extraversion, conscientiousness, and perhaps low levels of agreeableness, are modestly correlated with independently rated job performance among the U.S. presidents. Nevertheless, no study has examined the relation of psychopathic personality traits to leadership and job performance among the U.S. presidents. We hypothesized that certain features of psychopathy, especially those assessed by FD, would be associated with successful functioning, including overall presidential leadership effectiveness, but that other features of psychopathy, especially those assessed by IA and proxies of PCL-R Factor 2, would be associated with unsuccessful functioning, including poor presidential job performance, negative personal character and integrity, and ethical misbehavior. To test these hypotheses, we drew on an existing data set of personality items obtained from biographers and experts on each president (Rubenzer & Faschingbauer, 2004) and extracted estimates of psychopathy factors based on empirically established equations from the published literature. We then correlated these psychopathy scores with (a) indices from several recent (2008 – 2011) and largely and in some cases entirely independent panels of eminent historians who had rated each president on dimensions relevant to work performance and leadership, including overall job effectiveness, leadership ability, public persuasiveness, crisis management, vision, and domestic and foreign policy accomplishments; (b) an empirically derived composite developed by Simonton (1987) of six largely or entirely objective indices of presidential greatness, including war heroism, number of years served, and assassination; and (c) several other largely or entirely objective indicators of both presidential success and failure, including reelection, introduction of legislation and programs, Congressional impeachment resolutions, and rated negative presidential character (as assessed by largely objective behaviors indicative of dishonesty and unreliability). By examining largely or entirely objective indicators, we addressed the criticism that any associations between psychopathy traits and rated presidential performance are merely a function of shared subjective impressions of the presidents by different raters. We also evaluated the specificity of these findings to psychopathic personality traits, especially FD, per se. In particular, we examined the incremental validity of a number of theoretically relevant variables above and beyond FD in an effort to rule out rival hypotheses concerning the potential linkages between FD and presidential performance. In this respect, we adopted a “destructive testing” approach (see C. A. Anderson & Anderson, 1996) in an effort to ascertain how well the relations between FD and presidential performance survive covariance adjustments from “competitor” variables that provide alternative explanations. Specifically, because it is unclear whether personality traits contribute to the prediction of presidential performance above and beyond intelligence, which is an established predictor of such performance (Simonton, 2006), we examined the incremental validity of psychopathic personality traits beyond established estimates of each president’s intelligence. In addition, we examined the incremental validity of psychopathic personality traits above and beyond FFM traits, especially extraversion and openness to experience, which are positively associated with FD (Lilienfeld & Widows, 2005) as well as traits of ASPD, which as noted earlier overlap with those of psychopathy. We also examined the incremental validity of FD above and beyond rated need for power, which has clear-cut conceptual relations to interpersonal dominance and perhaps the FD dimension of psychopathy. As Winter (2005) observed, “power-motivated presidents. . .invest a great deal of energy in the job, and they enjoy it” (p. 561). Need for power has been demonstrated to be a robust predictor of presidential success (Winter, 2005). Finally, we examined the incremental validity of FD for presidential performance above and beyond Simonton’s (1987) six-element equation of largely or entirely objective historical indicators. As Simonton (2008) observed, multiple empirical efforts have failed to unearth any consistent indicators that predict presidential greatness above and beyond this equation. This lattermost incremental validity analysis provides an especially stringent test of the unique contribution of psychopathic personality traits to presidential performance. Method Raters Raters of presidents’ personality traits in this study were 121 experts recruited by Rubenzer and Fashingbauer (2004) to evaluate the personality of the 42 U.S. presidents up to and including George W. Bush; Barack Obama was not included because of the unavailability of FFM data on him from presidential experts (although there were 43 presidencies up to and including George W. Bush, there were only 42 presidents, as Grover Cleveland was elected president twice in nonconsecutive terms). Importantly, these experts were asked to rate their target president’s preoffice (see the Procedure section) personality traits using well-validated PSYCHOPATHY AND U.S. PRESIDENTS 491 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. personality measures (see the Measures of Personality, Psychopathy, and Covariates section). Because some raters completed ratings on more than one president, the total number of ratings was 177. These experts were American biographers, journalists, and scholars who are established authorities on one or a few of U.S. presidents. They had authored published biographies on each president or had been nominated by other presidential experts as particularly well informed regarding a given president. The number of expert raters per president ranged from zero to 13, with a mean of 4.2 (SD 2.9; Rubenzer et al., 2000). Measures of Personality, Psychopathy, and Covariates Revised NEO Personality Inventory (NEO PI-R) Form R. The NEO PI-R is a 240-item questionnaire that assesses the five major dimensions of personality (Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness) from the FFM (Costa & McCrae, 1992). Nested within each of the five domains are six facet scales, each containing eight items cast in nontechnical language and endorsed on a 5-point Likert-type scale. Support for the NEO PI-R’s construct validity is extensive at both the domain and facet levels (Costa & McCrae, 1992; Lynam & Widiger, 2001). As discussed in the section below, scores on all four psychopathy indices were derived from ratings on the NEO PI-R. In this study, raters (121 presidential experts; see the Raters section) completed Form R, an observer report version of the NEO PI-R “designed to be completed by family member, friend, acquaintance— or anyone who knows the person well” (Rubenzer & Faschingbauer, 2004, p. 5). In this sample, the internal consistencies (Cronbach’s alphas) of the five NEO PI-R domain scales ranged from .91 to .94. FFM-derived prototypes of psychopathy factors and ASPD. Using a rational/theoretical approach, Derefinko and Lynam (2006; see also Widiger & Lynam, 1998) mapped the 30 facets of the FFM onto the two major factors of the PCL-R. As noted earlier, PCL-R Factor 1 assesses the core interpersonal and affective features of psychopathy, whereas PCL-R Factor 2 assesses an antisocial and impulsive lifestyle. The scores on FFM Factors 1 and 2 (which parallel the corresponding two factors of the PCL-R) are weighted composites of several of the FFM facets, namely, those deemed relevant to psychopathy. For example, FFM Factor 1 is a weighted composite of FFM facets from the domains of neuroticism, extraversion, agreeableness, and conscientiousness, all reversed in scoring (see Derefinko & Lynam, 2006, Table 1, p. 265). These FFM factor scores display good validity; for example, both correlate highly (rs between .5 and .6) with total scores on the PPI and the Self-Report Psychopathy Scale (SRP; Hare, Harpur, & Hemphill, 1989) and exhibit significant positive correlations with their respective Factor 1 and Factor 2 scores on the PPI and SRP (Derefinko & Lynam, 2006). To assess ASPD, scores on the prototype developed by Miller et al. (2001) were used. These authors constructed an expertgenerated FFM prototype of psychopathy and the 10 DSM–IV–TR personality disorders by asking experts to rate the prototypical expression of each personality disorder on a 1–5 scale using the 30 facets of the NEO PI-R. Any FFM facet with a mean lower than 2 or higher than 4 was included in each disorder’s prototype. Scores that most closely match the expert-generated psychopathy prototype correlate significantly and positively with several laboratory tasks theoretically relevant to psychopathy (e.g., measures of temporal discounting and proactive aggression) and self-reported aggression (Derefinko & Lynam, 2006; Miller & Lynam, 2003). In this study, expert-generated psychopathy FFM prototypes of Factors 1 and 2 were used, which parallel the two broad factors of the PCL-R, as well as the FFM prototype for ASPD (see Lynam & Widiger, 2001). Factor estimates of FD and IA. To extract measures of FD and IA, we relied on regression-based formulas developed by Ross, Benning, Patrick, Thompson, and Thurston (2009, p. 80), which use the 30 NEO PI-R facets of the FFM to estimate scores on these two dimensions, heretofore referred to as FFM-FD and FFM-IA.2 Ross et al. found that these regression formulas, after double cross-validation within their sample, accounted for between 68% and 79% of the variance in FD and IA scores derived from the PPI. Intellectual brilliance. Intellectual brilliance estimates for each president were drawn from the work of Simonton (1986, 2006), who derived a measure of Intellectual Brilliance from an exploratory factor analysis of adjectives from the Gough Adjective Checklist (Gough & Heilbrun, 1965) completed by multiple independent judges who rated the presidents. Using scores on FFM openness to experience (which tends to be moderately correlated with measured intelligence), Simonton (2004) later used missingdata iterative methods to extrapolate Intellectual Brilliance scores for the presidents from Ronald Reagan onward. The Intellectual Brilliance measure consists of such adjectives as intelligent, wise, complicated, and insightful, and correlates highly with other estimates of the U.S presidents’ intelligence derived from biographical information (Simonton, 2006). Need for power. Ratings of power needs were derived from Winter (1987; see also Winter, 1973, 1983), who examined inaugural addresses from American presidents (available before 1981). These speeches were coded by two raters, who demonstrated category agreement over .85 on power imagery. Disagreements between raters were deliberated upon until resolved. Raw scores, used in the analyses here, were defined in terms of power images per 1,000 words. Outcome Measures of Presidential Performance Presidential performance surveys. To assess outcome variables relevant to presidential performance, we relied primarily on data from two recent, large, and widely publicized American surveys of presidential historians. First, data were used from a 2009 C-SPAN poll of 62 identified presidential historians who rated the presidents on 10 continuous dimensions of job performance (see http://legacy.c-span.org/Content/PDF/CSPANpresidentialsurveyPR021509.pdf). Fifty-four of these 62 historians were independent of those who rated the presidents on 2 The PPI also contains a subscale, Coldheartedness, that does not load highly on either FD or IA and hence is excluded from computation of these two factors. Analyses of FFM-estimated Coldheartedness did not yield significant associations with any of the primary presidential poll variables examined here with one exception: PPI Coldheartedness was significantly and negatively associated with C-SPAN Poll Pursuit of Equal Justice (2 5.42, p .020). In addition, Coldheartedness was significantly and negatively associated with Siena College Poll Ability to Compromise (see Footnote 3) (2 4.93, p .026). 492 LILIENFELD ET AL. This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. the NEO-PI-R and other personality items. The 10 dimensions of job performance were overall job performance, public persuasiveness, handling of crises, moral authority, economic management, international relations, administrative skill, Congressional relations, setting of an agenda, and pursuit of equal justice (pairwise rs across these dimensions ranged from .46 to .96, all ps .001). Second, data were used from a 2010 Siena College survey of 238 anonymous presidential historians who ranked the presidents on 21 dimensions of job performance (see http://www.siena.edu/ uploadedfiles/home/parents_and_community/community_page/ sri/independent_research/Presidents%20Release_2010_final.pdf). For the analyses reported here, we focused on 13 Siena College survey variables for which we had clear-cut predictions: overall ranking, overall ability, leadership ability, party leadership, integrity, executive ability, communication ability, domestic accomplishments, foreign policy accomplishments, handling of the economy, relationship with Congress, willingness to take risks, and avoiding crucial mistakes (pairwise Spearman rs across these rankings ranged from .18 to .97, all ps .05). To facilitate comparisons between the two surveys, the Siena College rankings were reversed in scoring for the analyses reported here so that higher ranks correspond to superior-rated job performance. As a third indicator, psychopathy scores with a composite measure of presidential greatness derived from the work of Simonton (2006, p. 515) were correlated. This greatness measure is a sum of standardized (z-scored) results from 12 independent surveys of overall presidential performance. Research demonstrates that independent surveys of presidential performance taken across the decades yield similar results, with correlations of overall rankings typically in the r .9 range or above (Simonton, 2006). As a final check on the findings from the C-SPAN and Siena College surveys of presidential performance and Simonton composite measure of presidential greatness, data from two additional recent smaller surveys of U.S. presidential performance from the United Kingdom were examined (see http://americas.sas.ac.uk/research/ survey/index.html). These two surveys have two major advantages: (a) The individuals who completed these polls are entirely independent of those who completed the NEO-PI-R and other personality measures on the presidents, and (b) they do not derive from U.S. historians, and hence offer a largely independent international test of the association between psychopathic personality traits and presidential performance. As a consequence, they should be relatively free of biases shared exclusively by U.S. historians. The first U.K. poll was a 2008 survey conducted by the Times of London that asked eight premier political and international reporters to rank the U.S. presidents in terms of overall quality (The Times of London, 2008). The second U.K. poll was the United States Presidency Centre (USPC) Survey conducted by the Institute for the Study of the Americas (2011) at the University of London. The raters in this survey were 47 U.K. scholars who were established experts in U.S. presidential and political history. They were asked to rate the U.S. presidents on five dimensions: vision/setting of an agenda (heretofore referred to as vision), domestic leadership, foreign policy leadership, moral authority, and long-term positive legacy. In addition, the poll yielded an overall ranking of the presidents in terms of quality. Two presidents (William Henry Harrison and James Garfield) were excluded from this survey because of their brief presidencies. Again, the ranked scores on these two surveys were reversed in scoring so that higher scores corresponded to more successful presidencies. Historical measures of presidents’ job performance and behavior. In addition to the aforementioned surveys of presidential performance, scores on an empirically established (regressionderived) formula developed by Simonton (1987) was examined to predict presidential greatness. This Simonton historical composite consists of a weighted sum of six largely or entirely objective variables of behavior: number of years served, number of war years as president, war heroism prior to becoming president, estimated intellectual brilliance (see below), scandals while in office (coded negatively), and victim of an assassination. Being the victim of an assassination is a well-established indicator of presidential greatness. Indeed, this dichotomous variable correlates positively with a variety of independent indicators of presidential greatness. As Simonton (1994) noted, Systematic analyses of all U.S. presidents reveal that successful assassination is one of the best things that can happen to a chief executive’s (necessarily posthumous) reputation. Getting assassinated adds about as much to a former president’s greatness rating as serving five years in office or leading the nation through four years in war (p. 76). Although at least some of the association between assassination and rated presidential performance is probably reputational (being the victim of an assassination probably leads historians to view a president as great in hindsight), it is probably also partly a function of the fact that presidents who were the targets of assassination were willing to make enemies by initiating bold and controversial changes (see also Simonton, 1994). Indeed, in this data set, the dichotomous variable of being assassinated was associated with rated willingness to take risks in the Siena College survey (point biserial r .18, p .019) and with ratings (on a 1–9 scale) by presidential historians on the variable of “shows moral courage” (point biserial r .20, p .008). In addition, six other largely or entirely objective indicators of presidential performance were examined: (a) reelection (Kenney & Rice, 1988), (b) winning an election by a landslide (i.e., by 55% or more of the popular vote; Kenney & Rice, 1988), (c) subject of one or more Congressional impeachment resolutions (Perkins, 2003), (d) initiation of new legislation and programs, (e) viewed by others as a world figure, and (f) tolerates unethical behavior in subordinates. Variables 1–3 were coded dichotomously, and were derived from the historical record. Variables 4 – 6 were rated on a 1–9 scale and estimated by the same 121 experts who evaluated each president on the NEO PI-R. As a consequence, these latter three variables are not strictly independent of the NEO-PI-R ratings from which psychopathy score estimates were derived. It was predicted that given its association with the successful features of psychopathy, FFM-FD and perhaps FFM Factor 1 would be positively associated with the Simonton composite sixitem index of greatness and Variables 1, 2, 4, and 5. In contrast, it was predicted that given their theoretical ties to adaptive behavior, FFM-FD and perhaps FFM Factor 1 would be uncorrelated or negatively correlated with Variables 3 and 6 but that given their ties to unsuccessful behavior, FFM-IA and FFM Factor 2 would be positively correlated with these variables. Presidential character. To supplement the largely or entirely objective historical indicators, a composite measure of negative presidential character consisting of various indicators of antisocial PSYCHOPATHY AND U.S. PRESIDENTS 493 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. and otherwise problematic behavior was analyzed. From the personality items administered to presidential historians, Rubenzer and Faschingbauer (2004) used a rational/theoretical approach to construct several measures of presidential character and integrity (with items scored on a 1–9 scale), one of which was deemed relevant to the analyses here. Character Scale 1 (Negative Character) comprises 20 items administered to the presidential historians that assess largely objective behavioral indicators, in particular “the types of behaviors that make the news as indicators of character or the lack of it” (p. 332). These items include bullying others; abusing positions of power held; stealing; frequent cursing; extramarital affairs; cheating on sports, taxes, or business; gambling; and frequent absenteeism. It was predicted that FFM-FD and FFM Factor 1 would be largely uncorrelated with this measure, but that FFM-IA and FFM Factor 2 would be positively correlated with this measure. The internal consistency of the Negative Character scale in this sample, as measured by Cronbach’s alpha, was .90. Procedure The 121 expert raters completed a 596-item questionnaire evaluating the personality and behavior of their respective president(s) of focus; this measure contained the NEO-PI-R (Costa & McCrae, 1992), a set of items designed to assess presidential character (Rubenzer & Faschingbauer, 2004), and other items that were not analyzed here because they were not directly pertinent to psychopathy. These experts rated their target president’s personality for the 5 years prior to his assuming office to minimize criterion contamination in analyses of the associations between personality and presidential performance. Results Interrater Reliabilities of Measures of Psychopathy and ASPD In this sample, the average pairwise interrater reliability correlations, estimated using generalized estimating equations (GEEs; see the Associations between psychopathy factors and surveyrated dimensions of presidential performance section) across presidential raters for FFM Factor 1 (which assesses the core interpersonal and affective features of psychopathy) and FFM Factor 2 (which assesses an antisocial and impulsive lifestyle) were .31 and .42, respectively; for the FFM prototype for ASPD, the average pairwise correlation was .62. These correlations are well within the range of correlations typically reported for interobserver agreement in personality. For example, Kenrick and Funder (1988, Table 2, p. 26) found that mean correlations for personality traits (e.g., dominance, sociability) across raters were mostly in the .30 –.50 range. The average pairwise interrater reliability correlations for FFM-FD and FFM-IA across presidential raters, again obtained using GEE, were .56 and .34, respectively. Correlations Among Psychopathy Measures The correlation between FFM Factor 1 and FFM Factor 2 was r .63 (p .001). Consistent with previous literature on the PPI factors (e.g., Benning et al., 2003; Miller & Lynam, in press), FFM-FD and FFM-IA were not significantly correlated (r .09, ns). The correlations between FFM-FD and FFM Factors 1 and 2 were r .16 (p .05) and .18 (p .05), respectively; the correlations between FFM-IA and FFM Factors 1 and 2 were r .59 (p .001) and .92 (p .001), respectively. Mean Psychopathy Scores of the Presidents We next compared presidents’ scores on the four major psychopathy variables with those of the normative sample on which NEOPI-R Form R had been completed. To do so, we computed scores on these four variables from the Form R facet-level normative data reported in the NEO-PI-R manual (see Costa & McCrae, 2000) and compared them with the scores on the 42 presidents from the present sample, in both cases using the formulas described earlier (see the Measures of Personality, Psychopathy, and Covariates section). Presidents scored higher on FFM-FD (M 0.32, SD 1.48) compared with the normative sample ( M 0.94). In contrast, their mean scores on FFM-IA (M 11.55, SD 2.45) were virtually identical to those of the normative sample (M

Is this the question you were looking for? If so, place your order here to get started!