Radical Humility

Measures of other concepts that are related to narcissism

The dark triad: narcissism, psychopathy, and Machiavellianism

Introduction

Since the early 2000s, researchers have often assessed narcissism in concert with two other traits that are regarded as malevolent: psychopathy and Machiavellianism.  Researchers often refer to these traits collectively as the dark triad (Paulhus & Williams, 2002).  In contrast to narcissism, Machiavellianism tends to describe people who are manipulative, whereas psychopathy tends to describe people who are callous, impulsive, and likely to commit crimes.

Despite the differences between psychopathy, Machiavellianism, and narcissism, Jonason et al. (2009) suggested that perhaps researchers should combine these traits to generate one index. According to these authors, all three traits revolve around an interpersonal style in which individuals exploit other people to generate more immediate gains.  One index, therefore, could measure this interpersonal scale. To generate this index, Jonason et al. (2009)

  • administered separate measures of psychopathy, Machiavellianism, and narcissism,
  • converted each of the three scores to standardised responses or z values,
  • calculated the average of these three standardised values.

Subsequently, researchers have developed and applied shorter variants of this index to gauge the Dark Triad. For example

  • the dirty dozen is a scale, comprising 12 items, that measures the dark triad (Jonason & Webster, 2010),
  • the short-D3, comprising 27 items, can gauge both the overall dark triad as well as each trait separately (Jones & Paulhus, 2014).

The dirty dozen scale

Jonason and Webster (2010) constructed the dirty dozen instrument to measure the dark triad as concisely as possible.  Of the 12 items, four items measure narcissism—primarily grandiose features and entitlement:

  • I tend to want others to admire me.
  • I tend to want others to pay attention to me.
  • I tend to seek prestige or status.
  • I tend to expect special favours from others.

Four of the items gauge psychopathy—especially features that relate to callous affect or interpersonal manipulation:

  • I tend to lack remorse.
  • I tend to be unconcerned with the morality of my actions.
  • I tend to be callous or insensitive.
  • I tend to be cynical.

Finally, four of the items measure Machiavellianism:

  • I tend to manipulate others to get my way.
  • I have used deceit or lied to get my way.
  • I have used flattery to get my way.
  • I tend to exploit others towards my own end.

Jonason and Webster (2010) conducted four studies to establish the reliability and validity of this instrument.  The first study, in which 273 undergraduate students completed this instrument and a range of other scales revealed that

  • responses to the Narcissism Personality inventory were more highly associated with the items of the Dirty Dozen that assess narcissism than with the items that assess psychopathy or Machiavellianism,
  • similarly, responses to the Mach IV (Christie & Geis, 1970), a measure of Machiavellianism, were more highly associated with the items of the Dirty Dozen that assess Machiavellianism than with the items that assess psychopathy or narcissism,
  • however, responses to the Self-Report Psychopathy Scale–III were highly associated with all three subscales of the Dirty Dozen—consistent with the notion that Machiavellianism and narcissism may indeed be facets of psychopathy. 
  • all three subscales of the Dirty Dozen were inversely associated with a measure of agreeableness too.

A subsequent study of 246 undergraduate students also confirmed the validity of this measure.  For example, a confirmatory factor analysis verified the three correlated factors: CFI = .92; RMSEA = .08.

The short dark triad or SD3

To overcome some purported shortfalls of the dirty dozen, Jones and Paulhus (2014) constructed an alternative measure of the dark triad.  Specifically, according to Jones and Paulhus,

  • because the dirty dozen comprises only 12 items—4 items for each trait—the instrument does not assess the diverse facets of narcissism, psychopathy, or Machiavellianism sufficiently,
  • to illustrate, narcissism, psychopathy, or Machiavellianism, as measured by the dirty dozen, tend to be more correlated with each other than with other measures of the same traits (Jonason & Webster, 2010; Rauthmann, 2013),
  • furthermore, Machiavellianism, as gauged by the dirty dozen, is positively associated with a tendency to discount the value of future rewards  (Jonason & Tost, 2010)—perhaps contrary to the assumption that Machiavellian individuals tend to be more strategic than impulsive (Jones & Paulhus, 2009).

The measure that Jones and Paulhus (2014) constructed, called the short dark triad or SD3, comprises 27 items, designed to assess the main facets of narcissism, psychopathy, or Machiavellianism (for a Spanish version, applicable to adolescents, see Penado Abilleira et al., 2024).  For example, to assess narcissism, the 9 items measure

  • leadership and authority, such as “People see me as a natural leader”,
  • exhibitionism, such as “I hate being the centre of attention” [reverse-scored],
  • grandiosity, such as “I know that I am special because everyone keeps telling me so” or “I have been compared to famous people”, and
  • entitlement, such as “I like to get acquainted with important people”.

To assess psychopathy, the 9 items gauge

  • anti-social behaviour, such as “I like to get revenge on authorities”,
  • erratic lifestyle, such as “People often say I’m out of control”, and
  • callous affect, such as “It’s true that I can be mean to others”.

Finally, to assess Machiavellianism, the 9 items measure

  • reputation, such as “It’s not wise to tell your secrets”,
  • planning, such as “You should wait for the right time to get back at people”, and
  • coalition building, such as “Whatever it takes, you must get the important people on your side”.

Jones and Paulhus (2014) conducted a sequence of four studies to refine and to validate the measure. Across these studies, as evidence of validity, the researchers discovered that,

  • narcissism, as measured by the SD3, was highly related to all facets of the Narcissism Personality Inventory but not as highly related to other measures of Machiavellianism or psychopathy,
  • similarly, Machiavellianism, as measured by the SD3, was highly associated with all facets of the Mach-IV (Christie & Geis, 1970) but not as highly associated other measures of narcissism or psychopathy,
  • likewise, psychopathy, as measured by the SD3, was highly associated with all facets of the Self-Report Psychopathy scale (Williams et al., 2007) but not as highly associated other measures of narcissism or Machiavellianism,
  • the dark triad, as measured by SD3, were highly related to ratings of these traits by informants.

Application of the Hogan Development Survey

Robert Hogan, together with collaborators, designed the Hogan Development Survey (Hogan, 2009) to assess dysfunctional personality styles that

  • impede the reputation or performance of people at work,
  • but do not correspond to mental health disorders or clinical populations.

The instrument comprises 154 items and 11 scales that, as factor analyses reveal, can be divided into three clusters:

  • moving against—in which individuals become assertive and dominant when stressed—comprising the scales called bold, mischievous, colourful, and imaginative,
  • moving away—in which individuals tend to withdraw from people when stressed—comprising the scales called excitable, cautious, sceptical, reserved, and leisurely, and
  • moving towards others—in which individuals tend to seek approval or establish alliances when stressed—comprising the scales called diligent and dutiful.

As Ferrell and Gaddis (2016) revealed, cited in Coleman and Dulewicz (2022), these 11 scales correlate with various measures of the dark triad, such as the Short Dark Triad scale.  Consequently, researchers can utilise the Hogan Development Survey to gauge the dark triad.  For example, to measure narcissism, researchers utilise the following scales:

  • cautious [reverse scored], in which a typical item is “Before I act, I usually think about everything that could go wrong”,
  • bold, in which a sample item is “Rules often don’t apply to people with my abilities”,
  • mischievous, in which a typical item is “I enjoy testing limits, even if it sometimes causes trouble”,
  • imaginative, in which a sample item is “I often think of ideas that others find unusual or unrealistic”.

To gauge psychopathy, researchers utilise the following scales:

  • sceptical, in which a typical item is “I often question the motives of people who try to be friendly with me”,
  • colourful, in which a sample item is “”,
  • as well as bold, mischievous, and imaginative.

Finally, to assess Machiavellianism, researchers utilise the following scales:

  • excitable, in which a typical item is “I sometimes lose interest in projects once the initial excitement wears off”,
  • reserved, in which a sample item is “I don’t feel the need to share much about my personal life with coworkers”,
  • leisurely, in which a typical item is “I usually agree to requests, but sometimes I don’t actually follow through”,
  • as well as sceptical, bold, and mischievous.

As evidence of this approach, Coleman and Dulewicz (2022) revealed that narcissism, psychopathy, and Machiavellianism, as derived from the Hogan Development Survey, were inversely associated with supervisor ratings of job performance, as hypothesised.   correlated factors: CFI = .92; RMSEA = .08.

Concerns about the dark triad: The breadth of psychopathy

Most researchers appreciate that each of these traits—psychopathy, Machiavellianism, and narcissism—uncover vital insights.  Nevertheless, scholars have questioned the practice in which researchers combine these traits to generate one index.  One concern emanates from the breadth of psychopathy.  Specifically, according to Glen and Sellbom (2015), psychopathy subsumes both narcissism and Machiavellianism.  Thus, if researchers want to assess an index of the dark triad, a suitable measure of psychopathy may be sufficient.   

To illustrate, when researchers assess the dark triad, they generally utilise scales, such as the Narcissistic Personality Inventory, that measure grandiose narcissism—the inclination of individuals to inflate their capabilities, achievements, and importance—rather than vulnerable narcissism.   Yet, many, but not all, measures of psychopathy also include questions that gauge this tendency of people to inflate these qualities.  For example

  • one of the items in the Psychopathy Checklist–Revised (Hare, 2003) is a “grandiose sense of self-worth”,
  • the Psychopathic Personality Inventory (Lilienfeld & Andrews, 1996) comprises a subscale, called Machiavellian Egocentricity, that encompasses narcissism as well as Machiavellianism and includes items like “Frankly, I believe I am more important than most people”,
  • similarly, measures of psychopathy tend to include subscales or items that gauge Machiavellianism.

Indeed, many scholars and researchers have discussed the overlap between narcissism and either psychopathy or antisocial personality.  For example, these traits comprise overlapping signs or symptoms, such as the inclination of these individuals to inflate their importance, to erupt in anger, to behave impulsively, and to manipulate other people.  Consequently, correlations between measures of narcissism and measures of antisocial personality are high (Jornkokgoud et al., 2023).  Indeed, some researchers conceptualise psychopathy or antisocial personality as an extreme variant of narcissism (e.g., Miller et al., 2017). 

Concerns about the dark triad: Negligible incremental validity

Glen and Sellbom (2015) also raised other concerns about an index that gauges the dark triad. Specifically, according to these researchers, this index should be calculated and utilised only if

  • the dark triad index is associated with another relevant measure, such as criminal activity, after controlling psychopathy,
  • the association between one trait, such as psychopathy, and another relevant measure should depend on one of the other traits, such as narcissism.

To assess these conditions, Glen and Sellbom analysed two archival datasets of over 900 inmates from two correction centres.  The dataset included the following measures to gauge the dark triad:

  • the Psychopathic Personality Inventory (Lilienfeld & Andrews, 1996) to gauge psychopathy and comprising 187 items and eight subscales, such as stress immunity, impulsive nonconformity, and blame externalisation,
  • the Narcissistic Personality Inventory (Raskin & Terry, 1988) to measure grandiose narcissism,
  • the Machiavellianism Inventory-Version IV (Christie & Geis, 1970) to assess Machiavellianism, comprising 20 items.

In addition, the database included measures of the potential causes, correlates, or consequences of the dark triad, such as

  • the Emotional Empathy Scale (Mehrabian & Epstein, 1972),
  • the Sensation-Seeking Scale (Zuckerman, 1979) to gauge disinhibition as well as the inclination to pursue thrilling, risky, and unfamiliar activities,
  • the Big Five Inventory (John et al., 1991) to assess extraversion, neuroticism, openness to experience, conscientiousness, and agreeableness,
  • the Emotionality-Activity-Sociability-Impulsivity Inventory (Buss & Plomin, 1975, 1984) to gauge temperament: traits that are regarded as inherent
  • the Behavioral Activation and Inhibition Scales (Carver & White, 1994) to gauge sensitivity to reward and punishment,
  • the Minnesota Multiphasic Personality Inventory 2 Restructured form to assess mental disorders.

The analyses uncovered findings that challenge the utility and necessity of the dark triad index. For example, Machiavellianism and narcissism were only modestly associated with each other (r = .15) but strongly associated with psychopathy.  This finding accords with the assumption that psychopathy subsumes Machiavellianism and narcissism.

In addition, after controlling psychopathy, a dark triad index was not associated with most of the measures, such as empathy, sensation seeking, and sensitivity to punishment.  Nevertheless, the dark triad index, after controlling psychopathy, was positively associated with some measures—such as agreeableness—and negatively associated with other measures—such as two facets of behavioural activation, but all in the opposite direction to the hypothesised relationships.  One exception was that, after controlling psychopathy, the dark triad index was positively associated with anger.  Overall, however, as the findings imply, the incremental validity of the dark trial is limited.  Indeed, psychopathy was often more associated with the outcomes than was the dark triad index.

Finally, the unique facet of the dark triad, after controlling psychopathy, did not moderate the association between psychopathy and the various personality traits.  As this finding indicates, the questions that assess narcissism or Machiavellianism are redundant if psychopathy is also measured.

The dark tetrad: narcissism, psychopathy, Machiavellianism, and sadism

In the mid 2010s, researchers inserted an additional measure or characteristic into the dark triad: sadism (Paulhus, 2014).  Sadism refers to the tendency of some individuals to enjoy acts in which they are cruel to other people, either physically or psychologically (Foulkes, 2019).  This everyday sadism is distinct from sexual sadism, a paraphilic disorder, referring to people who enjoy the suffering of other people in relationships that are not consensual (Mokros et al., 2019).  The dark triad, when coupled with sadism, has been labelled the Dark Tetrad (Chabrol et al., 2009).

Measures

Researchers have developed a range of instruments to gauge the Dark Tetrad (e.g., Paulhus et al., 2021).  For example, Thibault and Kelloway (2020) constructed the Dark Tetrad at Work Scale.  This scale comprises items that measure

  • narcissism, such as “I am much more valuable than my coworkers”,
  • Machiavellianism, such as “At work, people are only motivated by personal gain”,
  • psychopathy, such as “I don’t care if I accidently hurt someone at work.”, and
  • sadism, such as “It’s funny to watch people make mistakes at work”.

As evidence of validity, Thibault and Kelloway (2020) conducted a confirmatory factor analysis to verify the four distinct traits and revealed that all traits were significantly associated with workplace deviance, such as bullying, after controlling social desirability.

Concerns about whether sadism is distinct from psychopathy

The main concern that researchers have raised, however, revolves around the distinction between psychopathy and sadism.  That is, whether sadism offers insights that transcend psychopathy remains contentious.  The primary source of this contention is that

  • psychopathy and sadism are highly associated with each other, sometimes generating correlations that approach 0.6 (Bonfá-Araujo et al., 2022),
  • sadism may be significantly associated with some criminal behaviours after controlling psychopathy, but the percentage of additional variance that sadism explains is limited, seldom if ever exceeding 5% (Jonason & Zeigler-Hill, 2018; Jonason et al., 2017),
  • conceptually, psychopathy and sadism both revolve around a callous personality—although the tendency of individuals to derive enjoyment from these callous acts seems to be more prevalent in sadism.

To explore this overlap between psychopathy and sadism more comprehensively, Bonfá-Araujo et al. (2025) conducted a meta-analysis.  The meta-analysis uncovered 185 studies that

  • referred to the dark triad, to sadism and psychopathy, or to sadist and psychopath.
  • reported a correlation between sadism and psychopathy,
  • reported correlations between some other variable—such as narcissism, agreeableness, or honesty-humility—and both sadism as well as psychopathy.

The researchers used the meta-R package (Balduzzi et al., 2019) to undertake the meta-analysis, such as to generate funnel plots and to conduct the Egger’s Regression Test of funnel asymmetry to assess publication bias.  The researchers also utilised the metafor R package (Viechtbauer, 2010) to remove outliers if publication biases were identified.  As the meta-analyses revealed

  • the overall correlation between sadism and psychopathy was 0.56—although this value depends on which instruments researchers utilise to assess these traits,
  • psychopathy and, to a lesser extent, sadism were positively associated with narcissism and Machiavellianism,
  • psychopathy and, to a lesser extent, sadism were negatively associated with agreeableness, conscientiousness, and honesty-humility (Bonfá-Araujo et al., 2025).

Accordingly, psychopathy is more strongly associated with many key outcomes than sadism.  Although not definitive, this study questions the assumption that a Dark Tetrad is more useful than is a Dark Triad. finding indicates, the questions that assess narcissism or Machiavellianism are redundant if psychopathy is also measured.

The Mach-IV: A widespread measure of Machiavellianism

Overview

Researchers have developed several measures to assess Machiavellianism, defined as the degree to which individuals behave manipulatively, deceitfully, callously, and connivingly to fulfill their personal interests.  These measures include

The Mach-IV, developed by Christie and Geis (1970), is perhaps the most prevalent measure of Machiavellianism and comprises 20 questions that assess three factors: views, tactics, and morality.  The first factor, views, relates to the cynical belief that people are hostile and thus individuals need to prioritise their own needs to survive or thrive. Typical items include

  • “It is hard to get ahead without cutting corners here and there”,
  • “Generally speaking, men won’t work hard unless they’re forced to do so”,
  • “Most people are basically good and kind” [reverse-scored].

The second factor, tactics, relates to the willingness of these individuals to apply manipulative strategies, approaches, and behaviours, such as deceit and exploitation, to fulfill personal needs and goals.  Sample items include

  • “It is wise to flatter important people”,
  • “Never tell anyone the real reason you did something unless it is useful to do so”,
  • “Honesty is the best policy in all cases” [reverse-scored].

The final factor, morality, revolves around the tendency of these individuals to disregard morals that could impede progress and success. A sample item is “It is better to be humble and honest than important and dishonest” [reverse-scored].

Development of the Mach-IV

To develop the Mach-IV, Christie and Geis (1970) first wrote 71 items that seem to characterise the qualities that Niccolo Machiavelli delineated in The Prince and The Discourses.  The researchers then classified each item into three clusters, corresponding to the three subscales: views, tactics, and morality.  Next, the researchers instructed participants to indicate the extent to which they agree or disagree with these 71 items on a 5-point scale.  The participants included

As this preliminary study revealed, 50 of the 71 items were significantly and positively associated with the total scores on Machiavellianism, defined as the sum of all responses after reverse-coding, in all three samples. The researchers then extracted the 20 best items—that is, items that

  • were highly associated with the total scores on Machiavellianism,
  • sufficiently distinct from other items.

Limitations of the Mach-IV: Overlap with psychopathy

The Mach-IV, although utilised extensively, has attracted some criticism. Miller et al. (2017), for example, uncovered some compelling evidence to suggest that measures of psychopathy and the Mach-IV assess the same underlying characteristic.  In one study, 393 adults completed a series of scales online, including

Several patterns in the data indicated that psychopathy and Machiavellianism, at least as measured by the Mach-IV, may assess the same underlying tendency.  Specifically

  • when the responses to all measures of the dark triad were subjected to a confirmatory factor analysis, a model that combines Machiavellianism and psychopathy explained almost the same level of variance as a model that distinguishes Machiavellianism and psychopathy,
  • the model that distinguishes Machiavellianism and psychopathy generated a correlation of .90 between these two traits,
  • the Mach-IV and psychopathy—derived from three distinct measures—were, in general, positively correlated with the same personality facets, such as anger hostility, depression, and impulsiveness,
  • similarly, the Mach-IV and psychopathy were, in general, negatively correlated with the same personality facets, such as most facets of agreeableness, dutifulness, achievement-striving, self-discipline, and deliberation.

Conceivably, these findings might reveal limitations of the Mach-IV and other measures of Machiavellianism rather than demonstrate that Machiavellianism and psychopathy are, in principle, redundant.  Consistent with this perspective,

  • the findings indicated that both the Mach-IV and psychopathy are inversely associated with impulse control,
  • yet, most specialists in the field argue that Machiavellian individuals are not impulsive but strategic and deliberate, prioritising future goals over immediate needs (e.g., Jones & Weiser, 2014),
  • indeed, in a subsequent study (Miller et al., 2017), 36 relevant academics, when asked to rate psychopathic and Machiavellian individuals on 30 personality facets, indicated that Machiavellian people are more deliberate and disciplined but less impulsive.  

The five-factor model of Machiavellianism

Overview

To overcome some of these concerns about the Mach-IV, Collison et al. (2018) constructed and validated another instrument. Specifically, to develop this scale, Collison et al. utilised the specialist ratings, collated by Miller et al. (2017), to ascertain which facets of personality are especially high and low Machiavellianism. This procedure uncovered 13 relevant facets to characterise Machiavellianism.    Subsequent factor analysis suggested these facets can be divided into three clusters or subscales.  The first subscale, agency, comprises

  • achievement, such as “I am not an ambitious person”, [reverse-scored]
  • activity, such as “My friends would call me lazy”, [reverse-scored]
  • assertiveness, such as “In meetings, I typically let others do the talking” [reverse-scored],
  • competence, such as “People look to me to get the job done”,
  • invulnerability, such as “I am not easily flustered”, and
  • self-confidence or low self-consciousness, such as “I am not easily embarrassed”.

The second subscale, antagonism, comprises

  • selfishness or low altruism, such as “I view others as tools to be used and manipulated”,
  • low modesty, such as “I am better than others”,
  • manipulativeness or low straightforwardness, such as “Being honest all the time won’t lead to success”,
  • callousness or low tendermindedness, such as “I don’t worry about other people’s needs if they conflict with my own”, and
  • cynicism or low trust, such as “I have a great deal of faith in human nature” [reverse-scored].

The third subscale, planful-ness, comprises

  • deliberation, such as “I like to carefully consider the consequences before I make a decision”, and
  • order, such as “I like having everything in its own, proper place”.tor, morality, revolves around the tendency of these individuals to disregard morals that could impede progress and success. A sample item is “It is better to be humble and honest than important and dishonest” [reverse-scored].

Scale development

To design this instrument, for each of the 13 facets that purportedly characterise Machiavellianism, Collison et al. (2018) constructed a series of questions. To construct these questions, the researchers adapted the relevant items of the International Personality Item Pool.  The items were modified to be more characteristic of Machiavellianism or more extreme. 

Next, 430 participants completed this initial draft of 201 questions.  To decide which items to retain, the researchers

  • removed items that were highly correlated with other items, uncorrelated with other items, or limited in variability,
  • subjected the remaining items to item response theory, retaining the most informative items,
  • attempted to retain four items for each of the 13 facets or subscales as well as preserve an even proportion of positively worded and negatively worded items.

Initial validation of the scale

To validate the scale, another sample of 280 participants completed this five-factor model measure of Machiavellianism as well as a series of other relevant instruments.  These instruments included

  • the Mach-IV,
  • measures of the dark triad, including the dirty dozen and SD3,
  • measures of psychopathy and narcissism—specifically, the self-report psychopathy scale-III and narcissistic personality inventory,
  • a measure of the five main personality traits,
  • the UPPS impulsive behaviour scale (Whiteside & Lynam, 2001) to gauge facets of impulsivity, such as urgency, limited premeditation, limited perseverance, sensation seeking,
  • the crime and analogous behaviour scale (Miller & Lynam, 2003) to assess drug use, property crime, violent crime, and risky sexual behaviour,
  • measures of aggression, such as the reactive and proactive aggression questionnaire (Raine et al., 2006) and the self-report of aggression and social behaviour measure (Morales & Crick, 1999),
  • the aspiration index (Kasser & Ryan, 1993) to gauge extrinsic aspirations, such as wealth and fame, and intrinsic aspirations, such as meaningful relationships, personal growth, and community contributions,
  • behavioural activation scale, to assess drive, fun-seeking, and reward responsiveness, as well as the behavioural inhibition scale (Carver & White, 1994).

The results broadly validated the five-factor model measure of Machiavellianism as well as uncovered some insights about this trait.  For example, in contrast to other measures of Machiavellianism, this five-factor instrument was

  • only modestly associated with other psychopathy scales,
  • inversely related to the various facets of impulsivity, as gauged by UPPS impulsive behaviour scale and behavioural inhibition scale,
  • generally unrelated to measures of aggression, crime, and irresponsible behaviour—although the antagonism factor did tend to coincide with aggression (Collison et al., 2018).