Abstract
Those high on Dark Triad traits (narcissism, Machiavellianism, primary and secondary psychopathy) are more likely to engage in sexual harassment and less likely to empathize with others. Few studies have, however, considered the impact of Dark Triad traits on perceptions of sexually aggressive behavior performed by others. The present study investigated the relationship between Dark Triad traits and perceptions of sexual harassment. Heterosexual women (N = 142) aged 18 to 50 years (M = 20.86, SD = 5.62) completed the NPI-16 (Narcissistic Personality Inventory), Mach IV, Levenson Self-Report Psychopathy Scale, and Sexual Harassment Attitudes Questionnaire. Standard multiple regressions were conducted to investigate the extent to which Dark Triad traits predicted victim and perpetrator blame and attitudes toward victim responses to sexual harassment. Primary psychopathy was the only significant individual predictor such that women with higher levels of the trait were more likely to blame the victim and less likely to blame the perpetrator. In addition, primary psychopathy was related to higher endorsement of victim compliance, and lower likelihood of supporting confrontation of the perpetrator.
Introduction
Dark Triad traits (narcissism, Machiavellianism, and psychopathy) comprise related but distinct personality traits, each associated with emotional coldness, exploitation of others, and a lack of empathy (Paulhus & Williams, 2002). Previous research suggests that Dark Triad traits influence the perpetration of manipulative and aggressive sexual behavior in both men and women. These behaviors include sexual coaxing (i.e., the use of seductive tactics to obtain sex from a reluctant partner), coercion (i.e., the use of pressure, alcohol, drugs, or force to obtain sex from a person against their will), and harassment (i.e., unwanted and inappropriate behaviors such as comments, touching, and assault) (e.g., Figueredo, Gladden, Sisco, Patch, & Jones, 2015; Jones & Olderbak, 2014). The present study expands these findings to examine the influence of Dark Triad traits on perceptions of sexual aggression, that is, attributions of blame and appropriateness of victim responses.
Previous studies demonstrate that personality influences engagement in sexual harassment (e.g., Lee, Gizzarone, & Ashton, 2003). In particular, research that has investigated aspects of the Dark Triad personality traits separately has found that primary psychopathy (i.e., callous, unemotional predisposition) and narcissism (i.e., superiority and a sense of entitlement) are associated with the use of sexually coercive strategies of varying degrees (Blinkhorn, Lyons, & Almond, 2015; Khan, Brewer, Kim, & Centifanti, 2017; Muñoz, Khan, & Cordwell, 2011), as well as intentions to use sexual coercion in the future (Miller, Bouffard, & DeHaan, 2017). The few studies that have investigated the Dark Triad traits together have implied that the common core (i.e., lack of empathy, callousness) shared by the three traits explains the proclivity for sexual coercion and harassment better than the individual traits in isolation from each other. For example, psychopathy and the common core shared by the Dark Triad are significant predictors of expressing rape-enabling attitudes, that is, empathy for the rapist, lack of empathy for the victim, and acceptance of rape myths (Jonason, Girgis, & Milne-Home, 2017), as well as intentions to engage in sexual harassment (Figueredo et al., 2015; Zeigler-Hill, Besser, Morag, & Campbell, 2016). Overall, men and women high in the Dark Triad traits employ predatory sexual strategies, including higher incidences of sexual harassment, and this is likely to be due to their callousness and low empathy.
Victims of sexual harassment are at increased risk of depression, anxiety, and substance use (e.g., McGinley, Wolff, Rospenda, Liu, & Richman, 2016). Despite the detrimental impact of sexual harassment, experiences are often underreported. Reluctance to report victimization may reflect a range of factors including reputational concerns and fear of reprisal (Llewellyn, Karageorge, Nash, Li, & Neuen, 2018). Victims may therefore be dependent on informal support, and the reactions of others are expected to have a substantial influence on the manner in which an individual makes sense of their experience. There are, however, significant individual differences in how victims of sexual coercion are viewed by others (e.g., Sakallı-Uğurlu, Yalçın, & Glick, 2007), highlighting the importance of understanding the extent to which personality influences the quality of informal support. To date, researchers have focused on the extent to which the Dark Triad traits predict male and female perpetration of sexual coercion, but the perceptions of harassment and attitudes toward victims have received less attention.
The Dark Triad traits, especially psychopathy, are associated with traits and behaviors that would suggest a lack of concern for victims, such as low levels of empathy (Jonason, Lyons, Bethell, & Ross, 2013), inappropriate emotional responses (Lyons & Brockman, 2017), and moral disengagement whereby an individual convinces himself or herself that ethical standards do not apply to them in that situation (Egan, Hughes, & Palmer, 2015). Furthermore, Watts, Bowes, Latzman, and Lilienfeld (2017) report that psychopathy predicts lower empathy toward rape victims. Hence, Dark Triad traits may impact on the perceived acceptability of sexual harassment or related issues, such as willingness to intervene or offer support to the victim.
There are several reasons to predict that specific components of the Dark Triad have a differential association with lack of victim empathy and perpetrator blame. For instance, primary psychopathy (i.e., emotional coldness, callousness), rather than secondary psychopathy (i.e., risk taking and impulsivity), seems to be driving the Dark Triad influences on difficulties generating an appropriate emotional reaction in response to others (Wai & Tiliopoulos, 2012). More specifically, primary psychopathy has been linked to experiencing positive feelings and emotions in response to sad stimuli (Ali, Amorim, & Chamorro-Premuzic, 2009). The coldheartedness factor of primary psychopathy could relate to reduced motivations for justice and less condemnation of immoral behavior (Decety & Yoder, 2016), which could have a negative influence on empathy toward the victim. The callousness can lead to a lesser likelihood of positive public action (White, 2014), and utilitarian judgments (which endorse harmful acts to an individual to favor the majority) in moral dilemmas (Takamatsu & Takai, 2017), which could have an impact on willingness to intervene or offer support to the victim. It is clear that primary psychopathy is the most relevant Dark Triad trait when trying to understand issues related to intervening or providing support for a victim of sexual harassment. Hence, we investigate the influence of all Dark Triad traits and distinguish between primary and secondary psychopathy. We predict that those high on primary psychopathy will exert the greatest influence on perceptions of sexual harassment.
Researchers have often focused on men’s attitudes to sexual harassment and the likelihood of perpetration. Studies have shown, however, that Dark Triad traits may also predispose women to use more sexually coercive strategies too (e.g., Blinkhorn et al., 2015; Khan et al., 2017; Muñoz et al., 2011). Indeed, psychopathy in women is a stronger predictor of sexual coercion (Hoffmann & Verona, 2018) and acceptance of rape (Watts et al., 2017) than it is in men. Endorsement of aggressive sexual behavior by women higher on Dark Triad traits suggests that Dark Triad traits may also predispose women to be more supportive of coercion, with less empathy toward the victim. The influence of personality on women’s reactions to sexual harassment is important in terms of understanding the likelihood of reporting an offense, as well as reactions to female victims who disclose their experiences to other women. Informal support networks are often similar in gender (Ullman, Lorenz, Kirkner, & O’Callaghan, 2018), which means that women are more likely to seek support from other women. This help may not be forthcoming if the person to whom harassment is disclosed displays high levels of the Dark Triad. Therefore, the present study is focusing on a female sample in investigating attitudes toward victims and perpetrators of sexual harassment.
Method
Participants
Heterosexual women (N = 142) aged 18 to 50 years (M = 20.86, SD = 5.62) were recruited online via adverts placed on social networking sites such as Facebook and a University participation point scheme. Participants were most commonly single (55.6%) or in a serious relationship (34.5%) at the time of the study; relatively few (9.9%) participants were in a casual relationship.
Materials and Procedure
Participants completed initial demographic questions (e.g., age, relationship status) and a series of standardized measures online. These included the NPI-16 (Narcissistic Personality Inventory; Ames, Rose, & Anderson, 2006), Mach IV (Christie & Geis, 1970), Levenson Self-Report Psychopathy Scale (Levenson, Kiehl, & Fitzpatrick, 1995), and the Sexual Harassment Attitudes Questionnaire (Malovich & Stake, 1990).
The NPI-16 (Ames et al., 2006) is a 16 item measure of narcissism, developed from the original NPI-40 (Raskin & Terry, 1988). Participants respond to 16 statement pairs, selecting the statement that most closely corresponds to their own feelings. Example statement pairs include “I like to be the centre of attention” (narcissistic response) versus “I prefer to blend in with the crowd” (non-narcissistic response). Narcissistic responses are scored as 1 and non-narcissistic responses are scored as 0. The Mach IV (Christie & Geis, 1970) is a 20 item measure of Machiavellianism. Participants respond to each item on a 7-point scale (1 = strongly disagree to 7 = strongly agree) with 10 items reverse coded. Example items include “Anyone who completely trusts anyone else is asking for trouble.” The Levenson Self-Report Psychopathy Scale (Levenson et al., 1995) is a 26-item measure of psychopathy intended for noninstitutionalized populations. Seven items are reverse coded. The scale measures both primary psychopathy (16 items) and secondary psychopathy (10 items). Participants respond to each item on a 7-point scale (1 = strongly disagree to 7 = strongly agree). Example items include “For me, what’s right is whatever I can get away with” (primary psychopathy) and “I have been in a lot of shouting matches with other people” (secondary psychopathy).
The Sexual Harassment Attitudes Questionnaire (Malovich & Stake, 1990) contains two sexual harassment scenarios. These were modified for the present study to ensure participants responded to scenarios in which another woman was the target of harassment. For example “Suppose that a close woman friend of yours is attending classes on this campus. After class one day, a professor asks that she come to his office to discuss her grade with him. When she gets there he notes that she barely passed the last exam and is in danger of receiving a D for the course. He then tells her at length how much he enjoys having her in the class, leading up to a dinner invitation. He states that if they ‘get to know each other better’ he might be able to work things out so she can get a better grade.”
Participants are then asked to respond to a series of questions (12 per scenario) on a 9-point scale (1 = strongly disagree to 9 = strongly agree). For each scenario two items respond to victim blame, two assess perpetrator blame, and two measure no blame. Example items include “The student is probably hoping that getting to know the professor personally will help her get a better grade in the course” (victim blame), “The professor is using his status unfairly to pressure the student into dating him” (perpetrator blame), and “The professor probably meant no harm so it should not be taken too seriously” (no blame). The remaining questions address attitudes toward responses to the harassment. For each scenario, two items relate to compliance, two measure confrontation, and two describe ignoring the behavior. Example items include “See the professor on a social basis, if he is interested it may help her grade” (compliance), “Tell the professor that she is not interested in a personal relationship, and that this should have nothing to do with her grade in the course” (confrontation), and “Change the subject and try to forget about the conversation” (ignore).
Cronbach’s alphas were acceptable for the present study: narcissism (α = .81), Machiavellianism (α = .76), primary psychopathy (α = .90), secondary psychopathy (α = .75), victim blame (α = .63), perpetrator blame (α = .72), no blame (α = .82), compliance (α = .86), confrontation (α = .78), and ignore (α = .64), though lower sexual harassment alphas reflect the number of items (4) in each harassment subscale.
Results
Pearson’s correlations revealed significant relationships between the Dark Triad traits, the attribution of blame, and perceptions of victim responses to sexual harassment. These data are shown in Table 1. All of the traits had significant or near-significant positive correlations with victim blame, no blame, and compliance, and significant or near-significant negative correlations with perpetrator blame. In addition, primary psychopathy and Machiavellianism were significant negative correlates of confronting the perpetrator.
Descriptive Statistics and Correlations for Dark Triad Traits and Perceptions of Sexual Harassment.
Note. PP = Primary Psychopathy; SP = Secondary Psychopathy; MC = Machiavellianism; NC = Narcissism; VB = Victim Blame; PB = Perpetrator Blame; NB = No Blame; CO = Confront; CP = Comply; IG = Ignore.
p < .05. **p < .001.
A series of six simultaneous multiple linear regressions were conducted with Dark Triad traits (narcissism, Machiavellianism, primary and secondary psychopathy) entered as predictor variables and victim blame, perpetrator blame, no blame, confrontation, compliance, and ignore the harassment as criterion variables. The overall model predicted victim blame, F(4, 132) = 5.17, p = .001, and perpetrator blame, F(4, 132) = 6.38, p < .001, explaining 13.5% (R2 = .14, adj R2 = .11) and 16.2% (R2 = .16, adj R2 = .14) of the variance in victim and perpetrator blame respectively. In each model, primary psychopathy was the only significant predictor (victim blame: β = .36, t = 3.00, p = .003; perpetrator blame: β = −.50, t = −4.30, p < .001), such that those high on primary psychopathy were more likely to blame victims of sexual harassment and less likely to blame the perpetrator. The model also predicted no blame, F(4, 132) = 6.95, p < .001, explaining 17.4% of the variance (R2 = .17, adj R2 = .15). Primary psychopathy was the only significant predictor (β = .36, t = 3.12, p = .002), such that those high on primary psychopathy were more likely to not assign blame.
The overall model predicted attitudes toward compliance, F(4, 132) = 10.41, p < .001, and confrontation, F(4, 132) = 3.92, p = .005, in response to sexual harassment explaining 24.0% (R2 = .24, adj R2 = .22) and 10.6% (R2 = .11, adj R2 = .08) of the variance in attitudes toward compliance and confrontation respectively. Primary psychopathy was the only individual predictor (compliance: β = .46, t = 4.13, p < .001; confrontation: β = −.34, t = −2.79, p = .006), such that those high on primary psychopathy were more supportive of compliance and less positive toward confrontation. The model did not predict attitudes toward ignoring sexual harassment, F(4, 132) = .69, p = .600. These data are shown in Table 2.
Multiple Regression Results for Dark Triad Traits and Perceptions of Sexual Harassment.
Note. ANOVA = Analysis of variance.
Discussion
The present study investigated the influence of Dark Triad traits (narcissism, Machiavellianism, primary and secondary psychopathy) on attitudes toward sexual harassment. Primary psychopathy predicted both the attribution of blame and attitudes to responses to sexual harassment. Women high on primary psychopathy were more likely to assign blame to the victim or not assign blame for the harassment and less likely to blame the perpetrator. Furthermore, women high on primary psychopathy were more likely to endorse compliance with the harassment and less likely to support confrontation.
The tendency for those high on primary psychopathy to blame the victim of sexual harassment may reflect the emotional coldness and callous, self-centered behavior (Hare, 1996), which characterize primary psychopathy. Indeed those high on psychopathy are more likely to feel positive when others are suffering (Ali et al., 2009). The reluctance to blame the perpetrator for the harassment may indicate an overall endorsement of sexually aggressive behavior, consistent with the increased use of sexual coercion by those high on primary psychopathy (Muñoz et al., 2011). Primary psychopathy is related to low levels of pro-social emotions, such as guilt (Lyons, 2015; Prado, Treeby, & Crowe, 2016), which, together with empathy, form a duo of emotions influencing the feelings of responsibility for others (Zahn-Waxler & Robinson, 1995). The tendency to avoid any form of blame is consistent with the abnormal moral decision-making (Koenigs, Kruepke, Zeier, & Newman, 2011) and deficits in those areas required to make moral decisions (Glenn, Raine, & Schug, 2009) observed in those high on psychopathy. It is possible that the lack of blame for the perpetrator, and blaming the victim, is related to a host of factors relevant in regulation of emotions in moral behavior, and those high in primary psychopathy may have different ways of dealing with moral decision-making.
Victims may display a range of responses to sexual harassment (Knapp, Faley, Ekeberg, & Dubois, 1997). Women high on primary psychopathy were more likely to endorse compliance in response to harassment. Findings may reflect a strategic and competitive approach (Ross & Rausch, 2001) whereby participants believe victims should accept the harassment to gain advantage (or avoid disadvantage). The tendency to focus on such outcomes rather than victim distress may be facilitated by the aforementioned lack of empathy. The reluctance to advocate confrontation in response to sexual harassment may also indicate a strategic approach, with those high on primary psychopathy more likely to engage in instrumental than impulsive aggression (Bjørnebekk, 2007). Future research may also consider the influence of primary psychopathy on responses to the perpetrators of sexual harassment. For example, those high on primary psychopathy may be more willing to support the perpetrator (e.g., statement of support or false alibi).
Responses to the disclosure of sexual violence vary and may have important consequences for the victim (Ahrens, Cabral, & Abeling, 2009; Ullman, Starzynski, Long, Mason, & Long, 2008). The current study investigated attitudes toward sexual harassment and future research should consider the extent to which primary psychopathy impacts on behavioral responses to sexual harassment, and willingness to intervene particularly. A range of factors influence willingness to intervene such as the perceived morality of the behavior (Bowes-Sperry & O’Leary-Kelly, 2005), suggesting that those high on primary psychopathy may be less likely to intervene. Training programs may improve attitudes to sexual harassment (Antecol & Cobb-Clark, 2003) or bystander willingness to intervene (Moynihan & Banyard, 2008) though some people may be more responsive than others to such interventions (Franklin, Brady, & Jurek, 2017). In particular, research indicates that interventions that educate participants about the seriousness of sexual harassment are especially beneficial for those low on empathy (Diehl, Glaser, & Bohner, 2014). Hence, these educational interventions may have a greater impact on those high on primary psychopathy who are more likely to display low empathy (Wai & Tiliopoulos, 2012).
Findings are of course limited by reliance on self-report responses to hypothetical scenarios. Although hypothetical scenarios are frequently employed to investigate sexual harassment (e.g., Brassel, Settles, & Buchanan, 2019; Carlucci & Golom, 2016), future research should consider actual responses to disclosure of sexual harassment and the level or form of support provided. In the present study, we did not assess women’s personal experience of sexual harassment. Although previous research indicates that personal experience does not influence attitudes to sexual harassment (Mazer & Percival, 1989) or the attribution of blame (Malovich & Stake, 1990), future research should include this. In particular, it is important to consider both the actual experience of harassment and perceptions that harassment has occurred (Ellis, Barak, & Pinto, 1991). Furthermore, there is substantial cross-cultural variation with regard to the incidence and awareness of sexual harassment (Tang, Yik, Cheung, Choi, & Au, 1996). Future research may consider whether knowledge of sexual harassment influences the relationship between primary psychopathy and perceptions of sexual harassment or responses to harassment victims.
To conclude, the present study investigated the influence of Dark Triad traits (narcissism, Machiavellianism, primary and secondary psychopathy) on the attribution of blame in relation to sexual harassment and perceived appropriateness of victim responses. Women high on primary psychopathy were more likely to blame the victim, less likely to blame the perpetrator, and more likely to assign no blame. In addition, primary psychopathy was related to higher endorsement of victim compliance, and lower likelihood of supporting confrontation of the perpetrator. Future research should consider the influence of primary psychopathy on related responses such as willingness to believe that harassment has occurred and intervention when a bystander is being harassed.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
