The Association Between Different Types of Psychopathic Traits and Peer Harassment: The Moderating Role of Gender
Simon Dahlbäck & Marcus Pettersson Örebro Universitet Abstract
The purpose of the study was to examine whether gender moderated the association between psychopathic traits and peer harassment, as well as examining each psychopathic trait individually alongside the different harassment forms. Secondary data from the Seven School project was used which collected its data from seven middle-sized schools in Sweden. Besides descriptive and correlation analyses, a series of hierarchical regression models were used. The results show that boys in general scored higher than girls on both psychopathic traits and peer harassment. Also, impulsivity predicted peer harassment and so did gender in all outcome steps except for step 1 and 2 for personal harassment. However, gender didn’t moderate the effect between psychopathic traits and peer harassment except for sexual harassment and impulsivity. The conclusion of the study is that impulsivity is a major concern in peer harassment whilst grandiosity and callous-unemotional does not seem to be of major concern. Also, the moderating role of gender only had an effect on impulsivity and sexual harassment in males.
Keywords: Psychopathic traits, gender, peer harassment, sexual harassment, bullying
Handledare: Metin Özdemir Psykologi, kandidatkurs
HT2018
Associationen Mellan Olika Typer Av Psykopatiska Drag och Trakasserier: Den Modererande Rollen av Kön
Simon Dahlbäck & Marcus Pettersson Örebro Universitet Sammanfattning
Syftet med studien var att undersöka om kön modererade kopplingen mellan psykopatiska drag och trakasserier bland jämnåriga. Samt även att undersöka de olika psykopatiska dragen separat tillsammans med de olika trakasserierna. Sekundärdata från ”Sju skolor” projektet användes som hämtade in sin data från sju medelstora skolor i Sverige. Utöver descriptive- och korrelationsanalyser, utfördes även en serie av hierarkiska regressionsmodeller. Resultatet visar på att pojkar generellt får högre resultat på psykopatiska drag och trakasserier än flickor. Impulsivitet förutsäger trakasserier bland jämnåriga och det gjorde även kön i alla resultatsteg förutom steg 1 och 2 för personliga trakasserier. Däremot så modererar kön inte effekten mellan psykopatiska drag och trakasserier förutom sexuella trakasserier och impulsivitet. Slutsatsen av studien är den att impulsivitet är ett stort problem för trakasserier bland jämnåriga medan grandiositet och känslokallhet inte verkar vara ett större problem. Den modererande koppling mellan könen hade bara en effekt på impulsivitet och sexuella trakasserier bland män.
Nyckelord. Psykopatiska drag, Kön, trakasserier, sexuella trakasserier, mobbning.
Handledare: Metin Özdemir Psykologi, kandidatkurs
The moderating role of gender on the association between psychopathic traits and peer harassment
Harassment among adolescents have been increasing over the last decades which has created concerns in our society. In schools, neighborhoods, and in different kind of settings youth are being harassed by their peers. Harassment behavior take on different forms and some may be more harmful than the others. Personal harassment, discrimination and internet-based
harassment are just some of the many forms of harassment youth experience in their everyday life (Fanti & Kimonis, 2012). These types of harassment may trigger emotional and
psychological problems among adolescents. Adolescents who fall victim of peer harassment can experience both internalizing and externalizing problems. Loneliness, low self-esteem, poor academic performance and depression are some of the potential outcome’s adolescents may display when they are harassed by their peers (Fanti & Kimonis, 2012). Overall, there are many types of harassment and the outcome is almost always negative.
Bullying and sexual harassment are two of the most commonly studied types of harassment among young people. Bullying at school can be described as a phenomenon that involves punches and being treated in an unpleasant way. Fanti & Kimonis (2012) define bullying as “repeated physical, verbal, or psychological attack or intimidation that is intended to cause fear, distress, or harm to the victim that occurs within particular
interpersonal relationships and is characterized by an imbalance of power” (p. 617). It also involves ostracizing because of their heritage or being special. Bullying in childhood or adolescence is associated with negative consequences in adulthood. For example, several studies reported that being bullied in an early age predicted long-lasting and pervasive effects on children’s ability to function both socially and emotionally, and this continues into
among youths can give a negative effect on mental health later into adulthood (Beduna & Perrone-McGovern, 2018). Overall, bullying is a harassment form with long-lasting effects.
A second type of peer harassment is sexual harassment. Sexual harassment is according to Kaltiala-Heino, Kaltiala-heino, Fröjd and Marttunen (2016) defined sexual harassment “legally as gender-based discrimination which in workplaces or schools may seriously impair the victim’s performance at work or ability to participate in and benefit from education, and creates a hostile work/school environment” (p. 1193). Existing literature have categorized sexual harassment into three types: gender harassment, unwanted sexual attention and sexual coercion. Gender harassment involves harassment based on their gender, unwanted sexual attention can be a sexual behavior that the victim finds unpleasant or hostile and sexual coercion cover sexual assault or threats (Kaltiala-Heino et.al., 2016). At around the ages of 14-15, young adolescents first sexual experiences often occur, and with this, we also see a high level of sexual harassment (Pellegrini, 2002). According to Pellegrini (2002), almost 50% of the cases of sexual harassment reported in school, are proactive. Seeing as it has a proactive nature, it resembles bullying. Adolescents who had experienced sexual harassment from their peers reported that it made them less likely to go to school, more reluctant to talk in front of their class, less focused at school, and ultimately led them to achieve lower grades (McMaster, Connolly, Pepler, & Craig, 2002). Sexual harassment among adolescents may be a way to assert male dominance over females, as well as a tool to show and establish
masculinity in front of your male peers (McMaster et al., 2002). Previous studies have shown that sexual harassment was linked to both depression and delinquency among adolescents (Kaltiala-heino et al., 2016). According to previous studies, sexual harassment among youths are linked to negative outcomes, which may last into adulthood.
In addition to bully victimization and sexual harassment experiences, adolescents may be harassed due to their personal features such as the way they look or dress. For example,
some may say that you are ugly, look like a pig, that you are fat, or in general just dislike the way you look. They may be harassed verbally or by text written on note papers or the school toilet, or through rumors spread around the peers. These types of behaviors often do not involve any physical coercion and one-in-one contact between the perpetrator and victim. Thus, personal harassment, despite the similarities, is a unique form of harassment.
Different research traditions have asked: why do some youth harass their peers? These studies have focused on factors such as school, friends and individual characteristics as well as ethnicity (Gower, Gower, McMorris & Eisenburg, 2015; Bayram Özdemir, Özdemir & Stattin, 2016). Harassment in school can be linked to a way for the harasser to establish dominance over their victim as well as gaining popularity among friends. This may be due to someone’s individual characteristics, such as looks, weight, ethnicity (van Geel, Toprak, Goemans, Zwaanswijk & Vedder, 2016). One line of research that aimed to understand why youth harass their peers focused on psychopathic traits such as callous-unemotional traits and impulsivity (van Geel, et al., 2016). Other dimensions of psychopathic traits, grandiosity, however, has not been examined well. In the current study, we examined the role of callous-unemotional traits, grandiosity and impulsivity together in predicting different types of peer harassment: personal harassment, sexual harassment and bullying. In addition, we also examined gender has a moderating factor in the association between psychopathic traits and peer harassment.
Psychopathic traits and peer harassment
Psychopathy is very commonly seen as a severe personality disorder which is related to antisocial behavior (Orue & Andershed, 2015). It is a construct with multiple dimensions which has been used to explain antisocial behavior for a long time. It can be conceptualized as a distinct constellation of interpersonal, affective and behavioral traits that can be grouped together to describe a unique group of antisocial adults (Hare, 1999). There are three main
dimensions of psychopathy: Callous-unemotional traits is the first dimension. This trait is focused on an affective style which is characterized of callousness, a severe lack of empathy and remorse, and a very shallow emotionality. Scoring high on this trait means that you aren’t capable of feeling any true remorse or guilt (Orue, Calvete & Gamez-Guadix, 2016). It is a persistent pattern of having no regard for other people. Roughly, it can be translated into a general uncaring of other human beings (van Geel et al., 2016). Grandiose-manipulative trait is the second dimensions. This trait is characterized by arrogance, manipulative, lying and being extremely charming at times. This trait has a grossly inflated view of themselves and their importance to others. They are egocentric and see themselves as the center of the
universe, justified by obeying and follow their own rules, not the rules of society (Orue et al., 2016). The third trait is called the behavioral dimension and is characterized by impulsivity and irresponsibility. Impulsivity is referred to as a tendency to act on impulse and having no thoughts about what may happen in response to those actions in the future. The common response to their actions would be “I did it because I felt like doing it”. These impulses be a central role in a psychopath’s behavior, the need to achieve instant pleasure, relief or satisfaction. Having high impulsivity can lead to rash decisions, such as quitting your job, breaking up with family members or partners and disappearing for long periods of time without any word (Orue et al., 2016). Irresponsibility is considered as taking no responsibility for one’s actions. Obligations and commitments are just empty words, and while they might say something in the lines of “ I’ll never cheat on you again”, these are often lies, promises written in the wind. When someone who high on the third trait, behavioral dimensions, does something wrong, they consider it to either be someone else’s fault, or take no responsibility for it at all. It can be looked at like psychopaths aren’t deterred by the chance that their actions may cause problems or risk for other people (Orue et al., 2016). By combining these
people in terms of what is right and wrong. People who scores high on psychopathic traits will often feel like rules of society does not apply to them which will facilitate their involvement in violent and aggressive acts (Orue et al., 2016).
Young adolescents who score high on psychopathic traits have been found to be especially prone to use bullying as a form of establishing dominance and popularity,
especially within the classroom (van Geel et al., 2016). The main dimensions of psychopathy, namely callous-unemotional, grandiose-manipulative and the behavioral dimension, have been found to be the most influential in studies of adolescents and bullying. While these three traits together have been thought to define psychopathy, each of the traits separately have been found to have a relation to antisocial behavior. According to several studies, bullies are characterized by having a lack of empathy against their victims and having a dangerous and even reckless way of behaving (van Geel et al., 2016). Adolescents with conduct problems that also scores high on callous-unemotional traits are according to cross-sectional research more likely to bully others than adolescents scoring low on either conduct problems or
callous-unemotional traits (Fanti & Kimonis, 2012). Several studies have reported that bullies are at a higher risk of displaying impulsive behavior than adolescents who are not involved in bullying (O'Brennan et al., 2009; Olweus, 1995; Schwartz, Proctor, Chien, Juvonen, & Graham, 2001). Whereas Olweus (1995) also pointed out that bullies are highly characterized by a strong need to dominate others as well as being very impulsive.
At the moment, to the best of our knowledge, no study has examined the link between sexual harassment and psychopathic traits. With this said, however, one can speculate that we may expect there to be a link among these variables. From the three dimensions of
psychopathic traits, impulsivity might be the strongest link. Acting out on pure impulse may cause opportunistic seizing, which may very well be sexual harassment. However, this is pure
speculation and more research needs to be done on the connection between sexual harassment and the three dimensions of psychopathic traits.
The link between personal harassment and psychopathic traits is not a subject that have been researched and we did not find any research that discussed the link between these two variables. Although, our definition of personal harassment is somewhat very similar to bullying, because it involves harming others emotionally. With that being said, it is
reasonable to expect that people who engage high level of bullying would most likely engage high levels of personal harassment as well. It is also reasonable to expect that people who score high on psychopathic traits would likely engage in both of these types of harassment. Moderating role of gender
Research on aggression and harassment in adolescents goes way back in time, back to the 1950s. Most of these studies focused on young men and mostly focused on the assumption that men are more aggressive than females (Estevez, Povedano, Jimenez & Musitu, 2012). During the last 65 years there has been an increase in girls’ harassment and antisocial
behavior. However, despite there being more research, there are still contradictory findings on the gender differences. The research on gender ranges from frequency to the type of
aggressive or harassment behavior (Estevez et al., 2012). This may be due to instrument used in measuring these constructs. Researchers has found that when only direct or physical aggression or harassment is the core measure, boys tend to score higher than girls, but when both indirect and direct types of aggression and harassment is measure, different results seems to emerge (Estevez et al., 2012).
There is still an open discussion about the differences between boys and girls in regard to psychopathic traits and peer harassment. While it is still unclear if and how gender interacts with the psychopathic traits callous-unemotional, grandiose-manipulative and behavioral dimensions in predicting peer harassment, boys are overrepresented in being bullies, showing
greater conduct problem as well as aggression and scoring higher on psychopathic
measurements than girls (Fanti & Kimonis, 2012). In the literature on psychopathic, boys have been found to engage in more violent acts than girls (Wendt, Bartoli, & Arteche, 2017). Taking this into account, one may expect that there is a gender difference between boys and girls in the average amount of bullying and psychopathic scores (Fanti & Kimonis, 2012). However, Fanti and Kimonis (2012) reported that boys and girls scoring similarly high on psychopathic traits, will show similar behavioral and personality correlates, which may indicate that there is no moderating effect of gender on psychopathic traits and bullying.
Even though boys continue to dominate girls in the number of perpetrators of more serious acts of violence and aggression the total number of girls that show antisocial behavior is increasing, and so does the cost to society (Silverthorn & Frick, 1999). When boys are compared with girls, research generally suggest that boys engage in more acts of physical aggression and less acts of relational, interpersonal and social forms of aggression (e.g. forms of behavior or acts that are meant to damage relationships, friendships or acceptance in certain groups), whereas girls dominate the later three (Bjorkqvist, Lagerspertz, & Kaukiainen, 1992; Crick, 1995).
There is a limited number of studies that examine the link between several dimensions of psychopathic traits and different form of aggressive behavior, as well as examining the gender differences, or if gender has a moderating role. Also, studies who extend their construct of psychopathy to adolescents, most often focus on boys or have failed to test
gender differences in a way that correlates with the psychopathic traits (Marsee, Silverthorn & Frick, 2005). It is possible that the three dimensions of psychopathy that is the best predictor of antisocial behavior differs between the genders. Marsee et al., (2005) stated that callous-unemotional traits when there was an absence of impulsivity and conduct problems, better predicted antisocial behavior in girls than in boys. There is a big issue when looking into the
association between psychopathic traits and antisocial behavior in females (Marsee et al., 2005)
In the current literature, most of the data that were collected have been on either boys alone or mostly boys. This approach makes it harder to establish any comprehensive
examination on the role of gender (Fontaine, Barker, Salekin & Viding, 2008). Thus, in the current study, we examined the moderating role of gender in the association between psychopathic traits and peer harassment
The Current Study
In the current study, we aimed to examine the association between psychopathic traits and peer harassment, and also look into if gender plays a moderated role in this association. We will also examine the psychopathic traits individually alongside the three different types of harassment. Psychopathic traits were operationalized into three subdimensions. The subdimensions are Callous-unemotional, Grandiosity and behavioral dimensions. Associated with these psychopathic traits we will also examine three types of peer harassment; bullying, personal harassment and sexual harassment.
Our overall research question is: Does gender moderate the association between psychopathic traits and peer harassment? We hypothesize that there will be a gender difference in the number of psychopathic traits exhibited as well as the amount of peer harassment. We hypothesize that boys will in general score higher than girls on all of our study variables. To test our hypotheses, this study used a longitudinal data from the “Seven School Project”.
Method
This study used secondary data and the data came from a longitudinal study named “Seven School Project”. The data of from this project were collected from seven different school in a city in Sweden and at two different times, with one-year interval. The project collected data from students who were in grade 7, grade 8 and grade 9. However, this study uses students that were grade 7 and grade 8 since they were able to answer the questionnaires the following year. Grade 9 students were excluded because they changed school the
following up year and were not able to answer the questionnaire two times. The age of the students in the sample range from age 13 to 14 (M = 13.9, SD = .74). The sample included 1002 participants, with 520 boys and 482 girls. Of these participants, 798 completed all of the questions in the questionnaire both times.
Measures
Psychopathic traits. Youths psychopathic traits inventory (YPI) was used to measure psychopathic traits. The YPI includes a 10 subscales or psychopathic traits. Each
psychopathic trait includes 5 items each, which means a total of 50 items in the YPI
(Andershed, Kerr, Statin, & Levander, 2002). YPI assesses three psychopathic dimensions: Grandiosity, Impulsivity and Callous-unemotional. As said before, each of these psychopathic traits in the YPI contains several items and each item was rated on a scale from 1 to 4. In this scale, 1 refers to “Don't agree at all” and 4 refers to “Agree completely”. Cronbach’s alpha showed that Grandiosity was .82, Impulsivity was .72, and Callous-unemotional was .84.
Personal harassment. Study participants were first asked about being victim of personal harassment. The measure included 6 items (e.g., “Has anyone said things about the way you look, like fatso, skinny, big nose”, Has anyone written derogatory things about you on example walls, toilets, the blackboard”, Has anyone commented or made fun of you or the way you look in a derogatory way?). Following these questions, they were asked: Have you yourself at any occasion this semester said or done any of the things described above towards
others (Regarding looks and personal characteristics)? This item was used as the measure of personal harassment in the current study. The responses were rated on a 5-point Likert scale ranging between 1 and 5. From 1, which refers to “no”, to 5, which refers to “Almost every day”.
Sexual Harassment. Sexual harassment was originally assessed using 9 items (e.g., Has anyone commented on your looks or your body in a sexual way that you don’t like”, has anyone spread rumors of a sexual nature about you”, “Has anyone called you a whore, slut, bitch, hooker, pussy, fag, cocksucker, said I’ll fuck you, or things like that”, Has anyone showed you or given you offensive images, pictures, drawings, or messages (Text
messages?”)). Following up, the participants were asked: Have you yourself at any occasion this semester said or done any of things described above towards another person (it is about sexual things)? This item was measured the same way as personal harassment in the current study. The responses were rated on a 5-point Likert scale ranging between 1 and 5. From 1, which refers to “no”, to 5, which refers to “Almost every day”.
Bullying. Bullying was measured using 3 items (e.g., Have you said nasty things, mocked, teased anyone in an unpleasant way at school (this semester)”, “Have you beaten, kicked, or assaulted anyone in an unpleasant way at school or on the way to or from school”, “Have you participated in ostracizing someone (this semester)?”). Each item was rated on a 4-point Likert scale ranging between 1 “No, it has not happened” to 4 “Yes, it has happened several times a week”. Cronbach’s alpha for bullying at Time 1 was .70, and .77 at time 2.
Demographic measures. In order to answer the goal of the study and our hypothesis, some basic demographic information had to be collected. The participants were asked to fill in age, gender, and their grade level.
This study used pre-collected data from the Seven School Project. Parents were informed about the Seven School Project study in advance and they were asked to provide passive consent. A pre-paid envelope was available for those parents who did not want to give their consent to the project. If parents didn’t return the consent form, they were interpreted as approving of the study. The data was collected by trained research assistants, who were there to assist with answering questions and maintaining a calm atmosphere. Information such as being anonymous, the study being confidential, as well as the possibility to terminate the study at any time was given to the students. Those students who were not allowed by their parents, or simply didn’t want to participate in the study was given school assignments while their classmates answered the questionnaire. The time for completing the questionnaire was approximately 90 minutes on average. We had two waves of data at our disposal to be able to estimate overtime changes in harassment from year to year. Grade 7 and 8 students were assessed in the first wave of observations and then the same students were assessed again one year later, in grade 8 and 9 using the same questions. First, the inter-item reliability of the already existing items was checked to make sure that the data could be used properly. After making sure that each item had a reliable fit, we began by creating scale scores for the variables which were assessed using more than a single item. Once having the correct variables, analyses could be made. We started off with descriptive statistics to get an
overview of the data, which was then followed up with a correlation analysis to see how well the variables correlated with each other. Z-scores were examined along with histograms to spot any outliers the study might have had. In this sample, a few outliers were found and dealt with by computing new variables. This was done by taking the original variable minus the mean of the same variable. Afterwards, a series of three regression analyses using several models was conducted in order to see whether gender moderates the link between
first assessment, T1, in order to be able to predict changes in the outcome measures over a certain amount of time.
Statistical analyses
All the analyses were run using Statistical Package of Social Sciences (SPSS). We had two waves of data at our disposal to be able to estimate overtime changes in harassment from year to year. Grade 7 and 8 students were assessed in the first wave of observations and then the same students were assessed again one year later, in grade 8 and 9 using the same questions. First, the inter-item reliability of the already existing items was checked to make sure that the data could be used properly. After making sure that each item had a reliable fit, we began by creating scale scores for the variables which were assessed using more than a single item. Once having the correct variables, analyses could be made. We started off with descriptive statistics to get an overview of the data, which was then followed up with a correlation analysis to see how well the variables correlated with each other. Z-scores were examined along with histograms to identify outliers. We also inspected the distribution of the variables. The measures of peer harassment were skewed. Thus, we transformed these variables using log10 transformation. To compute the interaction terms between gender and psychopathic traits, we coded gender as 0=girls and 1=boys and centered the three measures of
psychopathic traits by subtracting their mean values. Then, we multiplied gender and each of the centered psychopathic trait variable to create the interaction terms. For the analyses, we first estimated the bivariate correlations among the study variables. Then, we examined the differences between boys and girls in the study variables. Afterwards, a series of regression analyses with multiple steps were conducted in order to examine whether gender moderates the link between psychopathic traits and peer harassment. In all of the regression models, we controlled for the first assessment of the outcome variable in order to predict the changes in the outcome measures over one-year period.
Results
To examine whether psychopathic traits were associated to peer harassment and gender, we used bivariate correlation analysis. The bivariate correlations showed that all the psychopathic traits were positively related to peer harassment (See Table 1). In addition, gender was
positively correlated with all the study variables suggesting that boys were more likely to score high on any of the study variables than girls.
Variable 1 2 3 4 5 6 7 8 9 10 11 1. Age 1 .04 -.03 .01 .02 .00 .08* .07* .00 -.03 -.02 2. Gender 1 .12*** .17*** .07* .11*** .13*** .17*** .32*** .16*** .10*** 3. Bullying T1 1 .37*** .51*** .32*** .38*** .28*** .27*** .26*** .35*** 4. Bullying T2 1 .27*** .51*** .27*** .46*** .22*** .23*** .29*** 5. Personal harassment T1 1 .23*** .44*** .23*** .17*** .24*** .24*** 6. Personal harassment T2 1 .21*** .44*** .16*** .19*** .22*** 7. Sexual harassment T1 1 .29*** .14*** .21*** .22** 8. Sexual harassment T2 1 .11*** .14*** .17*** 9. Callous-Unemotional 1 .50*** .43*** 10. Grandiosity 1 .48*** 11. Impulsivity 1 *p<.05, **p<.01, ***p<001
Boys Girls
Variables Mean (SD) Mean (SD) t p
1. Age 13.97 (.75) 13.90 (.73) -1.34 .18 2. Bullying T1 .08 (.13) .56 (.10) - 3.67 .001 3. Bullying T2 .07 (.12) .034 (.08) -4.98 .001 4. Personal harassment T1 .09 (.17) .06 (.15) -2.27 .02 5. Personal harassment T2 .10 (.17) .07 (.14) -3.22 .001 6. Sexual harassment T1 .06 (.15) .03 (.10) -3.94 .001 7. Sexual harassment T2 .07 (.15) ,02 (.09) -5.17 .001 8. Callous-Unemotional .15 (.48) -.16 (.44) -10.65 .001 9. Grandiosity .09 (.53) -.08 (.51) -5.20 .001 10. Impulsivity .06 (.54) -.05 (.53) -3.17 .002 *p<.05, **p<.01, ***p<.001
Table 2. Means, standard deviations, t-test value and p-value.
We ran an independent t-test to examine whether there were any gender differences in psychopathic traits and peer harassment (see Table 2). The results suggested, that there were gender differences in the majority of the study variables. Specifically, results showed that the largest gender differences among our variable were Callous-unemotional. The result
suggested that boys (M = .15 SD = .48) displayed higher levels of callous-unemotional than girls (Girls M = -.16 SD = .44). Also, boys displayed higher levels of Grandiosity (M = .09 SD = .53) than girls (M = -.08 SD = .51). Among the variables, girls only displayed higher levels of bullying in T1 than boys, (Boys M = .08 SD = .13, Girls M = .56 SD = .10).
Outcome: Personal Harassment at T2 Outcome: Bullying at T2 Outcome: Sexual Harassment at T2 Step 1 Step 2 Step 3 Step 1 Step 2 Step 3 Step 1 Step 2 Step 3
ß ß ß ß ß ß ß ß ß Step 1 Outcome at T1 .22*** .18*** .18*** .36*** .29*** .29*** .27*** .24*** .24*** Age –.01 .00 .01 .02 .03 .03 .06 .07* .07* Gender .09* .05 .05 .13*** .10** .10** .13*** .12*** .13*** R2 .06*** .09*** .096*** .15*** .19*** .19*** .10*** .12*** .13*** Step 2 Callous .03 .07 .02 -.02 -.03 .01 Grandiosity .08 .13* .07 .02 .05 .01 Impulsivity .12** .11 .15*** .16** .10** .03 R2 change .03*** .003 .04*** .00 .02** .01* Step 3 Callous*Gender -.04 .76 -.07 Grandiosity*Gender -.07 1.09 .06 Impulsivity*Gender .01 -.30 .12* R2 change .00 .00 .01* Total R2 .10*** .19*** .13*** *p<.05, **p<.01, ***p<.001
Table 3. The results of the regression models testing the association between psychopathic traits and peer harassment, and the moderating role of gender
Table 4. Interaction figure between gender and levels of impulsivity
Do psychopathic traits predict harassment of peers?
We run hierarchical regression model to estimate the overtime changes in personal
harassment among youth (see Table 3). In the first step, we entered the initial assessment of personal harassment, age, and gender into the regression equation. In the second step, we entered the three psychopathic traits (i.e., callous-unemotional, grandiosity and impulsivity) into the regression equation. In the third step, we added the interaction effects between gender and psychopathic traits.
The results showed that, 6.1% of the variations in personal harassment were explained by the variables in the first step, F(3, 823) = 17,84, p < .001. In the first step, only the initial assessment of personal harassment, ß = .22, p < .001, and gender, ß = .09, p = .011,
significantly predicted the changes in personal harassment. Specifically, personal harassment had a positive significant effect which means that boys engaged in more personal harassment than girls. Next, adding psychopathic traits into the regression model explained an additional
3.2% of the variations in personal harassment over time, R2
change= .03, Fchange(3, 820) = 9.60, p
< .001. Specifically, impulsivity positively predicted changes in personal harassment, ß = .12, p < .01, suggesting that youth who were high on impulsivity were more likely to engage in personal harassment of their peers. The other two dimensions of psychopathic traits were not statistically significant predictors of personal harassment. In the third step, we added the interaction terms between gender and psychopathic traits to test the moderating role of gender. Overall, adding the interaction terms did not increase the amount of explained variation in overtime changes in personal harassment, R2change= .003, Fchange(3, 817) = 1.01, p
> .05 .These results suggest that gender did not moderate the association between
psychopathic traits and personal harassment. Off note, when the interaction terms entered into the regression model, the unique effect of grandiosity became significant, even though it was not significant in the prior step. This finding could be an indication of suppression effect due to inclusion of interaction terms in the model. Thus, we decided not to interpret this specific finding further.
We ran another hierarchical regression model to estimate the overtime changes in sexual harassment among youths (see Table 3). Like the previous regression model, we entered the initial assessment of sexual harassment, age and gender in the equation. In the following step, we entered the psychopathic traits variables callous-unemotional, grandiosity and impulsivity in addition to the previous variables used in step 1. In the third and final step, we added the interaction effect of gender combined with our psychopathic traits.
The results showed that, 10.4 % of the variations in sexual harassment was explained by the variables in the first step, F(3, 812) = 31.34, p < .001 . In the first step, only the initial
assessment of sexual harassment, ß = .27, p < .001, and gender, ß = .13, p < .001, significantly predicted the changes in sexual harassment. Specifically, boys were more active in harassing their peers sexually than girls, ß = .13 , p < .001. Next, adding psychopathic traits into the
regression model explained an additional 1.5% of the variations in sexual harassment over time, R2change = .015, Fchange(3,809) = 4.59 p < .001. To be more specific, Impulsivity
positively predicted sexual harassment, ß = .10, p < .01. With these findings, it suggested that youths who were impulsive were more likely to sexually harass their peers. Sexual
harassment was not statistically significant and could not be explained by
callous-unemotional and grandiosity. When the interaction terms were added to the model in the third step, the amount of explained variation in sexual harassment increased significantly, R2
change= .009, Fchange(3, 806) = 2.82, p < .05. This finding suggested that there was a significant
moderating effect of gender on the association between impulsivity and sexual harassment, ß = .12, p < .05. Specifically, boys that displayed high levels of impulsivity were engage more in sexual harassment then boys that displayed low levels of impulsivity, and were statistically significant ß = .19, t = 3.42. Furthermore, girls did not increase in the amount of sexual harassment among peers depending on the levels of impulsivity, and were not statistically significant, ß = .03, t = .51 (see table 4). Finally, the unique effect remained significant in the final step.
To examine the predictors of bullying a hierarchical regression model was used (see Table 3). As before, the initial assessment of bullying, age and gender were entered in the first step. The three psychopathic traits were added in the second step of the regression equation. In the third step, we added the interaction effects between gender and psychopathic traits. Result of the regression model showed that 15.5% of the variance in bullying were significant explained by the variables in the first step, F(3, 831) = 50.95, p < .001. The initial assessment of bullying significantly predicted the changes of bullying, ß = .36, p < .001 , and gender, ß = .13, p < .001. More specifically, boys did bully their peers more than girls. In the next step, we added the psychopathic traits in the regression model. In this step results explained an additional 3.6% of the variation in bullying over time, R2change = .036, Fchange(3, 828) = 12.30,
p < .001. Impulsivity positively predicted the changes in bullying, ß = .15, p < .001. This suggested that youths that were high on impulsivity were more likely to bully their peers. Callous-Unemotional and grandiosity were not statistically significant predictors of bullying. In our third step, the interaction terms between gender and psychopathic traits were added to test the moderating role of gender. By adding these interaction terms, the amount of explained variation in overtime did not increase the changes in bullying, R2
change = .003, Fchange(3, 825) =
1.02, p > .05. In addition, gender did not moderate the association between psychopathic traits and bullying.
Discussion
By using longitudinal data collected from adolescents from different schools in Sweden, we first examined the association between psychopathic traits and peer harassment. More specifically, we looked at whether the three main dimensions of psychopathic traits, namely callous-unemotional, grandiosity and impulsivity, as well as three different kinds of peer harassment, namely personal harassment, sexual harassment and bullying. Then, we tested the gender differences to see whether boys or girls scored higher on each of the variables. Finally, we tested the moderating role of gender on the association between peer harassment and psychopathic traits.
The results of this study showed that impulsivity was the only significant predictor of the three dimensions of psychopathic traits. Previous findings suggest that people who score high on impulsivity may tend to bully more, due to the fact that they won’t think about the long-term consequences of their actions (van Geel et al., 2016). They may resort to bullying due to feeling like they have been provoked, or perhaps they simply just act without thinking while bullying others. However, van Geel et al., (2016) argued that perhaps some degree of impulsivity in adolescence in normal and might not strongly be related to bullying, whereas older youth might have these traits and an indicator of psychopathy and more strongly related
to bullying. Seeing as our age range is between 13 and 14 years old, one might argue that our results may be due to normality’s in adolescence, which may very well decrease as the person ages.
Previous findings on impulsivity and sexual harassment is to our knowledge
something that hasn’t been studied. However, Pellegrini (2002) said that as sexual harassment is proactive, it resembles bullying in a way. As people who tend to score high on impulsivity also tends to have higher risks of being bullies, one may argue that seeing as sexual
harassment has a resemblance to bullying, one high on impulsivity may also sexual harass more than others. Acting on pure impulses may cause a person to seize an opportunity without thinking about it, which may end up being sexual harassment.
Like impulsivity and sexual harassment, personal harassment is another subject that hasn’t been researched and we could not find any research that would discuss the link between personal harassment and impulsivity. We will however argue that the definition we have of personal harassment is one that is very similar to bullying, and with this it is quite reasonable to expect there to be a similar link between personal harassment and bullying. People who will engage in high levels of bullying will most likely also engage in similar levels of personal harassment.
As impulsivity seems to be a key factor to why some adolescents harass their peers, research on the link between impulsivity and personal harassment needs to be expanded as well as the research on the link between impulsivity and sexual harassment. These two harassment forms are serious in today’s youth and we need to be able to find a find to minimize or prevent it.
Unlike our expectations, grandiosity and callousness did not significantly predict peer harassment for adolescents. A meta-analysis done by van Geel et al., (2016) suggests that there is a sizable and significant relation between bullying and the psychopathic traits of
grandiosity and callousness. Adolescents who scores high on callousness are less affected by the fear and suffering from other people and are a lot more likely to believe that something positive will come from being aggressive, which may be a reason those who scores highly on callousness have a stronger inclination to bully their peers. On the other hand, adolescents who scores high on grandiosity may bully others to keep control, to feel like they are maintaining a sense of power over others. The worth of having a grandiose self-image is extremely important or to gain entrance to the popular kids. Fanti & Kimonis (2012) reports that scoring high on grandiosity would increase the likelihood of bullying others. They argued that adolescents who scores high on grandiosity have strong feelings of entitlement combined with a willingness to take advantage of younger and possibly weaker children for personal gain. The behavioral dimension in their study did not prove significant in predicting the engagement in bullying.
Also, an unexpected finding was the fact that girls scored higher than boys on bullying in the first assessment of bullying. Why girls tend to bully more in T1 is unclear and one may only speculate to why we have gotten these results as boys score higher on every other variable in the study. We can speculate that girls starting a new school may be nervous and trying to act up to show others that they are cool and someone people would want to hang with, but this is pure speculation. Perhaps more research on bullying in this age needs to be conducted to find out the real reason.
However, the results of our study do not support what previous studies have said about the link between peer harassment and callousness and grandiosity. The result of our study goes in line with the results Wendt et al., (2017) received on their study. Their results showed that callousness was not a statistically significant predictor of antisocial behavior, such as bullying and personal harassment. Both of our study’s findings goes against what most
research results are showing today, which is that the three dimensions of psychopathic traits seems to have a link to some sort of harassment.
A possible explanation to why our results differs from Fanti and Kimonis (2012) may be due to using other measures. Another possibility is the fact that the studies were conducted in different countries.
To conclude, it is still very relevant to consider callousness and grandiosity to be important risk factors for future problems, such as peer harassment or criminal activities.
We hypothesized that gender would moderate the association between psychopathic traits and peer harassment. The results provided evidence only for one of the harassment outcomes. Specifically, gender moderated the effect of impulsivity on sexual harassment. Previous studies did not explain the gender differences in the association between impulsivity and sexual harassment. However, we speculated and expected that there would be a link between impulsivity and sexual harassment. Being impulsive may cause opportunistic seizing, which one might convert into sexual harassment. This expectation comes in line with our finding, since boys did score higher on impulsivity than girl. Why boys with high impulsivity were more likely to sexually harass their peer could be because impulsivity characterized as taking rash decisions (Orue et al., 2016). Taking rash decisions could be an explanation why boys had a higher mean value in sexual harassment than girls.
Several limitations of the current study should be mentioned. As mentioned before, personal harassment was very similar to bullying it became an issue during our research search. Trying to find previous studies that examine personal harassment were almost non-existing which was one of this study’s biggest limitations. Bullying were dominant in previous research about peer harassment in the association with psychopathic traits.
Despite all its limitations, the current study also had some strengths that worth mention. First, the sample size consists roughly an equal number of boys and girls, 520 boys
and 482 girls. Whilst most other studies that touch on our subject has had most boys in their samples and since we test the moderating effect of gender, an equal number of boys and girls makes this study more reliable. Second, this study used longitudinal data, which provided this study with more information about our participants over time. With Longitudinal data it is easier to find potential changes in a person's outcome than a cross-sectional study. A third strength in this study is that it examines three different types of harassment and three types of psychopathic traits. With these variables, this study provides us with information about which type of harassment predicts which type of psychopathic trait. The study also examines the unique effects of each dimensions of psychopathic traits, which previous studies lack in information.
Conclusions and implications
Previous studies support the facts that youths that score high on our three psychopathic traits engage more in an antisocial behavior (van Geel et al., 2016), but it is not clear if gender have a moderating role. Our findings showed that boys were more likely to harass their peers as well as more likely to score higher on the psychopathic traits than girls.
In conclusion, this study did not support the fact that gender play a moderating role in the association between psychopathic trait and peer harassment. However, this study provides us the information that boys with high impulsivity will harass their peers sexually. Findings showed, it was statistically significant that boys who scored high on impulsivity would
engage more in sexual harassment. Girl who scored high on impulsivity did not predict sexual harassment and were not statistically significant. With this information, future research can focus on boys with impulsive difficulties in grade 7 and 8. On this bias, prevention program for the boys could be implicated to stop sexual harassment. It would also be interesting if future research followed these boy’s outcome during a longer period, a longitudinal design.
With this design, future research may find factors that increase or decrease boy’s sexual harassment behavior.
There is still lots and lots to learn about the moderating role of gender on the association between psychopathic traits and peer harassment and this in an important topic that will need further studies to be able to really comprehend the full meaning of the association. Our study has contributed to the literature to understand and explain whether there is a moderating role of gender. However, to truly be able to understand and explain this, we will need to use larger samples sizes and more extensive resources to conduct a larger study. Purely practical, more studies need to be done, not only to further explain the gender role, but also to extend the research to a qualitative perspective which whose purpose will be to take fourth hypotheses to further explain this role. Seeing as the moderating role of gender only had an effect on sexual harassment and impulsivity, is it true that there are no other moderating effects of gender, or are we just lacking the research on the subject?
References
Andershed, H., Kerr, M., Stattin, H., & Levander, S. (2002). Psychopathic traits in non-referred youths: A new assessment tool. In E. Blaauw, & L. Sheridan (Eds.),
Psychopaths: Current International Perspectives (pp. 131-158). The Hague: Elsevier. Bayram, Özdemir, S., Özdemir, M., & Stattin, H. (2016). What makes youth harass their
immigrant peers? Understanding the risk factors. The Journal of Early Adolescence, 36(5), 601–624. doi:10.1177/0272431615574887
Beduna, K. N., & Perrone-McGovern, K. M. (2018). Recalled childhood bullying victimization and shame in adulthood: The influence of attachment security, self-compassion, and emotion regulation, Traumatology. doi:10.1037/trm0000162 Björkqvist, K., Lagerspetz, K. M., & Kaukiainen, A. (1992). Do girls manipulate and boys
fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18(2), 117-127. doi:10.1002/1098-2337
Crick, N., & Grotpeter, J. (1995). Relational Aggression, Gender, and Social-Psychological Adjustment. Child Development, 66(3), 710-722. doi:10.2307/1131945
gender perspective. In A. M. Columbus (Ed.), Advances in psychology research., Vol. 99. (pp. 71–91). Hauppauge, NY: Nova Science Publishers. Retrieved from
https://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2014-34536-004&site=ehost-live
Fanti, K. A., & Kimonis, E. R. (2012). Bullying and victimization: The role of conduct problems and psychopathic traits. Journal of Research on Adolescence, 22(4), 617-631. doi:10.1111/j.1532-7795
Dimensions of Psychopathy and their Relationships to Cognitive Functioning in Children, Journal of Clinical Child & Adolescent Psychology, 37:3, 690-696, doi:10.1080/15374410802148111
Gower, Amy L., A., Gower, B., McMorris, M., & Eisenberg. (2015). School-level contextual predictors of bullying and harassment experiences among adolescents. Social Science & Medicine., 147, 47-53. doi:10.1016/j.
Hare, R. D. (1999). Psychopathy as a risk factor for violence. Psychiatric Quarterly, 70(3), 181–197. doi:10.1023/A:1022094925150
Kaltiala-Heino, R., Kaltiala-Heino, S., Fröjd, M., & Marttunen. (2016). Sexual harassment and emotional and behavioral symptoms in adolescence: Stronger
associations among boys than girls. Social Psychiatry and Psychiatric Epidemiology., 51(8),1193-1201. doi:10.1007/s00127-016-1237-0
Marsee, A, M., Silverthorn, P., & Frick, J, P. (2005). The association of psychopathic traits with aggression and delinquency in non‐ referred boys and girls, Special issue: Gender and psychopathy, 23:6, 803-817. doi:10.1002/bsl.662
McMaster, L. E., Connolly, J., Pepler, D., & Craig, W. M. (2002). Peer to peer sexual harassment in early adolescence: A developmental perspective. Development and Psychopathology, 14(1), 91–105. doi:10.1017/S0954579402001050
O'Brennan, L. M., Bradshaw, C. P., & Sawyer, A. L. (2009). Examining development differences in the social‐ emotional problems among frequent bullies, victims, and bully/victims. Psychology in the Schools, 46, 100–115. doi:10.1002/pits.20357 Olweus, D. (1995). Bullying or peer abuse at school: Facts and interventions. Current
Directions in Psychological Science, 4, 196–200.
Orue, I., & Andershed, H. (2015). The Youth Psychopathic Traits Inventory-Short Version in Spanish adolescents—Factor structure, reliability, and relation with aggression,
bullying, and cyber bullying. Journal of Psychopathology and Behavioral Assessment, 37(4), 563–575. doi:10.1007/s10862-015-9489-7
Orue, I., Calvete, E., & Gamez-Guadix, M. (2016). Gender moderates the association between psychopathic traits and aggressive behavior in adolescents. Personality and Individual Differences, 94, 266–271. doi:10.1016/j.paid.2016.01.043
Pellegrini, A, D. (2002) Bullying, Victimization, and Sexual Harassment
During the Transition to Middle School, Educational Psychologist, 37:3, 151-163. doi:10.1207/S15326985EP3703_2
Schwartz, D., Proctor, L. J., Chien, D. H., Juvonen, J., & Graham, S. (2001). The aggressive victim of bullying: Emotional and behavioral dysregulation as a pathway to
victimization by peers. In J. Juvonen & S. Graham (Eds.), Peer harassment in school: The plight of the vulnerable and victimized (pp. 147–174). New York, NY: Guilford Press.
Silverthorn, P., & Frick, P. J. (1999). Developmental pathways to antisocial behavior: The delayed-onset pathway in girls. Development and Psychopathology, 11(1), 101-126. doi:10.1017/S0954579499001972
van Geel, M., Toprak, F., Goemans, A., Zwaanswijk, W., & Vedder, P. (2017). Are youth psychopathic traits related to bullying? Meta-analyses on callous-unemotional traits, narcissism, and impulsivity. Child Psychiatry & Human Development, 48(5), 768-777 Wendt, G. W., Bartoli, A. J., & Arteche, A. (2017). Dimensions of youth psychopathy
differentially predict concurrent pro- and antisocial behavior. Revista Brasileira de Psiquiatria, 39(3), 267–270. doi:10.1590/1516-4446-2016-2143
