Adolescents’ Antisocial Breeding Ground:
Longitudinal associations of familial psychopathological risk and Big Five personality traits with adolescents’ antisocial behavior
Masterthesis Forensic Clinical Child and Adolescent Studies Graduate School of Child Development and Education University of Amsterdam L.M. Harder 10873880 Under the supervision of prof. dr. G.J. Overbeek Second assessment by dr. H.E. Creemers Amsterdam, July 2016
Abstract
This Dutch three-wave (i.e., 2005, 2006, 2007) longitudinal survey investigated to what extent familial psychopathological risk and adolescents’ Big Five personality traits are predictive for adolescents’ antisocial behavior. In addition, we examined whether Big Five personality traits would moderate the relationship between familial psychopathological risk and antisocial behavior. Self-report data were used from 774 Dutch adolescent students (Mage = 13.6 years, SD = .89, age range = 11-16 years). Logistic regression analysis showed that only for girls low levels of conscientiousness, and familial psychopathological risk were predictive for an antisocial development with low levels of agreeableness and conscientiousness further strengthening this latter relationship. Our findings indicate that in families where one or both parents may suffer from psychopathological problems, the behavioral development of
children should be acutely monitored. When interventions for these children’s behavior are administered one should be alert to gender differences regarding personality and etiology since the appropiate approach to intervene may be dependent on gender.
Keywords: adolescence, antisocial behavior, Big Five personality traits, psychopathology, familial psychopathological risk
Longitudinal associations of familial psychopathological risk and Big Five personality traits with adolescents’ antisocial behavior
Behavioral problems including antisocial behavior are the most common problem in youths under 18 (Bonin, Stevens, Beecham, Byford, & Parsonage, 2011; Scott, Knapp,
Henderson, & Maughan, 2001). In the Netherlands, more than 13% of adolescents between 11 and 17 years report behavioral problems and between 13 and 18 years almost 6% report antisocial behavior (De Looze et al., 2014; Rigter, 2013). Antisocial adolescents have lower chances at social succes, may develop a continuous pattern of misconduct throughout their life course, and have chronic health problems. Moreover, antisocial behavior is a strong precursor to psychiatric disorders in adulthood (Bardone et al., 1998; Bonin et al., 2011; Reef, Van Meurs, Verhulst, & Van der Ende, 2010; Rutter, Kim-Cohen, & Maughan, 2006).
Besides, the costs of antisocial behavior for society are tremendously high; antisocial
adolescents cost society upto 10 times more than adolescents without these problems (Bonin et al., 2011; Scott et al., 2001).
To prevent antisocial behavior it is essential to discover its causes, so that intervention in an early stage becomes possible and children have a fair chance on a healthy development (Foolen, Ince, De Baat, & Daamen, 2013; Sachse, 2013; Scott et al., 2001). Some of the most widely empirically supported predictors of antisocial behavior considers youths’ personality traits like low levels of agreeableness and conscientiousness (Caspi, Roberts, & Shiner, 2005; Jones, Miller, & Lynam, 2011; Klimstra, Akse, Hale III, Raaijmakers, & Meeus, 2010; McCrae, 2011; Widiger, Trull, Clarkin, Sanderson, & Costa, 2002). Also a familial risk of psychopathology in general has a substantial effect on the development of antisocial behavior (Baker, Jacobson, Raine, Lozano & Bezdijan, 2007; Ferguson, 2010; Frick et al., 1992; Moffitt, 1993; Patterson, DeBaryshe, & Ramsey, 1989; Rigter, 2013). However, the interaction between these predictors to the development of antisocial behavior has, to our
knowledge, not been studied yet – despite numerous studies (see for a review Van der Merwe & Dawes, 2007) indicating the significance of such individual and family based predictors to the development of antisocial behavior. Accordingly, the current study aims to examine to what extent familial risk for psychopathology in general and adolescents’ Big Five personality traits are predictive for the development of adolescents’ antisocial behavior, and to examine whether the relationship between familial psychopathological risk and antisocial behavior is more pronounced in youths with specific personality traits.
Antisocial behavior constitutes the deliberate exceeding of rules, laws or standards (Baker et al., 2007; Loeber, 1990; Rigter, 2013). As a result of changes in hormone and brainstructures during puberty, adolescents may behave antisocial, such as acting recklessly, delinquent, and disruptive (Crone, 2014; Rigter, 2013). Their antisocial behavior is (to a certain extent) normal, situational, temporary, and may be seen as part of a healthy
development (Crone, 2014; Moffit, 1993; Overbeek, Vollebergh, Meeus, Engels, & Luijpers, 2001; Rutter et al., 2006). For most adolescents antisocial behavior such as delinquency decreases after puberty, when they notice that prosocial behavior is more rewarding (Moffitt, 1993). However, particularly lower vocational educated adolescent boys reveal significantly more severe and persistent antisocial behavior (De Looze et al., 2014). Severe antisocial behavior entails among other things of lying, fighting, bullying, stealing and aggressiveness on a regular basis which disturbs the environment (Foolen et al., 2013; De Looze et al., 2014; Loeber, 1990; Rigter, 2013). This study focuses on this severe antisocial behavior and
controls for gender, age and education level.
Familial Psychopathological Risk
So, why does one adolescent develop into a socially well-adapted adult and the other does not, or less so? It appears that the familial environment in general seem to make a difference in the level of alignment. For instance if the environment of upgrowing children
includes parental antisocial believes, attitudes or behavior such as parental criminal activities, the risk of children imitating and developing antisocial behavior is higher (Frick et al., 1992; Loeber & Stouthamer-Loeber, 1986; Rigter, 2013; Van der Merwe & Dawes, 2007).
Moreover, because antisocial parents often ignore the negative consequences of their
antisocial acts, children could emerge into believing acting antisocial is normal and acceptable (Loeber & Stouthamer-Loeber, 1986; Van der Merwe & Dawes, 2007). A familial risk can also be described more specifically into familial psychopathological risk. This risk can be defined as mental health problems of the parents, which are present before the child develops any deviant behavior, and are exposed to the child on a regular basis which has shown to have a significant negative effect on their healthy development (Loeber, Van der Laan, Slot, & Hoeve, 2008; Rigter, 2013). Two familial psychopathological risks appear to be strong predictors of children’s antisocial development: genetic predisposition and ineffective parenting as a result of parental problems and inability (Baker et al., 2007; Ferguson, 2010; Frick et al., 1992; Patterson et al., 1989; Rigter, 2013).
With regard to genetic predisposition, twin studies showed a heritability of antisocial behavior within a range of 41% to 56% (Ferguson, 2010; Scourfield, Van den Bree, Martin, & McGuffin, 2004). That is, roughly half of children’s antisocial behavior could be due to genes inherited of their parents. However, genetic determination of complex human behaviors, such as antisocial behavior, is still a poorly understood area (Ferguson, 2010). Research shows there are no specific, single malfunctioning genes who are solely responsible for antisocial behavior (Ferguson, 2010). Yet, neural maldevelopment can be a consequence of a complex of ‘risk genes’ that co-act (Ferguson, 2010; Moffitt, 1993). An example of a genetic variant that increases the risk for externalizing behavior, is the catechol O-methyltransferase [COMT] gene (Ferguson, 2010). This gene is related to increased antisocial behavior, because it is associated with a maldevelopment on the prefrontal cortex in the area of inhibition control
and aggressiveness. Also strongly associated with antisocial behavior is the monoamine oxidase A [MAOA] gene, which codes for the degradation of dopamine, serotonine and norepinephrine in the brain and relates to self-control, sentence sensitivity and proneness to aggression (Chhangur, Weeland, Matthys, & Overbeek, 2015; Ferguson, 2010). If a low MAOA activity variant of the gene is present in an individual, and if this individual grows up in an antisocial environment, there is a significantly higher risk to exhibit antisocial behavior (Chhangur et al., 2015; Ferguson, 2010). Accordingly, genetic predisposition could be contributing to children’s development of antisocial behavior.
Familial risk for psychopathology may also be explained on the basis of ineffective parenting (e.g., poor parental supervision, inconsistent parental discipline, parental rejection, and a lack of parent-child involvement) as a result of parental problems and inability (Frick et al., 1992; Loeber & Stouthamer-Loeber, 1986; Patterson et al., 1989; Rigter, 2013).
Ineffective parenting could be based on so-called coercive family processes (Boendermaker, 2008; Patterson et al., 1989; Smith, Dishion, Shaw, Wilson, Winter, & Patterson, 2014). The coercive cycling theory implies that negative behavior from the parent and the child is being reciprocally extorted. In a coercive process, children learn to act disruptive (e.g., shouting, manipulating, lying, being aggressive) to get what they want, based on parents withdrawing behavior requests and being inconsequent in their follow through. Thus, children’s antisocial behavior is functional, remains intact and extends over time (Boendermaker, 2008; Patterson et al., 1989; Smith et al., 2014). With this ineffective parenting children do not learn an appropriate set of social skills or moral values that are necessary to become socially well-adapted adults. Especially parents with psychiatric illness or disorders (e.g., substance use disorder, antisocial personality disorder, depression) are more often exchanging negative behavior with their child(ren) and more often are unable to enact effective parenting practices. This is, because they are frequently emotionally unavailable, unable to maintain a daily
structure, and are focused on their own problems instead of being sensitive to their
child(ren)’s needs (Frick et al., 1992; Loeber & Stouthamer-Loeber, 1986; Patterson et al., 1989; Van der Maas, 2010). Moreover, if both parents suffer from psychopathological problems the likelihood that children develop antisocial behavior increases even by a third compared to only one parent with ineffective parenting practices due to psychopathological problems (Frick et al., 1992; Loeber & Stouthamer-Loeber, 1986; Patterson et al., 1989). Accordingly, also ineffective parenting strategies due to psychopathological problems of the parents could be contributing to children’s development of antisocial behavior.
Personality Traits
Just like antisocial behavior is at least partially genetically determined, personality traits are also partially heritable (Jones et al., 2011; McCrae, 2011). Personality traits can be defined as unique characteristics that inform how someone thinks, behaves and feels. Together they form one’s personality (Caspi et al., 2005; Jones et al., 2011). The Big Five Factor model is broadly used to describe individual personality differences and divides personality traits into five dimensions (Caspi et al., 2005; Jones et al., 2011; Klimstra et al., 2010). The dimensions are extraversion (e.g., traits related to enjoying social attention, being sensitive to rewards and sensation, dominance, being optimistic), emotional stability (e.g., traits related to relaxation levels, being imperturbable, unenvious), conscientiousness (e.g., traits related to responsibility, being careful, orderly, and ambitious), agreeableness (e.g., traits related to being cooperative, generous, polite, empathic) and openness to experience (e.g., traits as intelligent, creative, curious, insight-ful) (Caspi et al., 2005 Jones et al., 2011; Klimstra et al., 2010). Meta-analyses showed that the dimensions of the Big Five Factor Model are consistently found and that these dimensions are the most relevant to understand one’s personality in relation to antisocial behavior (Jones et al., 2011; Miller & Lynam, 2001).
The manifestation of antisocial behavior (i.e., severity, prognosis, presentation) could be depending on individual differences of specific personality traits (i.e., pathoplasty model) (Klimstra, Luyckx, Hale III, & Goossens, 2014). Previously longitudinal research and meta-analyses associate particularly low levels of agreeableness (e.g., not conforming to the law, unremorseful, aggressive, no empathic feelings) and low levels of conscientiousness (e.g., irresponsibility, inattention, impulsiveness) to antisocial behavior (Caspi et al., 2005; Jones et al., 2011; Klimstra et al., 2010; Miller & Lynam, 2001; Widiger et al., 2002). To a lesser extent a low level of emotional stability (e.g., anxious, irritable distress, hostility) and high levels of extraversion (e.g., sensation seeking) are associated to a higher likelihood with the development of antisocial behavior (Caspi et al., 2005; Jones et al., 2011; Klimstra et al., 2010; Widiger et al., 2002). At last, for the personality trait of openness to experience, there is less scientific consensus about how it is associated with antisocial behavior (Caspi et al., 2005; Jones et al., 2011; Klimstra et al., 2010).
Also in line with the pathoplasy model is the self-control theory by Hirschi and Gottfredson (1990). This theory connotes that the lower the level of self-control, which is strongly associated with less conscientious individuals (Jensen-Campbell, Knack, Waldrip, & Campbell, 2007), the larger the likelihood to any deviant behavior (Buker, 2011; Kuhn & Laird, 2013; Pratt & Cullen, 2000). A low self-control is an innate mechanism, engenders a lack of resistance against any temptation (e.g., stealing an unlocked car) and focuses on quick gratification (e.g., joyriding). Antisocial behavior is associated with a low self-control because of traits like a lack of empathy, impulsiveness, sensation seeking, delinquency, and disruptive behavior regardless social consequences. Also the lack of discipline to prosocial long-term rewards (e.g., academic achievement) is associated with a low self-control. The self-control theory is consistently supported by numerous empirical studies. However, from meta-analyses the key criticism is the notion that the development of antisocial behavior not solely roots in
self-control. In addition, other influences like genetics and (other) neurobiological factors as well as the environment (e.g., family, peers, education) must be considered (Buker, 2011; Kuhn & Laird, 2013; Pratt & Cullen, 2000). Nevertheless, one’s personality could be contributing to children’s development of antisocial behavior.
Familial Psychopathological Risk × Personality Interaction
So, how do familial psychopathological risk and personality structure in adolescents combine to produce antisocial behavior? To our knowledge, no studies have looked at the interaction between these two risk factors and antisocial behavior, despite of the added value that knowledge about these interactions could bring regarding the determination of risk factors and variance in treatment approach and responsivity (Baker et al., 2007; Chhangur et al., 2015). However, there is research, abstracted from the General Strain Theory, on
interaction effects between familial strain (e.g., a disorganized, unpredictable, not cooperative family life), personality, and the development of criminal behavior with familial strain as a pivotal element. That is, if someone experiencing strain from his familial situation,
personality traits relating to criminal behavior lead to more criminal behavior than the familial strain alone. However, if this strain is absent, the same personality traits on their own do not lead to criminal behavior (Agnew, Brezina, Wright, & Cullen, 2002; Jones et al., 2011). Seen from this perspective, the interaction between a familial and individual risk factor contributes to a higher risk of developing deviant behavior. Moreover, one’s personality might be helpful as a moderator to uncover the contribution of predictors (Chhangur et al., 2015; Jones et al., 2011; Klimstra et al., 2014; Rigter, 2013) assuming that specific personality traits are predicting deviant behavior. Based on all the reviewed literature one might hypothesize that higher familial psychopathological risk is more strongly related to the development of antisocial behavior, and even more if adolescents have a personality structure that is
conscientiousness and low levels of agreeableness). Similarly, one might assume that with a protective personality structure (e.g., characterized by high levels of conscientiousness and high levels of agreeableness), the association between familial psychopathological risk and the development of antisocial behavior might be significantly less strong.
The Present Study
The development of children to socially well-adapted adults is a complex and ongoing configuration and appears to be dependent on both familial and individual risk factors
(Loeber, 1990; Loeber et al., 2008; Rigter, 2013). Accordingly, the current study aims to examine to what extent a familial risk on psychopathology, and adolescents’ Big Five personality traits are predictive on the development of adolescents’ antisocial behavior. Moreover, we examine whether the association between familial psychopathological risk and the development of antisocial behavior is stronger if adolescents have particular personality traits like low levels of conscientiousness and low levels of agreeableness. To our knowledge, this familial psychopathological risk × personality interaction have not been studied yet. Awareness of these interaction may contribute to a possible better understanding of the development of antisocial behavior and might give indications for treatment effectiveness. In this study, we will test the interactions in a three-wave (i.e., 2005, 2006, 2007) longitudinal survey among 774 Dutch adolescents aged 11 to 16 years.
Method Sample
The participants in this study were Dutch adolescents. At T1 the sample consisted of 2475 students, at T2 still 1419 students (57%) participated, whereas 774 students (31%) were left at T3. Concerning the decrease of the sample size during the study, logistic regression analyses showed that no selective attrition between T1 and T3 appeared.The high attrition rate is attributed to the leaving of students as they graduated, and to a disability to retain students
after they changed classes. Unfortunate for this longitudinal research, changing classes during high-school is quite common in Dutch education. To still acquire an optimal sample size, only classes with at least seven students who also participated at T1 were included at T2 and T3. The final sample of 774 students consisted of 405 girls (52.3%) and 369 boys (47.7%). The
average age of adolescents at T1 was 13.6 years (SD = .89, age range = 11-16 years). Most of them (375; 49.4%) were enrolled in lower vocational education, 110 students (14.5%) were enrolled in mixed classes of lower vocational and higher general secondary education, 159 students (20.9%) participated in higher general secondary education, and 114 students (15%) in pre-university education. Most students (89.1%) came from intact, two-parent families and were native Dutch (97.6%).
Procedure
An existing dataset with a stratified sample from the SODA (SOcial Development of Adolescents) study was used (see Overbeek, Zeevalkink, Vermulst, & Scholte, 2010). This dataset was chosen because it accurately converged with the research questions. In the SODA study 28 high-schools were selected and approached within a 100-kilometre radius around the city of Nijmegen, The Netherlands, by sending them a letter with an introduction of the research and invitation for a follow-up phone call. After this phone call 23 schools (82%) chose to be involved in the study. In a joined decision of the research team and the school administration of each school, it was determined which and how many classes would participate. With regard to the requirement of informed consent, students and their parents were informed about the content and purpose of the study. They all gave permission to participate. From January to March 2005, students filled in the questionnaire during a regular lesson (45-50 min.) at school. They were supervised by a teacher and trained undergraduates who were responsible for administering the questionnaires. Students were told that their
information would be processed anonymously and would not be shared with other parties. Data collection procedures at T2 (2006) and T3 (2007) were identical.
Measures
Familial psychopathological risk. This variable was assessed at T1 and consisted of four items. Three items were assessed related to parents’ mental health: ‘Are your parents seeing – or have they been seen – a psychiatrist or psychologist’, ‘Are your parents incarcerated – or have they been incarcerated – in a psychiatric facility?’ and ‘Are your parents having – or had they – a drug and/or alcohol problem?’. The fourth item was about criminogenic environment and was assessed by the item: ‘Are your parents having problems on a regular basis - or had they problems – with the justice system?’. Correlation between the assessed items ranged from .05 to .25. Items were scored with a categorical distribution of 1=‘No, ’2 =‘Yes, one of my parents’, and 3=‘Yes, both parents’. The summed score formed a new continuous variable where higher scores indicated a stronger predisposition to antisocial behavior.
Big Five personality traits. This variable was assessed at T1 by the Quick Big Five [QBF]. This 30-item self-report personality questionnaire is a shortened version for Dutch research based on the 100-item Big Five factor model questionnaire (Goldberg 1992;Kiekens et al., 2015). The QBF assesses the five personality dimensions extraversion, agreeableness, emotional stability, openness to experience and conscientiousness. Each dimension is measured with six items. Examples are talkative (i.e., extraversion), cooperative (i.e.,
agreeableness), anxious(i.e., emotional instability), innovative (i.e., openness to experience) and accurate (i.e., conscientiousness). On a 7-point Likert scale adolescents indicated to what extent an item fitted their personality. High scores indicated a high presence and low scores a low presence of a particular personality trait (Kiekens et al., 2015). The QBF is applicable to adolescent populations, a reliable measure for the Big Five dimensions and there are
indications of sufficient construct validity (COTAN Documentation NIP, 2015; Overbeek et al., 2010; Scholte, Van Aken, & Van Lieshout, 1997).Also the convergent validity is adequate with sufficient correlations with criterium variables (e.g., depressive feelings) (Kiekens et al., 2015). In this sample Cronbach’s Alpha coefficients were all sufficiently reliable; .80 for extraversion, .78 for agreeableness, .78 for emotional stability, .65 for openness to experience and .81 for conscientiousness.
Antisocial behavior. This variable was assessed at T1, T2 and T3 and calculated based on scores of the variables delinquency and bullying. For delinquency the 13-item self-report questionnaire by Houtzager and Baerveldt (1999) was used. Each item presented a minor offense such as vandalism, (petty) theft, arsonism, burglary and fighting. Adolescents
indicated how many times they committed these offenses over the last 12 months on a 5-point Likert scale from 1 ‘none’ to 5 ‘more than 12 times’.
For bullying the 5-item self-report bully scale from the Bully/Victim Questionnaire [BVQ] by Olweus (Solberg, Olweus, & Endresen, 2007) was used. The definition of bullying which was given to the students was: ‘Bullying is saying bothersome and/or nasty things to a peer. Also bullying is threatening, hitting, kicking, incarcerating, socially excluding or something to that effect, of a peer. Bullying happens deliberate, on a regular basis and the bullied victim has a hard time defending himself. It is NOT bullying when equivalent peers are quarreling, fighting or teasing each other’. The 5-item bully scale assesses on a 5-point Likert scale from 1 ‘(almost) never’ to 5 ‘multiple times a week’, the frequency of students taken part in bullying per academic year and in the last 5 days. This latter item had different criteria for the Likert scale and ranged from 1 ‘never’ to 5 ‘5 times or more’. In addition, the frequency of covert (e.g., saying bothersome and/or nasty things to peers) and overt (e.g., threatening, hitting, kicking peers or something to that effect) bullying was assessed.
Accordingly, 18 items represented the total antisocial behavior scale. To indicate the presence of antisocial behavior dichotomous variables were designed with 0 being ‘antisocial behavior not present’ and 1 being ‘antisocial behavior present’. To fall in the category of youths with antisocial behavior, one had to have committed at least one offense in the last 12 months and have had bullied someone at least 2-3 times a month or more. This latter cut-off score is compliant to the regular basis criteria by the definition of being a bully. Cronbach’s Alpha coefficients of the antisocial behavior scale in this sample were all reliable; .74 at T1, .79 at T2 and .87 at T3.
Results
The sample consisted of 774 adolescents of which 133 (17.2%) met the criteria of antisocial behavior at T1. At T2, 126 (16.3%) and at T3 119 (15.4%) adolescents met these criteria. A small group of 46 adolescents met the criteria for antisocial behavior at T1 as well as at T3. Familial psychopathological risk was measured at T1. In total, 607 (78.4%)
adolescents reported no manifest predisposition to antisocial behavior. That is, they scored neither of their parents on any item for familial psychopathological risk. A moderate predisposition to antisocial behavior was present for 64 (8.3%) adolescents who had scored one of the parents on at least one item of familial psychopathological risk. Finally, 24 (3.1%) adolescents had a strong familial predisposition to antisocial behavior, scoring both parents on at least one item of familial psychopathological risk.
Mean differences between boys and girls for familial psychopathological risk at T1, antisocial behavior at T1 and T3, and each Big Five personality dimension at T1 were examined with independent t-tests. These results (presented in Table 1) indicated that the participants were generally characterized by a low familial psychopathological risk for antisocial behavior as well as by low expressions of antisocial behavior at T1 and T3. Despite the low expression level, boys reported more antisocial behavior at T3 than girls. Regarding
the Big Five personality traits conscientiousness, emotional stability, extraversion and
openness to experience, the participants can generally be characterized by an average level of these traits. Yet, girls overall reported to be more conscientious, but less emotionally stable than boys. Finally, participants were generally characterized with a more than average level of agreeableness with girls overall being more agreeable than boys.
To examine cross-sectional and longitudinal correlations between familial psychopathological risk at T1, each Big Five personality dimension at T1, and antisocial behavior at T1, T2, and T3 Pearson correlation tests were performed. Table 2 presents these correlations divided by gender. The analysis showed that reported antisocial behavior at T1 was positively correlated to familial psychopathological risk, but only for boys. Thus, boys who were reporting more antisocial behavior at T1 were characterized by higher familial psychopathological risk. What also applied only to boys was the positive correlation between reported antisocial behavior at T2 and reported antisocial behavior at T3. So, if boys reported more antisocial behavior at T2 they also reported more antisocial behavior at T3.
Regarding reported antisocial behavior and Big Five personality dimensions, antisocial behavior at T1 and T2 were not correlated to any Big Five personality dimension for boys. However, for girls conscientiousness and agreeableness correlated negatively with their reported antisocial behavior at T1. So, less conscientious and less agreeable girls reported more antisocial behavior at T1. Reported antisocial behavior at T2 correlated again negative with girls’ conscientiousness, but significantly positive with girls’ extraversion. This latter implied that more extraverted girls were more likely to report antisocial behavior at T2. For girls there were no significant correlations with antisocial behavior at T3 and Big Five personality dimensions. However, reported antisocial behavior at T3 did correlate negatively with boys’ emotional stability meaning that if a boy was less emotionally stable more antisocial behavior at T3 was associated. Finally, regarding Big Five personality dimensions
and familial psychopathological risk, the correlation analysis showed that conscientiousness correlated negatively with familial psychopathological risk in boys, while for girls
agreeableness correlated negative with familial psychopathological risk. Thus, less
conscientious boys and less agreeable girls reported a higher familial psychopathological risk. To analyze to what extent familial psychopathological risk and Big Five personality dimensions would be predictive of the development of adolescents’ antisocial behavior, we performed a logistic regression analysis. In step 1, familial psychopathological risk at T1 and each Big Five personality dimension was specified as a predictor of antisocial behavior at T3 controlling for adolescents’ previous level of antisocial behavior at T1, gender, age and education level. The model significantly predicted adolescents’ antisocial behavior (𝒳2 = 75.167, df = 12, p = <.001) and accounted for 22.3% of the explained variance in adolescents’ antisocial behavior. The analysis showed that familial psychopathological risk at T1 (OR = 1.47, p = .045) predicted adolescents’ antisocial behavior on T3. Thus, with more reported familial psychopathological risk adolescents had a 1.5 times higher likelihood to become antisocial. Also male gender (OR = 3.59, p = <.001) and reported antisocial behavior at T1 (OR = 2.93, p = <.001) were associated with a roughly 3 times higher likelihood for adolescents to report antisocial behavior at T3. Moreover, adolescents following
pre-university education had a lower risk (OR = 0.19, p = .011) than adolescents enrolled in lower vocational education to develop antisocial behavior. Finally, non of the Big Five personality dimensions significantly predicted adolescents’ antisocial behavior. Table 3 shows all odds ratios, confidence intervals, and probability values for each predictor of main effects in this first step of the performed logistic regression analysis for both genders.
Because of the gender differences found in our correlation analysis, we performed step 1 of the logistic regression analysis again with a split file for gender. The girls’ model
accounted for 24.7% of the explained variance of antisocial behavior in adolescent girls. The boys’ model significantly predicted their antisocial behavior as well (𝒳2 = 32.905, df = 11, p = .001) and accounted for 17.9% of the explained variance of antisocial behavior in
adolescent boys. The analysis for girls showed that higher familial psychopathological risk at T1 (OR = 2.43, p = .002) and a low levels of conscientiousness at T1 (OR = 0.90, p = .019) significantly predicted girls’ reported antisocial behavior at T3. Thus, with a higher familial psychopathological risk, girls had an almost 2.5 times higher likelihood to become antisocial and low conscientious girls had a significantly higher likelihood to become antisocial. Also girls’ age (OR = 0.37, p = .014) significantly predicted their antisocial behavior; an increasing age contributed to less reported antisocial behavior at T3. Finally, reported antisocial behavior at T1 (OR = 5.45, p = .008) gave an almost 5.5 times higher likelihood for girls to report antisocial behavior at T3. The analysis for boys showed that reported antisocial behavior at T1 (OR = 2.88, p = .002) significantly predicted their antisocial behavior at T3 as well. Thus, their previous reported antisocial behavior was associated with an almost 3 times higher likelihood to report antisocial behavior at T3. Moreover, boys following pre-university education had a lower risk (OR = 0.65, p = .010) than boys enrolled in lower vocational education to develop antisocial behavior. Finally, none of the Big Five personality dimensions or familial psychopathological risk significantly predicted boys’ antisocial behavior. Table 4 shows all odds ratios, confidence intervals, and probability values divided by gender for each predictor of main effects in this first step of the performed logistic regression.
To analyze the extent to which adolescents’ Big Five personality dimensions would moderate the relationship between familial psychopathological risk and the development of antisocial behavior, in step 2 of the logistic regression analysis we entered the interaction terms for familial psychopathological risk with each of the Big Five personality dimensions. Just as in step 1, we controlled this analysis for adolescents’ previous level of antisocial
behavior at T1, gender, age and education level first for both genders. This model significantly predicted adolescents’ antisocial behavior (𝒳2 = 83.574, df = 17, p = <.001) and accounted for 24.6% of the explained variance. The analysis showed that none of the interaction effects predicted adolescents’ antisocial behavior at T3. Table 5 shows all odds ratios, confidence intervals, and probability values for each predictor and interaction effects in this second step of the performed logistic regression analysis for both genders.
Again, because of the gender differences found in the correlation analysis and yet also, because of the gender-specific findings in step 1 of our analysis, we performed step 2 of the logistic regression analysis too with a split file for gender. The girls’ model significantly predicted their antisocial behavior (𝒳2 = 45.042, df = 16, p = <.001) and accounted for 35.1% of the explained variance of antisocial behavior in adolescent girls. The boys’ model
significantly predicted their antisocial behavior as well (𝒳2 = 38.025, df = 16, p = .002) and accounted for 26.2% of the explained variance of antisocial behavior in adolescent boys. The analysis for girls showed that agreeableness (OR = 1.31, p = .010) and conscientiousness (OR = 0.77, p = .023) interacted with familial psychopathological risk. A correlation analysis with a high and low group on these two personality traits was performed to outline the direction of these interaction effects. The results of both analyses together showed that if girls were exposed to a higher familial psychopathological risk and were less agreeable or less
conscientious they had a significantly increased likelihood to report antisocial behavior at T3. The analysis for boys showed that none of the interaction effects predicted adolescent boys’ antisocial behavior at T3. Table 6 shows all odds ratios, confidence intervals, and probability values divided by gender for each predictor and interaction effects in this second step of the performed logistic regression analysis.
Discussion
This Dutch three-wave (i.e., 2005, 2006, 2007) longitudinal survey investigated to what extent familial psychopathological risk and adolescents’ Big Five personality traits are predictive for adolescents’ antisocial behavior. In addition, we examined whether Big Five personality traits would moderate the relationship between familial psychopathological risk and antisocial behavior. We hypothesized that (1) low levels of the Big Five personality traits conscientiousness and agreeableness are predictive for adolescents’ antisocial behavior, (2) familial psychopathological risk in general predicts the development of adolescents’ antisocial behavior, and (3) low levels of conscientiousness and low levels of agreeableness increases the likelihood to become antisocial if there is a higher familial psychopathological risk. Our three hypothesis were confirmed, but only for girls and with the exception that low levels of agreeableness did not significantly predict an antisocial development.
Our assumption that low levels of conscientiousness and agreeableness were significant predictors of antisocial behavior appeared to be partially true and dependent on gender. Only low conscientious girls had a higher likelihood to develop antisocial behavior. In general this finding is in line with previously longitudinal research, meta-analyses, and the self-control theory since less conscientious individuals are associated with more antisocial behavior (Buker, 2011; Caspi et al., 2005; Jones et al., 2011; Klimstra et al., 2010; Kuhn & Laird, 2013; Pratt & Cullen, 2000; Widiger et al., 2002). However, the claim from the self-control theory that antisocial behavior solely roots in self-self-control, which is strongly associated with less conscientiousness (Jensen-Campbell et al., 2007), cannot be supported by our
results, because we did not find low conscientiousness in boys as a predictor of their antisocial behavior. In line with prior research and meta-analyses, this is an indication that also other influences must be considered to explain antisocial behavior (Buker, 2011; Kuhn & Laird, 2013; Pratt & Cullen, 2000). Furthermore, we did not find the claim that low levels of
agreeableness are predictive to adolescents’ antisocial behavior. This may be attributed to our finding of the above average mean levels of agreeableness in general with as a result maybe too little lower agreeable adolescents in our sample. Our results are particularly striking, because most research associates low conscientiousness and low agreeableness to antisocial boys (Rigter, 2013; Thijs, Van Dijk, Stoof, & Notten, 2014) and we did not find this.
Regarding our hypothesis that familial psychopathological risk is a predictor of the development of adolescents’ antisocial behavior, our findings showed that only for girls this was indeed a significant predictor. For girls this finding is in line with prior research and support the assumption that an antisocial development of children could be dependent on genetic predisposition and ineffective parenting as a result of parental problems and inability (Baker et al., 2007; Frick et al., 1992; Loeber & Stouthamer-Loeber, 1986; Patterson et al., 1989; Rigter, 2013; Van der Maas, 2010). Again, it is remarkable that we did not find this effect for boys also, because we did not find significant mean differences between boys and girls regarding familial psychopathological risk. Furthermore, girls’ increasing age showed to predict less reported antisocial behavior, but to boys we did not find such relation. However, for boys the education level was vital since boys enrolled in pre-university education appeared to have a lower likelihood to become antisocial than boys enrolled in lower vocational
education. These last two results are aside from the gender differences in line with prior research, because a lower education level is predictive of developing antisocial behavior and an increasing age is not (De Looze et al., 2014; Moffitt, 1993). In complete accordance with prior research were our findings that male gender is predictive to an increased likelihood for the development of antisocial behavior and that for both boys and girls antisocial behavior is moderately stable over time (De Looze et al., 2014; Moffitt, 1993). Yet, there were some differences in the latter in terms of girls’ previously reported antisocial behavior being a stronger predictor of their future reported antisocial behavior.
With regard to our expected interaction effect that low levels of conscientiousness and low levels of agreeableness increases the likelihood to become antisocial if there is a higher familial psychopathological risk, our findings showed that this is confirmed, but once again only for girls. For girls our findings are in line with prior research since interaction between a familial and individual risk factor result into a higher risk of developing deviant behavior than a familial risk alone (Agnew et al., 2002; Chhangur et al., 2015; Jones et al., 2011; Klimstra et al., 2014; Rigter, 2013). Thus, girls with a less socially well-adapted personality (i.e., less conscientious and less agreeable), are more sensitive to adverse familial circumstances like coercive processes and with this more likely to develop antisocial behavior (Belsky, 1997; Pitzer, Jennen-Steinmetz, Esser, & Schmidt, 2011). Once more, it is striking that we did not find any of these results for boys.
Gender-Specific Vulnerability for Risk Factors
In light of these findings, the question arises as to why we only did find that for girls familial psychopathological risk and being low conscientious were predictive to their
antisocial behavior? And why did we not find an interaction effect for boys? It appeared that research and explanations about these phenomena are scarce (Baker et al., 2007; Gorman-Smith & Loeber, 2005; Thijs et al., 2014). Yet, we could explain our findings with the vulnerability hypothesis that states that there are gender differences in susceptibility for risk factors of antisocial behavior (Steketee, Junger, & Junger-Tas, 2013; Tiet, Wasserman,
Loeber, McReynolds, & Miller, 2001). Firstly, concerning our finding on conscientiousness, it appears that adolescent girls are more vulnerable to a low self-control, which is strongly associated with less conscientiousness (Jensen-Campbell et al., 2007), than adolescent boys (Thijs et al., 2014). Research showed that the same level of low self-control appears to trigger girls more easily into antisocial behavior than boys (Thijs et al., 2014). This discrepancy might be due to gender-specific developmental brain processes and neural mechanisms
relating to inhibition control and emotion regulation, but more research on this subject is needed (Diekhof et al., 2012). Secondly, regarding our finding on familial psychopathological risk, it is possible that girls suffer more from psychopathological problems of their parents than boys, because they might be more sensitive to the influences of adverse family
circumstances (Steketee et al., 2013; Thijs et al., 2014; Pitzer, 2011), and more sensitive to the average genetic predisposition (Baker et al., 2007). Girls are more emotionally involved in the relationship with their parents and more often have a need for a close relationship with their mother than boys (Steketee et al., 2013; Pitzer, 2011). In this way girls are more affected by disruptions, deficits or negativity in these relationship caused by for instance ineffective parenting strategies. As a result, girls seem more likely than boys to cope with these adversities by developing antisocial behavior (Gorman-Smith & Loeber, 2005; Thijs et al., 2014; Steketee et al., 2013). However, a consistent view on this subject remains open, because other studies examining gender differences of antisocial behavior did not find gender-specific etiologies (Baker et al., 2007; Steketee et al., 2013).
Developmental differences of Antisocial Behavior
Another explanation of our findings could be derived from the dual-taxonomy model by Moffitt (1993) who distuinguished two categories of developmental differences of
antisocial behavior: a life-course-persistent and adolescence-limited development (Boendermaker, 2008; Moffitt, 1993). Life-course-persistent antisocial individuals are characterized by an early onset of antisocial behavior, neuropsychological deficits (e.g., low self-control), and adverse rearing circumstances (e.g., psychopathological problems of the parents) resulting in more severe and persistent antisocial behavior (Moffitt, 1993; Rutter et al., 2006). Adolescence-limited individuals are characterized by an adolescence onset of antisocial behavior, a tendency to imitate deviant peers, and being antisocial as a reaction on frustrations arising out of the maturity gap (an imaginary gap between biological and social
age like not allowed to vote, drink or drive a car). In contrast of the life-course-persistent antisocial adolescent, the adolescence-limited antisocial adolescent is still able to show prosocial behavior when this is more rewarding and has more control over his antisocial responses (Moffitt, 1993; Rutter et al., 2006).
Since we investigated personality structures and familial risk factors that could be assigned to the course-persistent category, it might be possible that we had mainly life-course-persistent antisocial adolescent girls and adolescence-limited antisocial adolescent boys in our sample. For the latter, their friends and intimate relationships could have a more substantial effect on their antisocial behavior than personality structure and familial factors. These differences in etiology could explain our gender-specific findings. If we assume that the girls in our sample showed more persistent antisocial behavior it is likely we found more effects since this behavior is more extreme. Moreover, research showed that it is common for boys to exhibit more antisocial behavior than girls (De Looze, et al., 2014; Rigter, 2013; Tiet et al., 2001). Yet, if girls are exhibiting antisocial behavior this may already be an indication of more severe and possibly more persistent antisocial behavior, designated as the
genderparadox (De Looze, et al., 2014; Rigter, 2013; Tiet et al., 2001). Besides, this is supportive of our findings that girls’ previously reported antisocial behavior more strongly predicted their future reported antisocial behavior than it did for boys, despite girls having lower mean levels of antisocial behavior. Accordingly, next to differences in vulnerability of risk factors, developmental differences may also account for our gender-specific findings.
Limitations and Strengths
Our study has a few limitations. Firstly, the study only addresses risk factors within the family. However, especially in adolescence environmental factors outside the family, like peer relationships and friendships have a substantial effect on children’s behavior (Loeber et al., 2008). Moreover, if the influence of peers is assessed maybe developmental differences of
antisocial behavior could be determined as well. Secondly, familial psychopathological risk items were only available at T1. If they would have been available also at T2 and T3, it would have been possible to analyze if the parents possibly develop (more) mental health problems in response to their children’s behavior and directions of effects would become more clear. Thirdly, the antisocial behavior scale only consisted of the variables delinquency and
bullying. Yet, antisocial behavior also entails behaviors like lying, mistreating and aggression (Rigter, 2013) and these were not assessed. Therefore the results might not be representative for the complete expression of antisocial behavior. A last limitation is that the data only consists of adolescents’ self-report questionnaires. Reliability of self-reports are generally sufficient, but the possibility of a memory or social desirability bias, inconsistent motivation, and the way the person is feeling at the moment, may all affect the outcome (Paulhus & Vazire, 2000). Moreover, if only adolescents’ perspectives are incorporated, important information may be missed since parents or teachers might have another perception about the functioning of the adolescent (Paulhus & Vazire, 2000).
Despite these limitations, the present study has several strengths as well. Strong aspects of this study are it’s longitudinal design and sufficient sample size (Bryman, 2012). In this way, generalizable causal inferences and changes could be discovered over time. In addition, because of implementing covariates such as age, gender, education level and due to controlling for earlier manifestations of antisocial behavior at baseline, this design resulted in a relatively small variance to be left explained. Hence, this design decreases the probability of finding significant results and also means that if significant results are found they have a relatively high effect size (Bryman, 2012).
Inferences
Our findings indicate the importance of studying gender differences in the context of the development of adolescents’ antisocial behavior. Researchers and clinicians should always
take gender differences regarding etiology and personality into account, because the
appropiate approach may depend on gender. In this way, more profound knowledge could be crystalized and treatment might be more effective. In addition, our findings indicate the importance of early intervention and/or monitoring in families where one or both of the parents are suffering from psychopathological problems. If these problems appear (i.e., one or both of the parents are in contact with a psychiatrist, with the justice system or having
substance problems in any way), it could be helpful to the behavioral development of their child(ren) to be part of some kind of ambulatory care services programme. If children from parents with psychopathological problems develop any pattern of misconduct or if other signs of maldevelopment appear, the programme perhaps can easily intervene and support the family so that children may have more opportunities to a healthy development. An example of such a programme in the Netherlands is the Children of Parents with Psychiatric Problems Kinderen van Ouders met Psychische Problemen [KOPP]-programme. Nevertheless, if parents receive an appropiate treatment for their own problems, as a result their parenting and subsequently their children’s behavior should improve (Kim-Cohen, Moffitt, Taylor, Pawlby, & Caspi, 2005), but it is important for clinicians to carefully monitor this process.
Our recommendation for future research on predicting adolescents’antisocial behavior is to replicate the results of this study with the complete expression of antisocial behavior and to use multiple informants, in to establish the robustness of effects. In addition, the research should include data on other important environmental agents like peers, should take gender-specific predictors into account, and possibly also contain a clinical (control) sample for more comparable and more extreme results of both genders. If these recommendations are
considered it becomes possible to yield a more comprehensive understanding on the gender-specific development of adolescents’ antisocial behavior.
Conclusion
This study contributed to the knowledge of risk factors of the development of
adolescents’ antisocial behavior. Remarkably, it appeared that gender differences are vital in this respect. However, more research is needed. In our study only for girls low levels of conscientiousness, and familial psychopathological risk were predictive for an antisocial development with low levels of agreeableness and conscientiousness further strengthening this latter relationship. Our findings indicate that in families where one or both parents may suffer from psychopathological problems the behavioral development of their child(ren) should be acutely monitored. When interventions for children’s behavior are indicated one should be alert to gender differences regarding personality and etiology since the appropiate approach to intervene may be dependent on gender.
References
Agnew, R., Brezina, T., Wright, J. P., & Cullen, F. T. (2002). Strain, personality traits and delinquency: Extending general strain theory. Criminology, 40, 43-72.
doi:10.1111/j.1745-9125.2002.tb00949.x
Baker, L. A., Jacobson, K. C., Raine, A., Lozano, D. I., & Bezdijan, S. (2007). Genetic and environmental bases of childhood antisocial behavior: A multi-informant twin study. Journal of Abnormal Psychology, 116, 219-235. doi:10.1037/0021-843x.116.2.219 Bardone, A. M., Moffitt, T. E., Caspi, A., Dickson, N., Stanton, W. R., & Silva, P. A. (1998).
Adult physical health outcomes of adolescent girls with conduct disorder, depression and anxiety. Journal of the American Academy of Child & Adolescent Psychiatry, 37, 594-601. doi:10.1097/00004583-199806000-00009
Belsky, J. (1997). Variation in Susceptibility to environmental influence: An evolutionary argument. Psychological Inquiry, 8, 182-186. doi:10.1207/s15327965pli0803_3 Boendermaker, L. (2008). De ontwikkeling van gedragsstoornissen. Retrieved from
http://www.nji.nl/nl/Ontwikkeling_gedragsstoornissen.pdf
Bonin, E-M., Stevens, M., Beecham, J., Byford, S., & Parsonage, M. (2011). Costs and longer-
term savings of parenting programmes for the prevention of persistent conduct disorder: A modelling study. Britisch Medical Journal, 11, 803. doi:10.1186/1471-2458-11-803
Bryman, A. (2012). Social Research Methods. Oxford, UK: Oxford University Press. Buker, H. (2011). Formation of self-control: Gottfredson and Hirschi’s general theory of
crime and beyond. Aggression and Violent Behavior, 16, 265-276. doi:10.1016/j.avb.2011.03.005
change. Annual Review of Psychology, 56, 453-484. doi:10.1146/annurev.psych.55.090902.141913
Chhangur, R. R., Weeland, J., Matthys, W., & Overbeek, G. (2015). Gene by environment research to prevent externalizing problem behavior: Ethical questions raised from a public healthcare perspective. Public Health Ethics, 8, 295-304.
doi:10.1093/phe/phv024
COTAN Documentation NIP (2015). Quick Big Five, QBF, 2009. Retrieved from http://www.cotandocumentatie.nl/test_details.php?id=560
Crone, E. (2014). Het puberende brein. Amsterdam, Netherlands: Prometheus.
De Looze, M., Van Dorsselear, S., De Roos, S., Verdurmen, J., Stevens, G., Gommans, R., Van Bon-Martens, M., Ter Bogt, T., & Vollebergh, W. (2014). HBSC 2013:
Gezondheid, welzijn en opvoeding van jongeren in Nederland. Retrieved from Health Behavior in School-aged Children website http://www.hbsc-nederland.nl/uploads/ publicaties/openbaar/HBSC_Rapport_2013.pdf
Diekhof, E. K., Keil, M., Obst, K. U., Henseler, I., Dechent, P., Falkai, P., & Gruber, O. (2012). A functional neuroimaging study assessing gender differences in the neural mechanisms underlying the ability to resist impulsive desires. Brain Research, 1473, 63-77. doi:10.1016/j.brainres.2012.07.010
Ferguson, C. J. (2010). Genetic contributions to antisocial personality and behavior: A meta- analytic review from an evolutionary perspective. Journal of Social Psychology, 150, 160-180. doi:10.1080/00224540903366503
Frick, P. J., Lahey, B. B., Loeber, R., Stouthamer-Loeber, M., Christ, M. A. G., & Hanson, K. (1992). Familial risk factors to oppositional defiant disorder and conduct disorder: Parental psychopathology and maternal parenting. Journal of Consulting and Clinical Psychology, 60, 49-55. doi:10.1037/0022-006x.60.1.49
Foolen, N., Ince, D., De Baat, M., & Daamen, W. (2013). Wat werkt bij gedragsproblemen en gedragsstoornissen? Retrieved from Nederlands Jeugdinstituut [NJI] website
http://www.nji.nl/nl/WatWerkt_Gedragsproblemen.pdf
Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4, 26-42. doi:10.1037/1040-3590.4.1.26
Gorman-Smith, D., & Loeber, R. (2005). Are developmental pathways in disruptive behaviors the same for girls and boys? Journal of Child and Family Studies, 14, 15-27.
doi:10.1007/s10826-005-1109-9
Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press.
Houtzager, B., & Baerveldt, C. (1999). Just like normal: A social network study of the relation between petty crime and the intimacy of adolescent friendships. Social Behavior and Personality, 27, 177-192. doi:10.2224/sbp.1999.27.2.177
Jensen-Campbell, L.A., Knack, J. M., Waldrip, A. M., & Campbell, S. D. (2007). Do Big Five personality traits associated with self-control influence the regulation of anger and aggression? Journal of Research in Personality, 41, 403-424.
doi:10.1016/j.jrp.2006.05.001
Jones, S. E., Miller, J. D., & Lynam, D. R. (2011). Personality, antisocial behavior, and agression: A meta-analytic review. Journal of Criminal Justice, 39, 329-337. doi:10.1016/j.jcrimjus.2011.03.004
Kiekens, G., Bruffaerts, R., Nock, M. K., Van de Ven, M., Witteman, C., Mortier, P., Demyttenaere, K., & Claes, L. (2015). Non-suicidal self-injury among Dutch and Belgian adolescents: Personality, stress and coping. European Psychiatry, 30, 743-749. doi:10.1016/j.eurpsy.2015.06.007
depression and children’s antisocial behavior: Nature and nurture effects. Archives of General Psychiatry, 62, 173-181. doi:10.1001/archpsyc.62.2.173
Klimstra, T. A., Akse, J., Hale III, W. W., Raaijmakers, Q. A. W., & Meeus, W. H. J. (2010). Longitudinal associations between personality traits and problem behavior symptoms in adolescence. Journal of Research in Personality, 44, 273-284.
doi:10.1016/j.jrp.2010.02.004
Klimstra, T. A., Luyckx, K., Hale III, W. W., & Goossens, L. (2014). Personality and externalizing behavior in the transition to young adulthood: The additive value of personality facets. Social Psychiatry and Psychiatric Epidemiology, 49, 1319-1333. doi:10.1007/s00127-014-0827-y
Kuhn, E. S., & Laird, R. D. (2013). Parent and peer restrictions of opportunities attenuate the link between low self-control and antisocial behavior. Social Development, 22, 813-830. doi:10.1111/sode.12028
Loeber, R. (1990). Development and risk factors of juvenile antisocial behavior and delinquency. Clinical Psychology Review, 10, 1-41. doi:10.1016/0272-7358(90) 90105- J
Loeber, R., & Stouthamer-Loeber, M. (1986). Family factors as correlates and predictors of juvenile conduct problems and delinquency. Crime and Justice, 7, 29-149.
doi:10.1086/449112
Loeber, R., Van der Laan, P. H., Slot, W., & Hoeve, M. (2008). Conclusions and
recommendations. In R. Loeber, N. W. Slot, P. H. Van der Laan, & M. Hoeve (Eds.), Tomorrow’s criminals: The development of child delinquency and effective
interventions (pp. 261-284). Farnham, England: Ashgate.
McCrae, R. R. (2011). Personality theories for the 21st century. Teaching of Psychology, 38, 209-214. doi:10.1177/0098628311411785
Miller, J. D., & Lynam, D. (2001). Structural models of personality and their relation to antisocial behavior: A meta-analytic review. Criminology, 39, 765-798.
doi:10.1111/j.1745-9125.2001.tb00940.x
Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological review, 100, 674-701. doi:10.1037/0033-295x.100.4.674
Overbeek, G., Vollebergh, W., Meeus, W., Engels, R., & Luijpers, E. (2001). Course, co- occurence, and longitudinal associations of emotional disturbance and delinquency from adolescence to young adulthood: A six-year three-wave study. Journal of Youth and Adolescence, 30, 401-426. doi:10.1023/A:1010441131941
Overbeek, GJ, Zeevalkink, H., Vermulst, A., & Scholte R. H. J. (2010). Peer-victimization, self-esteem, and ego resilience types in adolescents: A prospective analysis of person-context interactions. Social Development, 19, 270-284.
doi:10.1111/j.1467-9507.2008.00535.x
Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989). A developmental perspective on antisocial behavior. American Psychologist, 44, 329-335.
doi:10.1037/0003-066X.44.2.329
Paulhus, D. L., & Vazire, S. (2000). The self-report method. In H. T. Reis & C. M. Judd (Eds.), Handbook of Research Methods in Social and Personality Psychology (pp. 224-239). Cambridge, United Kingdom: Cambridge University Press.
Pitzer, M., Jennen-Steinmetz, C., Esser, G., Schmidt, M. H., & Laucht, M. (2011).
Differential susceptibility to environmental influences: The role of early temperament and parenting in the development of externalizing problems. Comprehensive
Psychiatry, 52, 650-658. doi:10.1016/j.comppsych.2010.10.017
theory of crime: A meta-analysis. Criminology, 38, 931-964. doi:10.1111/j.1745-9125.2000.tb00911.x
Reef, J., Van Meurs, I., Verhulst, F. C., & Van der Ende, J. (2010). Children’s problems predict adults’ DSM-IV disorders across 24 years. Journal of the American Academy of Child and Adolescent Psychiatry, 49, 1117-1124. doi:10.1016/j.jaac.2010.08.002 Rigter, J. (2013). Handboek ontwikkelingspsychopathologie bij kinderen en jeugdigen.
Bussum, The Netherlands: Coutinho.
Rutter, M., Kim-Cohen, J., & Maughan, B. (2006). Continuities and discontinuities in psychopathology between childhood and adult life. Journal of Child Psychology and Psychiatry, 47, 276-295. doi:10.1111/j.1469-7610.2006.01614.x
Sachse, H. (2013). Transitie jeugdzorg en jeugdgezondheidszorg: Wat gaat er gebeuren? Tijdschrift voor gezondheidswetenschappen, 91, 460-462. doi:10.1007/s12508-013-0150-0
Scholte, R. H. J., Van Aken, M. A. G., & Van Lieshout, C. F. M. (1997). Adolescent personality factors in self-ratings and peer nominations and their predictions of peer acceptance and peer rejection. Journal of Personality Assessment, 69, 534-554. doi:10.1207/s15327752jpa6903_8
Scott, S., Knapp, M., Henderson, J., & Maughan, B. (2001). Financial cost of social exclusion: Follow up study of antisocial children into adulthood. Britisch Medical Journal, 323, 191-194. doi:10.1136/bmj.323.7306.191
Scourfield, J., Van den Bree, M., Martin, N., & McGuffin, P. (2004). Conduct problems in children and adolescents: A twin study. Archives of General Psychiatry, 61, 489. doi:10.1001/archpsyc.61.5.489
(2014). Coercive family process and early-onset conduct problems from age 2 to school entry. Development and Psychopathology, 26, 917-932.
doi:10.1017/s0954579414000169
Solberg, M. E., Olweus, D., & Endresen, I. M. (2007). Bullies and victims at school: Are they the same pupils? Britisch Journal of Educational Psychology, 77, 441-464.
doi:10.1348/000709906x105689
Steketee, M., Junger, M., & Junger-Tas, J. (2013). Sex differences in the predictors of juvenile delinquency: Females are more susceptible to poor environments; males are influenced more by low self-control. Journal of Contemporary Criminal Justice, 29, 88-105. doi:10.1177/1043986212470888
Thijs, P., Van Dijk, I., Stoof, R., & Notten, N. (2014). Genderverschillen en deviant gedrag. Mens en Maatschappij, 89, 305-335. doi:10.5117/mem2014.3.thij
Tiet, Q. Q., Wasserman, G. A., Loeber, R., McReynolds, L. S., & Miller, L. S. (2001).
Developmental and sex differences in types of conduct problems. Journal of Child and Family Studies, 10, 181-197. doi:10.1023/a:1016637702525
Van der Maas, J. J. (2010). Diagnostiek van ouderfunctioneren: Een functioneel-contextueel perspectief (Doctoral dissertation, Free University of Amsterdam, The Netherlands). Retrieved from: http://dare.ubvu.vu.nl/bitstream/handle/1871/18782/dissertation.pdf Van der Merwe, A., & Dawes, A. (2007). Youth violence: A review of risk factors, causal
pathways and effective intervention. Jounal of Child and Adolescent Mental Health, 19, 95-113. doi:10.2989/17280580709486645
Widiger, T. A., Trull, T. J., Clarkin, J. F., Sanderson, C., & Costa, P.T. (2002). A description of the DSM-IV personality disorders with the five-factor model of personality. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the five-factor model of personality (2nd ed.) (pp. 89-99). doi:10.1037/10423-006
Table 1
Means and standard deviations
Boys Girls Variable M SD n M SD n Familial psychopathological risk 4.14 0.51 322 4.21 0.65 364 Antisocial behavior at T1 28.15 4.90 93 26.57 3.94 40 Antisocial behavior at T3 33.38 10.12 84 30.15* 5.44 35 Extraversion 29.47 6.72 313 28.90 7.22 338 Agreeableness 32.18 4.89 356 33.33** 4.30 391 Conscientiousness 24.81 6.74 348 26.34** 6.65 387 Emotional Stability 27.46 6.48 318 25.36*** 6.58 351 Openness to experience 27.52 5.64 356 28.04 5.28 391 Note. N = 560. * p < .05. ** p < .01. *** p < .001.
Table 2
Pearson Correlations
Variable 1 2 3 4 5 6 7 8 9
1. Familial psychopath. risk 1 -.073 -.119* -.029 -.078 -.043 .105 -.028 .078 2. Extraversion .063 1 .086 -.171** .554*** .051 .199 .429** .138 3. Agreeableness -.094 .090 1 .375*** .047 .361*** -.382* .032 -.234 4. Conscientiousness -.115* -.155** .399*** 1 -.104 .154** -.518** -.349* -.057 5. Emotional Stability .004 .613*** -.067 -.146** 1 -.135* .174 .289 .081 6. Openness to experience -.003 -.185** .439*** .191*** -.291*** 1 -.224 .271 .012 7. Antisocial behavior at T1 .267* -.017 -.098 -.107 -.089 .011 1 -.061 .271 8. Antisocial behavior at T2 .118 .069 -.111 -.217 -.001 -.157 .284 1 .453 9. Antisocial behavior at T3 .225 -.020 -.072 -.070 -.266* .047 .310 .361* 1 Note. Correlations for girls (n = 405) are presented in the upper right triangle of the table; correlations for boys (n = 369) are presented in the lower left triangle of the table. Psychopath. = psychopathological.
* p < .05. ** p < .01. *** p < .001.
Table 3
Main Effects of Predictors on Adolescents’ Antisocial Behavior for both genders
Predictor OR 95% CI for OR p Constant 1.45 .887 Covariates Age 0.80 [0.57, 1.12] .191 Gender 3.59 [2.00, 6.31] <.001*** Education level (0) .078 Education level (1) 0.74 [0.37, 1.65] .465 Education level (2) 0.93 [0.49, 1.76] .821 Education level (3) 0.19 [0.06, 0.69] .011* Antisocial behavior T1 2.93 [1.64, 5.21] <.001*** Independent Variables Familial Risk 1.47 [1.01, 2.14] .045* Extraversion 1.00 [0.96, 1.05] .880 Agreeableness 1.00 [0.94, 1.07] .970 Conscientiousness 0.96 [0.92, 1.00] .077 Emotional Stability 0.98 [0.93, 1.04] .521 Openness to Experience 1.00 [0.94, 1.05] .857 Nagelkerke R2 .223
Note. N = 560. CI = confidence interval; p = probability value; (0) = lower vocational education; (1) = mix of lower vocational and higher general secondary education; (2) = higher general secondary education; (3) = pre-university education.
Table 4
Main Effects of Predictors on Adolescents’ Antisocial Behavior divided by gender
Boys Girls Predictor OR 95% CI p OR 95% CI p Constant 0.89 .970 3304.35 .141 Covariates Age 1.02 [0.69, 1.51] .919 0.37 [0.17, 0.82] .014* Education level (0) .076 .687 Education level (1) 0.94 [0.36, 2.43] .892 0.38 [0.08, 1.96] .250 Education level (2) 1.05 [0.50, 2.19] .906 0.70 [0.16, 3.00] .631 Education level (3) 0.07 [0.01, 0.52] .010* 0.66 [0.12, 3.79] .643 Antisocial behavior T1 2.88 [1.45, 5.70] .002** 5.45 [1.56, 19.07] .008** Independent Variables Familial Risk 0.97 [0.57, 1.65] .905 2.43 [1.38, 4.28] .002** Extraversion 1.02 [0.96, 1.09] .498 0.97 [0.88, 1.06] .465 Agreeableness 1.01 [0.93, 1.09] .890 1.02 [0.90, 1.15] .779 Conscientiousness 0.98 [0.93, 1.03] .436 0.90 [0.82, 0.98] .019* Emotional Stability 0.96 [0.90, 1.02] .225 1.03 [0.93, 1.13] .610 Openness to Experience 0.98 [0.91, 1.05] .533 1.04 [0.93, 1.15] .541 Nagelkerke R2 .179 .247
Note. Boys N = 262. Girls N = 298. CI = confidence interval; p = probability value; (0) = lower vocational education; (1) = mix of lower vocational and higher general secondary education; (2) = higher general secondary education; (3) = pre-university education.
Table 5
Interaction of Predictors on Adolescents’ Antisocial Behavior for both genders
Predictor OR 95% CI for OR p Constant 0.01 .540 Covariates Age 0.80 [0.57, 1.14] .214 Gender 3.85 [2.12, 7.00] <.001*** Education level (0) .072 Education level (1) 0.72 [0.32, 1.63] .427 Education level (2) 0.91 [0.48, 1.73] .772 Education level (3) 0.19 [0.05, 0.67] .010* Antisocial behavior T1 2.84 [1.56, 5.17] .001** Independent Variables Familial Risk 4.27 [0.17, 105.04] .375 Extraversion 1.41 [0.91, 2.18] .129 Agreeableness 0.84 [0.56, 1.26] .386 Conscientiousness 1.10 [0.79, 1.55] .573 Emotional Stability 1.03 [0.67, 1.60] .886 Openness to Experience 0.88 [0.57, 1.37] .580 Interactions Familial Risk ∙ EX 0.92 [0.83, 1.03] .139 Familial Risk ∙ AG 1.04 [0.95, 1.15] .374 Familial Risk ∙ CO 0.97 [0.89, 1.05] .426 Familial Risk ∙ ES 0.99 [0.89, 1.10] .833 Familial Risk ∙ OE 1.03 [0.93, 1.14] .580 Nagelkerke R2 .246
Note. N = 560. CI = confidence interval; p = probability value; (0) = lower vocational education; (1) = mix of lower vocational and higher general secondary education; (2) = higher general secondary education; (3) = pre-university education; EX = extraversion; AG = agreeableness; CO = conscientiousness; ES = emotional stability; OE = openess to experience.
