Becoming a criminal:
Longitudinal associations from low-quality parent-child relationship and familial risk for psychopathology to adolescents’ delinquency
Masterthesis Forensic Child and Youth Care Sciences Graduate School of Child Development and Education University of Amsterdam M. H. Piels 10882049 Under the supervision of prof. dr. G. J. Overbeek Second assessment by dr. I.B. Wissink Amsterdam, July 2016
Abstract
This study examined to what extent a lower quality of the parent-child relationship would predict for the development of adolescents’ delinquent behavior. In addition, we examined whether this predictive relationship would be stronger for adolescents with a higher familial risk for psychopathology. Data was used from 774 adolescents (405 girls and 369 boys) between 10 and 17 year old who participated in a Dutch three-wave (i.e., 2005, 2006 and 2007) longitudinal self-report study. Multiple regression analysis showed that a familial risk for psychopathology was a significant predictor of the development of adolescents’
delinquent behavior, but that lower quality parent-child relationships did not predict more delinquency over time. Moreover, the interaction between the quality of the parent-child relationship and familial risk was significant, showing that the effects of familial
psychopathological risk on developing delinquent behavior is stronger if adolescents have a low quality parent-child relationship. Future longitudinal studies should show how the quality of the relationship with parents may provide a buffer of the effects of risk factors.
Keywords: delinquent behavior, parent-child relationship, familial risk,
psychopathology, adolescence
• U (10882049) heeft met succes uw document voor de opdracht Scriptie inleveren in cursus 2533S001.7014B457HY.AJ.EXCL.2015 verstuurd; het document wordt nu verstuurd naar Ephorus voor de plagiaatcontrole.
• De naam van het verstuurde document is MaschaPiels-7014B457GY-2..., u kunt het document
downloaden als u wilt controleren of het goed is doorgekomen.
• Het Ephorus ID is 69cb98ec-1292-4758-b500-e954f8989377 / 20160731T211700; maak eventueel een screenshot of copy/paste deze regels als bewijs van inleveren.
• U heeft deze opdracht al ingeleverd. zondag 31 juli 2016 21:17:43 uur CEST
Longitudinal associations from low quality parent-child relationship and familial risk for psychopathology to adolescents’ delinquency
Juvenile delinquency is a common problem in the Netherlands. Self-reports by adolescents show that about 40% of adolescents between 12 and 18 year old committed a criminal offense in 2010 (Van der Laan & Blom, 2011). This delinquency costs society a lot of money. Specifically, costs of care in juvenile justice institutions in 2007 were
approximately 306 million euros, which is just a fraction of all societal expenses necessary to deal with the effects of delinquency (Van Overveld, 2010). Based on increasing media coverage of rising crime rates in adolescents, societal concerns about issues of safety have increased (Boendermaker & Ince, 2008; Loeber, Van der Laan, Slot, & Hoeve, 2008). It is, therefore, necessary to examine what contributes to the development of delinquent behavior. Studies have already shown that a major predictor for the development of delinquent behavior is the quality of the parent-child relationship (Asscher, Wissink, Dekovic, Prinzie, & Stams, 2014; Hoeve et al., 2012; Loeber, 1990). In addition to the parent-child relationship, familial risk for psychopathology also appears to play a role in adolescents’ delinquent behavior (Ferguson, 2010; Loeber et al., 2008; Rigter, 2013). However, to our knowledge there is a lack of studies involving interactions between these predictors. In this study, therefore, our aim is to examine to what extent both familial risk for psychopathology and the quality of the parent-child relationship are predictive for the development of adolescents’ delinquent behavior. Also, we examine whether higher familial risk for psychopathology is more
strongly related to delinquency for adolescents with a lower quality relationship with parents. To a certain extent adolescent’s delinquent behavior may be considered a sign of normal, healthy and temporary development (Luijpers, 2000; Moffitt, 1993; Overbeek, Vollebergh, Meeus, Engels, & Luijpers, 2001). Adolescents tend to behave dangerously, disruptively and in a delinquent manner due to hormonal changes and changes in brain
structures (Crone, 2014; Luijpers, 2000). Adolescents’ development is characterised by increased arguing with parents about rules and by testing boundaries in different social
environments (e.g., in the family, with peers or in school). Sensation seeking – mainly in front of peers – and rule breaking and doing prohibited things are all more prevalent in adolescence (Luijpers, 2000; Overbeek et al., 2001; Weerman, 2007). Adolescent delinquent behavior (or “juvenile delinquency”) in this study is defined as a collective term for different types of behaviors that, according to established rules and laws, is punishable (Van der Laan & Blom, 2006). Studies have shown that boys report more delinquency than girls, a lower educational level is associated with more delinquent behaviors and that juvenile delinquency increases from early to mid-adolescence, shows a peak at age 17 and then steadily declines from middle to late adolescence, whereby most adolescents develop to responsible adults (Hirschi & Gottfredson, 1983; Luijpers, 2000; Moffitt, 1993; Overbeek et al., 2001; Van der Laan & Blom, 2006). Thus, in predicting delinquency it is important to account for these gender, age and educational differences.
Familial Risk for Psychopathology
How is it possible that in a small minority of youths, this generally normative risk behavior leads to serious and persistent behavior problems (Luijpers, 2000; Moffitt, 1993)? Besides age, gender and educational-related differences in adolescents’ development of delinquent behavior, familial psychopathological risk may be a key predictor of delinquency. A familial risk for psychopathology can be defined as a factor that has a negative influence on the healthy, mental development of a child (e.g., increases the likelihood of psychopathology and delinquent behavior), exists before the behavior problems are developed and is presented within the family (Boendermaker, 2008; Loeber et al., 2008; Rigter, 2013). Previous research shows that one of the main predictors of developing delinquent behavior was having a
showed that parental substance abuse and having a parent with psychological or psychiatric problems increased the risk of behavioral problems and developing delinquency in children (Loeber et al., 2008; Rigter, 2013; Van de Rakt, Nieuwbeerta, & De Graaf, 2006). In
predicting juvenile delinquency, a familial risk for psychopathology can be explained in two different ways, (1) genetic predisposition, and (2) parents’ poor childrearing practices (Loeber et al., 2008; Raine, 2008; Rigter, 2013).
As for the first explanation, based on genetic predisposition, twin and adoption studies show that delinquent behavior is partially genetic determined (Canton, Glenn, Raine, & Verkes, 2015; Moffitt, 2005; Van de Rakt et al., 2006). A variance of between 40% and 60% of antisocial behavior (e.g. aggression, violence and delinquent behavior) could be explained by children’s genes (Ferguson, 2010; Glenn & Raine, 2014). There is still much uncertainty about how inheritance occurs, yet it is clear that there is not a particular separate
malfunctioning gene solely responsible for the development of delinquency (Canton et al., 2015; Ferguson, 2010). Genes may particularly affect the likelihood of delinquency through their interaction and influence on hormones and neurotransmitters, which affect different brain structures such as the amygdala and prefrontal cortex (Canton et al., 2015; Ferguson, 2010; Glenn & Raine, 2014). These structures are important for, among other things, the establishment of social behavior processes, taking moral decisions, inhibitory control, the tendency towards aggression, and appropriately dealing with reward and punishment.
Maldevelopment in these structures as a result of genes, hormones and neurotransmitters (e.g., monoamine oxidase A [MAOA], catechol O-methyltransferase [COMT], dopamine,
serotonin, cortisol and testosterone) is associated with developing delinquent behavior (Canton et al., 2015; Ferguson, 2010; Glenn & Raine, 2014).
Besides genetic predisposition, familial risk for psychopathology might work through parents’ poor childrearing practices (e.g. lack of supervision, overprotection, inconsistent
discipline and rejection) (Loeber et al., 2008; Rigter, 2013). Research shows that parents’ psychological and/or psychiatric disorders are associated with ineffective parenting. Due to their own problems, parents may have less adequate supervision practices and show less sensitive parenting (Loeber et al., 2008; Rigter, 2013). Although more research is needed, it may be assumed that for good socialisation, an appropriate reward and punishment by parents is necessary and that children are at risk of developing delinquency if parents do not react consistently enough to their behavior (Loeber & Stouthamer-Loeber, 1986; Patterson, DeBaryshe, & Ramsey, 1989; Van den Berg, Cima, Klein Haneveld, & Van de Putte, 2015). So, if the child’s behavior is not being monitored and the parents do not give much attention, the likelihood to commit offences increases without disciplinary consequence from the parents. Without this consequence, children do not learn an appropriate set of social skills nor moral values that are necessary to become responsible adults (Loeber & Stouthamer-Loeber, 1986; Patterson et al., 1989; Van den Berg et al., 2015).
Parent-child Relationship and Delinquency
The development of delinquent behavior can be explained not only by familial psychopathological risk, but perhaps also by low quality of the parent-child relationship (Loeber et al., 2008; Rigter, 2013). There are two theories that support the influence of low-quality parent-child relationships on an adolescents’ development of delinquent behavior. First, the attachment-theory from Bowlby and second the coercion theory of Patterson. Both theories can explain why the parent-child relationship is a predictor for delinquency.
Bowlby (1988) indicates the importance of creating a stable, loving and close
relationship with primary caregivers (usually parents). This bond is described as attachment, which is characterised by sensitive and responsive communication, trust in the parent and given proximity by the parent (Armsden & Greenberg, 1987; Buist, Dekovic, Meeus, & Van Aken, 2004; Rigter, 2013). If parents do not respond sensitively and responsively to the needs
of the child, the child may develop insecure attachment, which is seen as a risk factor for the development of both mental disorders and delinquency alike (Goudena, 1994; Hoeve et al., 2012; Rigter, 2013). Because of an insecure attachment, children do not see themselves as worthwhile, and perceive the social environment as unreliable and unpredictable. This weakens the adolescents’ bond with important social figures in their environment and thus increases the risk of developing delinquency (Buist et al., 2004; Van IJzendoorn, 1997).
A second theory often used to explain delinquent behavior in adolescents is the coercion theory of Patterson (1982; Patterson et al., 1989; Rigter, 2013). This theory explains delinquent behavior in adolescents through a circular pattern of negative interactions between parents and adolescents. For instance, in such a circular pattern of negative interactions a child acts agitated and irritable, which leads parents to manifest negative emotions towards their child and react with controlling and coercive measures. In a coercive cycle, the child reacts with resistance and more deviant behavior and this pattern continues. Through this circular pattern parents and adolescents train each other to behave in such a way that the chance of children developing externalising, disruptive, and even delinquent behavior increases. The theory states that if parents do not break this pattern of negative interaction, adolescents will also exhibit this behavior toward peers and adults other than their parents (Patterson, 1982; Patterson et al., 1989; Rigter, 2013). This is supported by the failure model of Capaldi, which explains that disruptive and delinquent behavior and a lack of (social) skills mostly leads to rejection and receiving little to no support by important others (e.g. parents and peers), which in turn can lead to experiences of failure in social environments, what can again exacerbate aggravated behavior problems (Capaldi, 1992; Patterson & Capaldi, 1990; Rigter, 2013).
In line with these theories, the overall quality of the parent-child relationship is a notable risk factor that has had extensive empirical study (Asscher et al., 2014; Hoeve et al.,
2012; Loeber, 1990). Previous meta-analyses and longitudinal studies associated poor quality of the relationship (i.e. attachment) to parents with juvenile delinquency (Asscher et al., 2014; Bowlby, 1944; Hirschi, 1969; Hoeve et al., 2009; Hoeve et al., 2012; De Kemp, Scholte, Overbeek, & Engels, 2006). In addition, strong associations were found between specific parenting behaviors (i.e., low parental support, low monitoring and psychological control) and delinquency (Hoeve et al., 2009; De Kemp et al., 2006). It is known that such parenting behaviors are also related to a lower quality of the parent child relationship (Hoeve et al., 2012; Loeber et al., 2008; Rigter, 2013; Wissink, Dekovic, & Meijer, 2006). In predicting adolescents’ delinquent behavior, the quality of the parent-child relationship explained more of the variance than parenting behavior (Wissink et al., 2006).
Gene by Environment Interactions [G x E]
Where previous studies have either looked at the influence of familial risk or the quality of the parent-child relationship only, it seems more logical to look at their combined effect based on an interaction model (Van Beek & Canton, 2015). Most human behavior is the effect of complicated interactions between specific genetic and environmental factors
(Chhangur, Weeland, Matthys, & Overbeek, 2015). Such gene-environment interactions (GxE) may result in a higher risk in developing delinquent behavior. This means delinquent behavior may increase when children with heightened familial risk also perceive the quality of the relationship with their parents to be low (Chhangur et al., 2015; Guo, Roettger, & Cai, 2008; Loeber et al., 2008). Thus, for the present study we argue that familial risk may especially be detrimental in the context of low-quality family relationships. Understanding such interactions could gain insight to develop interventions and possibly prevent juvenile delinquency.
The Present Study
As reviewed above, the current literature shows a lack of knowledge about the extent to which the variance in adolescents’ delinquency is explained by combined familial risk and low-quality family functioning (Canton et al., 2015). Therefore, the aim of this study was to examine – apart from the main effects of a low-quality parent-child relationship and familial risk for psychopathology – whether the development of delinquent behavior would be predicted by the interaction of heightened familial risk and a low-quality parent-child relationship. We hypothesised that a familial risk for psychopathology predicts adolescents’ development of delinquent behavior and that a low quality of the relationship with parents predicts juvenile delinquency. We also expect that the familial psychopathological risk increases the likelihood of developing delinquent behavior if adolescents perceive a low quality parent-child relationship. In this longitudinal study, these questions were examined through self-report data from 774 Dutch adolescents between 10 and 17 year olds, collected over a period of three years (i.e. 2005, 2006 and 2007).
Method Sample
We used an existing dataset from the SODA [SOcial Development of Adolescents] study (see Overbeek, Zeevalkink, Vermulst, & Scholte, 2009) to analyse the research questions. This dataset was used because it precisely matched our research questions. The participants in this study were Dutch adolescents who participated in a three-wave
longitudinal survey study. The dataset comprises a total of 2475 participants, who completed the questionnaire at T1.At T2, 57% (1419 participants) was left and at T3, 774 participants
(31%) remained. The high attrition rate is attributed to participants that left school due to graduating and to changes in class composition by the Dutch school system, whereby participants couldn’t be traced. To acquire an optimal sample size, only classes with at least
seven participants – who participated at T1 – were included at T2 and T3. This sample contains
a total of 405 girls (52.3%) and 369 boys (47.7%). The age range, at T1, was between 10 and 17 years (M = 13.6, SD = .89), with the average age of 13.6 year for girls (SD = .86) and 13.7 year for boys (SD = .92). Most of the participants (97.6%) were from native Dutch
background and 89.1% came from families with two parents. 375 participants (49.4%) followed lower vocational education, 110 participants (14.5%) followed combined classes of lower vocational and senior general secondary education, 159 participants (20.9%) followed senior general secondary education and 114 participants (15%) pre-university education. Procedure
In the SODA study, 28 high schools in a 100-kilometre radius around the city of Nijmegen, The Netherlands, were selected and approached to participate by being sent an introductory letter about the study. In support of the letter there was a consultation by telephone. The participants were selected from the 23 high schools (82%) who decided to participate in the study. In collaboration with the school administration, the research team decided which and how many classes would be selected within each school. An informed consent procedure was used: both parents and participants were informed about the content and purpose of the study. All parents agreed to the participation of their children. The
questionnaires were administered to the participants by undergraduate students. Before these students collected the data, they were informed about the content of the questionnaire and the administration procedure in a classroom situation. In most classes, a teacher was also present to help keep order. From January to March 2005, participants filled in the questionnaire in their class during the time of a regular lesson (45-50 minutes). The participants were informed that they were not allowed to talk about their answers with other participants and assured that their information would be processed anonymously and would not be shared with others, such as parents or teachers. Each school received a research report on the social development of all
the participants (without providing information that identified individual participants), after finishing the data collection. For the data collection of T2 (2006) and T3 (2007) identical
procedures were used. Measures
Familial risk for psychopathology. This variable was measured at T1 and contained
four items that focused at the risk factors of parents on the development of delinquent
behavior in adolescents. The first three questions are related to the psychological background of parents (1) “Is your father or mother under treatment – or have they been treated – by a psychiatrist or psychologist?”, (2) “Is your father or mother incarcerated – or have they been incarcerated – in a mental institution?”, and (3) “Do your parents currently have – or have they formerly had – an alcohol and/or drug problem?”. The last and fourth question was about the criminogenic environment and was conducted by (4) “Do your parents have, on a regular basis, problems – or have they had any problems – with the justice system?”. Answering categories ranged from “Yes, my father”, “Yes, my mother”, Yes, both my parents” and “No”. Based on these categories, a familial risk index was composed, which increases from no manifest familial risk (1 = no, neither of the two parents), to a moderate familial risk (2 = yes, one of my parents), to a strong familial risk (3= yes, both of my parents), on at least one of the above-mentioned items. This summed score formed a new continuous variable, where higher scores indicated a stronger familial risk for psychopathology. Correlations between the familial risk items ranged from |.05| to |.25|.
Parent-child relationship. The quality of the parent-child relationship was measured at
each wave using the short version of the Inventory of Parent and Peer Attachment [IPPA] (Armsden & Greenberg, 1987; Nada-Raja, McGee, & Stanton, 1992). The IPPA is a self-report questionnaire which measures the perception in adolescents of the quality of the relationship with their parents (Armsden & Greenberg, 1987; Buist et al., 2004; Nada-Raja et
al., 1992). The IPPA contains 12 questions about the relationship with their mother, and 12 about the relationship with their father. In this study, we focused at the total score of all the 24 items. The items measure the perceived quality of the communication with their parents, mutual trust, understanding and respect, and the perceived feelings of alienation and isolation. Example items are “I tell my mother/father about my problems and troubles”, “My
mother/father accept me as I am” and “Talking over my problems with my mother/father makes me feel ashamed or foolish”. The items were scored on a 6-point Likert scale ranging from 1 (never) to 6 (always). Low scores indicated a low quality of the parent-child
relationship and high scores a high quality of this relationship (Armsden & Greenberg, 1987; Buist et al., 2004; Nada-Raja et al., 1992). In this sample, the reliability (Cronbach’s alpha) of both parents together was sufficient for all three waves, .74 for T1, .70 for T2 and .71 for T3.
Delinquent behavior. Delinquency was measured at each wave using a Dutch
self-report questionnaire containing 13 items about specific criminal activities (Houtzager & Baerveldt, 1999). Participants were asked whether they had engaged in minor criminal activities in the past 12 months. The questions are about different types of minor criminal activities, such as stealing, vandalism, burglary, arson, fighting and injuring someone. For instance, “Intentionally damaging school property, defacing or destroying (building, furniture, books, trees or shrubs)” and “Someone, with your hand or an object, beaten or injured that they had to see a doctor”. The items were scored on a 5-point Likert scale ranging from 1 (0
times) to 5 (more than 12 times). Low scores on the continuous delinquency scale indicated a
low delinquent behavior and high scores a high delinquent behavior (Houtzager & Baerveldt, 1999). Cronbach’s alpha coefficients of the delinquency-scale in this sample were all
Results
First, we analysed the mean scores on the independent variables. Table 1 shows that for all participants at T1, the reported mean level of the quality of the parent-child relationship was rather high, and that the mean levels of reported familial risk for psychopathology and reported delinquent behavior were fairly low. To examine if there were gender differences, the mean differences between girls and boys on the quality of the parent/child relationship at T1, the familial risk for psychopathology at T1, and adolescents´ earlier level of delinquency at T1 were examined with t tests, using gender as the independent variable. These t-tests showed that most of the participants – mean scores were comparable for boys (104.89) and girls (104.47) (t(739) = -.394, p = .694) – reported to have a high-quality relationship with their parents. Of all participants, 78.4% reported no manifest familial risk (i.e., in neither parent) and 10.3% did report familial risk in at least one parent. The mean score on familial risk at T1 was 4.18, which indicates that, overall, adolescents had a low familial risk for psychopathology. No gender differences were apparent on this familial risk index (t(684) = 1.682, p = .093). As for delinquency at T1, the total sample (n = 747) had an average of 14.24, which shows that adolescents at T1 generally exhibited low levels of delinquent behavior. Girls (n = 391) had an average of 13.73 and boys (n = 356) an average of 14.81. This difference was statistically significant. Boys reported a higher level (t(745) = -6.643, p < .001), as compared with girls.
Pearson correlations, presented in Table 2, were computed to examine longitudinal bivariate associations between adolescents’ reports of the quality of the parent-child
relationship, familial risk and delinquency. This was done separately for girls and boys. For both girls and boys, most correlations were significant, ranging in magnitude from |.10| to |.47|. Overall, lower quality of the parent-child relationship and higher familial risk were associated with more delinquent behavior. Also, we found a negative association between the
quality of the parent-child relationship and familial risk. This indicates that a lower quality of the parent-child relationship was related to a higher familial risk for psychopathology. In addition, there were significant moderately strong correlations between earlier delinquent behavior at T1 and subsequent delinquency at T2 and T3. This demonstrates that behaving delinquently at T1 was related to having higher levels of delinquency at subsequent
measurement waves as well.
Fisher z tests for comparing correlations between boys and girls showed that boys had a stronger association (r = -.24, p < .001) than girls (r = -.11, p = .028) between quality of the parent-child relationship and delinquency at T1 (z = 1.772, p = .038) (Lenhard & Lenhard, 2014). Boys also had a stronger association (r = -.21, p < .001) than girls (r = -.07, p = .203) between quality of the relationship with parents and delinquency at T2 (z = 1.977, p = .024). For girls, the association between earlier delinquency at T1 and delinquency at T3 (z = 2.099,
p = .018) was stronger than for boys (girls: r = .46, p < .001; boys: r = .33, p < .001). For all
other gender comparisons, the differences in correlations did not reach statistical significance (Lenhard & Lenhard, 2014).
To analyse the first research question, to what extent the quality of the parent-child relationship would be predictive for the development of delinquent behavior in adolescents, we performed a multiple regression analysis (see Table 4). The results of step 1 showed that only gender was significantly predictive of delinquency (β = .253, t(622) = 6.535, p < .001). This means that overall, boys reported more delinquent behavior than girls. Age (β = -.037,
t(622) = -.869, p = .385) and education level (β = -.065, t(622) = -.934, p = .351) were not
predictive of delinquent behavior. Step 1 explained 7.4% of the variance in the development of adolescents’ delinquent behavior. In step 2 we included adolescents’ earlier level of delinquency at T1 as a control variable. The results showed that delinquency at T1 (β = .400,
adolescents who had a higher score on delinquency at T1, also had a relatively high score on delinquency at T3. Step 2 explained 22.4% of the variance in the development of adolescents’ delinquent behavior. In step 3 the quality of the parent-child relationship at T1 and familial risk for psychopathology at T1 were included as predictors of adolescent’s delinquent
behavior at T3, controlling for adolescents’ earlier level of delinquency at T1, age, gender and education level. The analysis indicated that familial risk for psychopathology was
significantly (β = .078, t(619) = 2.152, p = .032) predictive of adolescents’ delinquent
behavior at T3. This means that adolescents with higher familial psychopathological risk, also showed more delinquent behavior at T3. In contrast, the quality of the parent-child
relationship was not significantly predictive (β = -.040, t(619) = -1.115, p = .265) of
delinquency at T3. Step 3 explained 23.2% of the variance in the development of adolescents’ delinquent behavior.
To examine the second research question, whether the association between low-quality of the parent-child relationship and delinquency would be stronger if adolescents had a higher familial risk for psychopathology, in step 4 of this regression analysis an interaction term between quality of the parent-child relationship and adolescents’ familial risk score was entered. The analysis showed that the interaction term significantly (β = -.096, t(618) = -2.457, p = .014) predicted adolescents’ delinquent behavior at T3. Specifically, this
interaction showed that, for adolescents with a low-quality relationship with parents, the variation in scores on delinquent behavior at T3 was more strongly associated with a higher psychopathological familial risk (r = .20, p = .001) than for adolescents with a high-quality relationship with parents (r = .04, p = .485). Step 4 explained 23.9% of the variance in the development of adolescents’ delinquent behavior.
Discussion
The aim of this Dutch three wave (i.e. 2005, 2006 and 2007) longitudinal study was to examine whether the quality of the parent-child relationship would be predictive of the development of adolescents’ delinquency. In addition, we aimed to examine whether this link between the quality of the parent-child relationship and the development of delinquency would be stronger for adolescents with a higher familial risk for psychopathology. We hypothesised (1) that a familial risk for psychopathology predicted adolescents’ development of delinquent behavior, (2) low-quality of the relationship with parents predicted juvenile delinquency, and (3) that the familial psychopathological risk increased the likelihood to develop delinquent behavior if adolescents had a low quality of the parent-child relationship.
Consistent with previous studies, our present analysis showed that boys reported more delinquency in comparison with girls, and that delinquent behavior was moderately stable over time (Ferwerda, 2011; Moffitt, 1993; Overbeek et al., 2001). Furthermore, our results show that a higher familial risk for psychopathology did indeed significantly predict the development of juvenile delinquency. In accordance with previous research, these results support the assumption that having parents with psychological or psychiatric problems, alcohol and/or drug problems and/or problems with the justice system indicates having more of a risk for developing delinquent behavior (Loeber et al., 2008; Rigter, 2013; Van de Rakt et al., 2006). How can this result be explained? Adolescents’ development of delinquency could have two different explanations. On the one hand it could be the effect of shared genetic predisposition, such as genes and psychopathological problems of the parents (Canton et al., 2015; Glenn & Raine, 2014; Ferguson, 2010). Research showed that genes are associated with maldevelopment in brain structures which could lead to delinquency (Canton et al., 2015; Glenn & Raine, 2014; Ferguson, 2010). On the other hand, this predictive effect of familial risk could be explained by poor childrearing practices. In line with previous studies, having
parents with psychological and/or psychiatric problems is associated with ineffective parenting, which puts the children at risk of developing delinquent behavior (Loeber & Stouthamer-Loeber, 1986; Patterson et al., 1989; Van den Berg et al., 2015).
Analyses also provides evidence in favour of the third hypothesis. Indeed, the results showed that a lower quality of the parent-child relationship and higher familial risk for psychopathology were associated with more delinquent behavior. The interaction term
between quality of the parent-child relationship and familial risk index significantly predicted adolescents’ delinquent behavior at T3. This interaction indicated that, for adolescents with a low quality of the parent-child relationship, a higher familial risk for psychopathology and the development of delinquency were more strongly associated. In other words, the effects of familial psychopathological risk on developing delinquent behavior are stronger if adolescents have a low-quality relationship with parents. Perhaps, familial risk affects adolescents’
delinquent behavior when the parent-child relationship is dysfunctional, but a higher quality of the relationship with parents seems to buffer the effects of familial psychopathological risk. This finding is in accordance with results from previous studies, also showing a buffer effect of the parent-child relationship (Cohen & Wills, 1985; Farrell, Barnes, & Banerjee, 1995; Windle, 1992). Specifically, these studies showed that social support of parents and a high quality of the parent-child relationship can buffer the effects of a specific risk factor. It is thus important to examine how such a buffer, or protective factors, specifically a high-quality relationship with parents, may buffer the effect of risk factors and predict positive outcomes (Loeber et al., 2008).
In contrast with the significantly predictive results, age and education level were not predictive of delinquency, this is likely due to a skewed distribution within these variables resulting in less variance. Remarkably the quality of the parent-child relationship also was not significantly predictive of delinquent behavior. According to earlier studies this is a major
predictor of the development of delinquent behavior (Asscher et al., 2014; Hoeve et al., 2012). Previous studies showed that a low quality of the parent-child relationship is especially considered to be predictive of delinquency (Bowlby, 1944; Hirschi, 1969). Perhaps the fact that we did not find evidence for such a predictive association in our study is that, overall, in this study participants reported a rather high mean level on the quality of the parent-child relationship – and thus, that the variance in our measure of parent-child relationship quality was relatively limited. Another explanation could possibly be found in the normative
character of delinquency in adolescence (Luijpers, 2000; Moffitt, 1993). That is, regardless of the quality of the parent-child relationship most adolescents may show this normal rule-breaking behavior. However, some adolescents show serious and persistent behavior problems (Moffitt, 1993). This may have affected the measured variance in delinquent behavior. Note that the questionnaire which measured delinquency, mostly measured small offences. This might be a confirmation that normal rule-breaking behavior was measured.
Another explanation could be found in the changes that occur during adolescence, such as the changes in the attachment relationship with parents. Adolescents become more independent and the level of attachment gradually decreases (Buist et al., 2004) Moreover, once a child becomes more independent the quality of the parent-child relationship simply becomes a less important determiner of child outcomes (Van Lieshout, Van der Meij, & De Pree, 2007). There might also be differences between the attachment relationships of the adolescent and either parent (Branje et al., 2012; Hoeve et al., 2012). When the parent and the adolescent have the same gender – it could be the case that they have a better quality of the parent-child relationship (Branje et al., 2012; Hoeve et al., 2012). In this study we analysed the quality of the parent-child relationship for both parents together, which made it possible that the predictive effect of the relationship with parents on developing juvenile delinquency was not found. A last explanation for the absent relationship between a low-quality
parent-child relationship and delinquency can be found in the research design. By analysing the development of delinquency over time, controlling for earlier manifestations of delinquency at baseline, we used a stricter test – leaving less variance to be explained by family variables. Limitations
Some limitations of the present study may be noted. First, a potential limitation is that we did not use a clinical or at-risk sample to examine delinquent behavior. By using a general population, the participants report fewer problems and instances of delinquent behavior, which may result in causing biased results, as this sample does not yield enough participants with severe problems to analyse associations of actual delinquent behavior (Asscher et al., 2014). Thus, with less variance on the variables it is less possible to find any significant effect. Second, only risk factors within the family are presented in this study. Due to the changes that occur during adolescence, environmental factors outside of the family (i.e. peer relationships) become more important and may have a more essential effect on the
adolescents’ development of delinquency (Loeber et al., 2008; Rigter, 2013). A last limitation is the use of only adolescents’ self-report questionnaires. Self-report questionnaires are usually sufficiently reliable and valid. However, the possibility of not filling in the questionnaire correctly, due to recalling information in memory, giving socially desirable answers and a lack of motivation, may influence and bias the outcome (Bryman, 2012; Thornberry & Krohn, 2000; Vazire & Mehl, 2008). Moreover, if only the perspective of the adolescent is incorporated a one-sided view may occur, since parents and teachers could provide more and other specific information about the adolescents’ behavior and therefore important information will be missed (Asscher et al., 2014; Hoeve et al., 2012). Also previous research shows that when data is collected from both adolescents and parents the findings indicate stronger associations (Asscher et al., 2014; Hoeve et al., 2012). This is also
accounted for studies that not only look at the quality of the parent-child relationship, but also at parental and psychological control (Hoeve et al., 2012).
Strengths
Despite the limitations of this study, the present study has several strengths as well. The longitudinal design and the relatively large size of the sample are strengths of this study (Bryman, 2012). In this way it is possible to show associations between family factors and delinquency over time. In addition, by using age, gender, education level and earlier
manifestations of delinquent behavior at baseline as covariates the variance is relatively small. So, the likelihood of finding significant results decreases – however, when significant results are found they have a relatively high effect size (Bryman, 2012).
Recommendations and Implications
From the limitations, a few recommendations for future research in predicting juvenile delinquency can be presented. We recommend replicating this study using additional
information from important environmental others – such as parents and teachers – on the behavior of the child. Besides, the research should include the quality of the relationship with peers to see if there are differences in the predictive effect, compared with the relationship to parents. Moreover, it also should contain a juvenile offenders’ sample as a clinical control sample. Further, it is important to examine how a high-quality relationship with parents, may buffer the effect of risk factors and predict positive outcomes. There are also a few
implications for practice. Because of the results, it is important to account for a higher familial risk and a low quality of the parent-child relationship in treatment and prevention of the development of adolescents’ delinquent behavior. When focussing on interventions, it is necessary to investigate if there are any familial psychological risk factors, such as parents with psychologic and/or psychiatric disorders, parents with problems with the justice system or with substance use. In preventing these children from becoming delinquent, interventions
could focus on reducing this risk or positively influencing the relationship with the parents. Another implication for interventions is to account for earlier delinquency of the adolescent and gender differences. There is a difference in developing delinquency and therefore prevention and intervention may need a different approach.
Conclusion
Based on the strengths, our study makes a clear contribution to the scientific knowledge on this subject. The results of this research have shown that familial risk for psychopathology is a significant predictor of the development of delinquent behavior. Particularly the effects of familial risk are stronger if adolescents have a low quality parent-child relationship. Our results indicate that it is important to monitor the behavioral
development of adolescents within families where one or both parents suffer from
psychopathology and/or show delinquent behavior, especially when a poor relationship with their parents exists. Interventions could focus on improving the parent-child relationship, thereby potentially decreasing the effect of the familial risk. Future longitudinal studies should show how the quality of the relationship with parents may provide a buffer of the effects of risk factors.
References
Armsden, G. C., & Greenberg, M. T. (1987). The inventory of parent and peer attachment: Individual differences and their relationship to psychological well-being in
adolescence. Journal of Youth and Adolescence, 16, 427-454. doi:10.1007/bf02202939 Asscher, J. J., Wissink, I. B., Dekovic, M., Prinzie, P., & Stams, G. J. J. M. (2014).
Delinquent behavior, poor relationship quality with parents, and involvement with deviant peers in delinquent and nondelinquent adolescents: Different processes, informant bias, or both? International Journal of Offender Therapy and Comparative
Criminology, 58, 1001-1019. doi:10.1177/0306624X13491389
Boendermaker, L. (2008). De ontwikkeling van gedragsstoornissen [The development of conduct disorders]. Retrieved from http://www.nji.nl/nl/Ontwikkeling_
gedragsstoornissen.pdf
Boendermaker, L., & Ince, D. (2008). Effectieve interventies tegen jeugddelinquentie: Geen tucht en discipline, maar oudertraining en therapie [Effective interventions for juvenile delinquency: No discipline, but parent training and therapy]. Retrieved from
http://www.nji.nl/nl/Effectieve_interventies_tegen_jeugddelinquentie.pdf
Bowlby, J. (1944). Forty-four juvenile thieves: Their characters and home life. International
Journal of Psycho-Analysis, 25, 107-127.
Bowlby, J. (1988). A secure base. Parent-child attachment and healthy human development. Retrieved from http://www.abebe.org.br/wp-content/uploads/John-Bowlby-A-Secure-Base-Parent-Child-Attachment-and-Healthy-Human-Development-1990.pdf
Buist, K. L., Dekovic, M., Meeus, W., & Van Aken, M. A. G. (2004). Leeftijdsgebonden veranderingen in kwaliteit van gehechtheid van adolescenten aan vader en moeder [Age-related changes in quality of attachment of adolescents with mom and dad].
Pedagogiek, 24, 246-261. Retrieved from
file:///C:/Users/Gebruiker/Downloads/244-267-1-PB.pdf
Canton, W., Glenn, A., Raine, A., & Verkes, R. J. (2015). Biologische verklaringsmodellen van psychopathie [Biological explanatory models of psychopathy]. In W. Canton, D. Van Beek, L. Claes, L. Gijs, I. Jeandarme, & E. Klein Haneveld (Eds.), Handboek
psychopathie en de antisociale-persoonlijkheidsstoornis (pp. 31-55). Utrecht, The
Netherlands: De Tijdstroom.
Capaldi, D. M. (1992). Co-occurrence of conduct problems and depressive symptoms in early adolescent boys: II. A 2-year follow-up at grade 8. Development and
Psychopathology, 4, 125-144. doi:10.1017/S0954579400005605
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
Crone, E. (2014). Het puberende brein: Over de ontwikkeling van de hersenen in de unieke
periode van de adolescentie [The adolescent brain: On the development of the brains
in the unique period of adolescence]. Amsterdam, The Netherlands: Prometheus. De Kemp, R. A. T., Scholte, R. H. J., Overbeek, G. J., & Engels, R. C. M. E. (2006). Early
adolescent delinquency. The role of parents and best friends. Criminal Justice and
Behavior, 33, 488-510. doi:10.1177/0093854806286208
Farrell, M. P., Barnes, G. M., & Banerjee, S. (1995). Family cohesion as a buffer against the effects of problem-drinking fathers on psychological distress, deviant behavior, and heavy drinking in adolescents. Journal of Health and Social Behavior, 36, 377-385. doi:10.2307/2137326
Farrington, D. P., Coid, J. W., Harnett, L. M., Jolliffe, D., Soteriou, N., Turner, R. E., & West, D. J. (2006). Criminal careers up to age 50 and life success up to age 48: New findings from the Cambridge study in delinquent development. Retrieved from https://www. researchgate.net/profile/Jeremy_Coid/publication/237523020_Criminal_Careers_up_t o_age_50_and_Life_Success_up_to_age_48_New_Findings_from_the_Cambridge_St udy_in_Delinquent_Development/links/00463532c18d17dffa000000.pdf
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
Ferwerda, H. B. (2011). Jeugdcriminaliteit [Juvenile delinquency]. In N. Duits & J. A. C. Bartels (Eds.), Jeugdpsychiatrie en recht: Wetgeving, zorgveld en praktijk (pp. 87-95). Assen, The Netherlands: Van Gorcum.
Glenn, A. L., & Raine, A. (2014). Psychopathy: An introduction to biological findings and
their implications. New York, NY: New York University Press.
Goudena, P. P. (1994). Ontwikkelingsopgaven en opvoedingsopgaven [Development tasks and education challenges]. In J. Rispens, P. P. Goudena, & J. J. M. Groenendaal (red.),
Preventie van psychosociale problemen bij kinderen en jeugdigen. Houten,
Netherlands: Bohn Stafleu Van Loghum.
Guo, G., Roettger, M. E., & Cai, T. (2008). The integration of genetic propensities into social-control models of delinquency and violence among male youths. American
Sociological review, 73, 543-568. doi:10.1177/000312240807300402
Hale III, W. W., Wijsbroek, S. A. M., Engels, R., & Meeus, W. (2006). Gegeneraliseerde angststoornis en de waarneming van ouders door adolescenten [Generalized anziety disorder and the preception of parents by adolescents]. Kind en Adolescent, 27, 62-69. doi:10.1007/BF03060981
Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press.
Hirschi, T., & Gottfredson, M. (1983). Age and the explanation of crime. American Journal
of Sociology, 89, 552-584. doi:10.1086/227905
Hoeve, M., Semon Dubas, J., Eichelsheim, V. I., Van der Laan, P. H., Smeenk, W., & Gerris, J. R. M. (2009). The relationship between parenting and delinquency: A meta-analysis.
Journal of Abnormal Child Psychology, 37, 749-775. doi:10.1007/s10802-009-9310-8
Hoeve, M., Stams, G. J. J. M., Van der Put, C. E., Semon Dubas, J., Van der Laan, P. H., & Gerris, J. R. M. (2012). A meta-analysis of attachment to parents and delinquency.
Journal of Abnormal Child Psychology, 40, 771-785. doi:10.1007/s10802-011-9608-1
Houtzager, B., & Bearveldt, C. (1999). Just like normal: A social network study of the relation between petty crime and the intimacy of adolescence friendship. Social
Behavior and Personality, 27, 177-192. doi:10.2224/sbp.1999.27.2.177
Lenhard, W., & Lenhard, A. (2014). Hypothesis Tests for Comparing Correlations. Retrieved from http://www.psychometrica.de/correlation.html. Bibergau, Germany:
Psychometrica
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). Retrieved from http://blackboard.uva.nl/
bbcswebdav/pid-5589829-dt-content-rid-5641063_1/courses/2533S001.7014 B463DY.T12.1.2013/Loeber_2008%281%29.pdf
Luijpers, E. T. H. (2000). Intentie tot exploratie, sociale binding en delinquent gedrag van
Nederlandse jongeren [Intention to exploration, social bonding and delinquent
behavior of Dutch youths]. Delft, The Netherlands: Eburon.
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
Moffitt, T. E. (2005). The new look of behavioural genetics in developmental
psychopathology: Gene-environment interplay in antisocial behaviors. Psychological
Bulletin, 131, 533-554. doi:10.1037/0033-2909.131.4.533
Nada-Raja, S., McGee, R., & Stanton, W. R. (1992). Perceived attachments to parents and peers and psychological well-being in adolescence. Journal of Youth and Adolescence,
21, 471-485. doi:10.1007/bf01537898
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, G. J., Zeevalkink, H., Vermulst, A., & Scholte, R. H. J. (2009). 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. (1982). A social learning approach: Vol. 3. Coercive family process. Eugene, OR: Castalia Publishing.
Patterson, G. R., & Capaldi, D. M. (1990). A mediational model for boys, depressed mood. In Rolf, J., Masten, A.S., Cicchetti, D., Nuechterlein, K.H., and Weintraub, S. (eds.),
Risk and protective factors in the development of psychopathology (pp. 141-163). New
York: Cambridge University Press.
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
Raine, A. (2008). From genes to brain to antisocial behavior. Current Directions in
Psychological Science, 17, 323-328. doi:10.1111/j.1467-8721.2008.00599.x
Rigter, J. (2013). Handboek ontwikkelingspsychopathologie bij kinderen en jeugdigen [Manual developmental psychopathology in children and adolescents]. Bussum, The Netherlands: Coutinho.
Van Beek, D., & Canton, W. (2015). Geschiedenis van de concepten psychopathie en antisociale-persoonlijkheidsstoornis [History of the concepts of psychopathy and antisocial personality disorder]. In W. Canton, D. Van Beek, L. Claes, L. Gijs, I. Jeandarme, & E. Klein Haneveld (Eds.), Handboek psychopathie en de
antisociale-persoonlijkheidsstoornis (pp.15-28). Utrecht, The Netherlands: De Tijdstroom.
Van de Rakt, M., Nieuwbeerta, P., & de Graaf, N. D. (2006). Zo vader, zo zoon? De intergenerationele overdracht van crimineel gedrag [Like father, like son? The intergenerational transmission of criminal behavior]. Tijdschrift voor Criminologie,
48, 345-360. Retrieved from http://crawl.prod.proquest.com.s3.amazonaws.com/
fpcache/7d4ba7acd0c2b3a4330a1913f65e9a73.pdf?AWSAccessKeyId=AKIAJF7V7K NV2KKY2NUQ&Expires=1466934518&Signature=lHhVlmaxC86xlRsyJ3BpPGSR NkI%3D
Van den Berg, A., Cima, M., Klein Haneveld, E., & van de Putte, D. (2015). Opvoedingsfactoren in de ontwikkeling van psychopathie en
antisociale-persoonlijkheidsstoornis [Education factors in the development of psychopathy and anti-social personality disorder]. In W. Canton, D. Van Beek, L. Claes, L. Gijs, I. Jeandarme, & E. Klein Haneveld (Eds.), Handboek psychopathie en de
antisociale-persoonlijkheidsstoornis (pp. 57-83). Utrecht, The Netherlands: De Tijdstroom.
Van der Laan, A. M., & Blom, M. (2006). Jeugddelinquentie: Risico’s en bescherming. Bevindingen uit de WODC-monitor zelfgerapporteerde jeugdcriminaliteit 2005 [Juvenile delinquency: Risks and protection. Findings from the WODC-monitor self-reported juvenile delinquency 2005]. Den Haag, The Netherlands: Wetenschappelijk Onderzoek- en Documentatiecentrum (WODC).
Van der Laan, A. M., & Blom, M. (2011). Jeugdcriminaliteit in de periode 1996-2010:
Ontwikkelingen in zelfgerapporteerde daders, door de politie aangehouden verdachten en strafrechtelijke daders op basis van de Monitor Jeugdcriminaliteit 2010 [Juvenile delinquency in the period 1996-2010: Trends in self-reported offenders, suspects and prosecuted offenders arrested by the police on the basis of the Juvenile Crime Monitor 2010]. Den Haag, The Netherlands: Wetenschappelijk Onderzoek- en
Documentatiecentrum (WODC).
Van IJzendoorn, M. H. (1997). Attachment, emergent morality, and aggression: Toward a developmental socioemotional model of antisocial behaviour. International Journal of
Behavioral Development, 21, 703-727. doi:10.1080/016502597384631
Van Overveld, C. W. (2010). Onderwijs en gedragsproblemen: Prioriteit voor preventie [Education and behavioraal problems: Priority for prevention]. Tijdschrift voor
Orthopedagogiek, 49, 119-129. Retrieved from
Weerman, F. M. (2007). Juvenile offending. Crime and Justice, 35, 261-318. doi:10.1086/501510
Windle, M. (1992). A longitudinal study of stress buffering for adolescent problem behaviors.
Developmental Psychology, 28, 522-530. doi:10.1037/0012-1649.28.3.522
Wissink, I. B., Dekovic, M., & Meijer, A. M. (2006). Parenting behavior, quality of the parent-adolescent relationship, and adolescent functioning in four ethnic groups.
Table 1
Means and Standard Deviations
Total Sample Girls Boys
Variable n M SD n M SD n M SD
Parent-child Relationship at T1 741 104.67 14.70 387 104.47 14.79 354 104.89 14.61 Familial Risk at T1 686 4.18 0.59 364 4.21 .65 322 4.14 .51 Delinquent behavior at T1 747 14.24 2.21 391 13.73 1.46 356 14.81*** 2.70
Note. M = mean; SD = standard deviation. Scale ranges for parent-child relationship at T1, familial risk at T1, and delinquent behavior at T1 are 24-144, 4-12, and 13-65,
respectively. *** p < .001.
Table 2 Pearson Correlations Variable 1 2 3 4 5 1. Parent-child – Relationship, T1 --- -.16** -.11* -.07 -.13* 2. Familial Risk, T1 -.13* --- .19*** .10* .21*** 3. Delinquent behavior, T1 -.24*** .23*** --- .42*** .46*** 4. Delinquent behavior, T2 -.21*** .19** .47*** --- .46*** 5. Delinquent behavior, T3 -.10 .14* .33*** .41*** ---
Note. Correlations for girls (n = 405) are presented in the upper right triangle of the matrix; correlations for boys
(n = 369) are presented in the lower left triangle. * p < .05. ** p < .01. *** p < .001.
Table 3
Main and Interaction Effects of Predictors on Adolescents’ Delinquent Behavior
Predictor Step 1 Step 2 Step 3 Step 4
Age -.037 -.059 -.062 -.066 Gender .253*** .165*** .175*** .177*** Education level -.029 .005 .007 .000 Delinquent behavior at T1 .400*** .380*** .379*** Parent-child relationship at T1 -.040 -.038 Familial risk at T1 .078* .036 Parent-child relationship at T1 x Familial risk at T1 -.096* N 627 627 627 627 R2 .074 .224 .232 .239
Note. R2 = coefficient of explained variance. Standardized betas are presented for both main and interaction effects. Gender was dummy coded (1 = girls; 2 = boys).
