There is a very close connection between delinquent behavior and other forms of deviant behavior in children and adolescents (White et al., 1990; Moffitt, 1993; Caspi et al., 1995). Criminal acts are frequently consequential of numerous difficulties in socialisation process. These difficulties generate various forms of behaviors potentially dangerous to child development and to his/her environment. Besides, children and adolescents, unlike adults, are subjects of intensive bio-psycho-social changes, whose dynamism offers considerable opportunities for socialisation; at the same time, they are subjected to influences that can direct their behavior in asocial, and antisocial direction.
Modes of communication that children/adolescents learn and develop in their family, undoubtedly have great impact on relationships in later life-course.
Intrafamily relationships are of great importance for prevention of behavior disorders as well as for their development (Marquis, 1992).
McCrae and Costa (1988) examined the relationship between adult’s assessment of their own personality traits (according to the "big five" theory) and the relationship with parents during childhood. The separate analyses were made for mothers and fathers, regarding the sex and the age of the subjects. Results showed that children raised up in loving atmosphere grow up to be well adjusted, open toward others, tolerant, co-operative, and creative. The highest correlation was observed with the neuroticism scale. The correlations were similar for mothers and fathers. However, the correlations were relatively low; therefore, authors concluded that parental influence in forming of personality is low, compared with the influence of the hereditary factors.
Studying effect of family atmosphere on later criminal conduct, McCord (1979) found significant effect of parenting on later child’s behavior. Family atmosphere showed even better than juveniles’ criminal records in prediction of later criminal conduct. Items describing parenting accounted for 30% variance in prediction of sentencing for stealing and homicide. The best predictors were parental control, maternal love and conflicts between parents.
Loeber and Dishion (1993) found that severe, non-consistent discipline and poor control predicted later delinquency even better than the socio-economic status. Henry et al. (1993) found that antisocial and "normal" children differ primarily in parental disagreement about parenting a five-year child, and in numerous changes of child’s caretakers. The conflicts between parents, child neglect and abuse are correlated with the child’s aggressiveness. If the child lives in extremely unfavourable family circumstances, there is a high probability for him/her to become a serious delinquent (McCord et all, 1963; Farrington, 1978, 1991).
The comparison between the children who were loved by their parents on one side and abused, neglected and rejected children on the other, shows that latter have significantly less favourable family circumstances, which especially refers to parent aggressiveness (McCord, 1983).
The children exposed to parental verbal and physical aggression show higher level of aggressiveness and greater inclination to delinquent behavior than children exposed only to verbal aggression (Vissing et all, 1991).
In search for a cause of child’s aggressive behavior, some authors especially emphasise the importance of family interactions (Patterson, 1976; Patterson et. al., 1984). Patterson (1976) maintains that family is a complex system in which child plays the role of the "victim" but also the one of the "architect" in building of aggressive patterns and disturbances of the family interactions.
Dumas et al. (1992) concluded that mothers of aggressive children are more aversive and authoritative than mothers of "normally" developed children. Further analysis showed that the mothers of aggressive children choose inconsistent and punitive patterns of communication. A good parent-child relationship diminishes the effect of developmental crisis and benefits the child’s positive mental development (Cohn et al., 1992).
Rankin and Kern (1994) confirmed the hypothesis about moderate, yet consistent relation between family attachment and juvenile delinquency; children equally attached to both parents showed significantly lower risk of later delinquent behavior than children attached only to one parent. Lyon et al. (1992) found an influence of insufficient parental control and poor family relationship on adolescents joining the juvenile gangs; Elliot et al. (1985) showed the relation between family disturbances and adolescents’ delinquent activities.
Deficient interpersonal relationships between family members are a crucial factor in juvenile alcohol abuse (Foxcroft et al., 1994).
A meta-analysis conducted by Denton and Kampfe (1994) showed that drug abusing adolescents most frequently report poor family relationships. They described their family environment as hostile; it lacks understanding, love, cohesion and cooperativeness. They also experience their parents as indifferent persons who care only about their own problems. The family communication is disturbed, rigid and unclear, without confidence, acceptance and understanding. On the other side, the addicts’ parents report a lack of self-confidence in the matters of upbringing; they feel that parent role demands too much self-sacrifice and self -denial.
The disturbed, "clinical" families with addicted or delinquent adolescent show lesser capability of problem solving, more defensive (including threats, punishment, control and domination) and less supportive communication (including the exchange of information and empathy).
A disturbed mother-adolescent relationship is characterised by insufficiencies in problem-solving skills, a greater proportion of "difficult" behavior and a less positive communication. In general, these findings suggest that adolescents’ problem behavior correlates positively with the aversive verbal behavior, and negatively with positive and functional parent behavior. However, the causality in this relationships remains unclear (Krinsley and Bry, 1992).
The importance of one belonging to his/her family considerably influences later life-course in many ways, independently of other family members experiences (Chubb and Fertmann, 1992). Many researchers found affective quality of family relationships during adolescence related with the measures of adolescent’s psychosocial competence (Bell et al., 1985; Grotevant and Cooper, 1985; Papini et al., 1989). These studies support the hypothesis by which the level of emotional involvement of children and parents (or other caretakers) reliably predicts adolescents’ problem behavior. The insecure relationship with caretakers during childhood also emerged as an important risk factor for later development of maladaptive behaviors (Cohn et al., 1992). If there is a high level of emotional closeness between family members, adolescents show a high level of self-confidence.
LeCroy (1988) found that the level of intimacy between father and adolescent predicts appropriate behavior of adolescents better than the level of intimacy between mother and adolescent.
When planning some shared activity, mothers of behaviorally disturbed adolescents more often react "negatively" on "negative" behavior of their children than mothers from control sample. Behaviorally disturbed adolescents are also more oriented to negativities of their mothers than adolescents from the control sample (Krinsley and Bry, 1992).
Depressive parents, comparing to non-depressive parents, showed significantly lower level of satisfaction with marriage quality, parent-child relationships and global family functioning (Kaslow et al., 1992).
The aim of present study was to analyse the differences between behaviorally disordered and "normal" adolescents in their perception of parent-child communication. Although there is little doubt about tendencies of the correlation, it is important to find out more about their specific contents.
Our general hypothesis was that behaviorally disturbed adolescents experience insufficient and more difficult communication with their parents than their "normal" peers.
Material and methods
The research was carried out on a sample of pupils of both sexes attending eight grades in 33 randomly selected schools in Zagreb area. In every school, we examined the pupils of one eight grade class, which was also randomly selected. The total of pupils examined was 1023 (513 girls and 510 boys, aged 14). However, only 968 subjects who lived with both parents were examined in this study.
To achieve a clearer distinction in communication space, we considered only the subjects whose standardised scores were placed on both ends of the first principal component of Behavior Disorders Questionnaire. The first subsample (N1 = 134) comprised the subjects who scored lower than -1 standard deviation; the second subsample (N2 = 140) comprised the subjects who scored higher than +1 standard deviation. The subjects from N1 subsample, therefore, rarely manifested behavior disorders, while the subjects from N2 subsample often manifested behavior disorders.
For the assessment of the behavior disorders we used the shortened and partially modified version of Behavior Disorders Questionnaire (see Table 2), constructed on Department of Behavior Disorders, Faculty of Defectology. The Questionnaire comprises the behavior forms whose frequent occurrence can eventually lead to deviant and delinquent way of life. Cronbach-alpha reliability coefficient of Questionnaire is .85.
Communication Questionnaire (see table 3) was also constructed on the same Department. For the purposes of this study, the shortened version was used (32 items, out of 60).
All items were ordinal 5-point Likert scales. In Communication Questionnaire we measured subjects’ agreement with each particular statement. The Behavior Disorders Questionnaire is a self-report scale; it measures frequency of 27 behavior disorders. The lowest result on scale marks disagreement with the statement, or absence of behavior disorders.
The first principal component of Behavior Disorders Questionnaire was extracted by the method of Hotelling’s principal component analysis (Cooley and Lohnes, 1971). Differences between two groups were analysed using one-way variance analysis (program SPSSWIN).
Table 1 and 2 show the results of principal component analysis. Only the first principal component was considered. It was defined mainly by the items that refer to external behavior disorders: negligence (laziness), lack of interests (boredom), obtrusiveness, untidiness, indiscipline during classes, defiance, lying, verbal and physical aggression (cursing and arguments, fights), smoking, drinking alcoholic beverages, unjustified skipping classes and whole school days, mingling with asocial persons, vagrancy and stealing. Therefore, our two subsamples differ primarily in manifestation of external behavior disorders.
Table 3 shows the results of one-way variance analysis. M1 and S1 are the means and standard deviations of the first subsample (subjects who did not report external behavior disorders); M2 and S2 are the means and standard deviations of the second subsample (subjects who did report external behavior disorders).
Results in Table 3 show that subjects without behavior disorders experience the communication with their parents as possessing higher quality. As much as 23 items showed statistically significant difference between the two subsamples. The interesting exception from the main trend are items 17 - 18 ("I believe I can tell my mother/father what I feel and think about things.") that appear to indicate the level of openness in parent-child relationship. The statistically significant difference was achieved only when item was related to father - behaviorally disturbed adolescents reported more open relationship. However, we could suppose that the subjects actually thought this statement referred to quarrelling with their fathers. It is also possible that subjects understood that the term "some things" meant something indecent, forbidden, (like sex, for example). It could explain our finding - the behaviorally disturbed adolescents scored higher on this item because they are more direct and defiant in conversation with their fathers, and also, more likely are ready to talk about "forbidden" themes.
The results on items 24 ("It’s difficult to talk with my father about anything."), 28 ("I talk with my father about my feelings."), and 30 ("I talk with my father about my friends.") could confirm our hypothesis - the behaviorally disturbed adolescents have difficulties in conversation with their fathers.
Further, items 15 - 16 ("When I do something wrong, my mother/father punishes me.") also slightly differs from the general trend; differences were found only in relation to one parent. These results imply that punishment coming from the mother does not distinguish adolescents, and, also, that fathers more often punish behaviorally disturbed subjects. The father’s role in punishing children is obviously more important than the mother’s. In our culture it is expected from father to be more rigorous, but it seems that exaggerated severity can result in behavioral disorders.
On the other side, the statement "My mother/father liked to play with me." differs our subsamples only when it refers to mother (mothers played less often with behaviorally disturbed children). This is probably the consequence of poorly developed, less emotional, "colder" relationship, with mother, which is in concordance with the results of some previous studies (Dumas et al., 1992; Harrist et al., 1994).
Our study was based on subject’s perception rather than on
real family situation. We are aware of the fact that the perception of the relationships
and actual parent-child relations in many cases are not the same. However, the
influence of the way of experiencing things on human behavior is probably greater
than the influence of reality itself; therefore, it could be more important
than reality (King, 1979).
LeCroy (1988) stressed an injustice of levelling the notions ’contact’, ’help’, and ’intimacy’ in family relationships studies, which manifests itself in the assumption that quantity of interactions is the same as quality, and material exchange is the same as emotional exchange. The results of our study also do not support such hypothesis. Our study showed that behaviorally disordered adolescents mainly differ from their non-disturbed peers by the general intrafamily context - openness, support, and mutual respect. Only a few items that marked some particular parental reactions did not show statistical differences. We suppose, if this context is favourable, then some sporadically negative parental reactions such as physical punishment (9, 10), occasional shouting (1, 2) or permissiveness (21, 22) are not related with the emergence of the behavior disorders.
In general, our results confirmed those of Rankin and Kern (1991) who stressed the importance of adolescent’s attachment to both parents. Our study suggests that two subsamples also experience differences in relationship with both parents (with only 3 exceptions); therefore, we can conclude that behaviorally disordered adolescents who manifest externalised disorders experience poorer relationship with both parents.
The positive relationship with both parents is important for "normal" development; it is a crucial condition for harmonical functioning of family, and, at the same time, the indicator of good relationship between the parents themselves.
Unfortunately, the causality of phenomena remains unclear; we can only suppose that behavior disorders emerge from interaction between child’s hereditary predisposition and the unfavourable family environment. This question could probably be answered by means of a long-term study, starting from the child’s earliest age.