Degree Name

Doctor of Philosophy


Faculty of Health and Behavioural Sciences


Adolescent dehnquency is a growing social problem affecting individuals, families, and communities. The cunent research comprised three studies, which sought to explore the contribution of family and personality factors to self-reported delinquency, and to discover the nature of the relationship between perceptions of parental discipline style and perceptions of the seriousness of delinquent behaviour. The fust study examined the links between perceptions of family relationships, perceptions of parental discipline style, locus of control, self-esteem and selfreported delinquency among Australian high school stadents (N = 177). The prediction that locus of control and self-esteem would mediate the effects of family processes on delinquency was tested using stmctural equation modelling. Although there appeared to be a good fit between the data and the proposed model, the amount of variance explained by the predictor variables was not large. Among females, the best predictor of low levels of self-reported delinquency was an inductive discipline style, whilst for males high levels of self-reported delinquency were best predicted by a punitive discipline style. Among males, positive family relations was a significant predictor of high self-esteem. No mediating effects of self-esteem and locus of confrol were observed. Given these results and the findings of previous research indicating links between family process factors, Eysenck's Psychoticism (P) factor, and delinquency, the second study investigated the relationships between perceptions of parental discipline style, perceptions of parental bonding, P, and selfreported delinquency among a sample of delinquent youth (N = 39). It aimed to determine the intervening effect of P on family process factors and self-reported delinquency. As expected, this sample of delinquent youth obtained significantly higher delinquency scores than mainstream high school students, as well as significantly higher P levels than established norms. Scores on the parental bonding instmment differed significantly in the expected direction from nondelinquent students. Self-reported delinquency was significantly related to an inductive parental discipline style and high P levels. As predicted, P mediated the effect of inductive parenting on delinquency. The third study aimed to assess the stracture of adolescents' and adults' perceptions of the seriousness of behaviours labelled as "delinquent", and to determine whether these perceptions vary across the sex of the respondent and sex of target (that is, the person engaging in the behaviour). A further aim was to examine whether these perceptions are linked to particular parenting discipline styles. The results indicated that, within a sample of high school students (N = 321) and their parents (N = 193), adolescent and adult perceptions of delinquent behaviours are multi-dimensional, possessing a particular stmcture. As expected, sex of target and sex of respondent were found to have some impact on adolescents' perceptions of offence seriousness. Parental discipline style was found to be especially important in predicting the perceptions of adolescent boys rather than girls, as well as some perceptions of parents. In exanuning the contribution of family and personality factors to delinquency, all the studies in this research found parental discipline style to be a key variable. The results of the three studies are discussed with reference to previous research, recommendations for intervention and clinical practice are made, and implications of the findings for further research are noted.

02Whole.pdf (4287 kB)



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.