This study investigated the effects of family structure on 13-18 year-old adolescents in Saudi Arabia. Comparisons were made between adolescents from polygamous and monogamous families in psychological well-being (self-esteem, satisfaction with life, depression), bullying and victimization. A series of investigations assessed the effects of family structure and several demographic variables on adolescents’ psychological well-being and behaviour. Also, the mediating role of parent-adolescent relationships measured by parent-adolescent bonding and father availability was investigated. A systematic review of previous research established that few studies had investigated mediating variables, such as demographic variables. In the first study, comparisons were made between 98 adolescents from polygamous and monogamous families. Results found that adolescents from polygamous families reported more problems in their psychological well-being, bullying and victimization than adolescents from monogamous families. The aim of the second study was to establish the validity of the Parental Bonding Instrument for use with adolescents in Saudi Arabia. The parental bonding instrument was validated for use in Saudi society with 301 participants aged 13-18 years. Results found that the ‘care’ dimension of the parental bonding instrument was valid for use in Saudi Arabia but the ‘overprotection’ dimension was not considered to be culturally valid because of different cultural patterns found in Saudi culture. The third study compared 266 adolescents from polygamous and monogamous families using the validated parental bonding instrument. The results found that adolescents in polygynous families reported lower ‘care’ scores than those in monogamous families. Also, comparisons by age group and gender found no effects of age or gender for father care, mother care, self-esteem, satisfaction with life, bullying or victimisation. A significant difference was found between age groups for depression. The fourth study was conducted with 500 adolescents using structural equation modelling to test the role of the parent-adolescent relationship measured by parental bonding on adolescent self-esteem, satisfaction with life, depression, bullying and victimisation. For polygamous families, parental care was a significant mediating variable between adolescent outcomes and the family variables of father availability and the position of the mother as the first or later wife. For monogamous families, although parental care predicted adolescent outcomes, family variables did not affect parental care. The fifth study was a qualitative analysis of interviews with 30 adolescents and 10 teachers on perceptions of father fairness, family functioning, attitudes toward polygamous marriage and academic achievement. Problems reported for polygamous families were lack of father fairness and family cohesion, emotional and behavioural problems, and poor academic achievement. In conclusion, this thesis is the first study to investigate the effects of polygamous family structure on adolescents in Saudi society and the first to provide a culturally validated measure of adolescent-parent attachment relationships. It was found that polygamy affects adolescent psychological well-being and behaviour, also adolescents’ perceptions of parental care and the fairness with which they feel that their father treats them have important effects on their relationship with their parents, their sense of well-being and their behaviour. The findings will be valuable for educators, counsellors and psychologists in Saudi Arabia.

A central aim of the present research was to investigate the short-term influences of video game play on aggression-related psychological states (including affect, cognitions, attributions and behavioural tendencies). More specifically, efforts were directed at establishing psychological effects of video games, which are causally related to game violence. A series of experiments examined short-term changes in adolescent players following various types of video game play. Experiment 1 identified a number of important game dimensions (i.e., characteristics of video game play) and explored their relationship to overall game enjoyment using path analysis. Of particular interest was the finding that violence did not strongly influence game enjoyment. In Experiment 2, increases in affective hostility and anger were reported after both types of video game play. Significantly greater increases after violent video game play provided support for a video game violence-hostile affect relation. However, the findings of subsequent experiments produced contrasting evidence showing that affective changes following video game play do not predictably vary as a function of game violence, but appear to be linearly related to video game pace. Game violence was more strongly implicated in cognitive effects of video game play. Evidence that game violence affects cognitions emerged on a variety of measures. These cognitive effects were seen as being reflective of aggression priming and short-term disinhibition processes. Finally, the extent to which short-term effects of violent video game play dispose players towards aggression was investigated using attribution and response tendency measures. Whilst the majority of the analyses failed to produce effects of game violence on attributions and response tendencies, an interesting interaction emerged involving game violence effects in females. The nature of the inteaction was viewed as being best explained by modelling processes, though disinhibition explanations were also viewed as being compatible. The findings were interpreted within existing social-psychological theories of media-elicited aggression. A number of video game effects could be accounted for using Berkowitz's cognitive neo-associationist framework, whilst other findings implicated the usefulness of Zillmann's excitation-transfer theory for understanding video game effects. Ultimately, the results were conceptualised using Anderson's General Affective Aggression model. Overall, the research was fairly successful in highlighting a number of short-term affective and cognitive states that can result from video game play. However, these effects were generally not manifested in behavioural tendencies towards others. The few findings that did implicate increases in aggressive behavioural tendencies were difficult to place within Anderson's framework, as they did not parallel changes at earlier stages of the model (i.e. affective and cognitive changes). Modelling and/or possibly disinhibition effects were viewed as the most appropriate theoretical concepts for explaining the findings relating to behavioural tendencies. The implications of the findings in relation to previous research on video game and media effects and limitations to the generalisability of the findings are discussed. Finally, several recommendations for future research are outlined.