Past research suggested the importance of school and family factors in accounting for bullying. The current research examines the possibility of reliably discriminating between bullies, victims, and non-bully/non-victim (nbnv) based on these two sets of variables. Family variables include authoritarian and authoritative parenting styles as well as family disharmony. School variables include liking for school, perceived control of bullying, and school hassles. Participants were parents and their children aged 9 to 12 years (N=562). Data were analyzed using ANOVA and discriminant function analysis. Two significant functions emerged, both of which appeared to be important in discriminating children according to their bullying status. Together they allowed for the correct classification of 70% of bullies, 55% of victims, and 82% of the nbnv. The overall rates were 68.5% and 69.3% in the initial and cross-validation classification, respectively. The main conclusion drawn from these findings is that family and school systems working together in a coordinated fashion may provide the most effective means of intervention. "Author's Abstract"