How is paragraph 8 effective in supporting the development of ideas in this text? These two studies do not examine how the influence of the peer group on bullying behaviors differs across sex, grade, or level of peer group status. A study by Espelage and Holt (2001) of 422 middle school students (grades 6-8), using a survey that included demographic questions, self-report, and peer-report measures of bullying and victimization, and measures of other psychosocial12 variables, examined the association between popularity and bullying behavior. Despite the finding that bullies as a group enjoyed a strong friendship network, the relationship between bullying and popularity differed for males and females, and also differed across grades. The most striking finding was the strong correlation13 between bullying and popularity among sixth-grade males, which dropped considerably for seventh-grade males and was not associated for eighth-grade males. Closer examination of peer cliques in this sample found that students not only “hung out” with peers who bully at similar rates but that students also reported an increase in bullying over a school year if their primary peer group bullied others (Espelage, Holt, & Henkel, in press).