Contributors to violent behavior among elementary and middle school children

Pediatrics. 1999 Oct;104(4 Pt 1):878-84. doi: 10.1542/peds.104.4.878.


Objective: To examine the relative contributions of exposure to violence, parental monitoring, and television-viewing habits to children's self-reported violent behaviors. The study hypothesized that: 1) children's exposure to violence would be associated positively with self-reported violent behaviors; 2) parental monitoring would be associated negatively with children's violent behaviors; and 3) the number of daily television-viewing hours and a preference for watching violent television shows would be associated positively with children's violent behaviors.

Methods: The study used a survey design with an anonymous self-report questionnaire administered to students (grades 3-8) in 11 public schools. A total of 2245 students participated in the study, representing 80% of the students attending the participating schools during the survey. The subjects were from 7 to 15 years of age; 51% were male, 57% were white, 33% percent were black, and 5% were Hispanic.

Results: Hierarchical multiple regression analysis of the total sample revealed that the combination of demographic variables, parental monitoring, television-viewing habits, and exposure to violence explained 45% of students' self-reported violent behaviors. Violence exposure and parental monitoring were the most influential contributors in explaining children's violent behaviors, accounting for 24% and 5% of the variance in violent behaviors, respectively.

Conclusions: All three hypotheses were supported. A significant association was demonstrated linking violence exposure, lack of parental monitoring, and television-viewing habits with children's self-reported violent behaviors within a diverse sample of elementary and middle school students. Our findings support the importance of parental monitoring of children and emphasize the need to identify and to provide services to youth who are exposed to violence.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Aggression / psychology*
  • Child
  • Domestic Violence
  • Female
  • Firearms
  • Humans
  • Male
  • Multivariate Analysis
  • Ohio
  • Parenting*
  • Regression Analysis
  • Risk Factors
  • Social Environment*
  • Television
  • Violence / psychology*
  • Violence / statistics & numerical data