Communication Research · 1994 · 859 citations · 34 references
Social SciencesBehavioral SciencesPsychologyExperimental PsychologySocial InfluenceViolence PreventionAggressionMedia EffectsTelevision AudiencesTelevision StudiesViolenceAntisocial Behavior AssessmentAntisocial Behavior DevelopmentChildhood Antisocial BehaviorFisher's Z TransformationMedia Violence EffectsSex Differences In Aggression
The authors conducted a meta‑analysis of studies on television violence and aggression, partitioning data by research design, viewer attributes, exposure variables, and antisocial behavior type, and performed additional tests to validate the results. The analysis revealed a positive, significant correlation between television violence and aggressive behavior, with effect sizes ranging from 0.19 in surveys to 0.40 in laboratory experiments, erotica emerging as a strong factor even without violence, and comparable effects for boys and girls in surveys.
A meta-analysis is performed on studies pertaining to the effect of television violence on aggressive behavior. Partitioning by research design, viewer attributes, treatment and exposure variables, and type of antisocial behavior, allows one to interpret computed effect sizes for each of the variables in the partitions. We find a positive and significant correlation between television violence and aggressive behavior, albeit to varying degrees depending on the particular research question. According to research design, we find Z Fisher values ranging from .19 for survey to .40 for laboratory experiments. Erotica emerges as a strong factor even when it is not accompanied by portrayal of violence. Additionally, the effect of television violence on the antisocial behavior of boys and girls is found to be marginally equal in surveys. A host of tests are performed to solidify these, and further results. Substantive interpretation is provided as well.
34
Statistical Power Analysis for the Behavioral Sciences
Keith E. Muller, Jacob Cohen · Technometrics · 1989 · 83.9K citations