Video games today are more popular and more widely accessible than ever. This includes those games with mature ratings or those that are considered more violent, such as first player shooters (FPS) games, or franchises that glorify crime, such as Grand Theft Auto (GTA). Because of their influence over growing adolescents, many have decried such violent games and blame them for influencing aggressive behavior in children and even ultimately inspiring mass shooters.

Several studies already exist on this topic, however not a lot of studies focused on factors that could affect adolescents in addition to video games. This study aimed to apply data science methodology to determine whether video games are a significant contributor to aggressive behavior in adolescents by utilizing neural network and logistic regression analysis on the Health Behavior in School-aged Children (HBSC) study to identify and examine the relationships between a multitude of factors in a child’s life.

The primary objectives of this project were to:

• Determine factors contributing to aggressive behavior (including hostile verbal and non-physical actions but focusing more on physical confrontations). Factors include, but are not limited to, video games, family and/or home circumstances, juvenile detention, etc.)
• Apply data science methodology to determine patterns in those factors to analyze the correlation of those factors with violent behavior and crimes in adolescents. Create and utilize predictive models that can be used for this purpose.
• Identify actionable insights that can be used by professionals who develop approaches to mitigate aggressive behavior in adolescents.