Viewing-aggression correlations can arise through bidirectional, spurious, and direct causal paths

Evidence

Correlations involving televised violence can arise through four different paths of influence. Viewing violence fosters aggression. Aggressive viewers are attracted to violent programs. The influence is a bidirectional one: Viewing violence fosters aggression; in turn, an aggressive proclivity increases the attraction to violent programs. In the fourth path of influence, a third factor may influence both aggression and viewing violence, creating a spurious causal relation between the two other factors. They are co-effects of the third factor rather than causally related. The third factor has no fixed content. It could be socioeconomic status, ethnicity, or intelligence. Multiple controls, therefore, must be applied to rule out third-factor causation. — Albert Bandura, Moral Disengagement, 720

Categorization

Category: Definitional Content

• XREF: Mirrors the four-factor confounding framework behind the later statistical correction for third-factor causation; also connects to the Milgrom chapter on how institutions solve information problems, since both concern inference from observational data.

Related

• Observed links between media violence and aggression have four rival causal explanations (similarity 0.94)
• No single cause explains aggression; multiple determinants always operate together (similarity 0.88)
• Viewing televised violence increases rather than reduces aggressive behavior. (similarity 0.88)
• Media violence shapes aggression potential far more than it triggers immediate acts (similarity 0.84)
• Televised violence does not increase viewership according to Nielsen data (similarity 0.83)
• Institutional incentives, not human nature, drive televised violence. (similarity 0.82)
• Aggressive behavior creates hostile environments through feedback from others (similarity 0.82)
• Media effects on aggression are common, not limited to the disturbed (similarity 0.82)
• Televised violence increases aggression despite industry catharsis claims (similarity 0.82)
• Televised violence reflects industry misbeliefs and constraints, not viewer demand. (similarity 0.82)
• Social science causality is established only through converging divergent methodologies (similarity 0.82)
• Media effects are common occurrences, not confined to rare psychopaths (similarity 0.81)
• Televised violence does not attract larger audiences (similarity 0.81)
• Childhood TV violence predicts adult aggressiveness decades later, net of controls (similarity 0.81)
• Violence producers may become as desensitized as viewers to onscreen violence (similarity 0.81)