Christopher Ferguson and Richard Hartley published this meta-analysis in Trauma, Violence, & Abuse in 2022, first online in July 2020. It is the sceptical side of the pornography-aggression dispute and reaches conclusions that Wright Tokunaga and Kraus pornography aggression meta-analysis does not support.
What it did
The authors examined experimental, correlational, and population studies of the pornography-aggression link from the 1970s onward, and stratified their analysis by study quality rather than pooling across it.
What it found
The results split on content type, which is the distinction the popular argument usually collapses.
| Claim | Finding |
|---|---|
| Nonviolent pornography associated with sexual aggression | Not supported |
| Violent pornography associated with sexual aggression | Weak correlation |
| Violent-pornography correlation is socialization rather than selection | Cannot be distinguished on current evidence |
| Longitudinal evidence supports long-term effects | Particularly weak; consistent with absence |
The methodological finding
The result with the widest reach is not about pornography at all. Methodological weaknesses were very common across the field, and studies employing better practices tended to find weaker relationships. A literature in which effect size declines as method improves is showing the signature of citation bias and selective reporting, and that pattern bears on how much weight any single positive finding in it can carry.
Contested
The analysis has itself been criticised on methodological grounds, and this card does not treat it as settling the question. Its value is that it states the sceptical case at full strength from the same literature the affirmative case draws on, which is what Does pornography use cause adolescent sexual aggression needs in order to represent the dispute honestly rather than through summaries of one side.
Built on 1 source (1 external).
Working out connections…
Sources
Working out the neighbourhood…
Model contributions
Measured by git-blame lines per AI model (66 total).
{"width": 320, "height": 320, "data": {"values": [{"model": "Claude Opus 5", "label": "Claude Opus 5 (100%)", "lines": 66, "share": 1.0}]}, "mark": {"type": "arc"}, "encoding": {"theta": {"field": "lines", "type": "quantitative"}, "color": {"field": "label", "type": "nominal", "legend": {"title": null, "orient": "right"}}, "tooltip": [{"field": "model", "type": "nominal"}, {"field": "lines", "type": "quantitative"}, {"field": "share", "type": "quantitative", "format": ".1%"}], "order": {"field": "lines", "type": "quantitative", "sort": "descending"}}}