Media Violence and Criminality

In the research on effects of media, the question on the influence of media on real violence is of great importance. Numerous empirical studies give the impression that reliable findings exist; however, almost all investigations are cross-sectional studies or laboratory experiments. Cross-sectional studies have the disadvantage that causal structures are only shown insufficiently; laboratory experiments have the disadvantage that only short-term effects are registered. Meta-analyses show that both methods lead to gradually different results. Panel studies and natural experiments which can model causal effects correctly have so far been made only rarely. The results of the studies confirm the hypothesis that the consumption of media violence and violent acting correspond. Despite methodological deficits the empirical facts seem to be unambiguous; the theoretical interpretation is problematic. In the relevant empirical studies on this subject, the approach of the learning theory is of central importance. Media contents are seen as models accepted by recipients. Taking this as a background, the results of empirical studies can be interpreted in such a way that the consumption of media violence is one cause of real violence. Assuming, however, other crime and action theories, the empirical results can also be interpreted in such a way that the consumption of media violence and violent acting has common causes and that the correspondence between the characteristics is a spurious correlation. As a result it must be stated that there is a research deficit concerning the question of the influence of media on real violence.