Evidence for publication bias in video game violence effects literature: A meta-analytic review

Violence in video games has come under increasing research attention over the past decade. Researchers in this area have suggested that violent video games may cause aggressive behavior among players. However, the state of the extant literature has not yet been examined for publication bias. The current meta-analysis is designed to correct for this oversight. Results indicated that publication bias does exist for experimental studies of aggressive behavior, as well as for non-experimental studies of aggressive behavior and aggressive thoughts. Research in other areas, including prosocial behavior and experimental studies of aggressive thoughts were less susceptible to publication bias. Moderator effects results also suggested that studies employing less standardized and reliable measures of aggression tended to produce larger effect sizes. Suggestions for future violent video game studies are provided. (c) 2007 Elsevier Ltd. All rights reserved.