Media effects research: Examining violent video game effects in a publicly available sample

Thumbnail Image
Miles-Novelo, Andreas
Major Professor
Anderson, Craig A
Gentile, Douglas
Kelly, Jonathan
Blakenship, Kevin
DeLisi, Matthew
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Is Version Of
There has been much debate surrounding the effects of violent media on aggression and violence. While numerous meta-analyses, reviews, and even taskforces from various scientific bodies have upheld the findings of a relationship between violent media and aggression (e.g., Anderson et. al, 2010; APA, 2017; ISRA, 2018), some studies find a null-effect and claim that such a relationship is non-existent (e.g., Przybylski & Weinstein, 2019; Hilgard et al., 2017; Ferguson, 2007). These claims are often spread by scholars on social media, gaining much attention – and have influenced policy and research around violent video game effects. One meta-analysis (Greitemeyer & Mugge, 2014) and several recent papers (Bushman & Anderson, 2021; Kim, Anderson, & Gentile, 2021; Wright, 2021) suggest that most of the failures to replicate are by one small group of scholars, and that poor research methods may be involved. One such failure to replicate established violent video game effects (Przybylski & Weinstein, 2019) has made its dataset publicly available, thereby offering a unique opportunity to examine the hypothesis that methodological and statistical problems underlie such replication failures. The aims of the present research were to examine 1) the methodology compared to previous investigations into video game use and aggression and 2) to see if the publicly available materials offer more sensitive ways to test this relationship. This was done through 2 studies. The first used the publicly available materials and examine issues within the original analysis, and found issues with statistical control, the measures used to assess aggression, and issues within the measurement of exposure to video game violence (VGV). By gathering new data on video game violence exposure, the second study aimed to create a more standard measure of VGV exposure as a way to more sensitively test the relationship between aggression and VGV exposure. The results demonstrate that conceptual misunderstandings of aggression, and poor measures regarding both aggression and video game violence exposure, led to the originally reported null result (Study 1). Furthermore, by using improved materials (including a more sophisticated coding schemes to assess exposure to violent video games) and sound statistical analysis (including correcting for overcontrol), the data does replicate the long-established relationship between violent video game play and aggression (Study 2).