Abstract
Recent meta-analytic findings (CitationSherry, 2007) suggest that the effect of violent video games on aggression tends to decrease the longer participants play. Therefore, it is likely that different mechanisms may be at work depending on the length of play. In this study, we employed a 4 (length of game play) × 2 (sex) design to examine the effect of length of video game play on aggressive outcomes. Findings suggest that length of play may be curvilinearly related to verbal and physical aggression, and that arousal may mediate this response. Implications for our current understanding of the relationship between video games and aggression are discussed, along with directions for further research.
Notes
1. The observant reader may notice that some diagnostic statistics for the CFA model are lower than would be expected in structural equation model with acceptable fit. Scholars have recently argued that an RMSEA closer to .10 is acceptable for CFA procedures, as traditional tests of model fit tend to be hypersensitive in CFA procedures and often lead to the erroneous rejection of valid models (CitationYu, 2002). Further, a CMIN/df of less >5 is generally considered to be an indication of adequate model fit (CitationArbuckle & Wothke, 1999).