Bivariate correlations, or correlations between two and only two variables, are used to assess the degree of association between two variables of interest. Such correlation coefficients do not indicate causality. Indeed, in many such cases, other important variables may explain the relationship between two variables of interest. In this case study, I offer the example that boys play more violent video games and are also more aggressive than girls. Thus, to some degree, correlations between violent video games and aggression may simply be a function of maleness rather than anything about violent video games themselves. Indeed, considerable research suggests that when other important factors are carefully controlled, bivariate correlations between video game violence and aggression are explainable as due to other factors rather than representing a unique relationship. However, this point is often lost in societal controversies over video game violence, and even some scholars may overemphasize bivariate correlations failing to explain how these correlations may not be as meaningful as they first seem. The current study examines the use and misuse of bivariate correlations in the field of video game violence as an example of what these simple statistics can and cannot tell us.
The Uses and Misuses of Bivariate Correlations: The Case of Video Game Violence Research