t differences between overall changes in aggressiveness and type of game played. Pearson’s correlation coefficients also failed to show a significant relationship between type of game played and change in level of aggression. Ballard and Weist (1996) used a one-way MANOVA test and found a significant relationship between levels of hostility and type of game. Hostility scores were higher after playing high-gore Mortal Kombat than they were after playing low-gore Mortal Kombat. Both were significantly high than those who played the billiards game. Sherry (2000) used the Pearson R on each of the 25 studies individually, he then subtracted out the variance due to sampling error. This resulted in a large standard deviation in the weighted effect size, suggesting extraneous variables at work. He also performed a Fisher Zr test, and was unable to reject the null hypothesis. In the words of Austin Powers, “whoopdy-doo, what does it all mean???” There are many factors that cloud the issue, but overall it seems that the evidence is not very convincing. In the Cooper and Mackie study (1986) the higher levels of play time with the aggressive toy was accounted for almost entirely by the female participants. One theory is that playing the aggressive video game empowered the girls (whom are normally shunned from doing aggressive things) to broaden their aggressive horizons; at the least it calls into question issues of generality. The age of this study is also a factor, Sherry (2000) points out that there is a trend for older video game studies to have greater effect sizes. An issue with all of the studies was the short amount of time participants played the games. Sherry (2000) found that playing time was a negative predictor of effect size. It seems that there is an initial spike in aggression that goes away after longer exposure. Tamborini et all’s (2000) findings are odd due to the fact that the observers re...
