Page 288 - Media Effects Advances in Theory and Research

10. EFFECTS OF MEDIA VIOLENCE 277

Controversy About the Effects of Media Violence

If most scholars agree that the research evidence tends to converge on the
conclusion that exposure to media violence causes aggressive behavior,
then why has scholarly and public debate on this topic produced so much
controversy? One important source of the dispute revolves around confu-
sion, even among researchers, about the concepts of statistical signifi-
cance, statistical importance, and social importance. When statistically
significant results show that media violence is causally related to aggres-
sive behavior, researchers can be confident that they have observed a rela-
tionship that is unlikely due to chance. Such a conclusion seems to point
to a clear conclusion of a media effect.
However, statistical significance says little about the strength of the rela-
tionship. In order to gauge the strength of the relationship or its statistical
importance, researchers usually appeal to some index of the statistical vari-
ance accounted for in aggressive behavior by knowing the level of expo-
sure to media violence. Studies on media violence and aggression are no
different than studies on other areas of human behavior in that they typi-
cally enable researchers to account for 10 to 15% of the variance in the
dependent variable. There are two aspects to this state of affairs that lead
some to minimize the overall magnitude of media violence effects on
aggression. First, the effect appears to be quite modest in that 85 to 90% of
aggressive behavior in most studies is left unaccounted for by media
exposure. Second, the extent to which the variance accounted for in
aggressive behavior in any given study can be used as a general guideline
for the nature of the relationship between these variables in the real world
is unclear. There is no easy way to map statistical indexes from isolated
studies to a general statement about the strength of the relationship in the
real world. Critics who minimize the relationship between media violence
and aggression emphasize the large proportion of aggressive behavior
that seems to derive from sources other than exposure to media violence.
Alternatively, others emphasize that given the multiplicity of causes for
any human behavior, to be able to account for 10 to 15% of the variance in
aggressive behavior in a given study by knowing only about media expo-
sure is quite impressive.
Added to the controversy about the magnitude of the effect size (statis-
tical importance) is the notion of social importance. Because media audi-
ences sometimes number in the millions, even very small statistical effects
can translate into very important social problems. If just one person in
several hundred thousand is influenced by a violent movie to commit a
serious act of aggression, the social consequences of several million view-
ers watching that movie might be dramatic. On the other hand, such a