Page 286 - Media Effects Advances in Theory and Research

10. EFFECTS OF MEDIA VIOLENCE 275

tion, some researchers have attempted to employ this method outside the
confines of the laboratory using methods appropriate for field research.
Berkowitz and his associates have conducted a number of field experi-
ments in institutions for delinquent boys (Leyens, Parke, Camino, &
Berkowitz, 1975; Parke, Berkowitz, Leyens, West, & Sebastian, 1977).
These experiments assessed physical and verbal aggression in boys who
had been assigned to watch media violence for several weeks and com-
pared their levels of aggression with similar boys who did not watch vio-
lence. The findings of these studies converged with laboratory investiga-
tions; boys who watched media violence were more likely to engage in
aggressive behavior.
The work of Williams (1986) is especially noteworthy in that she was
able to study changes in aggression that occurred naturally over several
years in a Canadian town that initially had no access to TV signals but,
over the course of the natural experiment, gained TV access. The results
of Williams’ research converged with the findings of the laboratory stud-
ies: increases in exposure to media violence lead to increases in aggres-
sive behavior. Unfortunately, because of the pervasiveness of TV signals
today, the possibility of gathering more evidence of this type is steadily
decreasing.
The experimental evidence on the causal impact of media violence
has been so consistent in favor of the conclusion that exposure causes
increased aggression that fewer experiments have been conducted in
recent years. However, one recent experiment that was reported by Zill-
mann and Weaver (1999) exposed participants to either four consecu-
tive days of gratuitious violence or nonviolence in the form of feature
films. As in earlier experimental results, their findings showed that the
participants who saw the violent films were more hostile in their behav-
ior subsequent to exposure. Unlike prior experiments, which tended to
show that participants would only show hostility toward a person who
had provoked them earlier, Zillmann and Weaver’s participants
showed such hostility regardless of whether they had been provoked
earlier or not.
Some researchers have attempted to study the possible facilitation of
aggression through exposure to media violence by recourse to the natural
experiment. Most notable among these attempts are studies by Phillips
(1979, 1983, 1986) and Centerwall (1989). According to Centerwall, prior
to television’s emergence in the United States, the national homicide rate
was 3 per 100,000. By 1974, the homicide rate had doubled. Centerwall
argues that this increase is directly linked to massive exposure to televi-
sion throughout the culture. He notes that essentially the same kind of
increase in homicides occurred in Canada. Moreover, he argues that
despite its similarities on nearly any variable of interest, homicides did