Page 151 - Media Effects Advances in Theory and Research

140 BANDURA

equal to, or outweigh nonmedia influences. Given the dynamic nature of
multifaceted causal structures, efforts to affix an average strength to a
given mode of influence calls to mind the nonswimming analyst who
drowned while trying to cross a river that averaged three feet in depth.
The view that the path of media influence is exclusively a filter-down
process is disputed by a wealth of knowledge regarding modeling influ-
ences. Human judgment, values, and conduct can be altered directly by
televised modeling without having to wait for an influential intermediary
to adopt what has been shown and then to serve as the diffuser to others.
Watt and van den Berg (1978) tested several alternative theories about
how media communications relate to public attitudes and behavior. The
explanatory contenders included the conceptions that media influence
people directly; media influence opinion leaders who then affect others;
media have no independent effects; media set the public agenda for dis-
cussions by designating what is important but do not otherwise influence
the public; and finally, media simply reflect public attitudes and behavior
rather than shape them. The direct-flow model from media to the public
received the best empirical support. In this study, the behavior was highly
publicized and could bring benefits without risks. When the activities
being advocated require the investment of time and resources, and fail-
ures can be costly, people are inclined to seek verification of functional
value from other sources as well before they act.
Chaffee (1982) reviews substantial evidence that calls into question the
prevailing view that interpersonal sources of information are necessarily
more persuasive than media sources. People seek information that may be
potentially useful to them from different sources. Neither informative-
ness, credibility, nor persuasiveness are uniquely tied to interpersonal
sources or to media sources. How extensively different sources are used
depends, in large part, on their accessibility and the likelihood that they
will provide the kinds of information sought.
Modeling affects the adoption of new social practices and behavior pat-
terns in several ways. It instructs people about new ways of thinking and
behaving by informative demonstration or description. Learning about
new things does not rely on a fixed hierarchy of sources. Efficacious mod-
eling not only cultivates competencies but also enhances the sense of per-
sonal efficacy needed to transform knowledge and skills into successful
courses of action (Bandura, 1997). The relative importance of interper-
sonal and media sources of information in initiating the adoption process
varies for different activities and for the same activity at different stages in
the adoption process (Pelz, 1983). Models motivate as well as inform and
enable. People are initially reluctant to adopt new practices that involve
costs and risks until they see the advantages that have been gained by
early adopters. Modeled benefits accelerate social diffusion by weakening
the restraints of the more cautious potential adopters. As acceptance