Page 26 - Media Effects Advances in Theory and Research

1. NEWS INFLUENCE ON OUR PICTURES OF THE WORLD 15

found strong negative correlations between the pattern of media salience
across these elections and the number of people who expressed com-
pletely ambivalent opinions about the candidates by checking the mid-
point of various rating scales used by the National Election Study
(Kiousis, 2000). Twenty of the 24 comparisons between the prominence of
the presidential candidates in the total news coverage—media salience—
and the proportion of the public without opinions were significant. Their
median value was .90. Similar results were found for the attribute
agenda, where 17 of 24 comparisons for the attribute of morality were sig-
nificant, and the median correlation between the pattern of salience in the
media for this attribute across the elections and the number of people
with ambivalent opinions was .80.
Mass communication effects can result from the sheer volume of expo-
sure, as we see in first-level agenda-setting. But, as both attribute agenda-
setting and priming show, closer attention to the specific content of mass
media provides a more detailed understanding of the pictures in our heads
and of subsequent attitudes and opinions grounded in those pictures.

SUMMING UP

More than 50 years ago, Harold Lasswell (1948) observed that mass com-
munication had three broad social roles—surveillance of the larger envi-
ronment, achieving consensus among segments of society, and transmis-
sion of the culture. Agenda-setting is a significant part of the surveillance
role because it contributes substantial portions of our pictures about the
larger environment. But the agenda-setting process also has implications
for social consensus and transmission of the social culture.
Evidence linking agenda-setting and social consensus was found by
the North Carolina Poll among demographic groups that are commonly
cited in public opinion polls as sources of differences rather than similar-
ities (Shaw & Martin, 1992). Comparison of the issue agendas for men
and women who infrequently read a daily newspaper was .55. But for
men and women who read a newspaper occasionally, the degree of cor-
respondence rose to .80. And, among men and women who read a
newspaper regularly, the issue agendas were identical ( 1.0). Similar
patterns of increased consensus about the most important issues facing
the country as a result of greater media exposure also were found in com-
parisons of young and old and black and white and were true for both
newspaper and television use. These patterns of increased social consen-
sus among demographic groups as a result of media exposure also have
been found in Taiwan and Spain (Chiang, 1995; Lopez-Escobar, Llamas,
& McCombs, 1998).