Page 43 - Media Effects Advances in Theory and Research

32 ZILLMANN

It would be premature, however, to conclude that base-rate informa-
tion is always inconsequential. Krupat, Smith, Leach, and Jackson (1997),
for instance, explored the influence of reliable quantitative information in
situations that gave it great diagnostic significance for a purchasing deci-
sion (specifically, the choice of a car), whereas a competing single exem-
plar seemed anecdotal. Under these conditions, the abstract, quantitative
information exerted dominant influence on judgment.
Base-rate information, then, can have informational utility (cf. Zillmann,
2000) that fosters attention and careful processing, ultimately giving such
information due influence on judgment. In principal terms, information
that relates to individuals’ immediate and prospective encounter of
predicaments or opportunities is thought to have utility for these individ-
uals, the degree of utility increasing with (a) the perceived magnitude of
threats or incentives, (b) the perceived likelihood of their materialization,
and (c) their perceived proximity in time. The reported findings suggest
that, in these terms, informational utility is rather high for purchasing
decisions and comparatively low for most issues presented in the news.

Effects of Exemplar Distributions

Effects of distributions of exemplars that differ in relevant characteristics
have been examined for series in which some exemplars are supportive of
a given issue, whereas others are opposed to it.
Brosius and Bathelt (1994), for instance, explored the perception of
public opinion concerning, among other things, the quality of a locally
produced wine. A radio broadcast featured interviews of wine drinkers
who either derogated or praised the wine. The ratio of negative and posi-
tive evaluations was systematically varied, however, ranging from no
unfavorable versus four favorable evaluations to the inverse situation.
Intermediate ratios were created as well (i.e., 1 vs. 3, 2 vs. 2, and 3 vs. 1). In
addition, the interviewer provided survey data on public opinion that
were either consistent or inconsistent with the distribution apparent from
the exemplars. The findings show that the respondents based their esti-
mates of the public’s liking and disliking of the wine on the exemplar dis-
tribution, even when the apparent distribution was contradicted by the
survey data. Fig. 2.1 displays the findings of this experiment.
An investigation by Daschmann (1999) in the realm of political commu-
nication yielded essentially the same results. The distribution of interviews
of voters presented in a print-news report was analogously manipulated
(i.e., the ratios of voters for candidate Avs. candidate B) and supplemented
or not with consistent or inconsistent survey projections of the upcoming
election. Respondents’ estimates of the vote were again a function of the
exemplar distributions, not of provided base-rate information.