Page 91 - Media Effects Advances in Theory and Research

80 SHRUM

judgments are constructed through the application of the availability
heuristic; that is, the magnitude of the judgments is positively related
to the ease with which an example can be brought to mind (Tversky &
Kahneman, 1973).

Testable Propositions

These general propositions can themselves be used to generate testable
propositions regarding the relation between television viewing and social
perceptions and the cognitive mechanisms that may mediate this relation.

Proposition 1: Television Viewing Influences Accessibility. Proposi-
tion 1 is a necessary condition for testing whether the availability heuris-
tic can explain cultivation effects. This proposition was tested by opera-
tionalizing accessibility as the speed with which judgments could be
constructed. Shrum and O’Guinn (1993) had participants provide preva-
lence and likelihood estimates of constructs frequently portrayed on tele-
vision (e.g., crime, prostitution) and measured the time it took partici-
pants to answer each question. If television information was more
accessible for heavy viewers than for light viewers, heavy viewers should
not only provide higher estimates than light viewers (a cultivation effect),
but should also construct their judgments faster (an accessibility effect).
The results of the study confirmed these hypotheses, even when control-
ling for individual baseline latencies, grade point average, and use of
other media. These same general relations have been replicated using a
variety of dependent variables, different operationalizations of television
viewing, and multiple control variables (cf. O’Guinn & Shrum, 1997;
Shrum, 1996; Shrum, O’Guinn, Seminik, & Faber, 1991).
Other studies have attempted to operationalize accessibility in a more
direct way. Busselle (2001) asked participants to recall an example of par-
ticular constructs, some of which were constructs frequently portrayed on
television (shooting, affair, doctor) and measured the time it took partici-
pants to recall the example. Busselle and Shrum (2000) used a similar
methodology, but also asked participants to indicate how easy it was to
recall the example. Heavy viewers were expected to be able to recall a
television-related example faster and easier than light viewers. The results
from both studies indicated no differences between viewers in speed of
recalling the examples, but subjective ease of recall was easier for heavy
viewers than for light viewers (Busselle & Shrum, 2000). The fact that
level of viewing was related to subjective ease but not to speed of recall
suggests that the conscious act of recalling an example may not be the
mechanism that is employed in the application of the availability heuris-