Page 97 - Media Effects Advances in Theory and Research

86 SHRUM

is a function of heuristic processing, then larger effects should be noted
under conditions that favor more heuristic processing (phone survey)
than under conditions that favor less heuristic processing (mail survey).
The results confirmed this speculation. Across five composite variables
representing perceptions of societal crime, societal vice (e.g., prevalence
of prostitution, drug abuse), marital discord, affluence, and the preva-
lence of particular occupations, the magnitude of the effects was signifi-
cantly larger in the phone survey condition than in the mail survey con-
dition for four of the five measures (as with Shrum, 2001, all but marital
discord).
Other evidence also supports the notion that ability to process informa-
tion has implications for the cultivation effect. Mares (1996) found that
people who tend to make particular kinds of source confusions (mistak-
ing fiction for fact) tend to exhibit a larger cultivation effect than those
who do not have a tendency toward those types of confusions. Thus, even
in instances in which people may be motivated to process information
(see Shrum, 1997), inability to properly process information (in this case,
accurately ascertain source characteristics) may facilitate a cultivation
effect.

Model Integration

The next step in model development is to integrate the testable proposi-
tions, and the implications of their supportive results, into a coherent con-
ceptual framework. This conceptual framework, which is presented in the
form of a flow chart in Fig. 4.3, specifies a series of links, or steps, which
lead from television viewing to the production of a cultivation effect. For
the most part, each link (designated by an arrow) represents a testable
proposition that has been empirically verified. As the figure indicates,
there are in fact a number of ways in which media exposure will not have
an effect on judgments (no cultivation effect), but only one way (path) in
which a cultivation effect will be produced.
In order to present as simple a model as possible, some misleading
aspects arise that should be clarified. One of the misleading aspects of Fig.
4.3 is that the links (Yes/No) and the outcomes (Effect/No Effect) are por-
trayed as dichotomous variables. In fact, it is more accurate to think of
each as a continuum, and movement along the continuum has implica-
tions for the size of the outcome. For example, rather than interpreting the
figure as “high motivation to process results in no cultivation effect,” it
may be better interpreted as “the higher the motivation to process, the
smaller the cultivation effect.”
This notion of a continuum is similar to the elaboration continuum that
forms the basis of the Elaboration Likelihood Model (ELM; Petty &