better equipped to relate this to a specific ballot proposition (see figure 1). In contrast,
when campaign intensity is low, voters have limited access to easily available
information and few incentives to make complicated decisions about the ballot proposal,
and they are thus more likely to vote randomly or depend more heavily on cognitive
short-cuts. Hence, a high intensity campaign is generally likely to produce a better model
fit, because voting behavior is more predictable. Following this line of argument, we can
formulate these testable hypotheses:
Hypothesis 1:
The higher the salience of the campaign, the more voters will rely on
their attitudes towards European integration when deciding in EU
referendums.
Hypothesis 2:
The higher the intensity of the campaign, the more information is
available to voters and the less likely individuals are to vote
in their attentiveness to political communication and
at this variance influences the process of opinion formation (see for example Alvarez and
model for why the political awareness of citizens has a low mean and a high variance. In
s will
costly for them to be
borne by other indiv
ansferable costs that must be borne by the individuals themselves (learning information
and relating it to attitudes and behavior) (Downs 1957:10). Some citizens are better able to
reduce these information costs than others, which implies that there will be significant
randomly. This, in turn, will increase the explanatory power of
factors such as attitudes and partisanship.
While we expect the campaign environment to affect how voters make decisions,
individuals also differ in their attention to political information, and these differences are
likely to influence patterns of voting behavior at the individual level.
Political awareness and voting behavior
Variation in individual exposure to political information is often operationalized in terms
of political awareness – that is, the extent to which an individual pays attention to and
understands political information (Zaller 1992:21). Many studies of opinion formation
have shown that people vary greatly
th
Brehm 2002; Zaller 1992; Converse 1975, 2000). Downs (1957) has provided a plausible
general, citizen
not be perfectly informed about political issues, because it is too
well-informed. These costs include transferable costs that can be
iduals (gathering, analyzing and evaluating information) and non-
tr
9
Convention