Page 55 - An Introduction to Political Communication Third Edition

POLITICS IN THE AGE OF MEDIATION

temperature of the air around it, so a public opinion poll becomes
part of the data upon which individuals calculate their future
political moves.

Voting behaviour
A second way in which the effects of political communication can
be measured is to observe patterns in actual voting behaviour.
Such evidence is clearly more tangible than opinion poll data, and
frequently contradicts the former (as in the 1992 general election,
when most opinion polls failed to predict a Conservative victory). It
is no less difficult to interpret, however. The relationship between a
party’s campaign and its eventual vote may not be apparent. Despite
the famous ‘Kinnock – the Movie’ party election broadcast (PEB) 5
shown during the 1987 campaign, and a communication strategy
widely viewed as superior to that of the Conservatives, the latter’s
actual vote on polling day was virtually identical as a percentage of
the national electorate to figures generated by opinion polls taken
at the beginning of the campaign (43 per cent). Labour’s support
rose by only 3 per cent from the beginning to the end of the
campaign, to give them a net gain of twenty seats on the 1983 result
(Butler and Kavanagh, 1988).
This could be interpreted in several ways. Perhaps the campaign
had no significant impact on the electorate (as opposed to the
commentators who almost universally praised it). Perhaps Labour’s
vote would have been even worse without the softening impact of a
good campaign. Perhaps voters recognised the quality of Labour’s
campaigning but regarded policies as more important than image,
and preferred those of the Tories. Any or all of these assertions could
be true, highlighting the deeper truth that even ‘objective’, empiri-
cally verifiable measures of voting behaviour (this is how people
actually voted) are subject to wide variations of interpretation.

Experimental research
The third method of assessing the effects of political communi-
cation shares with the first the fact that it relies on asking questions
of people. Numerous experiments have been conducted in which a
particular element of the political message is isolated before a subject
group. Their responses are then noted and conclusions drawn.
This laboratory-based approach is a much-used tool of
behavioural effects research, frequently employed, for example,

34