Page 53 - An Introduction to Political Communication Fifth Edition

Intro to Politics Communication (5th edn)-p.qxp 9/2/11 10:55 Page 32

POLITICS IN THE AGE OF MEDIATION
those of interpretation and contextualisation. Can a laboratory experiment,
no matter how sensitively prepared, really reproduce the complex political
environment in which individuals make their decisions? Can it compensate
for the weight of cultural and social resonances that will accompany a
political message in the real world?
To make these points is not to dispute the value of sensitively designed
empirical audience research in the study of political or any other type of com-
munication, but simply to highlight its limitations.

DOES POLITICAL COMMUNICATION WORK?
MICRO-EFFECTS

According to a MORI poll conducted during the British general election of
June 1987, the determinants of voting behaviour, particularly for the crucial
section of ‘floating’ or undecided voters who will ultimately decide the out-
come, are threefold. They are, first (and still, apparently, most importantly),
the image of party policy (44 per cent); second, the voters’ image of the party
leadership (35 per cent of choices in 1987 were attributed by respondents to
this factor); and finally, the ‘corporate’ image of the party itself (21 per cent)
(Worcester, 1991, p. 111).
Each of these aspects of a party’s identity have to be communicated,
suggesting at the very least that the ability and skill to communicate can be
important in influencing political behaviour and electoral outcomes.
Among the experiments conducted into the efficacy of political com-
munication at this level is Rosenberg and McCafferty’s study of the extent
to which ‘public relations experts [can] manipulate the public’s impression
of a political candidate’ (1987, p. 31). Their concern in this research was
with non-verbal aspects of communication, or the candidate’s ‘image’
defined in narrow, physical terms. As they put it, ‘we are interested in
exploring whether or not it is possible to manipulate an individual’s appear-
ance in a way that affects both voters’ judgments of the candidate and the
choice they make at the ballot box’ (ibid.).
To test the hypothesis that image does matter in shaping political
behaviour, Rosenberg and McCafferty selected a group of American uni-
versity students, whom they exposed to multiple photographs of a series of
fictional election candidates. The pictures differed in ways intended to
generate negative and positive responses, such as the inclusion or omission
of a smile. It was found that such changes affected ‘both the degree to which
an individual is perceived to be fit for public office and the degree to which
he is perceived to possess those qualities (competence, integrity and likable-
ness) that other research has shown to be relevant to voters’ evaluations of
political candidates’ (ibid., p. 37). Furthermore, even when subjects were
made aware of the respective candidates’ policies on important issues, image

32