Does Political Information Matter? An Experimental Test Relating to Party Positions on Europe
| DOI | 10.1111/j.1467-9248.2007.00698.x |
| Date | 01 March 2008 |
| Published date | 01 March 2008 |
| Author | Christopher Wlezien,James Tilley |
| Subject Matter | Article |
Does Political Information Matter?
An Experimental Test Relating to Party
Positions on Europe
James Tilley Christopher Wlezien
University of Oxford Temple University
This article focuses on whether the provision of ‘objectively’ correct information to voters about where
parties stand on an issue affects their placement of the parties, and ultimately their own position,on that
issue. Classic theories of how mass publics make voting decisions assume that voters are able relatively
accurately to place themselves and the parties on various issue dimensions.While these assumptions have
been challenged, it is generally assumed that the provision of new information makes voters’ placements
more informed.We explicitly test this idea using a survey experiment focusing on one political issue –
European integration.In the exper iment,all respondents were twice asked to place the three main British
parties and themselves on a bipolar scale of European integration. This was done towards the beginning,
and then at the end of the survey. Most respondents were also given information on the ‘informed’
positions of the parties, derived from expert survey placement. Our analyses indicate that individuals’
placements did change, and the tendency was related to both political sophistication and the inherent
difficulty of placing the party.Only less sophisticated voters updated their placements, and these changes
are concentrated on the placement of the Labour party,where the elite stance on Europe has been more
conflicted. Forall respondents we do not detect any corresponding changes in self-placement that would
be congr uent with ‘cueing’ effects.
Most would agree that the proper functioning of democracy requires an informed
public.1Citizens must have meaningful preferences and they must have reasonably
accurate information about what the incumbent government is doing, as they
otherwise cannot effectively reward or punish the government. To ensure
accountability and political competition, people also need information about
what other competing political parties and candidates prefer. It is not that people
need to know everything, but they do need to know something. They need to
have enough information to keep politicians honest; in effect, to correct ‘errors’
in their policy behaviour,that is, when their behaviour departs from what people
want.
Traditionally the scholarly view of the public has not been terr ibly flattering.
Early on, Walter Lippmann (1922) saw a public that was largely uninformed. He
argued that due to the sheer flow and increasing complexity of information, the
public could neither absorb nor truly understand what was happening around
them. Thus for Lippmann, rule by the people had become effectively obsolete.
The first mass surveys of voters in the 1940s and 1950s to a large extent
confirmed Lippmann’s arguments for it appeared that the public was badly
doi: 10.1111/j.1467-9248.2007.00698.x
POLITICAL STUDIES: 2008 VOL 56, 192–214
© 2008The Authors. Journal compilation © 2008 Political StudiesAssociation
informed about issues, policy and candidates (Berelson et al., 1954; Lazarsfeld
et al., 1944).Philip Converse (1964) extended these ideas to offer a more refined
view of the mass public. He depicted heterogeneity, where some people actually
have well-developed preferences and are attentive to, and therefore informed
about, politics.According to Converse,however,most people fall f ar short of this
ideal, and rely instead on more primitive cues, such as party and social group
identifications, or else short-term forces like economic performance. Taken
together,it would seem that the requirements for effective representative democ-
racy are not met, at least in a very broad way. At best, we appear beholden to a
handful of informed ‘elites’.
This, along with more recent investigations of electorates’ knowledge levels such
as Michael Delli Carpini and Scott Keeter’s (1996), is the starting point for much
of the work on deliberative democracy (see for example Fishkin, 1991; 1995;
Gastil, 2000).In this literature, representative democracy is considered to be in a
tenuous state.Public opinion is uninfor med,and most voters are largely incapable
of making decisions based on policy preferences. High incumbent re-election
rates indicate that sitting politicians have effectively stifled meaningful alternatives
for voters to choose from. It has engendered the misrepresentation, high levels of
non-voting and distrust that have been observed in the US, and increasingly across
the rest of the Western democracies. According to these scholars, the answer to
the civic neglect that they identify is deliberation, where citizens are brought
together to discuss issues face to face.2
Others claim that in fact things may not be as bad as these portrayals would lead
us to believe (Erikson et al., 2002; Soroka and Wlezien, 2005; Wlezien, 1995).
Recent research has argued that limited information need not prevent voters
from making ‘correct’ choices, as they can rely on cues or heuristics to overcome
the problems of little information (Lupia and McCubbins, 1998; Lupia, 1994;
Popkin, 1991; Sniderman et al., 1991). By either implicitly or explicitly defining
‘correct’ choices as the choices that would have been made in the presence of full
information, these authors have argued that cue taking leads to situations in
which parties/governments can be held to account.
This is not to gainsay the value of providing additional infor mation to voters.
After all, with more infor mation,there is reason to suppose that people’s opinions
will change; moreover with more ‘representative’ information opinions should
change for the better. Of course this is most likely where people have little
information to begin with, as there are presumably diminishing marginal returns.
Some prior research supports these suspicions ( Baum and Jamison,2006; Gilens,
2001; Kuklinski et al., 2000). Martin Gilens focused on the effects of providing
factual information to respondents. He shows that it makes a difference: when
given correct information many respondents changed their preferences. James
Kuklinski et al. show more specifically that although some factually misinformed
(on levels of welfare spending in the US) voters do update views when given new
information, this tendency is largely concentrated among the least politically
DOES POLITICAL INFORMATION MATTER? 193
© 2008The Authors. Journal compilation © 2008 Political StudiesAssociation
POLITICAL STUDIES: 2008, 56(1)
To continue reading
Request your trial