Context, Conversation and Conviction: Social Networks and Voting at the 1992 British General Election
| Published date | 01 December 1999 |
| DOI | 10.1111/1467-9248.00235 |
| Date | 01 December 1999 |
| Author | Charles Pattie,Ron Johnston |
| Subject Matter | Article |
Context, Conversation and Conviction:
Social Networks and Voting at the
1992 British General Election
CHARLES PATTIE*
University of Sheeld
AND RON JOHNSTON
University of Bristol
After some initial interest, analyses of contextual eects in British voting behaviour
have tended to downplay or ignore the role of face to face conversations between
electors. However, evidence from the 1992 British Election Study shows that
conversations with partisan discussants do act as a statistically signi®cant in¯uence
on voting. Those who discuss politics with supporters of a particular party are more
likely to switch their votes to that party, if they had not previously voted for it, and
less likely to switch to other parties. Conversations with family members are
particularly important, though talking to other discussants also plays a part.
Local context has long been recognized as an important dimension in
understanding voting: voters are in¯uenced by their immediate social and
geographical environments as well as by their own individual situations.1
However, the ways in which these contextual eects operate have proved
controversial. A variety of possible explanations has been evaluated, including
geographical variations in economic prosperity, varying political contexts, local
election campaigning, tactical voting, and inter-regional migration.2 However,
* Data from the British Election Study were gathered under ESRC grant by Anthony Heath,
Roger Jowell, and John Curtice. The data were made available to us by the ESRC Data Archive at
the University of Essex. Neither group is responsible for the analyses reported here.
1 J. Agnew, Place and Politics (London, Allen and Unwin, 1987); J. Agnew, `Mapping politics:
how context counts in electoral geography', Political Geography, 15 (1996), 129±46. For a dissenting
view, see I. McAllister and D. Studlar, `Region and voting in Britain: territorial polarization or
artefact?', American Journal of Political Science, 36 (1992), 168±99.
2 C. J. Pattie, D. Dorling and R. J. Johnston, `The electoral politics of recession: local economic
conditions, public perceptions and the economic vote in the 1992 British General Election',
Transactions, Institute of British Geographers, NS 22 (1997), 147±61; C. J. Pattie, R. J. Johnston and
E. A. Fieldhouse, `Winning the local vote: the eectiveness of constituency campaign spending in
Great Britain', American Political Science Review, 89 (1995), 969±83; D. Denver and G. Hands,
Modern Constituency Electioneering (London, Frank Cass, 1997); D. Denver and K. Halfacree,
`Inter-constituency migration and party support in Britain', Political Studies, 40 (1992), 571±80; D.
McMahon, A. Heath, M. Harrop and J. Curtice, `The electoral consequences of north-south
migration', British Journal of Political Science, 22 (1992), 419±43.
# Political Studies Association 1999. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main
Street, Malden, MA 02148, USA.
878
The 1992 British General Election
the political impacts of voters' social networks have been relatively neglected.3
Miller's widely cited comment that `people who talk together vote together'
notwithstanding, early interest in the in¯uence of conversations with family,
friends and workmates on voters' decisions has largely evaporated.4 In this
short paper, we re-examine the evidence for conversation as an in¯uence on
voting decisions by British electors.
Social Networks and Voting
On the face of things, the relative neglect of social networks is surprising. Most
people are deeply embedded in a variety of intersecting (and potentially cross-
cutting) social networks ± families, friends, neighbours, workmates, and so on.
In each of these social milieus, they are likely to encounter, and talk to, people
with a variety of dierent political views, some of which will agree with their
own, others of which will not. Through their conversations and interactions,
participants in these networks will therefore encounter a range of views and will
be potentially placed in situations where their own opinions are either reinforced
or are challenged.
Much research on the issue has concentrated on social networks in
neighbourhoods, and has employed indirect measures of the impact of social
networks on voting. A common strategy is to look at the relationship between
individual and local partisanship (as indicated by election results) or local
social conditions (measured from the Census).5 But whereas these partly
ecological studies do reveal patterns consistent with contextual eects, there is
little direct evidence of the electoral in¯uence of votes' conversations. Indeed,
one of the few pieces of research speci®cally designed to investigate this
directly suggested that few voters actually talked about politics with their
neighbours (to avoid con¯ict), and where political conversations between
neighbours did take place, they were either with people who shared the same
opinions (and hence only re-inforced pre-existing views), or were self-
cancelling.6 Indeed, some critics argued that the neighbourhood eect was
3 For exceptions, see: A. S. Zuckerman, N. A. Valentino and E. W. Zuckerman, `A structural
theory of vote choice: social and political networks and electoral ¯ows in Britain and the United
States', The Journal of Politics, 56 (1994), 1008±33; and R. Huckfeldt and J. Sprague, Citizens,
Politics and Social Communication: Information and In¯uence in an Election Campaign (Cambridge,
Cambridge University Press, 1995). For a more general discussion of political networks, see D.
Knoke, Political Networks: the Structural Perspective (London, Sage, 1990).
4 W. L. Miller. Electoral Dynamics in Britain since 1918 (London, MacMillan), p. 65. There was
some interest in conversations among voters in the late 1960s and early 1970s: see, for instance, K.
R. Cox, `The voting decision in a spatial context', Progress in Geography, 1 (1969), 81±117; M.
Fitton, `Neighbourhood and voting: a sociometric examination', British Journal of Political Science,
3 (1973), 445±72; P. J. Taylor and R. J. Johnston, Geography of Elections (London, Croom Helm,
1979).
5 E.g. Miller, Electoral Dynamics; M. Harrop, A. Heath and S. Openshaw, `Does neighbourhood
in¯uence voting behaviour ± and why?', in I. Crewe, P. Norris, D. Denver and D. Broughton (eds),
British Elections and Parties 1991 (London, Harvester Wheatsheaf, 1992); J. Curtice, `One nation?',
in R. Jowell, S. Witherspoon and L. Brook (eds), British Social Attitudes: the 5th Report (Aldershot,
Gower, 1988).
6 Fitton, `Neighbourhood and voting'.
# Political Studies Association, 1999
CHARLES PATTIE AND RON JOHNSTON
879
simply an artefact of residential segregation created by British housing
practices.7
It nevertheless seems premature to write othe neighbourhood eect, or
other social network eects, entirely. Numerous studies since 1980 continued to
show voting patterns consistent with the operation of local social networks:
people who live together do vote together to a considerable extent, even when
they come from dierent social backgrounds.8 Furthermore, the neighbourhood
is only one of the social networks within which people live, and it may not
necessarily be the most appropriate one for political conversation. We need to
explore other networks more extensively. If conversation with family, friends,
neighbours and workmates acts as a vehicle for...
To continue reading
Request your trial