project
Comparative Cross-National Electoral Research
1 July 2010 - 30 June 2014
PI/s in Exeter: Professor Jeffrey Karp
CI/s in Exeter: Professor Susan Banducci, Professor Nicole Bolleyer Professor Dan Stevens, Professor Jack Vowles
Research partners:
Funding awarded: £ 237,333
Sponsor(s): Economic and Social Research Council (ESRC)
About the research
Imagine that two individuals share the same characteristics, such as their age, education, social class, and gender. Also, assume that they hold identical political values. If these two individuals were deciding whether to vote in the next national elections or whom to vote for in that election, we would expect them to make an identical choice. But now, consider if they lived in two different countries with different party systems. In one country, the incumbent government is comprised of a single party which dominates the political landscape holding a large majority of seats and in electoral contests only faces two fragmented opposition parties. In the other country, two evenly matched parties compete for a majority of the seats in a highly competitive political landscape. How would these variations in political context affect their political behavior? Holding this thought experiment in mind, we can ask some important questions. Why do some countries have close elections and others do not? To explain these differences, we might ask questions about how individual preferences differ across countries. We might also ask how the electoral system and the structure of party competition influences citizen preferences and behaviour. For example, do citizens respond differently when there is competition between political parties and candidates? Both types of questions motivate comparative cross-national electoral research [CCNER] but only with a cross-national approach can we explicitly model and analyse the impact of the political context on individual attitudes and behaviour.
Over the past 60 years, a number of cross-national data collection efforts have produced data for the analysis of CCNER allowing researchers to investigate the types of questions outlined above. The availability of these large scale survey projects have allowed researchers to produce a number of useful observations about cross-national variations in political culture and political participation. More recent attempts to link institutions and electoral behaviour, such as the Comparative Study of Electoral Systems [CSES], has produced useful evidence about how institutional variations, such as the differences between proportional and majoritarian electoral systems, influence political preferences, participation and vote choice.
However, there can be problems in doing research using existing comparative cross-national data sets such as the CSES, the European Election Study [EES] or the European Social Survey [ESS]. First, sometimes we may lack data on relevant countries. How much does it matter to the conclusions of previous studies that some democracies are excluded from the analysis, and if it does, what can be done about it? Second, the political context at the time the data are collected may influence whether individuals respond to the survey and how they respond. For example, responses to questions about political preferences may be influenced by when the question is asked in the election calendar – the further the election away, the less stable preferences might be. Finally, cross-national research requires that we take into account that individuals are a part of neighbourhoods, communities and countries. Two individuals sharing the same geographic location are more likely to be similar to one another than two individuals from two different countries or two different communities.
This similarity needs to be taken into account during the analysis of data. Our research and training programme in CCNER will train academic researchers and students in methods that address the above problems. Our strategy is to focus on the question of the analysis of electoral competitiveness as it illustrates many of the problematic issues of this type of research. We plan eight workshops, a web portal, supplementary materials that will include macro data and analytical techniques, and publications that together will build capacity for cross-national research in the UK.
