Social Class and Party Choice in Northern Ireland's Ethnic Blocs

Abstract

The peace process in Northern Ireland has not diminished the acute ethnic electoral faultline between the majority Protestant British population, supportive of parties favouring Northern Ireland's continuing place in the United Kingdom and the minority Catholic Nationalist population, which backs parties harbouring long-term ambitions for a united Ireland. Within each bloc, however, there has been a dramatic realignment in favour of parties once seen as extreme and militant. The Democratic Unionist Party has emerged as the main representative of the Protestant British population, whilst Sinn Fein, having for many years supported the Provisional IRA's ‘armed struggle’ against British rule, has become the dominant party amongst Catholic Nationalists. As both parties have entered the political mainstream and advanced electorally, to what extent have they moved from their electoral near-confinement among the working class to enjoy broader cross-class support – and how?

Acknowledgements

We wish to offer grateful thanks to the two referees for their very helpful and constructive comments on the original submission of this article.

Notes

1. The 2001 Northern Ireland General Election Survey found that 28.7 per cent of voters graded the DUP between 0 and 4, on a scale of 0–10, where 0 = very left-wing and 10 = very right-wing, compared to 20.4 per cent of voters similarly ranking the UUP. Almost half (49.6 per cent) of voters graded the DUP's loyalism as a maximum 10, compared to only 20 per cent in respect of the UUP.

2. 2001 Northern Ireland Election Survey, http://www.essex.ac.uk/bes/data.html: 51.1 per cent claimed to belong to a social class, of whom 22 per cent said middle class, 28.5 per cent working class and 0.6 per cent ‘other’. The remainder claimed not to belong to a social class. Only 3.2 per cent of survey respondents declined to offer a national identification.

3. ARK. Northern Ireland Social Attitudes Survey, 1989–1996 (computer file). ARK http://www.ark.ac.uk/sol/ (distributor); Northern Ireland Life and Times Survey, 1998–2006 (computer file). ARK http://www.ark.ac.uk/nilt (distributor). Years not included are 1992 and 1997, due to a lack of survey in those years.

4. Given our focus on intra-ethnic bloc competition, other parties can reasonably be excluded. However, abstaining voters who prefer not to vote, rather than shift their support to the DUP or SF, could potentially be modelled across time in a multinomial logit model. Given the use of notional party support rather than actual vote, and the assumptions underlying such a two-bloc, three category contrast which would necessarily complicate the analysis, this is something to be explored in future research using electoral study data.

5. The main effects models are available from the authors on request.

6. A model looking at time interaction with age showed no evidence of this relationship changing significantly across time.

7. Recall that the survey question measures possible vote intention, rather than vote per se, and therefore may well under-represent actual voting.

8. The turnouts were: 2005 general election: Nationalist-held constituencies (8): 69.2 per cent; Unionist held constituencies (10) 58.6 per cent; 2007 Assembly election. Majority Nationalist candidates elected (7) 69.0 per cent; majority Unionist candidates (9) (58.2 per cent). Two constituencies did not yield an overall majority of unionist or nationalist elected candidates.