Indifference or Indignation? Explaining Purposive Vote Spoiling in Elections

Abstract

This article examines self-reported purposively spoiled votes using individual-level survey data from European Parliament election studies, 1989–2009. Three rival explanations, one centering on low political interest and compulsory voting, another on anti-party protest behavior and a further on anti-EU attitudes are evaluated to explain the puzzling behavior of turning out for elections while purposefully casting a spoiled vote. The results presented in this article give support to the second proposition. A perceived lack of party choices among the voting options and non-partisanship among voters otherwise satisfied with the democratic process seems responsible for, at first sight, the irrational choice of ballot spoiling. This behavior is best described as a mild form of anti-party protest voting.

Acknowledgements

We thank the anonymous reviewers, Dr Piret Ehin and especially the editor professor Rachel Gibson for very valuable comments on earlier drafts of the manuscript. All remaining errors are ours.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. Accessed November 20, 2013. http://www.france24.com/en/20120619-2012-06-19-2049-wb-en-webnews.

2. Accessed November 20, 2013. http://www.guardian.co.uk/politics/2010/may/06/general-election-2010-spoil-ballot.

3. Accessed February 5, 2011. http://www.themoscowtimes.com/mobile/news/article/448210.html.

4. European Election Study 1989, ZA2320; Eurobarometer 41.1 Post-European Election, June–July 1994, ZA2491; European Election Study 1999; European Election Study 2004; European Election Study 2009.

5. The vote choice question in the surveys offered a “spoiled vote” option as one answer after a list of parties/candidates. Respondents who reported the spoiling option were coded as vote spoilers and those who named a party/candidate were coded as respective voters. Respondents who reported not voting in the election on the electoral participation question were coded as non-voters. The exact wording of the participation and vote choice question has changed slightly between 1989 and 2009 and can be examined at http://eeshomepage.net/

6. The 1–10 scale used in 1999 and 2004 was simply split in the middle. The cases on the midpoint (5) on the 0–10 scale used in 2009 were randomly split, and one half assigned to the right and the other half to the left of the midpoint while the subsequent distribution was split in the middle as in 1999 and 2004.

7. Aggregate data were taken from the International Institute for Democracy and Electoral Assistance database. Accessed August 10, 2014. www.idea.int.

8. Table 1 shows that spoiling the ballot is a rare event. This can lead to biased logit coefficients and underestimation of the event probability (King and Zeng 2001, 693). Though pooling has given us a large enough case number not to be worried about this, just to verify that this is indeed not a problem here, rare event logistic regressions and simple logistic regressions were performed comparing spoilers to voters and spoilers to non-voters using the same independent variables as in the multinomial regression. No differences in coefficients up to the third decimal for the two types of regressions where observed. We are comfortable that the results are robust.

Additional information

Funding

This research was supported by the Estonian Science Foundation [grant number 7903], [grant number 8676] and Estonian Targeted Financing [grant number 0180128].