Heterogeneity and the calculus of turnout: Undecided respondents and the campaign dynamics of civic duty

• The paper analyses turnout propensities for decided and undecided respondents.
• It examines how the impact of various considerations vary over the election campaign.
• The influence of civic duty on turnout has a temporal component which is limited to undecided voters.
• The impact of civic duty on turnout is significantly larger for undecided voters in the final days of the campaign.
• The heterogeneous impact is primarily evident in the 2010 UK general election.

Accounts of turnout often maintain that citizens participate in elections because the expressive, instrumental and normative benefits associated with the act of voting outweigh the respective costs. Although the impact of those benefits has been empirically assessed in many studies, we know little about when and for whom this impact is stronger. To this end, this paper examines 1) how the effect of those benefits and particularly that of civic duty increases over the election campaign and 2) whether this increase can be attributed to voter heterogeneity. Survey respondents who have not yet decided how they are going to vote will be increasingly swayed to cast a vote on the basis of their civic duty and not other predictors of turnout. The empirical hypotheses are being tested by utilising recent rolling cross-section election studies from Britain. The results suggest that the influence of civic duty on turnout is stable for decided but increases for undecided voters the closer the election day looms.