Acceso usuarios
Richard Webber, Colin Rallings, Galina Borisyuk, Michael Thrasher
Ballot order positional effects, electoral advantage gained from a candidates surname location on the ballot paper, have been extensively studied. For ‘low-information’ elections especially candidates located at or near the top of the ballot paper are advantaged relative to their competitors. This examination of alphabetic bias uses data from local council election results in Britain. The examination of election contests featuring more than half a million candidates provides evidence of an alphabetic bias which increases as the number of vacancies increases. We calibrate the effects upon votes and vote share of alphabetical ordering of candidates. Comparing votes cast for last- and first-placed candidates in the ballot order demonstrates a clear advantage to those placed first. This increases in size as both the number of seats and competing candidates increases. Those located in the top half of the ballot paper are more likely to finish in the top half of the vote order. Subsequently, the distribution of surnames among elected councillors is clustered towards the top of the alphabet. Measures could remove the effects of alphabetic bias, such as randomising name order and introducing experiments in local council electoral practice, a cost-effective means for evaluating change.
