Researchers examining ballot order effects have typically focused on only one type of order (e.g. alphabetical, rotational, or rank order). They appear to reason as follows: if there is only one ballot, there must be only one ordering of candidates; paraphrasing the popular saying, ‘one ballot, one ordering’. However, they sometimes overlook the fact that they may be neglecting the phenomenon of ‘double ordering’ of candidates. Double orders may occur when two distinct orderings operate at the same time on the same ballot. For example, candidates arranged alphabetically (from A to Z) are additively numbered (e.g. from 1 to 10). This way, although the main order looks alphabetical, a secondary numerical order runs in parallel. In this paper, I examined a unique ballot paper design used in Polish mayoral local elections that incorporates two distinct ballot orders – one intentionally introduced, i.e. alphabetical order, the other a completely unnecessary element of the ballot paper, i.e. numerical order. I conducted a lab randomised experiment combined with a post-experiment questionnaire to disentangle the two orderings. I demonstrated that both orderings significantly favour candidates listed at the top of ballots, even though the alphabetical order operates more intensely.
No takes yet. Share an insight, caveat, or question.
Paweł Stępień (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: