Location, Location, Location

When election time comes around, there is endless debate over voter statistics: who's voting for who, voting trends within different social groups, etc. Undoubtedly, candidates are always looking for a way to swing the vote in their favour, and they will try a multitide of tricks to change the way people vote. But one aspect of voting patters has so far gone ignored: the location of polling stations. If the location of polling stations could affect the outcome of a vote, it could end up being a powerful way of affecting voting patterns. As it just so happens, new research has shown that it might just be the case.

A study recently published in PNAS¹ has shown that the location of a polling station can have a bias on the outcome of the poll. The authors predicted that a voting station in a school would have a higher proportion of voters supporting a raise in taxes to increase spending on education than polling stations set up in other areas. They then analyized data from one such poll, done in Arizona in 2000. The data showed that stations set up in schools had a higher ratio of voters in support of the tax increase than did other stations. They also performed a second study, using an online poll. Participants were given the same online poll, but the background images were changed between participants; some were shown images of schools and others were shown control images (offices, etc). The results echoed the analysis of the 2000 Arizona poll data: the polls done in a school enviornment were more likely to vote in favour of raising taxes to increase spending on education.

This means that the polling stations can be used to manipulate the outcome of any poll. Stations set up in churches will be more likely to vote against gay marriage. Set up a polling station in a laboratory to swing a vote in favour of stem-cell research. Since churches are often used as a polling station in rural comminuties, this could mean that liberal/democratic candidates are facing an unfair disadvantage.

There are obvious ethical implications of this. Polls are supposed to be fair and unbiased: using the polling location to your advantage - "contexual priming of polling locations", as the study authors call it - should be against the rules. The findings of this study beg for an unbiased, fair location for polling stations. In a time where voter support on hot-issue topics can be changed by as little as where candidates ate breakfast this morning, however, achieving a state of total non-bias is going to be a bit easier said than done.

--------------------------------------------------------------------------
1. J.Berger et al. Proc. Natl. Acad. Sci. USA doi:10.1073/pnas.0711988105;2008