photo: patrick dentler
The day of the US presidential election is approaching. There is an expected turnout of 80%. Both parties have worked very hard to register as many new voters as possible and companies such as MTV and Starbucks have been encouraging people to register to vote.
Starbucks is offering a free cup of coffee to those who vote on November 4th. How could this distort the results of the election?
Well, It seems pretty logical that the people who feel strongly about whether they are Republican or Democrat or have a strong preference for either Obama or McCain are the people who would vote anyway, regardless of incentives such as free coffee. So this promotion probably wouldn’t affect whether they would vote.
Swing and undecided voters, on the other hand, may not vote without an additional incentive such as free coffee. If, say, undecided voters mostly lean towards Obama – the incentive of free coffee at Starbucks would benefit Obama in the polls by encouraging the undecided voters to go to the polling station and to vote for him.
photo: aubrey arenas
It would certainly be an interesting research to see whether this promotion or other incentives may distort the results of the election. The “Starbucks Stores per Capita” differs immensely between each of the states. The District of Columbia for example has 1.18 Starbucks for every 10,000 people – nearly 22 times as many Starbucks stores per capita to Arkansas which has 0.054 Starbucks for 10,000 people. Swing state Virginia has the 11th highest “Starbucks per Capita” of the states. If the Starbucks promotion does have an effect on swing voter turnout, we would expect the biggest effects to be in a) the states with the highest concentration of Starbucks and b) cities (which are of course more liberal than small town America) where people are more likely to have a Starbucks nearby.
I’m exaggerating the effects of a free cup of coffee on the election results you say. Perhaps so. But research has shown it can be quite easy to “prime” people to affect who and what they vote for. For example, research found that people who used a church as their local polling station were less supportive of gay marriage.
Another piece of research looked at a 2000 ballot initiative in Arizona to increase spending on education:
The authors…divided the precincts between schools and non-schools, and found that voters who voted in a school had a marginal preference (3 points) for the initiative.
photo: HAMED MASOUMI
And when I spoke to some local activists for the Labour Party (UK) earlier this year, they suggested that the Gordon Brown calls an election before 2009. Not because they believe he is more likely to win: they believe that Gordon Brown losing the next general election is already a done deal. It’s because the local elections are also due to be held in 2009. And having the general election at the same time as the local one would mean Brown’s personal unpopularity would rub off onto the rest of the Labour party and their local councillors.
There are many subtle ways of affecting the results of an election. Could free Starbucks have a significant one?