Election prediction

Can polls tell us who is likely to win the election? Probably not, or, at least, only when the answer is so obvious that polling is unnecessary. It is not just a matter of finding out who is most popular. You need to balance the sample. A sample of all adults or all registered voters may give you one result while a sample of likely voters produces another. And defining likely voters is a difficult task. It is no good simply asking people. People don’t like to admit being apathetic and so even people with no intention of voting – or a slight intention, but no likelihood – will describe themselves as being very likely to vote.

Most polls presently show Mitt Romney with a small lead, though swing state polls are more even, and depend, in part, on which states the pollster defines as swing. One of the best checks to see if this is a properly balanced sample is to look at how the sample voted last time. If it shows that the people in your sample gave Barack Obama an eight point lead over John McCain then it is broadly representative of the 2008 electorate. (Obama won by seven points, but McCain’s voters averaged slightly older, and some will have died). But is the 2008 electorate representative of the 2012 electorate?

Anecdotal evidence suggests this year’s electorate will be somewhat different from last time. Republicans seem more motivated and Democrats somewhat less so. More Democrats than Republicans have voted early, but the margin on this occasion is much smaller than four years ago. And this year’s election is very different. 2008 was the first fully open election – with neither the president nor the vice-president on the ballot – since 1952. Strong early voting by Democrats was a sign of high enthusiasm. This time people who support the president can reasonably be expected to be sure of how they will vote. People who are against the president have, understandably, been more hesitant. They may not like Barack Obama, but some have been waiting to see what they think of Mitt Romney. And here the debates may have been critical. Only the first debate gave Romney a decisive victory, but in all three debates he was competent and presidential. He looked like someone who could cope. He went head to head with the most powerful man in the world and was not intimidated. He was, at least, the president’s equal.

So, who will actually vote? When the issue is the economy, there is some history of late deciders breaking for the challenger. The economy suggests the president will be sharply rejected. But, to this observer, the most telling statistic is that Mitt Romney is consistently leading among independents. Since most party supporters usually support their party, independents are usually the decisive factor. When they are very evenly split, the election sometimes goes against them. By the tiniest margin, independents favored John Kerry over George W Bush. But Republicans were more motivated than Democrats, and Bush won. By an equally slim margin independents favored Gerald Ford over Jimmy Carter. But Democrats then outnumbered Republicans by two to one, so Carter still won. 

With the nation fairly evenly split between the two parties, it is likely that the candidate favored by independents will win. That will be Mitt Romney.

Quentin Langley is a Senior Lecturer in Marketing at the University of Bedfordshire Business School as well as a freelance columnist published in the UK and all parts of the US. He blogs on social media and crisis communications at brandjacknews.com

%d bloggers like this: