More evidence that "independents" don't "swing" by @DavidOAtkins

More evidence that "independents" don't "swing"

by David Atkins

I was recently referred to a good academic article on "independent" voter turnout. The study was done on Canadian elections so many exceptional Americans will dismiss it out of hand, but even so the dynamics in Canada only serve to support the notion that "independent" voters don't shift party allegiance from election to election, so much as stay home out of apathy and to punish their preferred party for not doing its job. Those without access to online scholarly articles via ScienceDirect can only get the non-technical summary of the article, Estimating Voter Migration in Canada
Using Generalized Maximum Entropy
, in Electoral Studies 26(4), December 2007, pp. 756-771.:

Do the non-voters decide elections? Voters don't so much swing as bounce on a long elastic tether. They may abstain from voting when they're unhappy, but they don't necessarily switch parties. Studying patterns of voter migration in the three most recent federal elections in Canada and the three most recent provincial elections in B.C. provides significant evidence that voters maintain long-term political preferences, a kind of "tethered partisanship", and are less prone to float and drift between political preferences than often thought. Political strategists haven't yet taken this lesson to heart, as much political campaigning is still directed at the elusive "swing voter", and much less at getting the "affinity voter" sufficiently motivated and into the voting booth.

And here's the academic abstract (again, only available to those with university access to online publications):

As voters switch political preferences from election to election, understanding the magnitude of voter flows among parties and transitions between voters and non-voters is an essential element of political analysis. As exit polls are uncommon in Canada, voter migration can also be estimated using suitable statistical techniques. Backing out micro-level voter migration probabilities from macro-level election data is a problem of ‘ecological inference.’ This paper uses the method of generalized maximum entropy (GME) to estimate voter migration patterns for the two most recent Canadian federal elections (2004 and 2006) and two most recent provincial elections in British Columbia (2001 and 2005). The estimation results answer important questions about voter behaviour in Canada. These results will be of interest to political scientists, historians, and politicians, as well as econometric practitioners who wish to estimate voter migration.

The article concludes that voters maintain a tethered connection to their party of preference, don't tend to switch parties easily, and prefer to simply stay home to show their disaffection. Smartly, the article concludes that:

Politicians may conclude from this that the low-hanging fruit of political campaigning are often found among non-voters, especially when a party is recovering from an electorally unsatisfying result in a previous election. Given that much electoral campaigning is directed at political adversaries rather than non-voters, the results found in this paper may give politicians pause thinking about allocating their campaign resources. Estimation of voter migration flows may help parties understand where ‘their’ voters went, and how to win them back.

By "non-voters," of course, the article refers not to people who consistently refuse to vote, so much as the occasional voter who goes to the ballot box generally for the same party, but only occasionally.

This is very smart.

Pollsters looking to see how to "win back" so-called "independent" voters will often do focus groups with people who crossed party lines from one election to another--say, those who voted for Barack Obama in 2008 but then voted for Republicans in 2010. They then analyze the data they get from those people to tell Dem politicians like President Obama what they must do to "win back" those independents.

But this is the wrong way of going about it. Sure, those "switcher" voters are out there. But they're dwarfed in number by the people who hold an allegiance to the Democratic Party and progressive principles in general, and may have voted in the big presidential election of 2008, but failed to turn out to vote in 2010. That's a much bigger cohort--and not only is it bigger in size, it's more winnable and courting it doesn't create resentment and anger within the Party base.

Sure, pollsters want to go after the switchers, because winning a switcher is like winning two votes as opposed to the one apathetic vote. A switcher who votes consistently is also easier to get, theoretically, than the inconsistent voter who has to be brought to the polls through labor-intensive field operations.

But the "switcher" is likely to be a Fox News "Democrat," or simply a dramatically underinformed voter who may have had a good connection to Obama the candidate in 2008, but gets much their information from right-wing radio, from their pastor, or from Aunt Martha's crazy email blasts--or worse, a natural conservative Republican who had only "switched" to Obama out of distaste for Bush in the first place. A bias toward pleasing these "switchers" thereby leads to unnecessarily conservative policies that increase the likelihood of the truly tethered independents to stay home, as well as increase the likelihood of anger from the Party base.

Smart Democratic consultants would do well to do focus groups with Dem voters from 2008 who stayed home in the midterms, and aren't sure whether they're likely to come out in 2012. See what is driving their anger and apathy, and what they want in terms of policy and message. And then insofar as decisions are made based on focus groups and polls, tailor the message to those people. My suspicion? You'll find a lot of those very sorts of people at Occupy protests around the country.

h/t to Opendna on twitter for referring me to the article.

P.S. As a small side note, my fiancee has access to Science Direct and Google Scholar through her being a grad student. Back when I was at university in 2003 doing academic work, I would never have found out about this article. If I had, it would have come from reading the bibliography of a paper article I had copied. I would then have had to travel to campus to find the journal; if my university didn't have it (as was often the case in my obscure field), an interlibrary loan would be been necessary, necessitating a week of waiting to go back. Once it arrived, I wouldn't have been able to take it home; I had to pay for making copies of the article at the library itself, before giving it back to the staff librarian. If I missed a page in the copying process, too bad. I had to wait another week--probably too late for the paper deadline.

Yes, this is a 10 miles to school in 100 feet of snow story, and I'm only 30 years old. But today's upcoming scholars are so lucky to have Google Scholar and instant access to academic work online, they have no idea.