Friday, October 12, 2012

Winning the swing...


I have just returned from a few days in Wisconsin where I had the opportunity to meet some of the Republicans who are hoping to swing the state from blue to red. Its an interesting story. Wisconsin has voted Democrat in presidential elections for years but the margins in some recent elections have led Republicans to think that the state might be winnable.

Indeed Wisconsin was targeted in the 2004 election--as is mentioned in the fascinating book by The Victory Lab: the secret science of winning campaigns, by Sasha Issenberg. The book records how Republicans were faster on the uptake with the new science of microtargetting and this meant they could seek out Republican voters, or persuadable voters, in states that had formerly been assumed to be Democratic strongholds. Republicans still lost in 2004 but only by 0.4%, a tiny margin. Being able to find, and speak to, pockets of potential voters is now standard practice.

Of course in 2008, Mr Obama returned with an enormous majority in Wisconsin. But Republicans would like to think that much of this was down to his "celebrity" factor. And they are hoping that it has since vanished after four years in power. But I was in Madison on October 4th and caught an Obama rally with around 30,000. Clearly Mr Obama can count on huge support in cities like Madison and Milwaukee, particularly among the young. The question is, as ever, in a close race which side can pick off enough voters in the suburbs with their targeting techniques--which are undoubtedly even more advanced than they used to be. 

One of the most humbling things about the content of the Issenberg book is really the extent to which the science of what the campaigns do is actually so crucial and so unknowable. And much of what political journalists are writing about (myself included), the strategies, the targets, the choices politicians make to go to different places, are really just stabs in the dark at what the underlying real story might be. The campaign playbook is increasingly difficult to know with any accuracy because it will be data driven.

What campaigns do these days is increasingly based on evidence--evidence that we will not see during the campaign, and may not even know about for some time afterwards. After the 2004 election, some of those involved in the emerging science of micro targeting lied (or perhaps better to say mislead) about what they had done simply so that the other side would not figure out what they had done.

The other intriguing thing about this Issenberg book is that it shows how the internal campaign gurus of the old days, those that had a good story about what worked but no real evidence, were eventually replaced by hard data and it strikes me that the same revolution has not happened in journalism. We interpret the world, and the narratives we see, based on gut instinct and (for those who have been in the business for a long time) about what happened in previous elections.

For example, we journalists often moan about how campaign advertising has turned so negative. And we editorialise that this turns voters off--based on what we presume must be the more uplifting truth about humanity. Behind the scenes, though, negative campaigning is not some random decision made by some desperate campaign munchkin. There will be evidential support (support that journalists have never seen) which shows how many people are likely to be won, or lost, by a particular type of advertising. For heaven's sake, they've been testing messages since Kennedy had to decide whether to talk about his faith or not.

But we journalists think we know better, and we find a few real people willing to say that campaigns are more negative than they used to be and that it has turned them off politics. As if a couple of self-selected data points actually tells us anything about what is really going on, and when perhaps the story is that psychologists know negative messages work and political scientists can measure within fractions of a percent how many voters are won over. But we journalists do not have the data, so we prefer to write unflattering stories about campaigns... rather than unflattering stories about the flawed human beings that are being targeted.

Another point is that journalists are pretty upset that campaigns will rarely give on-the-record interviews but will is increasingly insisting on after-the-fact quote approval. Journalists assume that this is down to the fact that campaigns being overly cautious about gaffes. I don't wish to approve of the concept of after-the-fact quote approval but I do wonder that the reason campaigns insist on them is because an off-key message can have measurable effects these days.

All of this leads me to the uncomfortable conclusion that despite all our clever words we know far less about campaigns than we used to, than we ought to, and far less about the electorate than those inside the campaign themselves.  (See also this story from earlier this year.) Of course we have to string a story together about why one side is winning or losing, according to this poll or another. And after the fact we have to write about why one side has won or lost. But the more I learn about campaigns, the more I wonder whether journalists know a damn thing about what is really going on.




The swing states: Wisconsin 

Crossing the line 

A traditionally Democratic state is within Republican reach 

Oct 13th 2012 | from the print edition

PITY the confused inhabitants of Wisconsin. In June their Republican governor, Scott Walker, saw off an attempt to “recall” (ie, sack) him with the message that a new economic dawn had arrived. Now that a general election is under way, Republicans have made a hasty correction. Wisconsin’s economy is now struggling, thanks to the Democratic president. That is all politics as usual, of course, and local Democrats have undergone their own mirror-image conversion. But both parties sense they face a real fight for Wisconsin’s ten electoral-college votes. [More...]

2 comments:

  1. Paul Guinnessy8:23 pm

    There was an extremely good talk at the last AAAS meeting in DC on this very issue, particularly on how negative campaigning is effect, and how pointing out that your neighbors voted, increased the likelihood that you would vote in the next election.

    ReplyDelete
    Replies
    1. Voting is a social enterprise. If you tell people that everyone is doing it, and they know that their neighbours will find out that they have not voted, then people are WAY more likely to vote. If you want people to not vote then you send them mail bemoaning the fact that people don't vote much these days. We are social creatures and we have status anxiety.

      Delete