I’ve been reflecting on my differences with John Sides about the impact of gaffes, and I realize that explaining counterfactuals–an issue that John raises effectively in his recent posting– is very challenging. It’s challenging to explain to students, who tolerate our philosophizing about the world of “what if.” It’s much harder with journalists, who are attuned to a world of facts.
What brought this to mind was a query today from a reporter. He wrote:
If campaigns only swing the popular vote 1.1 percentage points (or whatever), is there a reason for them to grab our attention the way they do? That’s the part of my piece that’s the thinnest — what campaigns are for if they’re not (quite) for convincing independent voters, given how few bona fide independent voters there are.
Here is my reply. The third point about making the models “work” is stolen straight from my dear friend and colleague John Aldrich back when the presidential forecasting models were first becoming popular. I believed it then and I believe it now.
Campaigns are about a lot of things.
Campaigns are for establishing the mandate for the post-election. Whether or not voters actually chose candidates based on campaign appeals or campaign events, it’s almost inevitable that the winner(s) claim a post-election mandate based on what they said in the campaign.
Campaigns are for firing up the base and mobilizing voters.
Campaigns make the models “work.” This is a hard one to understand, but think of it this way: what incumbent would NOT focus on a strongly performing economy when there is a strongly performing economy? If the campaign were to magically disappear, then voters would be much less informed about the economy and would have to make decisions on another basis.
But of course no one has this magical power, and no campaign would fail to talk about their advantages.
Thus, the campaigns “matter” but it’s hard to find an independent effect because campaigns are translation mechanism that educate the citizens about “reality.” And the media cover campaigns because they (you) are also empowered with the responsibility to inform the citizenry, and if you did NOT cover campaigns, you’d be shirking your duty.
And then finally, campaigns matter because they can sway those who are undecided. That is a small proportion of voters, but it can be a critical proportion.
P.S. A nice place to start on counterfactuals is Philip Tetlock’s book “Counterfactual Thought Experiments in World Politics.”