Thursday, October 20, 2011

Three Lessons from the Third Berkshire

We lost the special election in the Third Berkshire District, albeit narrowly (one percentage point). But within the defeat, to misquote Winston Churchill*, there were several successes. By learning from these successes – and from the overall defeat – we can build stronger Green campaigns that will turn our candidates into legislators and our policies into law. What follows is not a complete post mortem, but a brief overview of three lessons we can learn about electoral communication.

1. Communication is an Exchange
An experienced professor once explained to me, in a nutshell, one of the reasons his service-learning program was so successful: “We meet our students where they are,” he said. Over the years he went on to show me what he meant by that phrase, and his praxis now informs my approach to electoral communication.

So how do we go about meeting voters where they are? Many GRP members care passionately about combating climate change, democracy, eradicating inequality and injustice, and transforming capitalism. Understandably, those are the issues we want to talk about. But those are not the issues most voters want to hear about. Or rather, those are not the issues they want to hear about from people running for State Representative. Even if their understanding of the role of a legislator is somewhat inchoate, regular voters know that State Representatives deal with state-level issues, such as the amount of money the state dedicates to their kids’ schools.

Does that mean we should ignore the big issues like climate change, poverty, and the iniquities and inequities of one-party domination? No. Should we talk down to people? Again, no. It simply means that we have to present our message in such a way that voters can perceive it through their pre-existing lenses. We can only learn about those pre-existing lenses by conversing with voters and really listening to what they have to say; not by commissioning phone polls, but by engaging in authentic one-to-one conversations and then reflecting on them.

From our conversations on the doorsteps and front porches, we learned that our target voters cared deeply about bringing jobs to the district and resented the fact that the last two state reps walked away from the task of representing them and straight into something more lucrative. So that is where our conversation with the voters went next. Here is the front of a postcard we sent to those target voters to let them know we were listening and shared their priorities.


2. Teach Republicans to Vote Tactically
A few days before the special election, we sent a mailing to 1,100 households with registered Republicans who had voted in the November 2010 election. Its message was simple (see the front of the postcard below) and it may have helped sway some Republicans. Nevertheless, on election day the Republican candidate received about 900 votes.  If just one hundred or so of those votes had come our way, we would have won.

It was obvious from the outset that the Republican candidate was not going to win. So, from a rational-choice perspective, it is unfathomable that 900 people would give him their votes. After all, these were individuals who did not want to see a Democrat win, but acted in a way that they should have known would increase the risk of a Democratic victory. For at least some of them, lack of information may have played a role in the decision. They may have perceived our bald assertion that the Republican could not win as self-serving propaganda. Presenting them with independent verification might have made a difference and encouraged them to vote tactically.

In future races where the Green has a better chance of beating the Democrat, we need to persuade Republicans to vote tactically. We need to learn how to deliver the two-horse-race message more effectively.

3. Known Unknowns
But among a few Republican voters, our message may have backfired and so enraged them that it redoubled their resolve to go out and vote Republican (a vote-and-be-damned attitude). How do I know this? I don’t, and I know that I don’t. It is, in the words of Donald Rumsfeld, a known unknown. This brings me to my final point: We need well-designed campaign evaluations.

Our party needs to invest in the kind of surveys and focus groups that will enable us to learn from our mistakes in a more rigorous, fact-based manner. Our analysis has to become less anecdotal and more analytical. Analyzing the effect of our campaigns on voting behavior is much more difficult than program evaluation in some other areas. For example, determining whether an advertising campaign aimed at reducing cigarette use among teenagers did, in fact, help reduce teenage smoking is relatively straightforward compared with learning what it was, in particular, that persuaded un-enrolled women voters aged 30-50 to vote for a particular candidate.

Voting certainly sends a message, but the message it sends is blunt, ambiguous, and inarticulate. Teasing out the intent of the voter presents a challenge, but the growing discipline of program evaluation offers us a choice of tools for doing just that. As a party, we need to use those tools so that we can learn what it is that voters are telling us when they engage in the act of voting.

*“We must be very careful not to assign to this deliverance the attributes of a victory. Wars are not won by evacuations. But there was a victory inside this deliverance, which should be noted.” Winston Churchill, 1940, regarding Dunkirk.