I’ve been campaigning since I was a precinct chair my senior year of high school. In that time, I’ve knocked on thousands of doors and made thousands of phone calls to undecided voters. There may be a few people in the world who appreciate being interrupted during dinner or during the Gopher football game on a pleasant fall Saturday afternoon, but for the most part, people are politely stilted at best, and outright dismissive on the bad end…and I’m not even counting the times I’ve been physically threatened. Doorknocking never seemed all that effective to me, and so it’s not surprising that a rare academic study found that no, doorknocking doesn’t really work all that well. The question is whether this data will make a difference in campaign practices.
Political campaigning is probably one of the most change-averse activities out there. It’s been common wisdom for decades that you simply need to put up lawn signs, you need to have poorly-trained volunteers knock on strangers’ doors, you need to make phone calls to undecided voters. In all the time I’ve been campaigning, I’ve probably genuinely persuaded a voter about as often I was told to “Get off my property!” As the study seems to show, this is especially the case for infrequent voters. With few exceptions, people who don’t vote have chosen that on purpose, and it’s probably because they hate politics. Having a stranger, or more accurately multiple strangers, knock on your door to give you terrible campaign literature isn’t likely to turn a hater into a voter.
What does work? That’s a very good question, one that I’m not sure anybody knows the answer to. If we really wanted to find out what works, we’d run the same kind of A/B testing that is common in the marketing industry, testing out various messages. If I were Grand Party Poobah, I’d find a few candidates that have similar districts, and run a bunch of tests: doorknocking versus none, micro-targeting versus none, text-filled encyclopædic literature versus something people would actually look at without immediately tossing, etc.
One idea in particular that I would love to see tried out is a personalized opt-out on campaign literature. Two, in fact: two QR codes on a mailing, one that says “I’m voting for this candidate, please stop annoying me!” and one that says “I would never vote for this candidate, please stop annoying me!” Snap the relevant code and the data is saved to the candidate’s database. Candidates would like it because it would provide useful data on voters, with probable higher response rates than phone calls. Voters would like it because it allows them some control over the process (of course, if you opt out, the candidate would actually have to adhere to it or risk even more wrath…). Sure, the cost would be higher because a personalized QR code would need to be printed on each piece, and backend campaign systems would have to support this, but it’s technologically feasible.
Alas, no candidates for office would ever subject their candidacy to randomized experimentation for the greater good. Which is too bad: I think it would be a blast to get some meaningful metrics. As a result, we are probably stuck with the same old campaign strategies, especially for local races that don’t have a dozen hired media consultants. And that means more banging on doors for little gain for the forseeable future.