Over the weekend, news spread that Republican presidential candidate Ted Cruz, seeking to drum up support during Monday’s Iowa caucus, had done something pretty sleazy: His camp had sent out official-looking mailers to Iowa voters with “VOTING VIOLATION” stamped ominously in red on them.
As you can see, the mailers “graded” voters on their past voting attendance and encouraged them to “CAUCUS ON MONDAY TO IMPROVE YOUR SCORE.”
In addition to the fact that there’s no such thing as a “voting violation,” meaning this mailing probably freaked out a bunch of Iowans who had done nothing wrong, Mother Jones makes a strong case that the scores were likely made up, anyway — amusingly, the Cruz campaign sent a mailer to a political-science professor at Iowa State who is very familiar with his own voting record, and he told MoJo that he was given a score of F despite having voted in five of the last six primary and general elections.
But setting aside the morality of this, it’s fair to ask: Would it be likely to work? This is, after all, a behavioral-economics-flavored attempt to nudge people’s behavior in one particular direction, toward voting, and there’s been a lot of research on this.
MoJo notes that “This direct-mail strategy is inspired by social science that shows that a citizen is more likely to vote if he knows his neighbors will be told whether he went to the polls.” Indeed, the mailer contains a line indicating that “A follow-up issue may be issued following Monday’s caucuses,” hinting at a wee bit of public shaming should someone fail to vote.
That’s true, but there’s an element of the mailer that may also nudge voters away from voting — even setting aside the ick factor surely spreading over Iowa this morning as people find out about the dishonest mailing. Look at the numbers in the above tweet: It suggests everyone in his immediate neighborhood is crappy at voting. There’s a chance that voters’ response to this wouldn’t be “Oh, I better vote,” but rather “Hmmm, I guess voting just isn’t that important to my local community, so I’m not going to bother.”
In my 2014 article on why awareness-raising is overrated, I noted the possibility for this sort of backfire effect during certain types of campaigns:
In the most unfortunate cases, raising awareness can have the opposite of its intended effect. … In one study famous to social scientists, visitors to Arizona’s Petrified Forest National Park appeared to be more likely to steal petrified wood when presented with information about the high frequency of other park visitors’ pilching, because the information “normalise[d] undesirable conduct,” as the researchers put it — if everyone else is stealing wood, who cares if I take some, too? (The same logic applies to promoting other sorts of environmentally conscious behavior.)
The authors of an International Rescue Committee literature review on preventing gender-based violence came to a similar conclusion: “Awareness campaigns [about gender-based violence] often propagate a descriptive norm that [violent] behavior is prevalent in the community, perhaps licensing violent behavior rather than activating behavior to reduce GBV,” they wrote. One of the co-authors, Laurie Ball Cooper, told me that the #YesAllWomen campaign could be an example of this. “If your focus is on how common the behavior is, you may actually reduce the likelihood of a bystander stepping in to to stop it, or you may reaffirm the perpetrator’s belief that they can do whatever the undesirable behavior is without repercussion,” she said. (Though she also pointed out that the campaign could have important goals beyond behavior change, like giving heretofore silent women a forum for relating what happened to them.)
Now, the Cruz mailer didn’t only describe neighbors’ voting behavior. There’s other stuff in there, too. Dr. Hana Shepherd of Rutgers, who recently co-authored an important paper on using social-norm nudging to fight bullying, told me in an email that the Cruz mailer is “basically pitting the power of social norms (in the negative sense, as you noted) against the power of public shaming. I don’t know any research that pits them against each other, probably because it would be impossible to get lying about your neighbors through an IRB … ”
So it’s unclear whether the shaming aspect would win out over the nobody’s-voting aspect, in other words. Not that that’s close to the biggest reason this mailer was a bad idea.