A new study released last week has shown that door-to-door canvassers can help change people's minds on the subject of transgender rights in particular. Two researchers from Stanford and University of California, Berkeley—David Broockman and Joshua Kalla—found that canvassers using a method developed by the Los Angeles LGBT Center helped significantly reduce bias against transgender individuals. FiveThirtyEight has the details:
The study used a common political science tool called a “feeling thermometer” to quantify emotions. Three days later, people who had experienced the L.A. LGBT Center’s persuasion technique showed an average 10-point increase relative to the control group in their positive feelings about transgender people, on a scale of 100. Three months later, that average 10-point increase in positive feelings persisted. For comparison, between 1998 and 2012, Americans’ positive feelings about gay and lesbian people increased by an average of 8.5 points. That change came about slowly, through a combination of cultural influence, explicit attempts at persuasion and implicit peer pressure. It’s considered to be one of the biggest success stories in the history of political persuasion, said Diana Mutz, director of the Institute for the Study of Citizens and Politics at the University of Pennsylvania. “It’s impressive that [Broockman and Kalla] were able to change minds in a short period of time and that it persisted,” she said.
The study was fascinating for a host of reasons, not least of which was that it found canvassers who were not transgender themselves were just as likely as trans canvassers to have positive results. (This conclusion counters the findings of a previous study that gained notoriety last year for being fraudulent. Broockman and Kalla actually stumbled upon the falsified results when they were trying to replicate the original study.) The fact that the researchers followed up with folks and were able to test the efficacy of the interaction was also pretty unique—reportedly, only about 20 studies in the last decade have taken a similar approach.
Here's some points of interest:
First, a description of the L.A. LGBT Center’s technique, which uses the Socratic method of successive questions to draw people into a conversation that can last as long as 20 minutes but averages closer to 10 minutes.
Canvassers are aiming for a conversation, in which they ask questions and the subject gets to talk. They don’t tell people ahead of time what conclusion they want to reach. There’s no sermon built in. The goal is that, by the end, subjects will have built up empathy with a group of people different from themselves. [...] Canvassers built empathy by leading cisgender people1 — those whose assigned-at-birth gender matches the gender they understand themselves to be — to think of times when they were judged by others and, then, to connect those feelings to how transgender people feel when they are judged.
Second, the survey was conducted in an unusually efficient and revealing way that both reduced the time and money necessary for canvassers to produce a sufficient sample size and allowed Broockman to examine the before/after results of the conversations.
He sent out mailers, inviting people to participate in a broad online survey. Only a couple of questions had anything to do with transgender rights issues. But, by taking the survey, people signaled to Broockman that they were more likely than the general population to open the door for a canvasser later. That meant fewer people had to be canvassed to reach the same power of result.
While it's possible that people willing to answer surveys might also have attributes that bias them or otherwise differentiate them from non-survey takers, the fact that people had essentially "opted in" by filling out a survey was true for both the control group and the treatment group. Perhaps, more importantly, the original survey gave Broockman a baseline view into their attitudes toward transgender people before the canvassing took place.
Third, because the researchers utilized the survey as a way to weed out people who were less likely to entertain canvassers who knocked on their doors, the research took considerably less effort and money to complete.
“If you wanted to do a study like this five years ago, it was going to be $2 million. [...] Now, we realized that we could do it for more like $25,000,” Broockman told me.
Interestingly, Broockman conducted a similar study related to people’s views of abortion and it failed to change attitudes. Exactly why is unclear. One possibility is that Americans’ views on abortion have become so entrenched over the last several decades, while transgender rights are still a relatively new political issue.