Study Shows Deep Canvassing Can Efficiently Change Anti-Trans Attitudes

Notes From A Queer Engineer_Rory Midhani_640Header by Rory Midhani
Feature image by Shutterstock


As Yvonne linked earlier, a recent study — “Durably Reducing Transphobia: A Field Experiment On Door-To-Door Canvassing” by David Broockman and Joshua Kalla — demonstrated that 10-15 minute conversations focusing on shared experiences were able to positively change one in 10 voters’ attitudes on trans people. Both cisgender and transgender canvassers were shown to be effective, and repeated re-measurement at the three month mark showed no reversal in attitude. The study is the first large-scale, real-world experimental effort showing that lasting opinion change is possible. Even better: all of the data is publicly available and it’s actually for real this time!

Real quick, let’s rewind. Last May, numerous scientists and queer activists (and queer activist scientists!) were dismayed by the retraction of a different, high-profile study that had indicated anti-gay prejudice could be dismantled through personal contact. Rachel wrote at the time:

A fairly well-known study in support of the “contact hypothesis,” which argues that knowing someone in the LGBT community increases the likelihood that a person will support LGBT causes, was redacted this week after it was revealed that its data was just totally, totally made up. The study by Michael J. LaCour of of UCLA claims to have sent canvassers door to door to talk to people about same-sex marriage for 20 minutes; after that conversation, survey respondents were asked about their attitudes towards it, and the data found that they were likely to feel more positively about it than they had previously. … It now seems that this was pretty much entirely made up — canvassers really were sent door to door, and they did talk about same-sex marriage, but people surveyed after weren’t actually asked about their opinions on the issue, and data wasn’t actually being collected from them regarding the study. LaCour is continuing to claim that the data is real, but admits that none of the funding for the study was used. It’s not clear yet how or if this will impact larger understandings of how LGBT activism should work — and it’s important to note that LaCour’s deception doesn’t mean that the contact hypothesis is false — but it’s a fascinating glimpse at how the sociopolitical currency of LGBT issues intersects with the world of academia.

Well intentioned though LaCour may have been, making up scientific results in support of a cause is an extremely risky (not to mention unethical!) endeavor. His study was, at best, a precariously placed stepping stone; at worst, a destructive distraction and embarrassing eroder of public trust. As any reader of this website is well aware, most LBT organizations aren’t exactly flush with cash. False scientific claims waste our community resources, siphoning money and support away from the people and projects where it would do the most good. I find this very hard to understand or forgive. So it’s a relief to see the level of transparency that researchers Broockman and Kalla [who uncovered LaCour’s “irregularities” while attempting a follow up study] have disclosed about their methods. 

Early last year, residents of Miami-Dade, Florida, were battling to keep transgender nondiscrimination ordinances they’d recently passed on the books. Concerned that the law would be overturned in favor of a discriminatory bathroom bill, South Florida’s largest and longest-serving LGBT organization, SAVE, partnered with the Los Angeles LGBT Center to canvass voters in conservative neighborhoods and influence their opinions on trans rights. The canvassing took place throughout Miami from January to June. Of these, Broockman and Kalla’s study looked at the last six canvassing runs, which took place in June 2015. During this time, 56 canvassers talked to a total of 501 voters. Although the below video isn’t part of the study data (this conversation took place in LA nine months later), it follows the same Leadership LAB-developed approach:

In the recording, you can see the canvasser asking a series of open-ended questions to prompt their conversation partner to think about a time they felt judged for being different. This is in line with the procedure canvassers in the Broockman and Kalla study were instructed to follow: establishing contact; encouraging “active processing” by soliciting opinions in a non-judgmental way; encouraging perspective-taking by soliciting personal experiences or sharing personal stories; then encouraging further active processing by asking for their opinion now, asking if and why the conversation changed their attitude. Canvass volunteers were given training before each canvass, including viewing video of past canvass conversations and participating in role play. Both first-time and experienced canvassers were effective, though experienced canvassers were especially effective.

Broockman compares the canvasser’s role to that of a therapist, calling it “a kind of theoretical cousin to cognitive behavioral therapy. It is about the canvasser’s skill in getting the voter to do mental work.” Essentially, the theory is that opinion shifts are more durable when people come to the conclusion themselves. Broockman and Kalla tested durability by looking at both longevity (re-measuring attitudes at three months) and resistance to attack (re-measuring after showing subjects anti-trans political attack ads). By both measures, the deep canvassing interventions were successful to a surprising degree. The paper compares it to public change in attitude toward gays and lesbians between 1998 and 2012 — a large, apparently lasting shift, all accomplished in a 10-15 minute timeframe!

Among the many interesting aspects of this story, one particularly nerdy footnote that caught my eye was the super-efficient way these canvassing routes were planned. Usually, canvassers knock on lots of doors and get turned away at almost all of them, making the whole endeavor very costly and time-intensive. For this study, however, they pre-selected canvassing participants by sending out mailers inviting people to participate in a broad online survey. Those who responded, the researchers hypothesized, would be more likely open the door for a canvasser later. Their prediction turned out to be correct in a big way. Broockman estimates that what LaCour had supposedly done for $1M, he and Kalla were able to accomplish for $25,000.

Again, given the limited budgets LGBT organizations typically work with, we need to make every dollar count. This is particularly true for trans advocacy. Today, anti-trans legislation continues to push across the country. Bigots actively conduct campaigns of misinformation to promote fear of trans people; meanwhile, trans people (mostly black trans women) are regularly being murdered. We need an intervention, fast. Perhaps the results of this remarkable study can serve as a starting point.


Notes From A Queer Engineer is a recurring column with an expected periodicity of 14 days. The subject matter may not be explicitly queer, but the industrial engineer writing it sure is. This is a peek at the notes she’s been doodling in the margins.

Laura Mandanas is a Filipina American living in Boston. By day, she works as an industrial engineer. By night, she is beautiful and terrible as the morn, treacherous as the seas, stronger than the foundations of the Earth. All shall love her and despair. Follow her: @LauraMWrites.

Laura has written 211 articles for us.

20 Comments

  1. Fascinating! I wish I could pull off the non-judgmental attitude when faced with homophobic and transphobic attitudes. I tend to start preaching and we know how quickly and easily people get defensive about their opinions and beliefs. Recognizing the role of active processing in opinion shifts does make me rethink my approach; suspending judgment tends to make me feel disloyal both as an ally and community member, but it clearly serves an important role in furthering the causes I believe in.
    Like if I was to meet the woman in the video and she told me all about how non-judgmental she was towards her niece while persistently misgendering her and then acting all clueless as to why her niece had stopped talking to her… Well. It would be so hard to stay open and not point out everything she was doing wrong. 🙂

    • Yeah that’s the stuggle for me, too! I think this highlights how well we can do if we try to empathize with the other (judging) person as we ask them to empathize with others (me, us, my friends). It’s tough though. Got to take care of ourselves emotionally, too.

  2. This is such a fascinating study, thank you for posting about this! I was dissapointed to hear about the LaCour controversy and I’m glad work like this is being done to prevent the underlying hypothesis from being discarded.

    As a practical tool to help effect change, though, I do have some worries, but maybe I am misinterpreting the data? The article here says that the canvassing reached 501 voters for a cost of $25,000, or $50 per voter. And it says 1 in 10 voters had a durable change of position. So that tallies out I think to $500 per vote-mind-changed. Am I doing that right? I’d love others thoughts on this. Even if you figure you can get the $500 number down quite a bit with practice, that is a very high starting point and LGBT groups are poor (especially post-Obergefell, sadly…..)

    It makes me happy to have data to confirm my own belief that most humans, no matter their political beliefs, are very able to feel empathy and tolerance for other humans if given a little help, but it saddens me that it is so hard to make that help/outreach happen.

    • I think this is a really good question, regarding financial efficiency. My guess is that expenses incurred in running a scientifically acceptable study of the practice and measuring results accurately are more than would be incurred by an organization to actually canvass. The initial development/study of the tool seems costly per voter. But now organizations can use that knowledge to run their own canvasses, and may be able to do so even on a tight budget, particularly if their canvassers are volunteers.

  3. That video was quite instructive… I mean, it was wonderful to watch a mind change. It’s hard to keep that non-judgmental engagement going, the kind that the canvasser is doing – but it’s a good reminder, for those of us who are really invested in changing minds, of how exactly to help that happen.

    Anyway, thank you, this is awesome! Would love more reporting on things like this.

  4. A bit of respect and empathy, and a willingness to listen without jumping all over the person whose opinion is so different from one’s own, goes a long way. I think that asking specifically why an interviewee holds an opinion and what worries them about the topic can bring out misconceptions so that the interviewee can look at the issue again and express an opinion IN THEIR OWN WORDS, perhaps seeing another aspect of the issue, and perhaps being open to gentle corrections of misconceptions. People are so swayed by mass media and social media and specific interpretations by their favorite authority figures (parents, preachers, etc)that they may not take time to work through why they hold certain opinions. This is an opportunity for them to stop and look at the issue for themselves and not regurgitate what they perceive as”common wisdom”. The interviewer’s willingness to acknowledge the complexity of both interviewee’s and interviewer’s emotions makes a huge difference.

    I have been on religious listservs where I have chatted respectfully with open-minded conservative individuals on a regular basis, and have seen their opinions change over time.

    This all goes back to Harvey Milk’s recipe for political and social change: “come out, whenever you can do so safely”.

Contribute to the conversation...

You must be logged in to post a comment.