Animals in industrial farms are treated appallingly. The conditions in which they’re held bother the public — that is, when they’re made aware of those conditions — so much that some US states, under pressure from the agriculture industry, have passed “ag-gag laws” making it illegal for whistleblowers and activists to release video of farm conditions.
At the same time, we’re finding out more about other negative consequences of eating meat. Animal agriculture is resource-intensive and a major contributor to climate change. In the close quarters on factory farms, animals get sick all the time, so it’s standard in the industry to give them antibiotics preemptively, which contributes to the growing threat of antibiotic resistance.
Farmed animal advocates have been working on these problems for a long time. One of the problems they’ve wrestled with is a pretty fundamental challenge: How do they persuade more people to become vegetarian? Over the past decade, they’ve done a lot of research to test which methods of advocacy work, and how and why people become vegetarians.
Studies have looked into the effects of different persuasion efforts: leafletting, online ads, undercover investigations of farms, street protests. What they have found, for the most part, is that none of the things advocacy groups were doing had detectable effects — partially because of constraints on the research, and partially because the underlying effects seem to be small and hard to measure (if there were effects at all).
The results, and what they mean, are still being hotly contested within the animal rights community. That disagreement underscores the steep challenge the animal rights community faces in trying to convince the rest of the world to give up meat. It also stands in sharp contrast to the repeated successes the advocacy movement has had in pursuing a different course: pressure campaigns against corporations.
Looking at the evidence to date, those campaigns aimed at companies — as opposed to persuasion efforts pitched at the individual level — may well be the more promising path to bringing change to our food system, even as advocates continue to research the most effective ways to change a person’s mind.
It’s exceptionally hard to persuade people to become vegetarian
One reason it’s so hard to persuade people to become vegetarian is that it’s hard to persuade people of things, period. Campaigning to change the minds of swing voters doesn’t seem to work. Exposing people to new facts that challenge their preconceptions doesn’t seem to work. So it’s perhaps unsurprising that it’s hard to persuade people to stop eating meat.
But the animal researchers I talked to were taken aback by how hard.
A popular tactic for spreading the word about conditions on factory farms is leafletting — handing out to passersby pamphlets that describe the conditions of farms, and urging people to look into it more. In 2015, more than 2 million of them were handed out by groups like Vegan Outreach, Mercy for Animals, and Compassionate Action for Animals.
In the past five years, there’s been a flurry of research directed at figuring out whether leaflets changed minds. A meta-analysis by Animal Charity Evaluators concluded that they pretty much don’t. This was a change of position for them; an earlier analysis, based on much more optimistic interpretations of the limited research, suggested that they did. But with more results in, there’s actually no evidence that leaflets persuade anyone to eat less meat. (There’s a little bit of evidence that suggests it might cause people to eat more meat.)
What about video ads? Initially, advocates saw promising results there. They estimated that watching videos — especially videos shot inside factory farms — reduced meat consumption a little, and that since ads were cheap and had wide reach, they might be worth it even though the gains were small.
But that was before advertisers everywhere learned that Facebook had been badly inflating metrics for video ads. Some of the case for these ads rests on the assumption that Facebook was reporting viewership truthfully — and there is now reason to believe those numbers were inflated 150 percent to 900 percent for two years.
And that wasn’t the only problem with the research: While it was easy to get numbers from Facebook about engagement with ads (though some of those numbers were likely wrong), it was always hard to get a reliable estimate of whether engaging with the ads actually reduced consumption, and the estimates on a lot of anecdata.
One organization, Mercy for Animals, asks people to pledge to go vegetarian. But it’s unclear whether people who pledge really go vegetarian, or how long they stay vegetarian, making it exceptionally difficult to estimate the cost-effectiveness of the video ads.
Research into complex areas with small effect sizes is hard for statistical reasons
In some respects, the picture is even more discouraging than that. The Humane League’s research division, Humane League Labs, has argued that there are several serious methodological hurdles to doing high-quality animal advocacy research.
First, effect sizes are small. It’s easy to design a study that will detect an effect that’s pretty large. It’s much trickier to design one that will detect a rare change. Most people won’t change their behavior at all in response to a leaflet. One person in 100, or maybe one person in 1,000, might. To accurately measure a small effect, you need a vastly larger sample size — which means a much more expensive study. If your study is underpowered, then you’re likely to turn up results that aren’t really there, and you can sometimes turn up results in the exact opposite of the direction of the real effect.
There are also concerns about what to measure. Lots of the research into animal advocacy ran aground on the extraordinary difficulty of getting truthful information about people’s diets out of them. “It’s incredibly hard to measure dietary behavior through self-report,” Jo Anderson, research director at Faunalytics, a major animal research organization, told me.
There’s social desirability bias — people give the answer they think researchers want to hear. There’s the fact that 60 percent of Americans who say they’re vegetarian on surveys also say that they’ve eaten meat in the past 24 hours. Researchers at other animal research organizations were also quick to call this one of their biggest woes.
Until they figure out how to measure actual dietary change, rather than self-reported dietary change, they can’t be too confident in any of their conclusions. The Humane League Labs have been looking at alternatives to self-reporting, like checking at grocery stores so they can measure how much meat gets purchased.
And ads and leaflets are in many respects much easier to research than other advocacy tactics. They’re distributed to a specific number of people. You can attempt to follow up with those people later. You can have a control group.
Studying the effects of protests is considerably more complicated. Advocates have largely looked at other protest movements, like the campaign for same-sex marriage or the Tea Party movement, and tried to figure out which lessons from those movements generalize.
So what does work? I asked Anderson at Faunalytics and she said. “There’s no real answer.” Her organization is exploring new directions for its research, trying more comprehensive interventions that hopefully have a larger effect size and are easier to study.
I asked Toni Adleberg at Animal Charity Evaluators the same question. She declined to answer too, telling me, “We need more research.”
These results are open to interpretation — and everyone has their own interpretation
Discouraging data, like the data these studies have produced, opens itself up to a lot of explanations. One I heard from advocates was that there are effects — we just aren’t conducting studies carefully enough to find them. If we fix the methodological issues with self-reporting and get big enough sample sizes, we’ll start turning up effects. From this perspective, the problem is that we’re not yet good enough at conducting research.
A second interpretation, which Anderson outlined for me, is that people go vegetarian approximately if lots of different conditions are satisfied all at once. They have to care about animals, learn about the positive effects of vegetarianism, get advice for recipes and strategies that’ll make it painless and healthy to go vegetarian, and then they need social support and encouragement to stay this way. Focusing on just one of those things probably won’t move the needle.
A third interpretation I heard was that the research was probably basically correct, and that animal advocates should accept its conclusions, painful as they are: Only a few people are open to reducing their animal product consumption. Most people just won’t, and you can’t persuade them. If that’s true, then advocates’ resources are probably better focused on fighting for political change and on developing plant-based alternatives to meat.
“One thing that we’ve been noticing lately,” Adleberg of Animal Charity Evaluators told me, “is that the animal advocacy movement directs the vast majority of their resources towards diet change and trying to convince individuals to go vegetarian, when there’s not that much evidence that it works.”
“All of the research collected by Faunalytics suggests that changing hearts and minds is not only a slow process, but it may also have an upper limit,” Che Green of Faunalytics wrote last year, in a call for “focusing more on influencing both corporate and government policies to create long-term, enforceable change for animals.”
Why targeting corporations may be the way to go
The picture for corporate campaigns is, indeed, more optimistic.
Organizations like the Humane League have had a very high success rate with securing concessions from a company they target with a welfare campaign. For example, they might demand that a restaurant chain source their eggs from farms that don’t cage the hens, and follow up with protests and public pressure if needed. The campaigns typically last a few months, and the executive director of the Humane League said a few years ago that it virtually always get results within that timeframe.
The Open Philanthropy Project called this string of victories a “major and unusual success story for rapid, large-scale change brought on by advocacy.” Often, of course, the concessions are about a policy change in the future, and follow-up is needed to make sure the change really happens.
Some of the concessions activists have won — for example, cage-free campaigns for chickens — are controversial. Activists disagree on how much they improve conditions for animals. But still, the string of wins here stands in sharp contrast to the string of murky results everywhere else.
The best way for an individual to help reduce factory farming remains to reduce their meat consumption. But if they’ve done that and they’re looking to do more, it may be surprisingly tough to convince others — so much so that their effort is better invested in large-scale reforms.
Animal advocates have mixed feelings about that. Ultimately, though, Anderson told me, “we want to be helpful to the maximum number of animals, whether that’s directly or indirectly” — and that requires a hard-headed and probing look at what works, what doesn’t, and what we don’t know yet.