I have previously expressed concerns about research in which the Humane Research Council (now Faunalytics) misinterpreted their data to make the case that we should be advocating for others to cut down on meat rather than promoting veganism. As I noted, their data actually showed the opposite of their conclusions and indicated that we should truly be promoting veganism rather than “reducetarianism.” Other flaws of the study were that there were no testable hypotheses based on theory and they misdefined veganism and measured it as diet only (rather than as an ethical stance against animal use). Also, they did not subject their findings to peer review, which is the standard practice in the scientific community.
So when I saw that a different group, the Humane League Labs, had put out a recent study that similarly concluded that we should be encouraging people to reduce animal consumption rather than completely eliminate it, I was naturally skeptical. I read the full report, though, to see if their data matched their conclusions. As I suspected, they did not.
The premise of this study is one that we have all seen before. They administered booklets that “discussed the cruelty of factory farming and the health benefit of removing animal products from one’s diet.” Next, the authors used eight different booklets that made different requests- some asking readers to “eat vegan”; some asking readers to “eat vegetarian”; some that encourage readers to “eat less meat”; and some that encourage readers to “cut out or cut back on” meat and other animal products.
As with the last study I critiqued, there are serious theoretical and methodological problems with this one. Once again, a major overarching problem with this research is that veganism is not properly represented. A vegan messaging approach would not only focus on “factory farming” but would discuss the ethics of using animals in all ways. It would also not focus on health. So again, if the authors are going to make any inferences about vegan messaging and its effectiveness, then truly vegan messaging should be employed.
I am not going to review the methodological problems in detail but the approach used would likely not be considered acceptable if it were to undergo scientific peer review. The major issues include a low response rate (less than half) at follow up and no accounting for missing data (i.e., missing data analysis); reliance on “change scores” that is a fairly weak data analytic approach; a lack of clarity on whether participants were randomized into conditions; and unequal group sizes.
What is far more concerning, however, is how the data were misinterpreted in a manner consistent with the worldview of this group. Those in the “control” condition reduced their meat and dairy consumption more than any other group. Moreover, the only statistically significant findings were those that demonstrated that those in the control condition reduced their consumption more than those who received different messages. In other words, the only “meaningful” finding from the main data analyses is that individuals reduce their meat and dairy consumption more when they are not asked to change anything than when they are asked to make some kind of change in their consumption. These counter-intuitive findings suggests that the methodological issues I previously reviewed may render this entire set of results to be questionable. In short, the findings don’t really make a lot of sense, and there is reason to be skeptical that we can take anything away from this at all.
The authors, on the other hand, have interpreted results that were not statistically significant to conclude that the message to “cut out or cut back on” meat and other animal products “might be the most effective approach” to get people to reduce animal product consumption. These conclusions are unwarranted given the actual findings, the lack of statistical significance of differences between groups (except for differences showing those receiving no message decreased consumption the most), and the methodological issues that call into question the validity of the data.
Pseudoscience is “a claim, belief, or practice that is presented as scientific, but does not adhere to a valid scientific methodology, lacks supporting evidence or plausibility, cannot be reliably tested, or otherwise lacks scientific status.” When a group frames a study and misinterprets flawed results to fit their preferred mode of advocacy, they are engaging in pseudoscience. Such practices appear to be all too common in the animal advocacy realm which is disappointing and potentially dangerous. The media and other groups report the conclusions from this research assuming it is valid. The organizations that conduct this kind of work can falsely claim that their form of advocacy is “evidence-based.” It is potentially harmful to animals to promote the notion of one form of advocacy as more effective than another based on flawed and seriously biased research. We can and must do better than this.
This article was originally published on the Vegan Publishers website at:
Casey is co-owner of Vegan Publishers, Professor of Psychiatry at Boston University School of Medicine, and staff psychologist at the National Center for PTSD in the VA Boston Healthcare System. He’s an internationally recognized researcher in the area of violence prevention, winning prestigious awards for his work from the International Society for Traumatic Stress Studies, the Institute on Violence, Abuse and Trauma, and the Centers for Disease Control and Prevention. He has published over 100 journal articles, book chapters, and scientific reports, and has a book forthcoming on trauma-informed violence prevention, published by the American Psychological Association.