Alina Salmen tells us about ACE’s Better for Animals, an evolving guide to the state of current evidence for different animal advocacy interventions. Alina explains how the resource came about, demonstrates how it can be used, and invites researchers to get involved in helping keep it accurate and up-to-date.
Alina, could you briefly introduce yourself — your name, current affiliation, and role?
Of course! I’m Alina, and I am a Senior Researcher at Animal Charity Evaluators (ACE). ACE’s mission is to identify and promote the most effective ways to advocate for animals. We recommend and give grants to charities doing high-impact work. As a secondary but important goal, we also want the animal movement to become more evidence-based and improve how we measure our individual and collective impact.
My background is in Social Psychology — I completed a PhD at the University of Kent, essentially testing hypotheses derived from ecofeminist theory using psychological methods. I do not run my own studies anymore, but this background still helps a lot in my day-to-day work, for example, when critically evaluating research, conducting literature reviews, or analysing impact data.
Last year, ACE launched “Better for Animals,” a resource providing evidence-based insights for effective animal advocacy. What gap in animal advocacy was it built to address?
Initially, the project was meant to serve ACE internally. We receive applications from groups doing vastly different types of work for animals, and to assess them we have to review research across a wide range of disciplines and topics. We used to do this fairly ad hoc, but we wanted a more systematic approach. Better for Animals is an evolving resource that tracks the state of evidence for different animal advocacy tactics, and helps us in our grant and charity evaluations.
At the same time, from conversations with advocates, researchers, and funders, it also became clear that such a resource was a gap for the wider movement. Research is often scattered across different platforms, or is too technical for advocates to understand. Results can be mixed or internally contradictory, and any single study tells you very little about what advocates should actually be doing. So we decided to turn this into something that is useful for us and the rest of the movement.
Could you walk us through the resource — how it came together, who built it, what does it cover, and how it is maintained and updated?
Better for Animals is essentially a literature review of the interventions that animal charities we work with most commonly use. My colleague Max Taylor and I consulted with a range of advocates, researchers, and funders to refine the project scope and methodology. We do not have the capacity for full systematic reviews and we can rarely include all research that is out there on a given topic. Nevertheless, we followed a research protocol to ensure a high level of quality and consistency and to help avoid cherry-picking the evidence as much as possible.
Our team started drafting the first batch of evidence reviews; we ran these past external experts again to refine the research protocol. We now have evidence reviews for 28 intervention categories, from books and podcasts to legal advocacy, policy outreach, and effective advocacy research itself.
For high-priority interventions, such as corporate outreach or vegan pledges, we also reached out to external reviewers. The best-case scenario is to have a researcher review the content for errors, misinterpretations, and omissions, and an advocate give input on implementation. Ideally, we would speak to experts and implementers much more. But we are a small team, and we spend most of our time evaluating charities and grant-making, so there are a lot of improvements we would still like to make given more time. That is why, when we launched Better for Animals publicly in September 2025, we published it as a living document that we improve and keep up to date by regularly incorporating new evidence.
How does the resource help readers make sense of the evidence behind different interventions?
Better for Animals has a section for each intervention type (e.g., classroom education, investigations, protests) – have a look at our table of contents! Each section provides a useful summary of the evidence, followed by an in-depth assessment of the intervention. We provide an evaluation of the quantity and quality of evidence available (which determines how confident we are in our conclusions), any cost-effectiveness estimates we found, and, importantly, a section on the conditions that make the intervention stronger or weaker. When asking whether an intervention is generally effective, annoyingly, the answer is usually “it depends”. A well-implemented protest might have a measurable positive effect on public opinion, but a poorly implemented one could massively backfire.
To illustrate some of this, can you give us an example of an intervention covered in Better for Animals and what you found?
Sure! Vegan pledges are a good example of how we synthesise evidence of varying strength to try and come to a balanced conclusion. These pledges ask participants to eliminate or reduce their animal product consumption, usually for a month. Veganuary is probably the most popular pledge campaign, and their own six-month surveys report over 90% success rates for sustained reduction in animal product consumption. However, these figures come from retrospective self-reports of Veganuary participants with a very low response rate. So the results are limited by issues with self-reporting dietary behaviour (people frequently misremember what they actually ate), on top of the potential for selective attrition (it is possible that only the most committed Veganuary participants responded to the follow-up survey).
Looking at participant purchasing data solves the self-report problem, and UK data does show real January reductions in animal products compared to previous months and a smaller reduction persisting for at least 6 months, but these patterns do not correspond to a dip in aggregate UK meat or fish sales.
Experimental data is still quite rare. In your own 2022 randomised pledge study, which is probably the most rigorous experiment to date, consumption dropped during the 28 days, but was back to baseline a month later. The problem with interpreting these results is that real-world participants self-select to participate in Veganuary or other campaigns, which means they probably differ from randomly assigned participants in important ways, such as starting motivation. As your 2022 study found, participants with a stronger starting motivation tended to reduce more during the pledge period. Those who already reduce their animal product consumption also tend to do better. So experimental research might not translate perfectly to real-world contexts.
Another big problem is that most studies do not measure or report reduction or replacement by type of animal product. Chickens and fishes experience much more suffering relative to volume than other animals (such as cows or pigs) – largely because more of them need to be killed to produce the same amount of meat. If participants primarily reduce their consumption of beef or pork, and maintain or even increase their intake of chicken or fish (which is plausible for those with primarily health or climate motivations), then studies reporting only absolute reductions could misrepresent the actual impact on animal suffering.
Overall, the evidence for sustained impact seems too thin for me to be excited about most pledge campaigns that largely aim for new, sustained dietary change in the general population. But I still would not write them off. Pledge campaigns have other goals too – for example, Veganuary is a hugely popular brand, which gives them some leverage for partnerships with companies keen to reach the Veganuary audience. So there are still a lot of open questions about pledges. That is our invitation to researchers reading this: Each review in Better for Animals ends by naming the specific gaps in research that could change our mind about an intervention. Most of the research on pledges comes from the US and the UK, so many of the gaps here are geographical. In terms of study design, an experimental pledge study with a sample that better reflects real-world participants, with species-level consumption data, and with long-term follow-ups is something that could move our verdict — and that is exactly the kind of feedback loop the resource is designed to create.
That is our invitation to researchers: Each review in Better for Animals ends by naming the specific gaps in research.
Across the intervention categories, where is the evidence thinnest? Where is research most needed?
Research is needed in many areas. But a few key areas are worth highlighting.
For one, our review of legal advocacy and litigation, which includes tactics such as suing companies for animal welfare violations, relies mostly on opinion pieces, historical case studies (e.g., litigation against the tobacco industry), and reports that are often written by current or former litigators, who might understandably present a more favourable view of the potential of legal work. Arguably, this kind of work has inherently slow feedback loops and large-scale wins are rare, but where they do occur, they can be transformative. Moreover, impact is often not dependent on winning. Just filing a lawsuit can open doors by securing meetings with officials, facilitating access to industry actors, or by generating media coverage and raising awareness of industry practices (as for example, the extensive broadcast coverage generated by Humane League UK’s judicial challenge of fast-growing chicken breeds). More studies and balanced, systematic reviews of legal advocacy are needed to better understand how it can be effectively leveraged to establish legal precedents and shift public discourse.
Perhaps easier to solve and of more interest to psychologists is that we have essentially no experimental studies on the effects of podcasts, whether that is on diet change, attitudes towards animals, or engagement in the movement. There is a huge mismatch between evidence and potential — the global podcast industry has grown enormously in recent years, with almost 5 million podcasts reaching almost 600 million listeners as of January 2026. In 2025, a single episode of the popular Dwarkesh podcast raised over $2 million for FarmKind.
Finally, there seems to be an overall imbalance in the outcomes researchers tend to look at. An enormous share of the available evidence is about individual diet change, while the question of how to get people to become and stay actively engaged in animal advocacy — and so grow our collective capacity — is much more neglected. In short, we need much more research on how to build strong movements. The agricultural counter-lobby outspends animal advocates by a wide margin, so questions about how to build the movement’s capacity and power deserve far more attention given what we are up against.
There seems to be an overall imbalance in the outcomes researchers tend to look at (…) we need more research on how to build strong movements.
You have pointed to some big gaps there. Of all the open questions Better for Animals has surfaced, which feel most important to you?
When it comes to selecting the highest-priority research questions, the key is to look not just at the biggest or most interesting gaps, but at what closing them could realistically achieve — weighing scale of impact and decision relevance. The questions that rise to the top are those where a lot of animals are at stake, where a clear answer would genuinely shift advocacy or funding decisions, and where acting on the current weak evidence might be doing real harm. There is a substantial number we have identified. I will briefly mention two that are especially relevant to social scientists, but if you are interested, browse Better for Animals and check out the open questions across all interventions!
First, plant-based defaults are among our better-supported tactics, and we can see they shift behaviour in the short term, but their bigger payoff might be longer-term: By making the plant-based option the normal, expected one, defaults could gradually reshape social norms around what people eat, with effects that might both last over time and spread beyond the people directly exposed. If that is real, it would make defaults considerably more cost effective than the short-term numbers suggest — but it is essentially unmeasured.
Second, there is a cluster of questions around the animals who should matter most by sheer number but receive the least attention: fishes, invertebrates like shrimps and insects, and wild animals generally. Wild animals make up around 99.9% of all animals alive at a given moment, but people care mostly about the comparatively tiny number of charismatic large mammals. There are lots of unanswered social science questions here, such as how we can shift how people weigh the welfare of individual animals against species- and ecosystem-level thinking. I am also curious about how people might discount suffering that occurs “naturally” or “in the wild”, and whether that tendency can be counteracted. Finally, the psychology of scope insensitivity suggests that the staggering numbers of fishes, invertebrates, and wild animals may be exactly what numbs people to their suffering. So, can we find ways to make suffering at that scale feel real rather than abstract, or should we avoid the scale-based framing altogether?
What researchers can do – how to get involved
Say a researcher reading this is convinced and wants to get involved. How can academic researchers actively contribute?
Plenty of ways! The simplest is feedback. Better for Animals is a living document — we update it several times a year — and it is explicitly not a systematic review, so we know there’s evidence that we have missed or misjudged. You can email me (Alina) directly or request comment access to the document itself and tell us what we have got wrong, what we have overlooked, or where our interpretation of a study does not match yours.
If you have a bit more time, sign up as a reviewer! If you have subject expertise in any of the intervention areas, email me with the topics you know well.
The biggest contribution probably comes from taking on the open research questions themselves — and here I will point you to some of the conclusions of my review on effective advocacy research itself. It is plausible that the single biggest determinant of a research project’s impact is not statistical power or sample size; it is the choice of question. A beautifully executed study on a question that no advocate can actually act on might earn a prestigious publication, but remain strategically irrelevant for the movement.
It is also worth making sure research reaches the people it is meant for, so that it stands a real chance of being used. If you can, consult or partner with advocates and other implementers early, ideally co-creating the question and design, so the findings actually match their context and the decisions they need to make. And communicate results proactively, in accessible and actionable terms, rather than leaving them buried in a paper — advocacy organisations rarely have the capacity to translate technical findings themselves. When publishing research with animal advocacy relevance, reach out to Faunalytics or PHAIR about having your findings covered.
Finally, you can donate. Building a more evidence-based movement is a core part of ACE’s theory of change, and our funds make this possible by supporting both our research and the effective organisations we recommend and fund.
Resources
For more about Better for Animals, we recommend reading Alina’s blog article for ACE.
Quick link to the Better for Animals resource: go here
Contact Alina Salmen by email
Interview questions and blog by Jared Piazza for PHAIR
Cover photo credit: Sarah Halliday
PHAIR Society 