A research program of Project Animal Freedom

The Animal Freedom Lab

Project Animal Freedom’s research arm, launching January 2027. It exists so that a region-sized goal rests on evidence instead of impressions, and so that what we learn belongs to the whole movement rather than to us.

Why an organization our size has a research arm

We are a small nonprofit with a goal that is not small: a fully vegan Midwest by 2056, across twelve states. Our Theory of Change explains how we think that happens, and it ends with an admission most strategy documents avoid. It states which parts of the model outside research supports, which parts we have directly observed, and which parts are still wagers we have not proven.

That admission is only worth something if somebody is working to close the gap. The Animal Freedom Lab is that somebody.

The Lab is Project Animal Freedom’s research arm, and it has a narrow remit. Measure the Midwest movement. Hold the standards that make our own claims checkable. Turn research that already exists into something an organizer can use on a Tuesday night. It launches in January 2027, and it starts small on purpose.

What the Lab does

Measure

The Lab runs MARI, the Midwest Animal Rights Index, which is the Measure pillar of MARA, the Midwest Animal Rights Alliance. MARI counts the animal rights and vegan advocacy organizations operating across the twelve Midwestern states, maps them through MARM, the Midwest Animal Rights Map, and analyzes where coverage is missing. Fieldwork opens in March 2027.

The first edition, MARI 2027, will deliver a published methodology, the map, a coverage and gap analysis, and a public-use dataset. It will deliver no composite score, whatever the response rate turns out to be, because the right to score has to be earned. The first full State of the Midwest Animal Rights Movement report follows in the next phase, launching at MARS II in 2028.

Curate

The movement does not lack research so much as it lacks translation. Faunalytics maintains a research library of thousands of studies. Stray Dog Institute runs a global State of the Movement survey tracking how the farmed animal movement allocates resources by region, animal, intervention, and outcome. Those are good, and we use them rather than duplicate them.

The Lab’s curation work is to find what already exists, summarize it plainly, and put it where organizers will actually meet it: inside toolkits, training sessions, chapter mentorship, and the shared MARA library, with citations attached so anyone can check the original. An organizer running a first march does not need a literature review. They need the two findings that change what they do on Saturday.

Set the standards

Most of what damages movement research is not fraud. It is loose language. So the Lab keeps definitions, and applies them to Project Animal Freedom first.

Views and impressions are gross exposures. Platform reach is a platform’s own unique-account count. Estimated local reach is a cross-channel estimate that can contain duplication. People reached is reserved for deduplicated counts we can document. Four different things, four different words.
Outputs are not outcomes. A march happening is an output. A law changing is an outcome. We will not blur them.
Association is not attribution. Where people choose their own level of involvement, we report relationships as associations unless a research design gives us a credible basis for saying more.
Absence of evidence is not evidence of absence. Where no qualifying organization has been identified in a metro, we say “no qualifying organization identified as of [date]” rather than asserting that none exists.

That last one is the whole epistemic standard of this project in one sentence, and it is the one the Lab will spend the most time defending. It is also the one that costs us rhetorically, because the careful version of a claim is always the weaker-sounding version. Where our own live pages state something more strongly than the evidence supports, the standard applies to us before it applies to anyone else.

Test our own model

Our Theory of Change names four hypotheses: that community improves retention, that community creates activists, that chapter infrastructure increases durability, and that playbooks developed in St. Louis transfer to other cities. Each one is stated as a hypothesis rather than a finding, and each one is observable.

That work starts before the Lab does. Beginning with STL VegMonth 2026, challenge participants tell us how they eat when they sign up and again one month and four months later. When the Lab opens in January 2027 it inherits that instrument, the follow-up, and the honesty of the reporting, which includes printing the response rate next to the result rather than underneath it.

What the Lab is not, and not yet

We hold ourselves to a rule across everything we build: no initiative may be described publicly at a stage above the one it has evidence for. Applied here, that rule produces a short and slightly unflattering list.

It is not running a program of original studies in 2027. The launch scope is curation, synthesis, standards, and measurement infrastructure, which includes MARI’s baseline census and its public-use dataset. That census does collect first-party data, so to be precise about a phrase we use internally: original research here means the later hypothesis-driven studies of movement outcomes, not the baseline measurement MARI is built on. The first of those studies, on what sustains new vegans, is planned for 2029 with contracted research leadership or an academic partner, and a permanent Research Director follows in 2030 once the methodology exists. If the funding ramp comes in slower, that study moves to 2030 and the Lab stays a curator through 2029. We would rather write that down now than explain it later.
It is not an academic institute. Nothing here goes through journal peer review. What it goes through is preregistration, published methodology, independent methodological review, and a public-use dataset, which is a different kind of accountability and a real one.
It is not independent of Project Animal Freedom. The Lab is a PAF program. Its integrity comes from structure rather than from distance, and the structure is described below.
It is not a neutral observer of the region. PAF is one of the organizations MARI counts, and we are building an alliance those same organizations are invited to join. That is a genuine conflict of interest. Saying we intend to be fair about it is not a safeguard.
It is not a communications department with footnotes. A finding that contradicts our model is the single most valuable thing the Lab can produce, because it is the only thing that can stop us spending a decade being confidently wrong.

How the Lab fits with MARA, MARI, MARM, and MARS

MARA organizes the movement. MARI measures it. MARM maps it. MARS brings it together. The Lab is who does the measuring, and the set of standards that measurement is held to.

Component	What it does
The Lab	Project Animal Freedom’s research arm. It runs MARI, curates existing research for the movement, and holds research and data governance.
MARA	The Midwest Animal Rights Alliance. The free, non-exclusive membership and coordination body that organizations join.
MARI	The research program, run through the Lab. MARA’s Measure pillar.
MARM	The Midwest Animal Rights Map, a MARI product.
MARS	The Midwest Animal Rights Summit, MARA’s planned annual gathering, where MARI findings are first presented.
MARGO	The Midwest Animal Rights Groups and Organizations: the annual Top Ten, voted by the alliance, and the only organization-level ranking MARI ever publishes.

One line to keep straight: the Lab is the who, MARI is the what. The alliance is consulted on the research and cannot negotiate it, and that separation is deliberate.

How the Lab fits into our theory of change

Our Theory of Change rests on one wager: thriving local vegan communities are the foundation every other form of animal advocacy stands on. From community, all else follows. It runs as five stages, and each stage makes a claim that can be checked.

Reach. The claim is that we put veganism in front of enough of the reachable middle to matter. That claim is only meaningful if the numbers behind it are defined, which is why the Lab keeps the definitions rather than letting each campaign choose its own.

Try. The claim is that seasonal challenges make the first step easy. The check is the follow-up survey: how people ate at sign-up, and how they eat one month and four months later, reported with the response rate attached.

Belong. This is the leaky bucket, and it is the most important claim we make. The check is whether challenge participants who engage more deeply with community are also more likely to stay vegan or keep reducing. Participants choose their own involvement, so this is reported as an association until a design supports more.

Act. The claim is that community produces activists. The check is what share of event and community participants later volunteer, organize, or lead. A share, not an anecdote about one organizer.

Multiply. Two claims sit here. First, that chapter infrastructure increases durability; the check is organizer retention, recurring-event frequency, and supporter growth. Second, that St. Louis’ playbooks transfer; the check is comparing outcomes in chapters that adopt our packaged programs against those that improvise locally, which also tells us which parts of a playbook were the load-bearing ones.

Read together, that is the Lab’s place in the theory of change: it is the instrument panel for a machine we intend to run for thirty years. Our model is a loop rather than a funnel, and a loop compounds. A compounding model that is right gets stronger every cycle. A compounding model that is wrong also gets stronger every cycle, and looks exactly the same from the inside. Measurement is how you find out which one you are running while there is still time to change it.

There is a funding argument here too. We ask funders to buy durable capacity rather than a year of programming. Durable capacity is a claim about the future, and the only honest way to make that claim is to publish, in advance, how it will be checked and what would count as it failing.

And there is the part we consider foundational. Project Animal Freedom has always held that recruiting, training, and above all sustaining activists is work that benefits the entire movement rather than the organization doing it. Research is the version of that commitment which travels furthest. A published methodology can be picked up by a group in a state we will never operate in, run by people who will never hear our name. That is not a loss. That is the point.

An organization that will not test its own thesis is not running a strategy. It is defending a preference.

Why this matters especially here

The Midwest is one of the industrial heartlands of American animal agriculture. Midwest states hold the country’s largest concentrations of hogs, egg-laying hens, and turkeys, and several of them have been central to the history of ag-gag laws restricting undercover investigation of agricultural facilities.

The movement working against all of that operates across twelve states without a shared regional map. We have not found a current, public, region-wide source that says which animal rights and vegan advocacy organizations operate here, where they are, what they work on, or where none has yet been identified. Advocates here have excellent national and global research available to them. What they do not have is a regional baseline.

That absence has consequences, and they run in one direction.

A funder who wanted to invest in Midwest animal rights advocacy tomorrow could not easily find out what they would be investing in, so the region competes for money with places that can describe themselves.
National organizations decide where to expand, and without a baseline those decisions default to impressions, personal networks, and whoever was visible most recently.
An organizer in Wichita or Fort Wayne cannot find out whether anyone nearby is doing this work, which is how people conclude they are alone when they may not be.
And the region gets described rather than measured. Description flows toward whoever is loudest. Measurement is the only thing that gives a quiet place a way to be counted.

This is exactly where being careful matters most. “Nobody is doing anything there” is a sentence that removes a place from a funding conversation, and it is a sentence nobody currently has the evidence to say. So the map will say “no qualifying organization identified as of [date],” the inclusion criteria will be published alongside it, and any organization can add itself, correct its record, or appeal a decision. A gap is a place to invest, not a place that lost.

The other reason this matters regionally is that the method is portable. The methodology publishes in full, including definitions, eligibility criteria, sampling, limitations, and version history. A Great Plains version, a Southern version, or an Appalachian version would be a success of this project rather than competition with it. The Midwest needs the measurement because industrial animal agriculture is enormous here while the advocacy infrastructure set against it is thin and, so far, unmapped. A region carrying this much of the industry should not be among the last to get a usable baseline.

Governance and integrity

Because PAF is inside the movement the Lab measures, the protections have to be structural rather than aspirational.

Research and data governance for MARI sit with Project Animal Freedom and the Lab rather than with MARA, so that the organizations being measured are consulted on the research without being able to negotiate it. Members can advise, correct errors, and tell us who we missed. They cannot vote out a variable because it makes the movement look weak.
The methodology is preregistered on a timestamped public repository before any data is collected. Inclusion rules, analyses, thresholds, and how we handle missing data all exist before the results do.
An independent methodological review group of outside researchers reviews the design and audits how difficult inclusion cases are handled. At least one reviewer has no stake in the alliance and need not be an animal advocate. Reviewers are named in the published methodology and MARI outputs. The group does not administer the eligibility test and does not replace the published appeals process.
Criteria are published before the assessment period they govern, never written after the data arrives, and they are applied identically to every organization including Project Animal Freedom. PAF takes no part in voting on its own placement in anything MARI publishes.
One functional test governs everything: an organization is included if advocating an end to animal use or exploitation, or promoting veganism toward that end, is a substantive organizational function. It is the same test that decides MARA membership, applied identically to every organizational form. Inclusion turns on activity rather than self-description.
Assessment and exposure are different things. MARI will eventually maintain an internal assessment of organizations in the region. What publishes from it is narrow: the annual Top Ten MARGOs, voted by the alliance against published criteria, which an organization can decline to be named in. Nothing below it is published, an organization’s own assessment goes to that organization rather than to the public, case studies naming an organization require consent, and factual errors can be corrected and decisions appealed.
What organizations tell us in confidence stays confidential, under a published consent and data-use policy. The survey does not field before those rules are public.
The public-use dataset carries public directory fields plus research variables that are publishable, de-identified, or aggregated in line with that policy. It is not confidential raw responses, and no organization’s individual answers are posted.
Organizations screened out by the functional test are recorded internally so the Lab can estimate coverage and audit its own decisions. That record is never published as a named list; screened-out organizations appear only in aggregate counts.
Response rate and coverage are reported as headline findings rather than buried caveats. If our coverage is thin, that will be the first thing you read.

Who actually runs it

Today the Lab is small: a volunteer operation inside a volunteer-led organization, with outside expertise brought in where the work requires it. The independent review group is seated at launch. Contracted research leadership arrives in 2029 and a permanent Research Director in 2030, as funding allows.

We would rather tell you the Lab is small and bound by published rules than imply a staff we do not have. The rules are the part that does not depend on our size.

What funding the Lab buys

MARI’s single largest determinant of quality is response rate, and response rate is bought with fieldwork hours. Somebody has to find every organization across twelve states, send the invitation, send it again, and then pick up the phone.

We can run the first edition on volunteer labor. It will work, and it will produce lower coverage than the same project with a paid fieldwork coordinator for three months. Funding the Lab does not make the report prettier. It makes more of the Midwest visible in it, which is the only thing that makes the findings usable.

Read about MARI Get the Midwest update

Frequently asked questions

Is the Animal Freedom Lab a separate organization?

No. It is a program of Project Animal Freedom, in the same way MARA is. It has no separate legal existence, no separate board, and no separate finances.

What is the difference between the Lab and MARI?

The Lab is the arm. MARI is the largest program that arm runs. The Lab is the who; MARI is the what. The Lab also does work that is not MARI, including curating existing research and holding the measurement standards Project Animal Freedom applies to its own claims.

Is the research peer-reviewed?

Not in the journal sense, and we will not describe it that way. It is preregistered before fieldwork, the methodology and survey instrument publish in full, an independent methodological review group reviews the design, and the release includes a public-use dataset so the work can be checked and reused.

Does the Lab do original research?

Not as a program of studies in 2027. MARI’s baseline census does collect first-party data, so the honest answer has two halves: the Lab’s launch scope is curation, synthesis, standards, and the measurement infrastructure MARI needs, rather than a portfolio of original studies. The first hypothesis-driven study, on what sustains new vegans, is planned for 2029 with contracted research leadership or an academic partner. Under a slower funding ramp it moves to 2030.

Would you publish something that makes Project Animal Freedom look bad?

That is a large part of what the Lab is for. Our Theory of Change already states which parts of our model are unproven. Coverage and response rate lead the reporting, we distinguish what we observed from what we can attribute, and Project Animal Freedom is assessed on the same criteria as every other organization.

Will the Lab rank my organization?

Internally, in time, yes. Publicly, only the annual Top Ten MARGOs, voted by the alliance, which you can decline to be named in. Your own assessment goes to you rather than to the public, you can correct factual errors in it, and you can appeal. None of it exists yet; the scored index is the last thing MARI builds.

Can we use your methodology in another region?

Yes, free, and we would be glad if you did. It publishes in full, including definitions, eligibility criteria, sampling, limitations, and version history. Tell us how it went so the next version is better.

Why call it a lab?

Because a lab is where you test the things you might be wrong about. We picked the word deliberately, and it commits us to something.

How is the Lab funded?

Project Animal Freedom underwrites it, and we are seeking grant and donor support for the research program, currently expected to be the most resource-intensive component of the regional work.

Take part

If you run an animal rights or vegan advocacy organization in the Midwest, the most useful thing you can do is answer the census when it opens in March 2027, and tell us now about the groups near you we are likely to miss. The organizations hardest to find are usually the ones the map most needs.

If you are a researcher, a methodologist, or a graduate student looking for a regional dataset for which we have not found a public equivalent, we would like to hear from you before the methodology freezes rather than after it publishes.

And if you fund this movement, the Lab is the part of our work that makes the rest of your Midwest giving legible. We would welcome that conversation before the first edition, not after it.

Sources

Faunalytics research library. faunalytics.org
Stray Dog Institute, State of the Movement. straydoginstitute.org
USDA NASS, Census of Agriculture and livestock and poultry production summaries, for hog, egg-layer, and turkey concentration by state.
Ceryes, C. A. & Heaney, C. D. (2019), “Ag-Gag” Laws: Evolution, Resurgence, and Public Health Implications, New Solutions: A Journal of Environmental and Occupational Health Policy. The Animal Legal Defense Fund ag-gag tracker works as a second, advocacy-side reference.

A research program of Project Animal Freedom. Last updated September 2026.
Spot something wrong on this page? Email info@projectanimalfreedom.org and we will correct it.