Plenty of people share a grievance. Almost none of them end up on a street corner holding a sign about it. That gap — between how many people are angry and how few of them show up — is what Most People Who Agree With You Won’t Protest: Predictors of Marches and Demonstrations from Six Decades of Research set out to measure, and the answer isn’t about how angry anyone was.

Shelley Boulianne (Mount Royal University and Lund University) and Jennifer Earl (University of Delaware) pulled together 181 survey-based studies and more than 1,800 statistical estimates, spanning research from 1965 to 2023, and asked a blunt question: across six decades of protest research, what actually predicts whether someone marches? PsyPost covered the findings on September 20, 2026, and the headline result deflates a lot of assumptions about how movements grow. Grievance barely moved the needle. Organizational membership and social ties did almost all the work. Émile Durkheim said as much in 1912, without a single regression table to back him up — he just watched what assembly does to people and wrote it down.

The Quick Verdict

Question	What we found
What was studied?	What predicts individual participation in marches and demonstrations, across six decades of survey research
Who ran it?	Shelley Boulianne (Mount Royal University, Lund University) and Jennifer Earl (University of Delaware)
Where and when?	Science Advances, covered by PsyPost, Sep 20, 2026
Sample	181 survey-based studies, 1,800+ estimates, 1965–2023
Weak/inconsistent predictors	Grievances, relative deprivation, political trust, political efficacy
Strong/consistent predictors	Political interest, ideology, organizational membership, education, age, urban residence, digital/social media use
What Durkheim already argued	The Elementary Forms of Religious Life (1912): shared assembly generates the emotional charge — “collective effervescence” — that individual belief or sentiment never can on its own

The Quick Version

A 2026 meta-analysis of 181 studies and six decades of protest research found that anger, grievance, and even belief in a cause are weak, unreliable predictors of who actually shows up to a march. What predicts turnout is organizational membership, a personal invitation, and how socially connected you are. Durkheim called this in 1912: collective action isn’t powered by what’s in any one person’s head. It’s powered by the electricity that only exists when people physically assemble.

What the Study Actually Found

Boulianne and Earl weren’t running a new survey. They were doing something more useful for a question this old: aggregating everything the field already knew. Six decades of it. Protest research has accumulated a lot of contradictory single studies over the years — one paper finds grievance predicts turnout, the next finds it doesn’t — and a meta-analysis is the tool built to settle exactly that kind of argument. Pool 181 studies, extract 1,800-plus estimates, and see what actually holds up once the noise cancels out.

What didn’t hold up: grievance. Relative deprivation. Political trust. Political efficacy — the belief that your participation could actually change anything. All four showed up as weak and inconsistent across the pooled data, which is a rough result for the intuitive theory of protest that most of us are carrying around without noticing. That theory goes: people get angry enough, or convinced enough, and then they act. Jennifer Earl’s framing of the result, per PsyPost, cuts right through it: “If you compare people who did participate and people who did not, a lot of people who didn’t participate share the same concerns.” Sharing the concern was never the bottleneck.

What did hold up, consistently, across six decades: political interest, ideological orientation, organizational membership, education, age, urban residence, and digital or social media use. Shelley Boulianne’s summary is the one line worth remembering from the whole paper: “The consistent predictors are political interest and social connections.” Not conviction. Connection.

Why Organizations Beat Anger

1. They deliver a direct ask. Belonging to a labor union, a church, a community group, a campus organization means someone, at some point, is going to specifically invite you to the thing. Earl’s other line from the coverage is almost embarrassingly simple: “If you want people to participate, specifically invite them to — it’s simple but very powerful.”
2. They put you inside a network that’s already mobilized. You don’t have to independently discover the protest, decide it matters, and talk yourself into going alone. Somebody in your existing circle is already going, and you’re going with them.
3. They convert private opinion into public commitment. An opinion held quietly at your kitchen table costs nothing and changes nothing. An opinion voiced inside a group that’s about to act on it is a different kind of thing entirely — harder to back out of, and reinforced by everyone around you doing the same.

Line those three up and they stop looking like three separate findings. They’re one finding, restated: protest runs on infrastructure, not intensity.

What Did Durkheim Mean by Collective Effervescence?

Here’s the concept stripped down to its working parts.

In The Elementary Forms of Religious Life (1912), Émile Durkheim argued that when people physically assemble around a shared focus — a ritual, a ceremony, a cause — something happens that can’t be produced by any individual thinking or feeling alone. He called it collective effervescence: a heightened emotional charge generated by the assembly itself, which then binds the group together and fuels shared action. The energy isn’t imported into the room by each person’s private conviction. It’s generated by the room.

Durkheim wasn’t writing about protest marches. He was watching Australian Aboriginal religious ceremonies through ethnographic accounts, trying to explain why sacred feeling seems to spike specifically when people gather, and drops when they disperse. But the mechanism he described doesn’t care what the gathering is for. A funeral, a football match, a revival meeting, a demonstration — Durkheim’s claim is that assembly itself, not the private beliefs of the people in it, is doing the emotional work.

This wasn’t a one-off idea for him, either. Fifteen years earlier, in Suicide (1897), Durkheim made structurally the same argument in reverse. He wasn’t asking what makes people act together — he was asking why some people’s private despair turns fatal and others’ doesn’t, and his answer was, again, not individual psychology. It was social integration. People embedded in dense social ties had lower suicide rates regardless of their personal circumstances; people cut loose from those ties — a condition he called anomie — were at far higher risk, independent of how bad or good their individual lives actually were. Two books, two completely different behaviors, one recurring argument: what predicts what a person does isn’t what’s inside them. It’s how tightly they’re woven into everyone else.

Why the Study and the Philosophy Rhyme

The connection here isn’t “a dead French sociologist predicted a regression coefficient.” It’s narrower and, I think, sturdier than that: Durkheim’s claim was structural in both books — individual sentiment, whether it’s grief or grievance, is a bad predictor of individual behavior, because behavior is downstream of social integration, not belief. Boulianne and Earl’s finding that organizational membership beats political efficacy is that exact claim, tested against 1,800 estimates instead of a stack of ethnographic notes and mortality tables.

It’s also worth sitting next to what Simmel argued about social networks — that the shape and density of someone’s connections predicts outcomes theories about individual attitude never quite manage to. And it lines up with MacIntyre’s argument that happiness itself runs on community rather than individual pursuit: both cases, the unit that matters isn’t the person and what they privately believe or want. It’s the structure they’re embedded in.

There’s a gentler cousin of this idea, too, in Empedocles’ account of shared awe merging the self — the finding that experiencing something together, rather than separately, changes what the experience does to you. Durkheim’s version is the same mechanism turned toward politics instead of romance: a room full of people doesn’t just contain more feeling than an empty room. It generates a different kind of feeling, one that a person sitting alone with the same opinion never produces.

What This Doesn’t Mean

It doesn’t mean grievance is irrelevant, or that the people in the streets don’t actually care about what they’re protesting. Nobody’s claim here is that protesters are indifferent to their cause — the study measured what predicts turnout, not what predicts conviction. Plenty of people who never show up are just as convinced as the people who do. That’s the whole point: conviction alone was never going to be a reliable predictor, because almost everyone has it, and only a small fraction act on it.

It also doesn’t mean organizing is the only thing that matters, or that solo conviction never leads anywhere. Digital and social media use showed up as one of the consistent predictors too — the same technology this site has covered as eroding some forms of community turns out to be a mobilization tool for others, connecting people to networks that didn’t exist for them a generation ago, no union card required.

And it doesn’t mean movements should stop trying to persuade people or making the moral case for their cause. It means persuasion is necessary but not sufficient. Someone has to also do the unglamorous work of showing up at the door.

What Actually Follows From This

1. If you care about something, join something. Not because caring alone won’t count, but because caring alone rarely translates into action — for you or for anyone else who shares your view. The study’s finding applies to you too. Organizational membership predicts your own turnout better than the strength of your opinion does.
2. If you want people to act, ask them directly. Earl’s line bears repeating because it’s the most actionable thing in the whole paper: specific invitation works, and it’s simple. A shared post is not the same as a text that says “I’m going, come with me.”
3. Don’t mistake widespread agreement for a movement. A lot of people quietly sharing your concerns is not the same thing as a movement, and treating it that way — assuming the anger alone will eventually organize itself — is exactly the assumption this study spent six decades of data undoing.

Durkheim never got to see a march organized over a group chat, and he certainly never ran a meta-analysis. But he watched people gather, and he noticed that something happens in the gathering itself that doesn’t happen anywhere else — not in a private grievance, not in a strongly held belief, not in anger nursed alone at a kitchen table. Boulianne and Earl just handed him the six-decade dataset he never had access to. Eighteen-hundred estimates, and the number keeps landing in the same place his ethnography did in 1912: the room does what the sentiment inside any one person, by itself, never could.

---

This post discusses a 2026 meta-analysis and a 1912 work of sociology. It isn’t an endorsement or criticism of any specific protest movement or cause.