What the data actually shows
The best-known evidence comes from Solomon Asch's conformity experiments in the 1950s. Participants were asked to judge which of several lines matched a reference line — an easy task with an obvious right answer — but were placed in a group of confederates who deliberately gave the wrong answer aloud first. A meaningful share of participants went along with the obviously incorrect group answer at least some of the time, even though, when tested alone, almost everyone got it right. The exact figures vary by version of the study, but the general pattern is robust: visible group pressure shifts what people are willing to say.
Morton Deutsch and Harold Gerard (1955) proposed a now-standard distinction between two kinds of influence. Informational influence is conforming because you treat others as evidence about reality — reasonable when the situation is ambiguous and others may know more. Normative influence is conforming to be liked or accepted and to avoid the discomfort of disagreement, even when you privately think the group is wrong. Asch's line task, where the right answer was obvious, points largely to normative pressure.
Later work complicates the simple story. Asch himself found that conformity dropped sharply when even one other person broke from the majority, suggesting that the pressure is about being a lone dissenter rather than about the group's sheer size. Cross-cultural reviews, including a well-known meta-analysis by Bond and Smith (1996), found that conformity in Asch-style tasks varied across societies and appeared somewhat lower in more recent decades — so the tendency is real but not fixed in strength.
Why this feels different from how it actually is
Conformity feels like something other people do, not us. Most of us picture ourselves as the independent thinker who would have given the right answer in Asch's room. This is itself a well-documented bias: people reliably overestimate how much they would resist group pressure and underestimate how strongly an unfamiliar social situation can steer them.
It also feels different because the pull is often invisible from the inside. When we conform for informational reasons, it doesn't feel like caving — it feels like updating toward what seems true. And when we conform for normative reasons, the discomfort of disagreeing is immediate and physical, while the cost of going along is vague and deferred, so going along can feel like the natural choice in the moment.
Finally, the word itself carries a verdict. "Conformity" sounds like a flaw, while the same behaviour described as "cooperation," "fitting in," or "learning from others" sounds healthy. Much of what gets labelled conformity is the unremarkable social glue that lets people drive on the same side of the road, follow shared norms, and coordinate — not a personal weakness.
Conformity feels less like a fixed trait and more like a situation people fall into.
What the research says to do about it
If the goal is to think more independently in a group, the most reliably supported lever is a single ally. Asch's own variations showed that the presence of just one other dissenter dramatically reduced conformity. Seeking out or becoming that first voice changes the situation far more than trying to be braver alone.
Separating the two kinds of influence helps in practice. When a situation is genuinely ambiguous and others may know more, leaning on the group is reasonable informational influence and often the right call. When you privately disagree but feel the pull to stay quiet, that is more likely normative pressure — and naming it to yourself is a documented way to take some of its force away.
At a group level, the research suggests building in structured space for dissent — for example, gathering views independently before discussion so people are not simply anchoring on whoever spoke first. The point is not to eliminate conformity, which keeps groups functional, but to keep it from quietly overriding what people actually know.
What the research says does not help
Assuming you are immune does not help, and the evidence suggests it may make you more susceptible, because you stop noticing the situational pull. The people in Asch's studies were not unusually weak; they were ordinary, and most of us would feel the same pressure in the same room.
Treating all conformity as a problem to be stamped out is also unhelpful. A great deal of going along with the group is sensible coordination or reasonable deference to people who know more. The research does not support a blanket goal of always resisting the group; it supports knowing which kind of influence you are responding to.
Trying to override the discomfort of disagreement through sheer willpower is unreliable on its own, because normative pressure is powerful and largely automatic. Changing the situation — finding an ally, gathering views privately first — tends to do more than simply resolving to be more contrarian.
The people in Asch's studies were not unusually weak; they were ordinary, and most of us would feel the same pressure in the same room.
What this looks like in real life
The obvious wrong answer
In Asch's experiments people judged which line matched a reference — easy alone, almost everyone got it right. But surrounded by confederates who confidently gave the wrong answer aloud first, a meaningful share went along with the group at least some of the time. The task wasn't ambiguous, so this points to normative pressure: the pull to not be the lone dissenter.
The one who speaks up first
You privately think a meeting's plan is wrong but stay quiet — until one other person voices a doubt, and suddenly saying so feels possible. That mirrors Asch's finding that a single dissenter dramatically lowers conformity. Becoming or seeking out that first voice changes the situation far more than trying to be braver alone.
Real numbers in context
Precise conformity rates are easy to overstate, so they are best held loosely. In Asch's line-judgement studies, a substantial minority of responses went along with the obviously wrong majority, and a majority of participants conformed on at least one trial — but most people did not conform on most trials, and a notable share never conformed at all. The headline is the existence and reliability of the effect, not a single percentage.
Bond and Smith's 1996 meta-analysis pooled well over a hundred Asch-style studies across many countries and found that conformity varied with culture and appeared to have declined somewhat over the decades the studies spanned. So the tendency to conform is broad and well-replicated, but its strength is shaped by context and era rather than fixed — a reminder to treat any specific figure with caution.