Psychology › Section L · Social psychology › Topic L.4

Conformity, compliance, and obedience

Would you give an obviously wrong answer because everyone else did, or keep pressing a shock switch because a scientist told you to? Some of psychology's most famous studies tested exactly that, and their findings are still being re-examined.

Conformity, compliance and obedience

Conformity is changing behaviour or beliefs to match a group, even though no one asks you to. Compliance is going along with a direct request, often without private agreement: you can agree in public while still disagreeing in private. Obedience is following an order from someone seen as an authority. This topic covers the classic studies of each, then how groups change effort and decisions.

Two reasons to conform

Morton Deutsch and Harold Gerard (1955) distinguished two motives. Informational social influence is conforming because we think the group knows better. It is strongest when a situation is ambiguous, and it often produces private acceptance, a real change of belief. Muzafer Sherif (1935) showed it with the autokinetic effect: a still point of light in a dark room seems to move. People judging the movement in groups drifted towards a shared estimate and kept it when later tested alone. Normative social influence is conforming to be liked or to avoid rejection. It often produces public compliance without private belief.

Asch's line judgements

In the 1950s Solomon Asch asked groups of students to say aloud which of three lines matched a standard line. The answers were obvious, and people judging alone almost never erred. But only one person in each group was a real participant; the rest were confederates, people working for the researcher, who gave the same wrong answer on most trials. Real participants went along with the wrong majority on about a third of those trials. About three quarters conformed at least once, and about a quarter never did. Because the task was unambiguous, this was mostly normative influence.

What makes conformity rise or fall

Conformity rose as the unanimous majority grew to about three people, then levelled off; larger groups added little. Unanimity mattered far more. When one confederate gave the right answer, conformity fell to a small fraction of its usual level. Writing answers privately also reduced it. Rod Bond and Peter Smith's 1996 meta-analysis found that conformity in the United States had declined since the 1950s and was generally higher in collectivist cultures. Influence also runs the other way: Serge Moscovici showed that a consistent minority can shift a majority's judgements, while an inconsistent one cannot.

Milgram's obedience experiment

In 1961 Stanley Milgram, at Yale, began studying how far people would obey an authority, prompted partly by the trial of the Nazi official Adolf Eichmann. Participants, told the study was about punishment and learning, took the role of teacher. They were told to give a learner, really an actor, an electric shock for each mistake, rising in 15-volt steps to 450 volts. No real shocks were given. When the teacher hesitated, the experimenter used scripted prods such as 'Please continue'. In the best-known early version, 26 of 40 men (65%) went all the way to 450 volts, many of them showing extreme stress as they did.

Obedience depends on the situation

Milgram ran more than 20 variations, and full obedience ranged from almost nobody to over 90%. It fell when the victim was closer: with the learner in the same room, and further still when the teacher had to press the learner's hand onto a shock plate. It fell when authority was weaker, for example when the experimenter gave orders by telephone. It fell sharply when two other 'teachers' refused to continue. It rose when the participant only passed on instructions while someone else pressed the switch. Small, gradual steps also made each shock feel little worse than the last.

Milgram today: replications and critiques

Jerry Burger (2009) ran a partial replication with men and women, stopping for ethical reasons at 150 volts, the point where the learner first demands to be released. About 70% were willing to continue past it, a little below Milgram's figure for the same point, and the difference was not statistically significant. Gina Perry's archival study, Behind the Shock Machine (2013), found that the experimenter often went beyond the script, that many participants doubted the shocks were real, and that many left without being told the truth. Later analyses found that the most order-like prod, 'You have no other choice', usually produced refusal, not obedience.

How groups change effort and decisions

Social facilitation is better performance on easy or well-practised tasks, but worse on hard or new ones, when others are present; Robert Zajonc (1965) argued that an audience raises arousal, which strengthens whatever response is dominant. Social loafing is reduced effort when contributions are pooled and unmeasured; making each share identifiable reduces it. Deindividuation, lost self-awareness in anonymous crowds, mainly makes people follow the local group's norms, good or bad. Group polarisation is the push of like-minded discussion towards a more extreme shared view. Groupthink (Irving Janis, 1972) is a cohesive group valuing harmony over realistic appraisal, as in the failed 1961 Bay of Pigs invasion; its signs include self-censorship and an illusion of unanimity.

Worked example: the silent meeting

At a planning meeting, six colleagues praise a launch date you think is unrealistic, and the manager has already said she loves it. 1. You stay silent partly to avoid looking negative: normative influence. 2. You also wonder whether they know something you do not: informational influence. 3. Saying 'sounds good' while still doubting it is compliance without private acceptance. 4. A cohesive team, a leader who states her view first and no outside voices are conditions for groupthink. 5. Predict: if one colleague voices doubts, you become much more likely to speak, as Asch found with a single dissenter. 6. Fix: the manager could appoint a devil's advocate and hold back her own view.

Worked example: predicting obedience

A hospital rule says nurses must not give an unusual dose on a telephone order. Use Milgram's variations to predict when a nurse might comply anyway. 1. Authority: a senior doctor who is physically present raises obedience; an unknown voice on the phone lowers it. 2. Victim distance: a patient the nurse can see and talk to makes the possible harm vivid, which lowers obedience. 3. Peers: a colleague who openly refuses makes refusal easier for others. 4. Gradual steps: a series of slightly larger requests is harder to resist than one large request. 5. Check: these are predictions from laboratory studies, not observations of real wards.

Common mistakes to avoid

First, assuming Asch's participants came to believe the wrong line was right: most complied in public while privately disagreeing. Second, saying conformity keeps rising with group size: it levels off after about three. Third, quoting 65% as 'Milgram's result': it is one condition among more than 20, which ranged from almost nobody to over 90%. Fourth, presenting Milgram as either fully confirmed or debunked: Burger's replication and Perry's archival doubts both matter. Fifth, confusing social facilitation (performing while others are present) with social loafing (pooled, unmeasured effort).

Key takeaways

Conformity matches a group, compliance answers a request and obedience follows an authority. Informational influence (Sherif) yields private acceptance; normative influence (Asch) often yields public compliance only. Asch's participants conformed on about a third of critical trials; one dissenter cut this sharply. Consistent minorities can shift majorities. In Milgram's best-known version, 26 of 40 men went to 450 volts; obedience rose with close, legitimate authority and fell with a nearby victim or defiant peers. Burger (2009) found similar results up to 150 volts; Perry (2013) documented departures from the script. Groups also bring social facilitation, social loafing, deindividuation, polarisation and groupthink.

Flashcards

Flashcard

Conformity

Changing behaviour or beliefs to match a group

No one has to ask.

Test your understanding

Question 1 of 5

In Asch's line studies, which change most sharply reduced conformity?

Question 2 of 5

In the best-known early version of Milgram's study, how many of the 40 men went all the way to 450 volts?

Question 3 of 5

What did Jerry Burger's 2009 partial replication of Milgram's study find?

Question 4 of 5

During a fire alarm in an unfamiliar building, you follow the crowd because you assume they know the way out. This is:

Question 5 of 5

A skilled pianist plays better in front of an audience, but a beginner plays worse. Which idea explains both?

← Previous topicNext topic →