Lesson 5 of 8 · 17 min
Asch (1955): conformity
Describe the aim, procedure, findings and conclusions of Asch's (1955) line-judgement research on conformity, how majority size and a partner changed it, and evaluate the study.
What you will be able to do
- Describe the aim, design, participants and procedure of Asch (1955).
- Describe the findings: the error rate on critical trials, individual differences, majority size and the effect of a partner.
- Explain the findings with normative and informational influence.
- Evaluate the study: control, validity, generalisability, ethics, reproducibility and its historical context.
Solomon Asch worked in the United States in the years after the Second World War, when the power of propaganda and mass movements, and in the early 1950s the pressure on Americans to show loyalty during the anti-communist investigations, made people ask how easily ordinary people go along with the crowd. Earlier research had shown people agreeing with a group about something ambiguous. Asch expected something different: that when the right answer was obvious, people would trust their own eyes. He set up a task in which the group was plainly wrong.
Key idea
Asch (1955) at a glance
Aim: to investigate whether, and how often, people would conform to a unanimous majority that gave an obviously wrong answer.
Participants: 123 male college students from three institutions (not Asch's own, Swarthmore College).
Design: a laboratory experiment. Participants judged lines either alone (control) or after a unanimous group of confederates gave a wrong answer (experimental), so the IV was the presence of the wrong majority and the DV was how often the participant gave the majority's wrong answer.
Procedure. A group of seven to nine young men sat in a classroom for what they were told was an experiment in visual judgement. Only one was a real participant; the rest were confederates who had been told beforehand what to say. On each trial the group saw a standard line and three comparison lines, and each person said aloud, in turn, which comparison line matched the standard. One comparison line was the same length; the other two differed from it by between three quarters of an inch and an inch and three quarters (about 2–4.5 cm). The real participant answered second last, after hearing all but one of the others.
There were 18 trials. On 12 of them, the critical trials, the confederates all gave the same wrong answer.
The people before you answer aloud. On most trials they agree on a line that is clearly wrong. Notice what it feels like to answer last.
Findings.
- Judging alone, people made mistakes less than 1% of the time.
- Under group pressure, participants went along with the majority's wrong answer on 36.8% of critical trials.
- People differed a great deal. About one quarter never agreed with the wrong majority. Some agreed almost every time. Altogether about three quarters agreed with the majority on at least one critical trial.
- Asch varied the size of the majority. With one opponent, errors were 3.6%; with two, 13.6%; with three, 31.8%. Larger majorities added little (see the graph).
- A partner who gave the right answer reduced errors to one fourth of the level under a unanimous majority. When the partner switched to the majority partway through, errors rose to about the level of a unanimous majority. Even a dissenter who gave a different wrong answer (more wrong than the majority) cut errors to 9%.
Errors rose steeply up to a majority of three, then levelled off (Asch 1955).
Using the graph, which statement best describes the effect of majority size on conformity?
Common mistake
Which number is "the" result?
36.8% is a percentage of critical trials (answers), not of people. "75% of people conformed" means three quarters went along at least once; most of their answers were still correct. Quote 36.8% for the overall rate and add that about a quarter never conformed.
Conclusion. Asch found the tendency to conform "so strong that reasonably intelligent and well-meaning young people are willing to call white black", which he called a matter of concern. He also stressed the other side: most answers were correct, a quarter of participants never yielded, and one ally was enough to break the majority's power. Unanimity mattered more than size.
Why did people conform? Interviews afterwards suggested mostly normative influence: many participants knew the majority was wrong but did not want to stand out or be ridiculed. Some came to doubt their own eyes and assumed the group must be right, which is informational influence.
In one variation a confederate partner gave the right answer on the first six trials and then joined the majority. What happened to the real participants' errors, and what does it show?
Worked example
Identify the variables and design in Asch's study
In Asch's (1955) main study, participants judged line lengths either alone or after a unanimous group of confederates gave a wrong answer. In one variation he changed the number of confederates from 1 to 15.
(a) Identify the IV and the DV of the main study. (b) Identify the experimental design, and explain why the alone condition was needed. (c) State an operational hypothesis for the majority-size variation.
Step 1: (a) IV and DV
IV: whether a unanimous majority gave a wrong answer (present or absent). DV: the percentage of critical trials on which the participant gave the majority's wrong answer.
Step 2: (b) Design
Different people were in each condition, so it is independent groups. The alone (control) condition showed that people almost never misjudged the lines on their own (under 1% errors), so the errors in the group condition (36.8%) can be put down to the majority, not to a hard task.
Without a control, high error rates could just mean the lines were hard to judge.
Step 3: (c) Hypothesis
"The percentage of critical trials on which participants agree with a wrong majority will differ according to the number of confederates in the majority (1 to 15)."
Try it yourself first, then check the working one step at a time.
Answer
(a) IV: presence of a unanimous wrong majority; DV: percentage of critical trials answered with the majority. (b) Independent groups; the alone condition (under 1% errors) shows the errors were caused by the majority. (c) The percentage of critical trials answered with the majority will differ with majority size (1–15).
Method
IV: what the researcher changed. DV: what was measured, operationalised.
Design: are the same or different people in each condition?
A hypothesis names the IV and DV and is testable.
Watch out for
Giving "conformity" as the DV without saying how it was measured.
| Aspect | Strength or limitation |
|---|---|
| Control | Strength: a standardised task with fixed confederate answers, and a control condition (judging alone, under 1% errors) shows the errors were caused by the majority. |
| Ecological validity | Limitation: judging lines among strangers is trivial and artificial; conformity may differ when the issue matters or the group are friends. |
| Sample | Limitation: male American college students in the 1950s, so the results may not generalise to women, other ages or other cultures. |
| Ethics | Limitation: participants were deceived about the aim and the confederates, could not give fully informed consent, and some were visibly stressed. Debriefing reduced, but did not remove, the harm. |
| Reproducibility | Mixed: conformity has been found in many countries, but at different levels. A meta-analysis of 133 Asch-type studies in 17 countries (Bond and Smith, 1996) found conformity was higher in collectivist cultures and had declined in the United States since the 1950s. |
| Historical context | The 1950s United States valued fitting in, so the study may partly reflect its time; a British replication around 1980 with engineering students found almost no conformity (Perrin and Spencer, 1980). |
Evaluate the external validity of Asch's (1955) findings. (3 marks)
Summary
Aim: whether people conform to a unanimous majority giving an obviously wrong answer.
Procedure: 123 male college students; groups of 7–9, all but one confederates; 18 line-judgement trials, 12 critical; answers given aloud.
Findings: under 1% errors alone; 36.8% of critical-trial answers agreed with the wrong majority; about a quarter never conformed.
Majority size: 3.6% with one, 13.6% with two, 31.8% with three; little change beyond. A partner cut errors to a quarter; when the partner deserted, errors returned.
Explanations: mainly normative influence; some informational influence.
Evaluation: well controlled, but artificial, a narrow 1950s sample, deception; conformity varies across cultures and times (Bond and Smith, 1996).
Finished the lesson?
Try the 3 quick checks above and the lesson completes itself, or mark it complete now.