Lesson 7 of 8 · 19 min
Haney, Banks and Zimbardo (1973): the Stanford Prison Experiment
Describe what Haney, Banks and Zimbardo (1973) concluded about status, roles and power from the Stanford Prison Experiment, and evaluate it with the critiques of Le Texier (2019).
What you will be able to do
- Describe the aim, participants and procedure of Haney, Banks and Zimbardo (1973).
- Describe the findings and the researchers' conclusions about status, roles and power.
- Evaluate the study with reference to demand characteristics, data collection and the experience of participants (Le Texier 2019).
- Explain why the study's results have not been reproduced, and how its context shaped it.
In the early 1970s, American prisons were under attack: most people released went on to reoffend, and reports of brutality were common. The usual explanation was dispositional: prisons are violent because the people in them, prisoners and guards alike, are violent people. Philip Zimbardo and his students at Stanford University, funded by the United States Office of Naval Research, set out to test a situational explanation instead.
Key idea
Haney, Banks and Zimbardo (1973) at a glance
Aim: to investigate whether the prison environment itself, rather than the personalities of the people in it, produces the hostile behaviour of guards and the passivity and distress of prisoners, by placing ordinary, healthy people in a simulated prison.
Participants: 75 men answered a newspaper advertisement for a study of "prison life" paying $15 a day. After questionnaires and interviews, the 24 judged most stable, most mature and least involved in antisocial behaviour were chosen: healthy male college students, mostly white and middle class, who were strangers to each other. Half were randomly assigned to be guards and half prisoners; the analysis is based on 11 guards and 10 prisoners.
Design: a simulation with one variable, the randomly assigned role (guard or prisoner), observed over time through video and audio recordings, direct observation, shift reports, mood questionnaires and interviews.
Procedure. The "prison" was a corridor in the basement of Stanford's psychology building, with three small cells (about 1.8 m × 2.7 m) and a tiny unlit closet for solitary confinement. Everyone had agreed to take part for up to two weeks.
- Prisoners were "arrested" without warning at their homes by real Palo Alto police officers, charged, handcuffed, searched, fingerprinted and blindfolded. At the prison they were stripped, sprayed with a "delousing" spray and dressed in a smock with an ID number, no underwear, a cap made from a stocking and a light chain locked round one ankle. They were called only by their numbers and stayed 24 hours a day.
- Guards wore khaki uniforms and reflecting sunglasses that made eye contact impossible, and carried whistles and wooden batons. They worked three-man, eight-hour shifts and went home in between. They were told to keep "the reasonable degree of order within the prison necessary for its effective functioning", and that physical punishment was forbidden.
- Zimbardo acted as the prison superintendent, and an undergraduate research assistant as the warden.
The uniforms were designed to deindividuate both groups: to make each person anonymous and part of a group, and to make the prisoners feel humiliated and dependent.
Step through the study as the researchers reported it. Then switch to the critiques.
Findings, as the researchers reported them.
- Interactions between guards and prisoners became mostly negative, hostile and dehumanising. Commands were the most common kind of speech; prisoners became passive while guards took the initiative.
- Guards' aggression escalated day by day, even after prisoners had stopped resisting. The counts of prisoners, which first took about ten minutes, stretched to several hours.
- Five prisoners had to be released early: four with extreme emotional depression, crying, rage and acute anxiety, starting as early as the second day, and one with a stress-related rash.
- Guards differed: some were "tough but fair", some went far beyond their role in "creative cruelty and harassment", and a few were passive.
- Prisoners spent about 90% of their private conversation on prison conditions, not their lives outside.
- The study was ended after six days of the planned two weeks. The researchers reported that the remaining prisoners were delighted and most guards seemed distressed that it had stopped.
Conclusions: status, roles and power. The researchers concluded that the situation, not the participants' personalities, produced the behaviour: the participants had been screened as normal and assigned at random, yet guards became abusive and prisoners broke down. They called it the pathology of power. Being a guard brought status, a group identity and the power to control other people's lives, and "many of the guards showed in their behavior and revealed in post-experimental statements that this sense of power was exhilarating". Taking on the role of guard or prisoner changed how people behaved toward each other, and the most hostile guards became models whom others on their shift copied.
Why did the researchers argue that the guards' behaviour was caused by the situation and not by their personalities?
The critiques. The study was attacked from the start: Erich Fromm (1973) pointed out that about two thirds of the guards did not act sadistically, and Banuazizi and Movahedi (1975) gave 150 students a description of the study and found that 81% correctly guessed the hypothesis and 90% predicted the guards would be oppressive and hostile, so participants could have known what was expected of them.
In 2019 the French researcher Thibault Le Texier published a study of the experiment's archives at Stanford and interviews with 15 of the participants. The syllabus asks you to know three of his criticisms:
- Demand characteristics. The guards were not left to invent their behaviour. At their orientation they received precise instructions about how to treat the prisoners, and the warden had been given the job of eliciting "tough-guard" behaviour, which he pushed for on later days. The guards were asked to play a part but were not told that they were subjects themselves, and they knew what results Zimbardo wanted.
- Data collection. Of about 150 hours of the study, less than 15% was recorded (about 6 hours of video and 15 hours of audio), mostly the dramatic events. The researchers themselves admitted some data were open to "selective sampling", and Le Texier argues the conclusions had been written in advance.
- The experience of participants. Prisoners were not free to leave. The plan sent to Stanford's ethics committee said they would "be led to believe that they cannot leave, except for emergency reasons", and when two prisoners asked to leave on the second day they were refused: Zimbardo told his team the only grounds were medical or psychiatric. Le Texier judges it highly probable that Prisoner 8612, the first to be released, was acting out a breakdown to get out, although Zimbardo disputes this. Le Texier also found that participants were almost never completely immersed: many said they never forgot it was an experiment.
For each claim in the published account, read what the archives show.
Classify each finding from later research on the Stanford Prison Experiment under the criticism it supports.
Note
Can it be reproduced?
The study was run once, with about 21 participants and no control group, so its results have never been properly reproduced. When Reicher and Haslam (2006) ran a prison study for the BBC under ethical oversight, with guards given no instructions on how to act, the guards did not become tyrannical; group identity and leadership mattered more than roles alone. Most psychologists now treat the Stanford Prison Experiment as a dramatic demonstration that raised important questions, not as evidence that roles alone make people cruel.
Worked example
Evaluate the Stanford Prison Experiment
Haney, Banks and Zimbardo (1973) concluded that the roles, status and power of a prison situation, not personality, made guards abusive and prisoners passive.
Evaluate this conclusion, with reference to the critiques of Le Texier (2019). (5 marks)
Step 1: State the conclusion and its basis
The researchers screened 24 healthy students and randomly assigned them to roles, so they argued that differences in behaviour must have come from the situation.
Step 2: Critique 1, demand characteristics
Le Texier (2019) found in the archives that the guards were given precise instructions about how to treat prisoners, and the warden worked to elicit "tough-guard" behaviour. The guards knew the results that were wanted. Their behaviour may reflect what they thought was expected, not the power of the role.
Step 3: Critique 2, data collection
Less than 15% of about 150 hours was recorded (6 hours of video, 15 hours of audio), mostly dramatic events, so the evidence of escalating abuse may be unrepresentative.
Step 4: Critique 3, experience of participants
Prisoners could leave only for medical or psychiatric reasons, and the archives suggest at least one breakdown was acted to get out. The distress may partly reflect being unable to leave, and some evidence is unreliable.
Step 5: Weigh it up and judge
The study had a single run, about 21 participants and no control group, and the BBC prison study (Reicher and Haslam, 2006) did not find spontaneous tyranny. Judgement: the conclusion is weakly supported. The study shows that situations can influence behaviour, but it cannot show that roles alone produced cruelty.
Try it yourself first, then check the working one step at a time.
Answer
The conclusion is weakly supported. Demand characteristics (guards instructed and briefed to be tough), selective data collection (under 15% recorded) and participants' inability to leave (and a possibly acted breakdown) offer other explanations; with one run and no control group, the study does not show that roles alone cause abuse.
Method
State the claim and the evidence behind it.
Apply each criticism: what it is, the evidence, and what it does to the conclusion.
End with a clear judgement.
Watch out for
Listing criticisms without saying how each weakens the conclusion.
Saying the study proves people become evil in uniforms.
Exam tip
Writing about the study in the exam
For "describe the findings", report what Haney, Banks and Zimbardo concluded (status, roles and power produced hostile guards and passive, distressed prisoners; five prisoners released; ended after six days). For "evaluate", go further: name the critique (demand characteristics, data collection, experience of participants), give the evidence (Le Texier 2019) and say what it does to the conclusion: the behaviour may reflect what participants thought was expected, not the power of roles.
Summary
Aim: whether the prison situation, rather than personality, produces guards' hostility and prisoners' passivity.
Procedure: 24 screened, healthy male students (from 75 volunteers, $15 a day) randomly assigned to guard or prisoner; realistic arrests; uniforms that deindividuated; a basement prison; planned for up to two weeks.
Findings: hostile, dehumanising interactions; escalating guard aggression; five prisoners released early; ended after six days.
Conclusion: the "pathology of power": status, roles and power in the situation shaped behaviour.
Le Texier (2019): demand characteristics (guards instructed and briefed), data collection (under 15% recorded, selective) and the experience of participants (not free to leave; at least one breakdown possibly acted).
Never reproduced; the BBC prison study (2006) found no spontaneous tyranny. Treat the conclusion with caution.
Finished the lesson?
Try the 2 quick checks above and the lesson completes itself, or mark it complete now.