Obedience (Edexcel A-Level Psychology): Revision Notes
Milgram's Research Into Obedience
Aim
Stanley Milgram designed this research to examine whether ordinary individuals would follow instructions to administer potentially harmful electric shocks to another person. His work was influenced by earlier research into conformity (Asch's studies), which he considered to lack real-world application. Milgram sought to create conditions that closely resembled genuine 'obedience to commit harm' scenarios.
Milgram's research emerged from a desire to understand the mechanisms of obedience that had enabled atrocities during World War II. Unlike Asch's conformity studies, Milgram wanted to examine situations where following orders could result in serious harm to others.
He also aimed to identify the circumstances that would either increase or decrease obedience levels, leading him to conduct multiple variations of his original design.
Method
Sample
Milgram recruited participants through a newspaper advertisement seeking male volunteers for a study on memory and learning. He advertised all occupations and offered $4 compensation. From 296 responses, Milgram selected participants through direct mail invitations, ensuring diverse representation across occupations, educational backgrounds and ages. The final sample comprised 160 participants, distributed equally across four initial experimental conditions.
Procedure
Participants received advance notification about their appointment at Yale University's Interaction Laboratory. The study employed two confederates: a biology teacher acting as the experimenter, and a middle-aged accountant playing the learner.
Upon arrival, participants encountered someone they believed was another volunteer but who was actually a confederate. The experimenter explained that the research investigated whether punishment enhanced learning effectiveness. This briefing served to justify the forthcoming procedure, particularly the use of electric shocks. The experimenter emphasised that whilst the study involved human participants of varying ages, it had not previously been tested on such subjects.
The allocation of roles appeared random through a rigged draw, ensuring the genuine participant always became the 'teacher' whilst the confederate became the 'learner'. Both lots contained the word 'teacher', guaranteeing this outcome.
This deception was crucial to the experimental design, as it ensured all participants experienced the same condition whilst believing they had been randomly assigned.
The learner was then seated in an electric chair with arms secured. The experimenter explained this restraint was necessary to prevent excessive movement during shock administration. An electrode was attached to the learner's wrist, with conductive gel applied to prevent skin irritation. The teacher was informed that whilst shocks would be painful, they would not cause permanent tissue damage—a statement designed to convince the teacher that genuine shocks would be administered.
Apparatus
The shock generator was positioned in an adjacent room. This device featured a panel with 30 switches arranged horizontally, increasing in 15-volt increments from 15 to 450 volts. Labels above the switches ranged from 'Slight Shock' through 'Moderate Shock', 'Strong Shock', 'Very Strong Shock', 'Intense Shock', 'Extreme Intensity Shock' and 'Danger', with the final two switches marked 'XXX'.
Visual and auditory feedback accompanied each switch activation: a light illuminated and a buzzer sounded. Additional features included a blue light and voltage meter with an oscillating needle when activated. These features were designed to enhance the apparatus's credibility and make the experience more psychologically realistic for participants.
To establish credibility, participants received a 45-volt sample shock.
Learning task
Participants were instructed to teach the learner a list of word pairs by reading them aloud. They would then present the first word alongside four possible options, and the teacher was to administer a shock if the learner gave an incorrect response. The learner responded using four switches visible to the teacher on the shock generator. Critically, the sequence of learner responses was predetermined.
Instructions
The participant teacher sat before the shock generator with instructions to administer progressively stronger shocks for each error, reaching 450 volts before repeating that level. The experiment incorporated systematic verbal encouragement to promote obedience.
Verbal prods:
- Please continue.
- The experiment requires that you continue.
- It is absolutely essential that you continue.
- You have no other choice, you must go on.
Special prods:
In response to specific concerns from teachers, such as whether the procedure was harmful or whether the learner wished to continue, the experimenter employed special prods:
- Although the shocks may be painful, there is no permanent tissue damage, so please go on.
- Whether the learner likes it or not, you must go on until he has learned all word pairs correctly. So please go on.
These standardised verbal prods were crucial to maintaining experimental control whilst encouraging participants to continue despite their discomfort.
During a pilot investigation, Milgram discovered that participants required minimal encouragement to administer maximum shock levels. Consequently, he introduced a remote-feedback condition (Experiment 1), where the learner banged on the wall at 300 volts and refused to answer after 315 volts. Of 40 participants studied, 26 continued to the shock generator's maximum level, administering 450 volts for three attempts before the experiment concluded.
Results from key experiments
Experiment 2: Verbal-Feedback Condition
In this variation, participants heard the learner's responses as shocks intensified:
- 75 volts: Grunts
- 120 volts: Declarations that the shock was painful
- 135 volts: Painful groans
- 150 volts: Shouted 'Experimenter get me out of here!'
- 270 volts: Agonised screams
- 300 volts: Screamed and refused to continue
- 330 volts: No verbal feedback and no word task response
The experimenter informed the teacher that no response should be treated as incorrect, justifying further shocks.
In Experiment 3 (proximity), the learner occupied the same room as the teacher. In Experiment 4 (touch-proximity), the learner was required to place his own hand on a shock plate to receive the shock. At 150 volts, the learner refused, and the teacher was instructed by the experimenter to physically force the learner's hand onto the plate.
Debrief
Following the experiment, each participant was reunited with the learner in a friendly atmosphere and informed that no genuine shocks had been administered. Both obedient and dissenting participants received reassurance that their behaviour was normal and defensible. Some participants were interviewed extensively, and all received a written report about the investigation alongside a follow-up questionnaire enabling them to express their feelings after reflection.
Main findings
Questionnaire
Milgram investigated whether people's predictions matched the actual experimental outcomes. He invited 39 psychiatrists, 31 college students and 40 middle-class adults to a lecture on obedience and the experiment's nature, then presented them with a shock generator panel diagram. All were asked to predict how far up the generator they believed they would proceed before dissenting.
Five participants claimed they would refuse to administer any shocks, and most stated they would withdraw earlier in the experiment; none anticipated passing the 300-volt threshold.
This dramatic discrepancy between predicted and actual behaviour highlights how difficult it is to anticipate our own responses in situations involving authority and obedience.
Milgram hypothesised that participants might have responded in socially desirable ways, so he asked a sample of psychiatrists, students and middle-class adults to plot their predictions for 100 average Americans. Psychiatrists predicted that most Americans would not persist beyond the 150-volt level, with only 4 per cent continuing to the end.
Experiments 1, 2, 3 and 4
| Experiment | Description | Obedience rate |
|---|---|---|
| Experiment 1 | Remote-feedback condition: all participants administered 300-volt shocks | 26 out of 40 (65%) continued to 450 volts for three attempts |
| Experiment 2 | Verbal-feedback condition | 25 out of 40 (62.5%) went to the 450-volt level |
| Experiment 3 | Proximity condition | Fell to 40% |
| Experiment 4 | Touch-proximity condition | 30% |
Throughout these experiments, participants displayed visible signs of distress, including protests, nervous twitch behaviours and anxiety. These reactions suggested that participants genuinely believed they were administering real shocks, despite later criticisms questioning this assumption.
Conclusion
Milgram concluded that when the learner could not be heard or seen, participants demonstrated greater willingness to follow orders causing harm. When the learner could be seen, participants showed more empathy towards the learner, resulting in reduced obedience.
Variation studies
Milgram repeated his experiment to examine under what conditions greater or lesser obedience would emerge, helping explain the unexpectedly high obedience levels to authority. His subsequent experiments varied from the original, with most conducted in Yale University's basement. He also modified the learner voice-feedback to include more verbal protests ('Let me out of here!') and comments about the learner's heart condition ('My heart is bothering me now').
Telephonic instructions (Experiment 7)
To establish whether the experimenter's physical proximity influenced obedience levels, Milgram designed three experiments altering the relationship between teacher and experimenter. In Experiment 5, the experimenter sat a few feet from the teacher; in Experiment 6, the experimenter's nature was altered from authoritative and technical-looking to gentle and unaggressive. In Experiment 7, Milgram examined the effect of physical distance between experimenter and teacher. After providing initial face-to-face instructions to the teacher, the experimenter departed the room and continued giving instructions via telephone.
Experiment 7 Results: Telephonic Instructions
Milgram discovered that participants willing to administer the maximum 450-volt shock decreased from 65 per cent to 22.5 per cent in Experiment 7.
Key observations:
- Participants continued administering lower shocks rather than increasing voltage
- Many deceived the experimenter about this, reassuring him they were increasing shock levels
- When the experimenter returned to the laboratory, participants' initial disobedience rapidly transformed to obedience
This demonstrated that the physical presence of authority significantly influenced compliance levels.
Rundown office block (Experiment 10)
Milgram recognised that Yale University's institutional context might encourage participants to display higher obedience levels than in less prestigious environments. Follow-up interviews conducted after the original experiment series revealed many participants referenced the integrity of Yale institution as providing confidence to participate in the study, which they would not have undertaken in less respected settings.
He relocated his experiment to a rundown commercial office building in Bridgeport, Connecticut. Participants were recruited through mailshot and paid for their time. Dissociated from Yale University, participants were informed that the study was being conducted by Research Associates of Bridgeport, a private company conducting research for commercial industry. The same laboratory procedures were followed as in Yale University's basement, though the building was rather poorly furnished.
Experiment 10 Results: Rundown Office Block
Milgram found a slight reduction in obedience of 48 per cent, concluding that the less reputable context reduced the study's legitimacy.
He also reported that participants questioned the company's credentials upon arrival (though these claims were made during debrief). This suggests that environmental context and perceived legitimacy play a role in obedience, though perhaps less significantly than the presence of authority figures.
Ordinary man gives orders (Experiment 13)
In experiments varying the learner's and experimenter's roles within the study, Milgram examined the impact of power relations on obedience. Experiments 12–15 altered the power, position and action of the actors during the experiment. In Experiment 13, the experimenter's role was played by an ordinary man, rather than an experimenter wearing a grey laboratory coat as portrayed in previous variations. This variation was designed to examine the role of authority and status on obedience.
Three people arrived at the laboratory; two were confederates of the study. A rigged draw determined who would become teacher, learner and experimenter. The first confederate received the learner role, the second confederate was assigned the experimenter role with the task of noting times from a clock whilst seated at a desk, and the genuine participant received the teacher role. The experimenter followed instructions to strap the learner into the electric shock chair but did not inform the teacher what shock levels to administer during the study. The experimenter received a phone call to leave the room and departed with a comment to the teacher to continue getting the learner to persist with the word pairs until mastered perfectly.
To ensure some instruction to increase shock levels was provided, the learner stated that a good approach to conduct the study would be to increase the shock level each time he made an error in learning the words. Throughout the experiment, the learner reiterated this instruction. In this variation, all instructions were provided by an ordinary man.
Experiment 13 Results: Ordinary Man Gives Orders
Milgram found that 80 per cent (16 out of 20 participants) withdrew before the maximum shock level, resulting in a 20 per cent obedience level.
However, Milgram described this experiment as strained, with the learner required to make considerable efforts to persuade the teacher to continue administering shocks in the experimenter's absence.
Experiment 13a: Subject as a Bystander
A further experiment was staged to resolve some of this difficulty. In this adaptation, when the teacher refused to administer shocks at the learner's insistence, the learner stated that he would administer the shocks himself if the teacher was unwilling to do so. The learner moved to the front of the shock generator and instructed the teacher to record the duration of shocks administered.
Results:
- All 16 participants protested this situation
- Five physically restrained the learner or interfered with the shock generator
- The remaining 11, despite their protests, allowed the learner to self-administer a 450-volt shock level
- This resulted in a 68.75 per cent obedience level (or in this case, passivity with some protest)
Evaluation
Methodological issues
Strengths:
Milgram's experiments demonstrated high levels of standardisation and control. Each participant received identical briefing and experienced the same verbal prods, feedback and apparatus. Behavioural data was gathered about participants' time taken to press each switch and switch depression duration. Both objective quantitative data and qualitative observations were collected about participants' behaviour during the research, making this research highly credible in terms of being scientific.
Weaknesses:
One criticism directed at the experiment series was that the sample of New Haven men and small numbers of women were not representative of the general population. Participants were recruited by advertisement and mailshot, resulting in a volunteer (self-selecting) sample. This could suggest that participants were more compliant or more authoritarian in character.
When debriefed, Milgram noted that participants participated for various reasons, not one single overriding factor that could account for high obedience levels. Milgram also noted that participants who administered the greatest shock levels tended to blame the learner for their pain rather than themselves or the experimenter; a typical trait of an authoritarian character. However, other research into volunteering characters has tended to find them less, not more, authoritarian.
The "Playing Along" Criticism
Another criticism of Milgram's experiments was that participants did not genuinely believe they were administering harmful or serious electric shocks, and were merely playing along with the game.
Milgram dismissed this claim and made two arguments against it:
- The observed anxiety of teachers throughout the shock administration duration was evidence itself that they believed the shocks were genuine
- Only two of the 40 participants in Experiment 2 thought that the study was a hoax
Milgram believed that these participants were probably defending their own behaviour so that they were not perceived as cruel.
Laboratory research, such as this, is often criticised for being unrealistic and not representative of real behaviour. We can certainly argue that the task lacked mundane realism as it was a unique situation that one would not ordinarily encounter. Milgram maintained that the process of obedience was the same regardless of location or task.
Interestingly, the Hofling et al. (1966) field study of the doctor-nurse relationship found far more nurses willing to obey a doctor's instructions than Milgram found in his experiment. Perhaps in real life, obedience is greater than Milgram's experiments would predict. Or perhaps the Hofling et al. findings could be explained by the strict hospital hierarchy and the legitimate status of doctors at the time (compared to the grey laboratory-coat wearing experimenter used by Milgram).
More recently, a virtual reality experiment has been developed by several researchers to examine obedience in computer simulation conditions. Slater et al. (2006) used virtual characters as victims and participants were aware, therefore, that the shocks being administered were not genuine. Consistent with Milgram's experiment, they found that participants who could see the animated victim were less obedient than those who only communicated with the victim via text. This raises the question of whether obedience can be studied under laboratory conditions to good effect and without the ethical implications associated with using live confederates.
Gina Perry's Archival Analysis
Gina Perry conducted archival analysis of Milgram's reports, video and audio tapes, revealing some substantial issues with Milgram's findings and conclusions:
- She found that Milgram overstated the obedience rate of his research, equating to 65 per cent, which did not represent the high levels of people who dissented throughout his variation studies
- She claims that many more participants revealed that they knew the study was a hoax and that no genuine shocks were being administered
- She claimed that Milgram's procedures were not as standardised as he claimed. Analysis of video and audio recording showed that there was a great deal of improvisation through the experiments by the experimenter, resulting in what she believed to be direct coercion to make the teacher continue with the shocks
Gina Perry published a book on her findings, Behind the Shock Machine: The Untold Story of the Notorious Milgram Psychology Experiments (2012).
Ethical issues
Diana Baumrind (1964) heavily criticised Milgram's experiments on ethical grounds. She expressed considerable concern for participant welfare and argued that the stress caused was deliberate. Milgram responded by stating that the anxiety induced by the experimental conditions was not deliberate or anticipated. He had discussed the experimental procedures extensively with colleagues and none had anticipated the participants' responses.
Although research outcomes cannot ever be predicted with reliability, it does not explain the fact that Milgram conducted 18 variations to his study, involving 636 participants. Although participants' reactions could not have been foreseen at the beginning, they certainly could have been predicted once the experiments were underway.
Milgram justified the anxiety he caused to participants by describing it as 'momentary excitement', which in his view was not the same as harm.
Every experiment Milgram conducted could also be criticised for involving considerable amounts of deception:
- Participants thought it was a study of memory and learning, not obedience
- They were deceived into believing the drawing of lots was genuine when in fact it was rigged
- They believed that the confederate learner was a genuine participant
- They thought the shocks administered were real
This deception was a necessary element for the procedure of the study, but would be ethically problematic by today's ethical standards. Moreover, it could have caused additional stress and embarrassment for the participants when they were debriefed.
Milgram, however, went to considerable lengths to ensure that participants did not feel embarrassment. He fully debriefed the obedient participants by explaining that their actions were normal, and the dissenting participants were assured that their decision making was justified. He also ensured a friendly reconciliation between the participant and the learner and followed up with a full written report for all participants and a follow-up questionnaire for them to express their feelings about the experiment after some time.
Milgram's post-experimental questionnaire seemed to confirm that participants did not have any negative feeling about their participation, with 84 per cent having said that they were glad to have participated.
Although Milgram clearly offered participants the right to withdraw from the experiment, some argue that their right to leave was violated by the verbal prods used by the experimenter. Milgram briefed participants advising them that they could leave at any time without adverse consequence, and they could even take the money incentive. It is true that Milgram did not tie his participants to their chair. He did, however, enter them into a contract of trust and incentivised their participation with money, and the verbal prods used directly challenged any participant's attempt to leave the situation.
In defence of Milgram, the verbal prods were an essential requirement to ensure orders were given that demonstrate obedience, and as 35 per cent (or more) of his participants did end the experiment, it could be argued that the prods merely dissuaded withdrawal.
Milgram vehemently defended his series of experiments, arguing that no one would have been so concerned with the ethical issues associated with the research if they had not found such high levels of obedience from the ordinary man.
Wider issues: gender
It is notable that the majority of Milgram's experiments involved only male participants. In fact, he only ever conducted one experiment involving women. This may lead us to conclude that the research is androcentric and therefore the findings can only be applied to explaining male obedience.
However, Milgram did not find any notable difference in obedience rates between men and women, although women did experience greater anxiety. Further research into obedience (Burger, 2009) also indicates that women are equally obedient, indicating that the only difference between the genders is that of empathy.
Key Points to Remember:
- Milgram's research demonstrated that 65% of ordinary people would obey orders to administer potentially lethal electric shocks to another person in the original experiment
- Obedience levels decreased when:
- The experimenter's proximity decreased (telephonic instructions: 22.5%)
- The setting was less prestigious (rundown office: 48%)
- An ordinary person gave orders (20%)
- The methodology was highly standardised and controlled, but has been criticised for lacking ecological validity and involving a self-selecting sample
- Serious ethical concerns were raised, particularly regarding deception, psychological harm, and the right to withdraw, though Milgram conducted thorough debriefing procedures
- The research has been criticised by Gina Perry for overstating obedience rates and lacking the standardisation Milgram claimed