Milgram's 1963 Obedience Study: What It Found and What Came After
Key Points
- In Stanley Milgram's 1963 paper, 26 of 40 men obeyed to the end, pressing the last switch on a shock generator marked up to 450 volts. The learner never received a shock.
- Fourteen Yale psychology seniors had predicted that, on average, 1.2 percent of people would go to the last switch.
- Across the 30 or so variants Milgram ran, the share going to 450 volts ranged from 0 percent to nearly 100 percent, according to a 2012 essay in PLOS Biology.

In 1963 the psychologist Stanley Milgram reported on a study he had run at Yale. In it, 26 of 40 ordinary men kept giving what they believed were electric shocks to a stranger, all the way to the last switch on the machine, as a man in a gray coat told them to continue. The paper, "Behavioral Study of Obedience," appeared in the Journal of Abnormal and Social Psychology1, and the experiment behind it is still argued over. Here is what the original paper says, and what later work has added.
How the experiment worked
The subjects were 40 men aged 20 to 50 from New Haven and nearby towns, recruited by a newspaper advertisement and direct mail1. The study ran at Yale. They believed they were joining a study of memory and learning there, and they were paid $4.50 just for coming. Each was cast as the "teacher." The "learner," strapped into a chair in the next room, was really a 47-year-old accountant trained for the role, and the experimenter, in a gray technician's coat, was played by a 31-year-old who taught biology at a high school.
The teacher was told to give the learner a shock for every wrong answer, one step higher each time. The shock generator had 30 switches running from 15 to 450 volts in 15-volt steps1, labelled in groups from "Slight Shock" up to "Danger: Severe Shock," with the last two marked only "XXX." The learner never received a shock. The only real one was a 45-volt sample given to each teacher before he began. When a subject hesitated, the paper says, the experimenter used a fixed series of four "prods," starting with "Please continue" or "Please go on" and ending with "You have no other choice, you must go on."
What happened
Before running the study, Milgram asked 14 Yale seniors, all psychology majors, to predict how 100 Americans would behave. Their estimates of how many would go to the end ranged from 0 to 3 percent, with a mean of 1.2 percent1.
In fact every one of the 40 subjects kept going to at least 300 volts, the point where the learner pounded on the wall and stopped answering. Fourteen defied the experimenter at some point after that, and 26 obeyed to the end, pressing the generator's last switch1. Many were visibly distressed. The paper describes subjects who sweated, trembled, stuttered and dug their fingernails into their flesh. Separately, 14 of the 40 showed definite signs of nervous laughter, and three had full-blown, uncontrollable fits of it.
Afterward each subject was interviewed. The paper says a friendly reconciliation was arranged with the learner and that steps were taken so subjects would leave in a state of well-being1. The Australian psychologist Gina Perry, who investigated Milgram's data and writings, later argued that participants were not properly debriefed4.
What later work has added
In a 2012 essay in PLOS Biology, the psychologists Alexander Haslam and Stephen Reicher note that the 65 percent figure (26 of 40 going to 450 volts) comes from Milgram's baseline study, and that it was not especially typical of the 30 or so variants Milgram ran. Across them, the share going to 450 volts ranged from 0 percent to nearly 100 percent, and across the studies as a whole a majority of subjects did not go that far2. Compliance fell when the study was moved from Yale's laboratories to commercial premises2. One such variant, according to Wikipedia's account of the experiments, took place in a modest office in Bridgeport, Connecticut, under the name Research Associates of Bridgeport4.
Haslam and Reicher argue that people's willingness to follow an authority depends on how far they identify with it and believe it is right2, rather than on passive, unthinking conformity. That is their interpretation of the experiments.
In a partial replication run in 2006 and published in 2009, the psychologist Jerry Burger copied a later version of Milgram's study, Experiment 5, in which the learner first protests aloud at 150 volts3. To protect participants, he stopped each session there. He found obedience rates only slightly lower than Milgram's, with 70 percent carrying on after the 150-volt switch until the experimenter stopped them, against 82.5 percent who continued past that point in Milgram's comparable condition. Men and women did not differ3. Milgram described 19 variations of the experiment in his 1974 book "Obedience to Authority: An Experimental View"4.
Trivia
- What was the highest voltage marked on Milgram's shock generator? 450 volts1.
- How many of the 40 subjects in the 1963 paper went to the last switch? 261.
- What share did the Yale psychology seniors predict, on average, would go to the end? 1.2 percent1.
- How much were subjects paid for coming to the laboratory? $4.501.
- In which Connecticut city did a variant run in a modest office away from Yale? Bridgeport4.
- What was the title of Milgram's 1974 book? "Obedience to Authority: An Experimental View"4.
Sources
- Stanley Milgram, Behavioral Study of Obedience, Journal of Abnormal and Social Psychology, vol. 67, no. 4, 1963, pp. 371 to 378 (copy hosted by the University of Miami). Accessed October 23, 2026.
- S. Alexander Haslam and Stephen D. Reicher, Contesting the “Nature” Of Conformity: What Milgram and Zimbardo's Studies Really Show, PLOS Biology, November 20, 2012. Accessed October 23, 2026.
- Jerry M. Burger, Replicating Milgram: Would People Still Obey Today?, American Psychologist, vol. 64, no. 1, 2009. Accessed October 23, 2026.
- Milgram experiment, Wikipedia. Accessed October 23, 2026.