Behavioural Study of Obedience


This study, conducted over five decades ago, is probably one of the most controversial studies in the history of psychology. Milgram, the researcher, was influenced by the horrific conditions and events that occurred in Nazi concentration camps. The inhumane practices of Hitler and the Nazis resulted in the systematic murder of millions of people. But why did those soldiers obey the orders to kill? Adolf Eichmann, one of the main figures of the Holocaust, said the following when he was on trial for his part in the massacres:

It was not my wish to slay people… Once again I would stress that I am guilty of having been obedient… At that time obedience was demanded, just as in the future it will also be demanded of the subordinate.

Some historians believed that German people were naturally more obedient, regardless of what they are asked to do. This theory is actually an example of a ‘dispositional hypothesis‘. Milgram originally set out to test this hypothesis. He said that anybody who was put into the same situation as those Germans would have behaved in the same way; they would have obeyed authority. Milgram’s theory is an example of a ‘situational hypothesis‘. This experiment was actually his pilot study.

Stanley Milgram’s magic hands

A newspaper advertisement and mail advertisement were sent out. The advertisement asked for volunteers to take part in a study on “memory and learning” at Yale University (the setting). The final sample consisted of 40 males aged 20 to 50 years old, all from New Haven and surrounding areas. The participants had different occupations, such as high school teachers and engineers. They were paid $4.50 for participating, but were to be given this money simply for coming to the laboratory, even if they chose not to take part.

The participants, when they arrived at the psychology laboratory, were given an introduction on the relationship between punishment and learning. They were in two’s: one participant and one ‘victim’. The victim was an actor who was 47 years old and came across as likeable. They were then given further information, including:

So in this study we are bringing together a number of adults of different occupations and ages. And we’re asking some of them to be teachers and some of them to be learners. We want to find out just what effect different people have on each other as teachers and learners, and also what effect punishment will have on learning in this situation. Therefore, I’m going to ask one of you to be the teacher here tonight and the other one to be the learner. Does either of you have a preference?

The participant then took a slip from a hat to decide his role. This was rigged so that he would always get the slip declaring him to be the ‘teacher’, and the actor automatically assumed the role of the ‘learner’.

The learner sat in an ‘electric chair’ with straps to “prevent excessive movement” during electric shocks that would be given from a shock generator, the participant was told. An electrode was attached to the learner’s wrist and electrode paste (really just cream!) was applied to “avoid blisters and burns”. The learner was told (while the teacher was in earshot!) that “although the shocks can be extremely painful, they cause no tissue damage”. All these tricks basically made the experiment seem much more real to the participant. The final trick was to give the participant a sample shock of 45 volts (from an external battery, since the shock-generator wasn’t actually supposed to work).

The participant was then seated in an adjacent room and asked to read word pairs to the learner. The learner had to memorise the pairs. The teacher then tested the learner by giving him one of the words in a pair with four other words. The learner had to choose the word that was previously paired with the first word. If his answer was correct, the next word on the list was read. If it was incorrect, the teacher had to state the correct answer and administer an electric shock (starting from 15 volts). For every wrong answer, the shock level would increase by 15 volts (15, 30, 45… you do the math). The learner actually gave a predetermined set of answers with about three wrong answers for every right answer.

At 300 volts, the learner would start hitting the wall and shouting. He stopped answering questions. Here, the participant usually looked at the experimenter for guidance. They were told to interpret no answer as a wrong answer, meaning that the learner would have to be shocked again. The learner’s pounding continued after 315 volts, after which the learner went completely silent. If the participant again asked for advice, he would be met with a sequence of firm standardised ‘prods’:

  1. “Please continue.”or “Please go on.”
  2. “The experiment requires that you continue.”
  3. “It is absolutely essential that you continue.”
  4. “You have no other choice, you must go on.”

If the participant asked about permanent physical injuries, a different prod was used: “Although the shocks may be painful, there is no permanent tissue damage, so please go on.” If the participant said that the learner did not want to continue, another prod was used: “Whether the learner likes it or not, you must go on until he has learned all the word pairs correctly, so please go on.”

The experiment ended if the participant refused to continue after the four official prods, or if 450 volts were administered.

A participant who stopped before reaching 450 volts was classed as a “defiant participant” while those who went up to 450 volts were classed as an “obedient participant”. The experiment was filmed and observers watched through a one-way mirror.

After the experiment, the participants were debriefed and an open-ended questionnaire was given. Some psychometric tests were taken to make sure there was no emotional harm. The teacher and learner were reunited in order to prove that the learner was an actor who was not really in danger. The participants were told that their reaction was normal, another measure taken to ensure they left the experiment without mental discomfort.

So how far did the participants go?

  • All 40 of the participants obeyed up to 300 volts.
  • 5 men refused to continue after 300 volts.
  • 4 men refused to continue after 315 volts.
  • 2 men stopped at 330 volts.
  • 1 man stopped at 345 volts.
  • 1 man stopped at 360 volts.
  • 1 man stopped at 375 volts.

That means 26 participants were obedient and administered a complete round of electric shocks, while only 14 were defiant. So 65% obeyed and 35% terminated the experiment before reaching 450 volts.

There are lots of other important things here. For example:

  • When the experimenter was the one to end the experiment (at 450 volts), many of the participants sighed in relief or shook their heads in what seemed like regret.
  • Most of the men were sweating, shaking, stuttering or fidgeting during the study.
  • 14 participants had nervous fits of laughter.

The participants who had nervous laughter fits (some were pretty extreme!) were keen to point out that they weren’t sadistic and that their laughter wasn’t a sign that they enjoyed shocking the learner.

Most of the participants had been fully convinced that the experiment was real.

Type of research method
This was a laboratory experiment. It was conducted in a controlled environment with electronic equipment (the shock-generator). However, it has also been described as a “controlled observation”.

Independent variable
The IV was the ‘prods’ given by the experimenter.

Dependent variable
The DV was the degree of obedience (how high the voltage went) shown by the participant.


  1. High level of control: Since the experiment was conducted in a laboratory, there was a lot of control over extraneous variables (variables which aren’t part of the experiment but might affect it, such as the weather).
  2. High level of experimental realism:Looking at the participants’ reactions (i.e. full-on freak-out), we can see that the men actually believed the experiment was really happening and that they were really electrocuting another person.
  3. Replicable: This experiment can be (and has been) replicated. It was conducted in a laboratory and was strictly standardised (like the prods) so another researcher can redo the procedure.


  1. Not generalisable:The participants were all males and all from the same area, meaning we can’t apply the results of this study to females or people living in other countries.
  2. Selection bias: The men responded to an advertisement which may mean that they had a common characteristic that wasn’t present in people who didn’t respond to the study… if that makes sense.
  3. Lack of ecological validity: This experiment was conducted in a laboratory, an artificial setting. This means the participants knew there was an experimenter present and that they were part of a study. They may have behaved unnaturally and acted different if they had been in a natural, non-experimental setting (though I’m not sure who would normally electrocute people even in an everyday setting, unless you’re an evil-mastermind-person-thing).
  4. Demand characteristics:This is a fancy way of saying that the participants were behaving in a way they felt was being asked from them; they were responding to a social cue. Someone may do this to look socially desirable or ‘normal’. This point links to the point above.

Ethical issues

  1. Informed consent: Milgram did get informed consent from the participants.
  2. Deception: The participants were deceived pretty much throughout the whole experiment. They were deceived about the nature of the study, the role of the actors and the false-danger.
  3. Confidentiality: The identities’ of the participants was kept private.
  4. Emotional or physical harm: The men were physically unharmed but emotionally they were pretty messed up! However, it was only short-term and interviews taken after the experiment showed that there was no sign of long-term emotional damage.
  5. The right to withdraw: Milgram said the participants could withdraw from the study at any time (and still get paid) but some researchers think that the prods made participants feel like they had to continue with the study.
  6. Debriefing: All participants were debriefed after the experiment and given lots of reassurance. They were told the real reasons behind the study and got to meet the actors.

Reference: Milgram, S (1963). Behavioral Study of Obedience. Journal of Abnormal and Social Psychology.