Switches were labelled as follows: - ‘slight shock’, ‘danger’, ‘intense/severe shock’ and ‘XXX’. During the experiment the confederate made increasingly more errors. The confederate pretended the shocks were real by crying out in pain and claiming he had chest pains. Milgram and his team urged the participants to continue despite all of this, and the experiment was only terminated if the participant refused to go on four times. Participants were visibly traumatised and 3 had uncontrollable seizures.
The results of the experiment were that 100% of the participants went up to ‘severe shock’ level and 65% continued to the maximum 450 volts level; they obeyed the instructions to the letter.
Milgram’s experiment has been criticised for the ethical issues it raises and in terms of its validity.
One of Milgram’s fiercest critics was Baumrind (1964) who argued that the rights and feelings of Milgram’s participants had been abused, and that inadequate measures had been taken to protect them from stress and emotional conflict.
Milgram defended himself by arguing that Baumrind’s criticism assumes that the experimental outcome was expected, inducing stress was not an intended or deliberate effect of the experimental procedure.
Milgram has also been criticised for deceiving his participants in his experiment.
The deception began even before the participants were selected, with the newspaper advertisement giving false information. They were then deceived when drawing lots for ‘teacher’ and ‘learner’ roles as this was fixed, as the confederate was always the ‘learner’. The biggest deception of them all was the fact that the electric shocks the participants thought they were administering weren’t even real (!).
The performance was very convincing and the apparatus used was highly realistic.
Milgram’s defence was that without the deceptions the experiment would not have worked.
Using deception, as Milgram obviously did, means that informed consent is not possible. This is another ethical issue that Milgram’s experiment raises.
Many of the participants were visibly shaken by their experience. This raises the issue ‘protection of participants’.
Initially poised businessmen were reduced to twitching, stuttering wrecks in under twenty minutes (!), (but yet they still administered the shocks up to maximum level!!!!!).
Milgram argues that the reactions of the participants were characteristic rather than exceptional responses to the experiment.
The BPS states that at the start of any psychological investigation researchers should inform the participants of their right to withdraw from the research at any time.
Firstly, Milgram refers to his participants as ‘volunteers’. Secondly, Milgram does admit that once participants have agreed to take part, there is an obligation not to withdraw. The whole point of the research was to encourage the participants to stay as seen due to the fact that ‘prods’ were used and also due to the experimenter making it very difficult for the participants to withdraw (using the prods).
The BPS also states that the participants should be debriefed after an investigation (i.e. given sufficient information to understand the purpose of the research and any deception should be revealed).
In terms of this aspect of ethical conduct, Milgram comes out well. After each experiment participants were interviewed and given a battery of psychological tests to ensure no harm had came to them. They were also told their behaviour during the experiment was completely normal and that they shouldn’t blame themselves or be embarrassed. They were told the shocks weren’t real and every effort was made to reduce any tension that arose as a result of the experiment. The debrief was careful and thorough.
Also, short-term and long-term effects can be looked at. In the short-term many participants were extremely distressed but does this amount to any psychological harm? And in the long-term did the participants experience any lasting harm?
Milgram had discussed his experiments at length with his colleagues beforehand and none had expected the results they got. The ‘extreme tension’ that resulted wasn’t anticipated nor intended.
Milgram admits the obvious short-term distress but argues that there is no evidence of long-term harm; mainly due to the fact that he hired a psychiatrist to interview 40 of the participants a year later and there was no indication of any harmful effects.
Milgram’s study is also criticised in terms of its validity; both internal and external.
In terms of its internal validity, did it measure what it intended to measure? If it actually measured obedience to authority then it is internally valid.
Milgram rejects the criticisms of the participants seeing through the experiment and playing along, by claiming that with few exceptions participants were convinced of the reality of the experimental situation. A questionnaire a year after the experiment itself showed that 56.1% fully believed they were administering the shocks. 24% had doubts but on balanced believed, and only 2.4% were certain the shocks were not real.
Milgram claims his experiment was internally valid but can his findings be generalised? I.e. was his experiment externally valid?
Both the setting and the relationship involved in the experiment are very different from the world outside the laboratory therefore they can’t really be generalised.
But in terms of gender the experiment can be generalised. This is because Milgram repeated the experiment with 40 female participants rather than 40 male participants.
It is not possible to say whether Milgram’s findings can be generalised to other societies and also that culture has an effect because although when replicated in other countries obedience differed, this may well have been due to differences in the replication.
Also, it is not possible to say whether Milgram’s findings can be generalised in other time periods due to the fact that changes in ethical standards stop researchers conducting this type of research nowadays. It is therefore not really possible at all to define the external validity of Milgram’s experiment.
Interestingly, in follow-up experiments results were very different when certain aspects of the experiment were changed. For example, when the experiment was conducted in a run-down office building and the experimenter was not wearing a scientist’s lab coat obedience fell to 47.5%.
The original experiment was carried out at Yale, a prestigious Ivy League university and therefore the participants perceived it as very important. They were convinced of the integrity of the experimenter due to the status of the university.
Milgram’s research has been branded as unethical for what others see as a deliberate deception, a lack of informed consent, a disregard of the right to withdraw, and a failure to protect participants from psychological harm, yet Milgram has defended himself well on all these points (as mentioned in the above).
His final defence is the fact that the significance of his findings justifies the methods he used. Yet his experiment is STILL criticised on ethical grounds and is often held up as an example of how psychological research with human participants shouldn’t be done.
It undoubtedly helped highlight the ethical dimension of psychological research, his and others experiments are amongst the most controversial in the area of social influence and in the whole of psychology and for these reasons are often cited in the context of the ethics of research BUT given the present ethical guidelines, it is very unlikely that Milgram’s experiment could be conducted today.