Social Influences

Kevin Wren

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781134663569
• Edition: 1
• Topic: Psychologie
• Subtopic: Geschichte & Theorie in der Psychologie

Social Influences looks at how we perceive ourselves and others and how this can influence our behaviour. It includes stereotyping and prejudice, obedience and conformity, collective behaviour and leadership.

1

Obedience

Introduction

Obedience research

Milgram's experiment

Criticisms of Milgram's experiment

Other obedience studies

Field experiments

Some explanations of obedience

Summary

Introduction

One aspect of behaviour brought about through social influence is obedience, i.e. obeying direct instructions or orders. As car drivers we obey the signals directed at us by traffic police, e.g. to turn or slow down. School pupils respond to the instructions directed at them by their teachers and as soldiers we obey the orders of our commanding officers. In some situations our obedience is taken for granted and in turn we seldom question our own obedience, e.g. responding to the instructions of health workers and carers. As with other aspects of human behaviour it has been the destructive, rather than the constructive, aspects of behaviour that has commanded the attention of both psychologists and non-psychologists alike.

Obedience research

It would be considered unethical to perform an experiment to investigate the obedience behaviour of soldiers during situations such as the Holocaust of World War II. We are left, therefore, with few alternatives other than to rely on the eyewitness evidence of people like Holocaust victims. The problem with this type of evidence is that very often these participants are psychologically and physically damaged by their experiences. Eitinger and Støm’s study (1973) of ex-concentration camp inmates revealed a higher than normal incidence of both physical and psychological illness. Similarly, interviews with those who perpetrated the violence are difficult, with interviewees being deliberately uncooperative or being very selective in what they wish to remember (Browning, 1992).

Milgram’s response to reading about incidents such as described above was to conduct a number of experiments to investigate destructive obedience. He wanted to investigate whether some individuals or cultures are simply more obedient than others, or whether everyone is obedient in certain situations. If so, what are the key elements of these situations?

Milgram’s experiment

His baseline experiment is described in article 1 of the key research summaries in chapter 6. (Read this now.)

Milgram’s original experiment demonstrated clearly, and shockingly, that ordinary individuals are obedient. He conducted a further eighteen experiments in which he varied some of the parameters in order to investigate further the situational factors which might have produced such a significant result. In all but one of the variants the participants were male, non-graduates between 20 and 50 years of age. In one other study, in which females were the participants, he found a similar level of obedience.

These were some of the key findings:

• The presence of another/others who are seen to disobey the experimenter reduces the level of obedience. In the case of two disobedient participants, obedience rates dropped to 10%. (Demonstrating the influence of group pressure: in this case positively.)

• The closeness of the teacher to the learner increased the likeliness of disobedience in so far as when the ‘learner’ was visible obedience rates dropped to 40%.

• Obedience was less likely if the experimenter was not in the room but issued orders over a phone link. When the experimenter was absent, teachers often administered shocks of a lower dosage than ordered.

• Re-locating the experiment from Yale University to a less prestigious office had some impact on the results in that obedience rates dropped from 65% to 48%.

• When participants were placed in the role of bystander to another’s obedience, 93% remained on task and did not intervene.

• A small reduction in obedience rates was observed when the participants and the experimenter agreed prior to the experiment to release participants if requested. The experimenter subsequently ignored this agreement.

A very interesting variant was where obedience to authority was demonstrated to have a beneficial effect. In a series of trials the learner demanded that the experiment continued despite indicating obvious pain on the pretext that he wanted to show that he could ‘take’ the punishment. When the experimenter ordered the teacher to stop, they obeyed instantly. This seems to show that people continue to obey authority for good or evil and further supports the major influence of an authority figure.

Milgram’s (1963, 1965, 1974) basic study and variations demonstrate that obedience is dependent on intricate situational factors. Participants who would harm another under the direct orders of an authority figure (the experimenter) would be able to disobey the experimenter if with a defiant co-participant, or if the experimenter had left the room.

The implications of Milgram’s baseline condition and many of the variations are best expressed in Milgram’s own words:

if an anonymous experimenter could successfully command adults to subdue a fifty-year-old man, and force on him painful electric shocks against his protests, one can only wonder what government, with its vastly greater authority and prestige, can command of its subjects.

(Milgram, 1965:75)

Progress exercise

AnsMwer the following either verbally or in written note form.

1 Briefly outlin Milgram’s baseline experiment.

2 Why were Milgram’s results so surprising?

3 List three situational factors Milgram changed in subsequent experiments.

4 In relation to 3, describe the results of these studies and what they seem to tell us about destructive obedience?

Criticisms of Milgram’s experiment

Ecological validity

Milgram’s studies have been questioned on the grounds of ecological validity. In research it is important that the claims you make for an experiment and the data generated actually examine what you say they examine. Thomas Kuhn (1962), a scientist, historian and philosopher, says that modern science experiments only show what modern scientists claim they show. In other words all experiments can be questioned on the grounds that they do not really test what the experimenter claims the experiment tests. In the case of Milgram, therefore, is his study an examination of destructive obedience such as was witnessed in real-world examples like Auschwitz, an example of Nazi obedience? There is not always a simple ‘yes it does’/‘no it doesn’t’ answer to this question. Many social psychologists have noted the different quality of human relationships in real life and in a laboratory. What we may be seeing in a social science laboratory experiment is only what happens in a laboratory and therefore it is difficult to generalise from the laboratory results to real life. In this case, the laboratory experiment is said to lack ecological validity. Others point out that limiting variables so that they can be manipulated and measured easily is not comparable with real-world situations such as German concentration or death camps. Briefly, anecdotal evidence (Wiesel, 1958; Levi, 1986) and evidence from social scientists (Cohen, 1954) indicate that Auschwitz was a complex society, as well as an infamous Nazi death camp. As a large collection of individuals, therefore, it developed its own unique set of social norms and values. Many would say that to try and emulate this or part of it, i.e. destructive obedience, in a laboratory setting would mean that many significant features of the ‘real’ situation are lost.

Internal validity

But studies can be invalid for other reasons, such as the way in which data was collected or the behaviour of the experimenter. This is called internal validity.

Rosenthal (1966), Orne and Holland (1968) and others have demonstrated the effects of demand characteristics. These are features of the experimental procedure or setting which bias results. An example of this would be an experimenter who, as a result of a poorly designed set of standard instructions, invited participants to behave in a particular way or who recruited participants from a sample which was not random.

In relation to the above, two aspects of Milgram’s procedure have come in for criticism:

• The fact that Milgram paid his ‘volunteers’ and recruited them via a newspaper advertisement is not considered to be random sampling.

• The experimenter contradicted his participants both verbally and non-verbally on a number of occasions, such as by telling participants that a shock of 345 volts would not cause tissue damage when the switch was clearly labelled as very dangerous.

On these alone, some psychologists would condemn Milgram’s results as internally invalid.

Ethics

These experiments are unethical to the point that they probably would not be sanctioned today. Bruno Bettelheim described them as ‘vile’ and went on to say that they were of little worth. Today’s APA and BPS guidelines, for example, state clearly that:

• At all times a participant’s full and informed consent must be obtained prior to the start of an experiment.

• The participant must be at liberty to withdraw at any time.

• The participant’s health and mental well-being must be safeguarded.

Milgram’s experimental procedures seem to have broken all three of the above guidelines. His own description of one of his participants 20 minutes into the experiment seems to contravene at least two:

he was reduced to a twitching stuttering wreck who was rapidly approaching a state of nervous collapse. He constantly pulled on his earlobes, and twisted his hands. At one point he pushed his fist into his forehead and muttered: ‘Oh God, let’s stop it.’

Counter-arguments

Milgram felt that his debriefing procedure, involving a friendly reconciliation with the victim (actor), was sufficiently thorough to ensure that each participant left the laboratory ‘in a state of well-being’. In a follow-up study Milgram, assisted by a psychiatrist, discovered that very few participants felt they were harmed by their experiences. Only 1.3% of participants felt that they were sorry or very sorry to have taken part in the experiment, whereas 83.7% were glad to have taken part (Milgram, 1974). Even so, modern psychologists would say that Milgram’s efforts to obtain naïve participants were unethical because the participants were not fully informed and every opportunity was taken to persuade the participant to continue when some were clearly apprehensive.

On the other side of the debate, social psychologists like Elms (1982) have described such research as ‘morally significant’. He feels the risks were worth taking in order to investigate a very contentious aspect of behaviour. Others would say that we could not turn our back on aspects of human behaviour just because they are distasteful and are difficult to investigate. And as Aronson (1995) observed: ‘the ethics of any experiment may be less open to question when the results tell us something pleasant or flattering about human nature, and more open to question when they tell us something we’d rather not know.’

Progress exercise

Answer the following either verbally or in written note form.

1 Summarise the ethical concerns that Milgram’s study brought into focus.

2 Summarise some of the points listed above which were criticisms of Milgram’s procedure.

Other obedience studies

Cross-cultural studies

Since Milgram’s original study and despite the ethical issues it raised there have been many variations of his baseline experiment conducted in other countries and cultures including Austria, Australia, Britain, Germany, Italy, Jordan, the Netherlands and Spain. On inspection though, the majority have been performed in predominantly ‘Western’ societies. They have produced a range of results, e.g. from over 90% obedience in Spain and the Netherlands to a low of 16% in Australian females. It must be pointed out that not all the studies were exact replications and therefore cannot be seen as wholly confirming Milgram’s results. On the other hand, the various replications seem to generally reflect the range of scores Milgram obtained (Miller, 1986). I will examine two, a study of children by Shanab and Yahya (1977) and a part replication study conducted by Kilham and Mann (1974) in Australia.

Shanab and Yahya studied Jordanian children between the ages of 6 and 16 years. The experimenter was female. The results of the experiment showed that 73% of children administered the maximum shock to same-gender peers. This is an important variation since it seems to show that gender may well be a factor in how we behave in a situation where we are ‘expected’ to behave in a destructive way. This study represents an 11% increase in the levels of obedience Milgram obtained with adults. This seems to confirm the reality that children are very obedient. In the light of the number of children, past and present, who have played and are playing an active part in warfare, it is worrying. In recent history children have been implicated in war crimes, e.g. as fighters for the Khmer Rouge in Cambodia and in the recent civil war in Rwanda. However, it is difficult to generalise too much from this study since the cultures of Jordan, Rwanda and the West are very different.

The Australian study was a modified versi...