By Dan

Social psychology can be defined as “the scientific investigation of how the thoughts, feelings and behaviours of individuals are influenced by the actual, imagined or implied presence of others” (Gordon Allport, 1935).

unit 1: social psychology

Section A: Obedience
1.1 Obedience
1.2 Milgram’s Study of Obedience (1963)
1.3 Evaluation of Milgram’s Study of Obedience
1.4 Variations of the Milgram Experiment
1.5 Meeus and Raaijmakers (1986)
1.6 Agency Theory
1.7 Hofling et al. (1966)

Section B: Prejudice
1.8 Social Identity Theory as an Explanation of Prejudice
1.9 Tajfel et al. (1970, 1971)
1.10 Sherig et al. (1954)
1.11 Reicher and Haslam (2006)
1.12 Asch (1951, 1952, 1956)

Unit 1 Key Issues
Unit 1 Revision Notes
Unit 1 Questions

1.1 obedience

To obey someone means to follow direct orders from an individual more often than not in a position of authority. There are three types of obedience in general:

  1. compliance – following instructions without necessarily agreeing with them (an example of this might be wearing a school uniform – although you don’t want to, you comply with the rules and do anyway because it causes you no harm)
  2. conformity – adopting the attitudes and behaviours of others, even if they are against an individual’s own inclinations (an example of this might be the Nazis during the Holocaust, they were instructed to do what they did, and some of them may not have wanted to do it but conformed to the rules anyway)
  3. internalising – this is carrying out orders with agreement

The term destructive obedience refers to the idea of an individual following the orders which they consider to be immoral, which will cause them a lot of distress and regret. This often occurs with conformity.

Taking the example of the holocaust further, think of Adolf Eichmann. He was the officer probably most responsible for what happened during the Holocaust, and he always said that he only did what he did because he was carrying out orders. Whether or not it was true, this is an example of how obedience can work, and it was particularly frightening because it makes people wonder if they would do the same thing if it ever happened again and they were in his position. This thought is what has encouraged numerous psychologists to carry out studies into the nature of obedience, probably the most famous of which being Stanley Milgram, who was specifically curious about potential replications of the holocaust, because he wanted to test to see if the Germans in particular were different to other people, by testing obedience on other people.

1.2 milgram’s study of obedience (1963)

Aim: To investigate how far people will go in obeying an authority figure


1 Volunteers responded to an advertisement in a paper for an experiment at Yale which investigated the effects of punishment in learning, they were paid $4.50 for participating

2 Via a fixed lottery, the subjects were chosen to play the role of teacher (the confederate or accomplice), and an actor, posing as another volunteered participant became the learner

3 The learner was strapped into a chair and had electrodes attached to him, and the teacher was informed that the shocks would result in no permanent damage. To prove the equipment was working, the subject (teacher) received an initial 45 volt shock themselves

4 The teacher is taken next door to the shock generator room where the they are told to administer a shock to the learner of increasing severity for each incorrect answer he gives using a word game based on memory, over an intercom

5 The actor frequently gave wrong answers and would receive a shock for each one, each time the voltage would increase by 15 volts. After each shock, a recording of a painful scream was played back to the teacher over the intercom

6 After 300 volts there was silence from the learner – he was either unconscious or dead

7 The experiment came to an end when the teacher refused to continue or they reached the full voltage (450V)

8 After the experiment finished, the teacher was fully debriefed about the true nature of the experiment and was reintroduced to the learner, who had come to no harm

The Subject

Milgram chose 40 males between the ages of 20 and 50 with a wide range of jobs from the New Haven Area. The use of males prevented interference on the basis of  reluctance towards intersexual abuse

The Learner

The learner was a 47-year old American-Irish actor who acted as ‘Mr Wallace’ – a mild-mannered and likeable accountant. He was an average person

The Experimenter

The experimenter watched the teacher as he administered the shocks, and if the teacher hesitated because they found it uncomfortable, he would use one of his standardised prompts from “please continue” to “you must go on.” He was a 31-year old dressed in a grey lab coat to give the appearance of an important, authoritative figure. He would be impassive during the experiment. The experimenter would not force the teacher to continue, but would sternly encourage them to carry on


Levels of obedience expected
When psychology students and professional psychologists were asked what percentage of the people participating in the experiment would go right through and administer the highest voltage of shock (450 volts – lethal), the answers ranged from 1 to 3; the mean value was 1.2

Levels of obedience obtained
When the study was carried out:

  • 65% of participants continued to the maximum shock level of 450 volts
  • Not one participant stopped the experiment before 300 volts

According to Milgram himself, the degree of tension reached extremes for some subjects as some were “observed to sweat, tremble, stutter, bite their lips, groan and dig their fingers into their flesh.” What is interesting is how these quite clear signs of body language show that the study was making them uncomfortable, and even though they were under no obligation to continue (the experimenter wasn’t forcing them to continue), most subjects obeyed the experimenter throughout the entire 450 volts, simply because he appeared to be a figure of authority.

“One sign on tension was the regular occurrence of nervous laughing fits… Full-blown, uncontrollable seizures were observed for 3 subjects. On one occasion we observed a fit so violently convulsive that it was necessary to call a halt to the experiment. In post experimental interviews, subjects took pains to point out that they were not sadistic types, and that the laughter did not mean they enjoyed shocking the victim.” Milgram, 1963


Generalisability refers to the idea that the findings can be applied to the target population as a whole
Reliability refers to the idea that repeating the experiment would obtain similar or identical findings
Application refers to the idea that the findings can be useful in a real-life application in society
Validity refers to the idea that results should measure what they initially were supposed to measure
Ethics refers to the idea that an experiment should be carried out whilst taking into consideration ethical grounds

In terms of generalisability, the test subjects were all males within a specific age group. So the data obtained from the experiment cannot necessarily apply to a whole plethora of people. However, Milgram purposely chose not to use all college students, but instead wanted a range of men with varied jobs to get a good range of data. His experiment was reliable, because the experiment was repeated a number of times, and different variations of the studies went out. Milgram experimented changes in gender and nationality. Other psychologists (Sheridan and King, 1972) even tried altering the species, using animals as the learners (victims).

Can the findings from Milgram’s experiment be applied to society and be useful in everyday situations? The supposed experiment which the subjects believed they were originally signing up for would have been, experimenting on the effect of punishment on learning, in terms of memory and forgetfulness. However, what uses did the findings from the data have that are implemented today?

Milgram’s study was well standardised and obedience was accurately operationalised as the amount of voltage given – so the study was experimentally valid. However, two psychologists, Orne and Holland (1968) said that they believed the subjects knew that they were not causing the learners any harm. Because the experiment was an artificial test, and because the test subjects were aware that they were being studied, it was argued that the study lacked “mundane realism” and was therefore not ecologically valid. However, one might argue that because the subjects were not actually aware of what the real study was investigating, the nature of the subjects was more natural, as they were less suspecting that it was their part being investigated, even if the environment of the university was not a natural place.

You might also say that because the test subjects were completely unaware of the true nature of the experiment, it was not an ethical study. This may also be the case because the experience the subjects went through may have a negative effect on them post-investigation when they realised how they behaved.

1.3 evaluation of milgram’s study of obedience

The main measure of how reliable a psychological study is will more often than not be its replicability. Milgram used a standardised procedure for each participant – for example, the same script was used by the learner and experimenter; the same rooms were used during the experiment; and identical equipment was used each time. This ensured that all the participants had a similar experience, so there was no bias in the experiment. The strong controls meant that the studies could be repeated, to test whether the findings were reliable – and the experiment was, indeed, repeated by Milgram himself, among other psychologists, afterwards.

Real World Application

Milgram’s work was of practical value because it showed that individual’s have a tendency towards destructive obedience. He believed that, by showing this, his work had wider benefits to society as it could avoid such incidents in the future, as the one which triggered Milgram’s investigations – the Holocaust.

The study helps us to understand how historical events such as this could happen, where people obeyed orders against the moral code they normally lived by.


1 The participants had to complete an artificial task by asking the learner to remember word pairs and then administer an electric shock whenever they didn’t remember correctly. Many theories suggest that most participants felt protected from their actions because they assumed whatever happened at Yale was fine and so trusted the study. Thus, it could be argued the experiment lacked experimental validity

However, Milgram tried to ensure the participants thought the situation was real, for example, by giving them a 45 volt shock at the start. The obvious stress experienced by participants implies that most did believe that what was happening was real, so this would suggest that in fact there was some experimental validity in his method

2 The study took place in a laboratory in Yale University, a very well-respected university with an extremely popular reputation. This is an unnatural setting for most people, which suggests that normal behaviour wouldn’t necessarily be usual. This means that the experiment lacked ecological validity

3 As Milgram’s sample of participants consisted of adult males from a range of backgrounds, it could be said that the experiment had some population validity, but only for American male adults

However, Milgram later repeated the study in a large number of variations (see 1.4 Variations of the Milgram Experiment), and many other psychologists have repeated the experiment. What was noticed is how the results tended to produce similar patterns (the number of participants who continued to the full 450V shock when it was all women in the experiment was almost the same as with the original men’s experiment), and so you might say it did in fact have definite population validity


The biggest criticism of Milgram’s study has always been on ethical grounds. There are 5 important guidelines to consider: informed consent, deceit, right to withdraw, debriefing and competence. Here, you will see in-depth analysis of each of these guidelines.

Informed Consent – In the study, the participants were not given the full details on the true nature of the experiment, so it initially sounds as though the experimenters did not gain correct informed consent, but you have to consider that had the participants been aware that the electric shocks were not real, the results gathered would not have been a clear indication of their obedience and behaviour because they would have known that the
consequences of their actions were not real. Milgram therefore could not ask for informed consent but did try to be ethical so asked participants if they would like to take part in such a study and they did – this is presumptive consent. Another way of remaining ethical is to ask the participants before the study if they agree to take part, but inform them that sometimes deception is necessary – this is prior consent

Deception – There was a severe amount of deception in Milgram’s experiment, but (as before) this was all necessary for the results of the experiment to be valid. Examples of the deception used include faking the shocks, leading participants to believe they were given the teacher role by chance, telling them it was for a study on memory and forgetfulness, telling them the learner and experimenter were real and not actors, and many more

Right to Withdraw – There is a lot of controversy over ethics regarding the right to withdraw. Whilst the participants were free to leave and were not being forced to continue, they were strongly encouraged to carry on by the experimenter, and the experimenter even had a script with lines to tell the teacher such as “the experiment requires that you continue” which almost made the subject feel they had to go on. When the participants said that they wanted to stop, they were strongly urged to continue, thus it might be argued they did not have a true right to withdraw, making the study unethical

Debriefing – Because the experiment was very stressful for the participants and it involved a lot deception, the debriefing process was essential. Additionally, the participants would have come to realise that had the fake “memory improving” experiment been real, they would have administered lethal shocks to random strangers, showing them they had the capability to commit murder. Therefore it is important for them and the experimenter to fully evaluate the experiment to ensure they are in a safe mental state before going home

Competence – Milgram knew the possible implications of the study; understood the ethical guidelines, did not feel the need to get advice from others; was suitably qualified as a scientist who had his PhD for three years; made sure that nobody would come to any immediate harm as a result of the experiment; adhered to the Data Protection Act and easily and correctly stored the data. However, the participants became distressed, making the experiment less ethical as a whole, but the fact that Milgram was competent to run the experiment and knew what he was doing means it wasn’t necessarily unethical as a whole

1.4 variations of the milgram experiment

In Milgram’s book, Obedience to Authority: An Experimental View (1974), he outlines 19 different variations of the original study of obedience, some of which were previously unreported. Each of the variations had one thing in common; they all led to a reduction in obedience. Some of the variations are listed below:

Evaluating the Variations

One of the strengths of the variations is its strong controls. This means that the studies are replicable and so reliability can be tested. Having strong controls means that there is a lack of bias, which allows you to draw more accurate conclusions about cause and effect.

Of course, the most important weakness to consider, which is similar to the original experiment, is how unethical the variations were. Again, there was a lot of deception involved in each experiment, and there is always a certain risk when dealing with subjects in such a way that could cause them distress, as finding out what the true nature of the experiment is might cause them.

Also, the experimental validity and ecological validity (and the population validity) are all questionable. The results can not necessarily be applied to the population as a whole, because throughout, it was essentially all people from the same categories used as subjects (20 – 40 year old men); although in one variation of the experiment all women were used instead of men. The results of that experiment were not significantly different from the original study, although women seemed to communicate higher experiences of stress than the men did.

1.5 meeus and raaijmakers (1986)

Aim: To investigate destructive obedience in the everyday situation of a job interview

Wim Meeus and Quinten Raaijmakers wanted to replicate Milgram’s original study but wanted to improve on two initial problems they saw within the study had they repeated it in exactly the same way:

  • Milgram’s participants were assured that there would be no permanent damage to the “learners”
  • The form of punishment would have been ‘old-fashioned’ according to Meeus and Raaijmakers


The aim of their experiment was to assess how the participants would handle destructive obedience in the everyday situation of a job interview, specifically, to see to what extent people would obey orders to psychologically abuse a job interviewee.

1 There were three people involved: a university researcher, a “job applicant” (who was an actor, similar to the role of the learner in the Milgram experiment), and a participant, who would issue the abuse

2 The applicant was following a script, and had to pass a test of 32 oral multiple-choice questions to “get the job”

3 The participants were told the job required ability to handle stress, so they had to cause the applicant stress during the interview by psychologically abusing them. This was essential because it gave the study motive to get the participant involved, had they not been told this information they would have been curious as to why they were being asked to verbally abuse the interviewee

4 Participants were also informed that it was part of an investigation to find out the relationship between psychological stress and test success, and the applicant didn’t know about the research (of course, none of this was true)

5 After the interview had begun, the participant would have to make a series of 15 negative comments about the interviewee’s performance and personality during the test, every time they would get an answer wrong

6 The applicant would occasionally interrupt the interview to object about the participant’s negative comments, but the participant was instructed to ignore these objections and continue with the interview

7 The “stress remarks” led the applicant to failing the test and so did not get the job

The Subject

Altogether, 39 participants were used, which consisted of both males and females between the ages of 18 and 55. These were split into two groups: a control group of 15 and an experimental group of 24. Each of the applicants had at least Dutch High School Education (equivalent education of GCSEs) and were recruited by a newspaper ad, offering $13 for the 30 minute experiment

The Stress Remarks

A TV monitor instructed participants when to make each stress remark about the job interviewee. The applicant had electrodes stuck on his skull to make the participant believe his stress levels were being measured. If the participant refused to continue making the comments, the experimenter had 4 prods to say to the participant. The questions for the interview were asked in four sets:

  1. Set A: no stress remarks made by the participant, to provide baseline measures
  2. Set B and C: five stress remarks were made for each of these sets
  3. There were ten errors in the last three sets

Participants in the control group chose when to make the comments and when to stop the experiment

The remarks started at question 9 (“your answer is wrong”) and went up to the end (“I think you’d be better suited for a lower function job”)

The study was conducted at a university, similar to Milgram’s study, and the experimenter was a well-dressed man of about 30 years who was friendly but stern, and who was present in the room with the participant, but the applicant was in another room and had to listen to questions via a speaker.

Results of Experiment

22 of the 24 participants in the experimental group obeyed until the end, making all 15 stress remarks. Most had some discussion with the experimenter, but continued when ordered to do so. Follow-up questionnaires showed that the participants did not like the experiment and were upset by the procedure

  • 73% of the total participants believed the experiment to be real, and only 4% thought it was a hoax, the remaining 23% were not sure
  • In terms of who the participants blamed for the applicant not getting the job, 45% blamed the experimenter; 33% blamed themselves, and the other 22% blamed the applicant
  • Within the control group, who could choose when to make comments and when to stop, none of the participants made the stress comments

As with the Milgram experiment, Meeus and Raaijmakers decided to conduct other variations of the original experiment to see how the results were affected by altering some of the variables:

Variation 1 – the experimenter ordered the participants to make the stress remarks and then left the room for the experiment (22 participants used)

Variation 2 – two other actor participants (confederates) present, who rebelled – the first quit after the tenth stress remark, and the second then did the same, so the experimenter ordered the real participant to continue alone (19 participants used)

The table below shows a comparison of the results from the three experiments:


These were the three main conclusions reached by Meeus and Raaijmakers which tried to explain why they achieved higher levels of obedience than Milgram did with his study:

1 physical violence has more of an immediate impact than psychological harm
the participants could hear the cries of the learner in Milgram’s study, but the real impact of psychological abuse only tends to become evident later (i.e. after they became upset and did not get the job)

2 consent levels were different
the participants’ consent to take part in the experiment carried more weight as they knew they were going to harm the applicant verbally and had agreed to participate; in Milgram’s study, the participants had not explicitly agreed to administer physical harm to the learners

3 the victim was more dependent on the outcome
in Meeus’ and Raaijmakers’ study, the applicant had to continue with the test to get the job, even if they objected to the stress remarks, whilst the learner in Milgram’s study could refuse to answer as there was no gain from continuing

Evaluation of the Dutch Study of Obedience

The main strengths of the of the Meeus and Raaijmakers experiment were:

  • The study builds on Milgram’s study by focusing deliberately on two areas that Meeus and Raaijmakers saw as needing attention. They used similar variations to Milgram to see if the levels of obedience fluctuated in the same way. Their study, therefore, is all the more useful because the findings can be compared with those of Milgram
  • Due to the attention to detail, the study is replicable and can be tested for reliability. There are controls, which mean that the details are clear and the study can be judged carefully. A study with good controls is easier to draw cause-and-effect conclusions from

Some of the weaknesses of the study are shown below:

  • The study is an experiment, and is therefore artificial. The need for controls, such as an applicant taking a test in a laboratory, means that the findings may not be valid. The situation is not very realistic and this might have affected the results
  • Although the findings were compared with Milgram’s findings, which is useful, there are differences between the two studies which make such comparisons difficult. One difference is that the studies were in different cultures (even though they are both western cultures); another is that the studies were twenty years apart, which could have affected obedience levels

The table below shows a comparison of the results between the main Milgram and Meeus and Raaijmakers studies to make these comparisons evident:

1.6 agency theory

In Milgram’s studies of obedience, participants who obeyed to the end tended to say that they were only doing what they did because they were being ordered to do so by a member of authority and would not have done it otherwise. They said that they knew what they were doing was wrong. The participants felt moral strain, in that they were aware that following the order was immoral, but they felt unable to disobey. Moral strain arises when people become uncomfortable with their behaviour, because they feel that it is wrong and goes against their better values.

In the Milgram study, all the participants obeyed until the shock level reached 300 volts. It was as if, having simply agreed to take part, they were in an agentic state. This meant that they were the agents of the experimenter and so obeyed his orders. Being in an agentic state is the opposite of autonomy. Being in an autonomous state is being under one’s own control and having the power to make one’s own decisions.

Milgram used the idea of being in an agentic state to put forward his agency theory.  This is the idea that our social system leads to obedience. If people see themselves as individuals, they will respond as individuals in an autonomous state in a situation.

For example, in a threatening situation, many people avoid aggression and turn away. This is likely to happen because avoiding aggression avoids being hurt and will lead to survival. Evolution theory suggests that avoiding aggression leads to survival. Early humans had a better chance of survival if they lived in social groups, with leaders and followers. A tendency to have leaders and followers may also have been passed on genetically. A hierarchial social system, such as the one Milgram’s participants were used to, requires a system in which some people act as agents for those above them. According to the agency theory, the agentic state is what led to the participants to obey in Milgram’s study.

Milgram suggested that not only was this system of obedience present as a survival strategy, but also because we are taught that it is the correct way from a young age. Obedience is hammered into children by their parents, and also there are very strict hierarchial systems in place in schools – it is clear who has the power, and so children learn exactly the same lessons there.

In the agentic state, people do not feel responsible for their actions. They feel that they have no power so they might as well act against their own moral code, as happened in Milgram’s basic study. In the variation in which the victim was nearer to the teacher, and the teacher had hold the victim’s hand to the shock plate, there was less obedience. This suggests that the participants had to take greater responsibility for what they were doing.

Evaluation of the Agency Theory of Obedience

These are some of the strengths of the agency theory:

  • The agency theory explains the different levels of obedience found in the variations to the basic study by explaining the relationship between the level of responsibility felt by the participant and the levels of obedience obtained
  • The theory helps (or tries to at least) explain the issue that triggered Milgram’s research into obedience, the holocaust. Probably the main officer responsible for the holocaust was Eichmann, who said he was merely obeying orders, and agency theory suggests why he, and so many others, would obey to such a degree
  • The theory offers similar explanations to events such as the My Lai massacre

However, one of the weaknesses of the theory is that there are other possible explanations for obedience, such as social power. French and Raven (1959) proposed five different kinds of power:

Legitimate power is held by those in certain roles, usually those of authority; Milgram’s role would have had legitimate power

Reward power is held by those with certain resources; Milgram may have had reward power as he way paying the participants

Coercive power is held by those who can punish another; Milgram gave the participants a small shock, so he may have felt he could punish them

Expert power is held by those with knowledge; the participants would have seen Milgram as someone with knowledge

Referent power is held by those who are able to win people over; the participants would not have seen Milgram to hold this type of power

Also, one of the biggest criticisms of Milgram’s agency theory is that it is just a description and not an explanation. Many people view the theory as more of a description of how society works than an explanation. It suggests that the participants obeyed because they were agents of authority. However, obedience is defined as obeying authority figures, so a theory explaining obedience should offer more detail into why it is that people follow orders against their better judgement under given situations.

1.7 hofling et al. (1966)

Aim: To investigate the levels of obedience shown by nurses to doctors in hospitals

Hofling et al. (1966) decided to investigate the reactions of nurses to orders from a person who they believed to be a doctor. They decided to test how far they would be willing to obey the doctor in unusual and unethical practices. The study took place in a hospital, and so was a field study.


Hofling et al. wanted to study the doctor-nurse relationship. They wanted to specifically look at health care, and many of the involved researchers were medical personnel. In particular, they were interested to see how nurses would respond to a doctor giving them orders which went against their usual professional standards, as this was an occupational issue

The Orders

To make the orders contrary to the nurses professional standards, some of the doctor’s requests were:

  • asking the nurse to give an excessive dosage of medicine (would actually be a placebo)
  • transmit the order over the phone (against hospital policy)
  • use an unauthorised drug (either one not on the ward stock list or one not yet cleared for use)
  • have the order given to the nurse by an unfamiliar voice


The situation for the main study involved 12 wards in public hospitals and 10 wards in private hospitals. Questionnaires were distributed to graduate nurses at a separate hospital in order for usage as a matched control. The questions in the questionnaire asked the nurses what they would do in the situations the nurses experienced in the real study, to see what ordinary nurses believed they would do. The same questionnaire was also given to some student nurses to see how less-experienced nurses would respond to the same situations on paper

Procedure of the Main Study

Pill boxes were central props in the study, each labelled “Astroten, 5mg capsules. Usual dose, 5mg. Maximum daily dose, 10mg.” The boxes contained placebo capsules and were placed on the wards. The doctor would give the nurse the orders via phone, and this would follow a script. Standard responses to potential questions were prepared. The caller, a supposed doctor the nurses had not heard of before, was always courteous yet self-confident. Researchers would always monitor the phone calls to check the tone was appropriate

There was an observer on each ward, who would stop the experiment:

  • if the nurse had the medication ready and moved towards the patient’s bed
  • the nurse refused and ended the conversation
  • the nurse began to contact another professional person

The observer would then interview the nurse to obtain more information, and also offered “psychiatric first aid”

The experiment was run on medical, surgical, paediatric and psychiatric wards from 7pm to 9pm, when administration of medication does not normally happen, and doctors are not normally present, so the nurses would have to make their own decisions

The Phone Call

Circumstances to end the phone call:

  • participant complies
  • participant refuses
  • participant insists on referring to someone else
  • participant becomes upset
  • participant is unable to find the medication
  • the call lasts longer than ten minutes

The Interview

After the incident, a nurse-investigator would follow up within half an hour and request a follow-up interview. The interviews were unstructured (but the nurse-investigator would have had the tape recording of the call, as well as the observer’s report). Information asked for was:

1 Unguided narrative (what happened…?)
Emotions (what are your feelings…?)
Discrepancies (are you sure it happened that way…?)
Any similar incidents (has this happened before…?)
Retrospective view (what do you feel about it now…?)
Biographical data (what is your age, religion, etc…?)


Questionnaires were sent to graduate and student nurses. The participants were closely matched for age, sex, race, area of origin, marital status and experience at work. Twelve graduate nurses were given the questionnaires with a doctor explaining the whole imaginary scenario to them. The nurses were not only expected to answer what they would do, but also what they predicted the majority of other nurses would do in the same situation. The same questionnaires were handed out to 21-degree programme nursing students

An example of the question might have been: “You are the only nurse on the ward. Now will you please give Mr Jones a stat dose of 20mg – that’s four capsules – of Astroten? I will be up within ten minutes and I will sign the order for them then. Write down what do you do?”

Results of the Main Study and Questionnaire Research

The researchers drew the following conclusions:

  1. None of those asked thought that nearly all the nurses would obey in the experiment. However, the obedience showed the strength of the doctor-nurse relationship, and how a patient can suffer as a consequence. The researchers say that instead of two “intelligences” – the doctor and the nurse – working for the patient, one of them seems to be non-functioning
  2. The nurses were affected by the study: they were upset that they had been observed without their permission and also that their specific behaviour had been noted
  3. Nurses think that they will defend their patients and are proud of being professionals. However, the reality seems to be different (the evidence of this is the discrepancy)
  4. The nurses appeared to trust the doctors, which may be a valuable trait. They were willing to act promptly and efficiently, again a valuable trait. However, this study suggests nurses need to be encouraged to use their own intellectual and ethical resources

The researchers behind the experiment concluded that there was definite potential for nurses to be encouraged to question and think more clearly about orders, especially in these types of circumstance, without being disloyal or discourteous to doctors.


The experiment took place in a hospital, where nurses would not feel out-of-place. Also, they were unaware that they were being observed by researchers, therefore normal behaviour would have occurred. This gave the experiment ecological validity. Nurses were going about their usual work (psychologists soon discovered that these “stranger doctor” phone calls were not an unusual experience for the nurses) and because it wasn’t strictly unusual for something against the rules to happen, the experiment was very realistic, and certainly true to life: therefore having experimental validity.

The study was replicable, i.e. could be repeated many times to find similar or identical results. It was replicable because of such strong controls on the experiment. Examples of these controls include the phone call following the same script, the type of drug and how much to be “prescribed”, the voice and tone of the caller and the place to put the fake pill boxes – all kept the same throughout. Replicability is a good test for reliability, therefore the study is reliable.

However, there are numerous faults with the experiment in terms of ethical issues. The main issue is that the nurses were being observed and their actions were being noted without their permission. This upset the vast majority of the nurses, and even angered a few of them, as they felt themselves it was very unethical. On the other hand, the counterpoint of this argument is that this withholding of information was necessary to maintain experimental validity. Another ethical issue breached by the experiment, tying into the lack of information to the nurses, is the lack of informed consent. This also meant that they had no specific right to withdraw from the study.

Extraneous variables (those other than the ones you’re testing) could have also intervened with the data. For example, the study could have actually produced results for a different reason, i.e. as the study was done in 1966 when it was practically all male doctors and female nurses, it could have simply produced results identifying the female-obeying-male relationship, rather than the nurse-obeying-doctor relationship. The experiment could also be said to be ethnocentric in that it was only tested in one area, so you cannot guarantee the results would be identical if the same study was carried out elsewhere. The experiment may therefore lack population validity (generalisability).

1.8 social identity theory as an explanation of prejudice

The word prejudice derives from ‘pre’ meaning ‘before’ and ‘judice’ meaning ‘justice’. The idea of prejudice refers to the judgements made by other people based on their membership to a particular group, rather than their individual nature. Discrimination refers to treating others differently according to their group membership due to prejudice.

Prejudice consists of three elements:

Social Identity Theory

Social identity theory is one of a number of theories that suggest prejudice can be explained by our tendency to see ourselves as part of a group. We therefore view others as either part or not part of the same group as us. Thus people are judged as being “us” and “them”. It is seen as part of human nature to view oneself as part of one or more groups, there are our in-groups – this leads us to discriminate against out-groups for no logical reason, i.e. there does not have to be any conflict or competition for ill feelings to develop.

Tajfel et al. (1970, 1971) conducted a series of lab experiments called the minimal group studies which led Tajfel and Turner (1979) to propose that there are three cognitive processes in deciding whether someone is part of the in-group or out-group, leading to the development of prejudice:

  • Social categorisation – the process of deciding which group you belong to: you see yourself as part of that group, where any group will do and you see no need for conflict between yours and other groups
  • Social identification – identifying yourself with the in-group more overtly, this is when you begin to take on the norms and attitudes of other group members within of the group
  • Social comparison – one’s self-concept becomes wrapped up with the in-group that self-esteem is enhanced by the perception that the in-group is better than the out-group

For more information on Tajfel’s minimal group studies, see 1.9 Tajfel et al. (1970, 1971)

According to social identity theory, there are three variables contributing to in-group favouritism:

1 the extent to which individuals identify with the in-group
the extent to which there are grounds for making comparison with the out-group
the relevance of the comparison group in relation to the in-group

The ideas of in-group favouritism and out-group prejudices have been confirmed in a number of studies…

  • Tajfel et al. (1970, 1971) conducted the minimal group studies in which boys of the ages of 14 and 15 were split into groups and had the chance to reward each other by giving them money, or punish them by taking money away from them, even though they didn’t win or lose anything themselves in making the decision, in-group favouritism soon became apparent as the boys gave more to their own group members and punished others
  • Lalonde (1992) studied a hockey team with poor performance and asked them about it, and the players claimed that it was down to other teams using “dirtier” tactics – however, Lalonde observed several of the team’s matches and concluded that the opponents’ teams were not using “dirtier” tactics, and so he had come across in-group bias from the poor team
  • Reicher and Haslam (2006) conducted their own variation and improvement on the famous Stanford prison experiment in which the prisoners had a chance to be promoted to guards, and guards were the superior figures in the study – the guards showed a lot more closeness and definitely had in-group favouritism

Evaluation of Social Identity Theory as an Explanation of Prejudice

A range of studies have shown support of the idea that people are willing to see their group as better in some way than other groups (as shown in the above examples). Tajfel, for example, replicated his experiment with a variation to prove that his findings were reliable. There is also a practical application, in that the theory helps to explain a wide range of social phenomena.

Social identity theory doesn’t take into account other factors which might be influencing behaviour, for example Dobbs and Crano (2001) have shown that under some circumstances there is much less in-group favouritism than suggested by Tajfel. The theory also doesn’t explain why there are individual differences in the level of prejudices shown. There are also other possible explanations of prejudice which might offer a fuller account of prejudice, for example the realistic conflict theory which sees social identity theory as only part of the explanation. It suggests that it is not just the creation of two groups that leads to prejudice, but that they need to have a goal in sight for conflict/prejudice to develop.

1.9 tajfel et al. (1970, 1971)

Aim: To test the idea that prejudice and discrimination can occur even without group history

Tajfel carried out a number of studies to develop and test social identity theory. Tajfel et al. wanted to test the idea that prejudice and discrimination can occur between groups even if there is no history between them, and no competition. Having found prejudice between such minimal groups, Tajfel et al. wanted to investigate further into the possible causes.

Experiment 1: Estimating Numbers of Dots

For the first of two experiments, 64 boys aged 14 and 15 were used. They were all from a comprehensive school in Bristol. They all knew each other very well and were split up into eight groups of eight boys each. The experiment was run in a laboratory. The experiment was designed to establish in-group categorisation (formation of the groups) and to assess the effect on behaviour of the group formations. To form the two groups, the boys were taken into a lecture room where forty clusters of varying numbers of dots were flashed onto a screen. They were asked to write down how many dots they thought there were each time on a score sheet. After they had estimated the number of dots:

  • in condition 1, they were told that people constantly overestimate or underestimate the number
  • in condition 2, they were told that some people are more accurate than others

Their judgements were then scored by one of the experimenters, and they were then randomly split into groups. They were told, in condition 1, that one group was the overestimators, and the other the underestimators; and in condition 2, they were told that one group was the better group at making judgements, and the other group worse.

The boys were told that the task used real money for rewards and punishments. They would know the code number of each boy and which group they were in, and would have to decide whether or not to allocate money to the other boys. They had to choose how much to reward or punish another boy in either their own group or the other group.

The experimenters showed the boys the type of matrix they would be using (similar to the above example), each one with 2 rows of 14 numbers. Those which were positive figures would represent amounts potentially rewarded to the boys; the negative numbers would be the amounts to be taken away from them. The boys could not allocate money to themselves, and had to work through a booklet of matrices.

The experimenter would call out “These are the rewards and punishments for member XX of your group” or “These are the rewards and punishments for member XX of the other group”. They had to decide which pair of numbers to allocate to the boys, because one number from each pair would affect one boy and the other affecting another.

The boys had to make decisions about the rewards and punishments they would impose. They had three types of decision: ‘in-group/in-group’, ‘in-group/out-group’ or ‘out-group/out-group’. If the boys allocated as much as possible to one boy, they were given a score of 14 (because there were 14 decisions for each row on each matrix). If they allocated as little as possible, the score was 1. For reach decision they were allocating to two boys. Therefore, a fair score would be 7 because this would mean that they had allocated rewards (or punishments) equally.

When decisions involved two boys, one from each group (an in-group/out-group decision), the average score was 9 out of 14. When boys were making in-group/in-group or out-group/out-group decisions, the average score was 7.5

It seemed that decisions about boys in the same groups were fairer than decisions when one boy was in the same group as the boy making the judgements and one boy was in the other group. A large majority gave more money to their own groups and showed in-group favouritism. This was found in all trials of this study.

Experiment 2: Klee and Kandinsky Preferences

This second experiment involved three new groups of 16 boys per group. The boys were shown twelve slides, showing paintings by foreign artists Klee and Kandinsky, six of each artist. The boys had to express a preference for one of the painters. The paintings were not signed, so that, in actual fact, the boys could be randomly assigned the groups, as again they had nothing to do with their choices, even though they were led to believe this was not the case.

The first experiment showed that forming groups led to in-group favouritism. The experimenters wanted to investigate this further by examining the factors leading to the boys making their decisions. They chose to investigate:

  • maximum joint profit – what was the most the two boys represented by each matrix would ‘receive’ from the boys?
  • maximum in-group profit – what was the most the boys would give to their in-group members?
  • maximum difference – what was the most difference between an in-group and out-group member benefiting the in-group members?

As in the first experiment, there were the same three conditions when making the choices. There were matrices as before, and again a choice was made of one pair of ‘rewards and punishments’. The experimenters could see if the boy had chosen the highest possible for his own group member, the lowest possible for a member of the other group, or a decision that was the lowest for both (or other similar patterns).

Maximum joint profit did not seem to guide the boys’ choices. Maximum in-group profit and maximum difference in favour of the in-group worked against maximum joint profit. If the boys had a choice between maximum joint profit for all and maximum profit for their in-group, they acted on behalf of their own group. Even if giving more to the other group did not mean giving less to their own group, they still gave more to their own.


  • Out-group discrimination was found and is easily triggered
  • There is no need for groups to be in intense competition, this goes against the realistic conflict theory
  • In the two experiments, all the boys needed was to see themselves as in an in-group and out-group situation, and discrimination ensued
  • People acted according to the social norms that they had learnt, such as favouring the in-group
  • The boys responded to the social norms of “groupness” and fairness and in general kept a balance between the two
  • In real life “groupness” may override fairness, for example, if the group is more important than counting dots, or choosing a preference between Klee and Kandinsky
  • Given the side effects of discrimination that were found in these experiments, teams in schools may not be a good idea

1.10 sherif et al. (1954)

Aim: To study the origin of prejudice arising from the formation of social groups

Sherif carried out research into groups, leadership and the effect groups had on attitudes and behaviour. The Robbers Cave Study built upon his previous work. He thought that social behaviour could not be studied properly by looking at individuals in isolation. He recognised how social organisation differs between cultures and affects group practices, so he claimed that groups have to be understood as part of a social structure. The Robbers Cave Study used two groups of young boys to find: how the groups developed; if and how conflict between the groups arose; and how to reduce any such friction. Three terms defined according to Sherif are:

small group – individuals share a common goal that fosters interaction; individuals are affected differently by being in a group; an in-group develops with its own hierarchy and a set of norms is standardised
– a product of group interaction that regulates member behaviour in terms of expected or ideal behaviour
– a social unit with a number of individuals who are interdependent and have a set of norms and values for self regulation; individuals have roles within the unit


22 young boys, aged 11, who did not know each other prior to the study. All from Protestant Oklahoma families to eliminate family problems and match the kids as much as possible. They were also matched based on a rating, including their IQ, from their teachers and were finally reassessed and matched , including issues such as sporting ability, before the experiment began. A nominal fee was charged for the children to attend the camp and they were not informed that they were being used for a piece of research in order to obtain “true” results

The Camp

The experiment is called the Robbers Cave Study because it took place in a camp at Robbers Cave State Park, Oklahoma. The location was a 200-acre Boy Scouts of America camp completely surrounded by the State Park. The site was isolated and keeping the two groups apart (at first) was easy because of the layout of the site, as shown in the diagram

Data Collection

There was a wide range of data collection methods:

  • observer – participant observer allocated to each group for 12 hours a day
  • sociometric analysis – issues such as friendship patterns were noted and studied
  • experiment – boys had to collect beans and estimate how many each boy had collected
  • tape recordings – words and phrases used to describe their own group were studied

The observers were trained not to influence the boys’ decisions but to help them once a decision was reached

Three Stage Experiment

  1. The two groups were formed and set up norms and hierarchies (to see how they became in-groups)
  2. The two groups were introduced and competition was set up, as a tournament (to test for friction, name-calling and hostility to the out-group)
  3. The two groups were set goals that they needed each other to achieve

Stage 1: in-group formation
The two groups were kept apart for one week to help the formation of group norms and relations. They had to work as a group to achieve common goals that required cooperation. Data was gathered by observation, including rating of emerging relationships, sociometric measures and experimental judgements. Status positions and roles in the groups were studied. There is much detail about how hierarchies within each group developed. The measurements were thought to be both valid and reliable because different data collection methods produced similar results. For example, in the bean-collecting task, the boys tended to overestimate the number of beans their own group members had collected and underestimate the number collected by the other group (the number of beans was actually the same).

Stage 2: inter-group relations, the friction phase
After the first week, the two groups were told about one another and a tournament was set up with competitive activities. Points could be earned for the group and there were rewards. As soon as they heard about each other, the two groups became hostile. They wanted to play each other at baseball, so they effectively set up their own tournament, which was what the researchers wanted.

The aim of the experiment was to make one group frustrated because of the other group, to see if negative attitudes developed. Adjectives and phrases were recorded to see if they were derogatory and behaviour was observed as previously. The researchers introduced the collecting the beans experiment: the boys had to collect beans and then judge how many each boy had collected. This was to see if the boys overestimated the abilities of the in-group members and minimised the abilities of the out-group members. As was mentioned before, this was the case.

Stage 3: inter-group relations, the integration
The researchers wanted to achieve harmony between the two groups, which they did by introducing superordinate goals. This meant that the groups would have to work together to achieve the goals. At first, they introduced tasks that simply brought the two groups together so that they could communicate. They then introduced the superordinate goals, which included:

  • fixing the water tank and pump when the water supply was threatened
  • a truck that would not start, so they had to pull together to try and start it
  • pooling resources so that they could afford a film that they all wanted to watch

The researchers measured the use of derogatory terms and used observation and rating of stereotyping.


Stage 1: in-group formation
By the end of the first Stage, the boys had given themselves names: the Rattlers and the Eagles. The groups developed similarly, but this was expected due to how carefully they had been matched. Any differences present were most likely due to the different decisions they had to make based on their cabins being located in different areas. For both groups, status positions were settled over days five and six of the first week, and a clear group leader was in place.

The Rattlers often discussed the situation of the Eagles, saying things such as “They had better not be swimming in our swimming hole”. Although the Eagles did not refer to the Rattlers so often, they wanted to play a competition game with them. It seems that even only knowing another group existed was enough reason for hostility to develop, even though neither group had been introduced yet.

Stage 2: inter-group relations, the friction phase
As soon as the groups found out about each other, they wanted to play baseball in a group competition: and so both groups had naturally moved onto Stage 2. The Rattlers were excited, and discussed such issues such as protecting their flag. The Eagles weren’t as excited, but made such comments as “we will beat them”. The Eagle selected as baseball captain for the baseball competition became the group leader of the Eagles for all of Stage 2, even though he was not the group leader at the end of Stage 1.

When the two groups first met, there was a lot of name calling. There is evidence collected, including what the boys said, who they were friends with and practical issues (such as the burning of a flag). It was found that there were clearly negative attitudes towards the out-group members.

Stage 3: inter-group relations, the integration
During the initial contacts of this Stage, the hostility remained. There were comments such as “ladies first” and when they watched a group movie together, they sat separated in their individual groups. After seven contact activities, there were superordinate goals set up:

1 The staff turned off the valve to the water pump and placed two large boulders over it. The children were informed that vandals had damaged it in the past. They worked together to fix the damage and rejoiced in common when they were successful

2 The second goal was to watch a movie together, but both groups had to chip in to pay for it. They eventually agreed to go halves even though one group had fewer members than the others. However, this agreement showed that the two groups cooperated to arrive at one final decision which they both were happy with

3 The boys all went on an organised trip to Cedar Lake, where the truck suddenly ‘developed’ a problem meaning the boys had to use the tug-of-war rope to try and pull it out and get it started

It was noticeable how friendships differed between Stage 2 and 3. More out-group members were chosen as friends by the end of Stage 3, which is evidence that friction was reduced by the superordinate goals outlined.


Most of the hypotheses put forward by the researchers at the beginning of the study were confirmed. Some of the conclusions drawn from the experiment include:

  • The groups developed social hierarchies and group norms, even though they were not stable throughout the study
  • Each group had a clear leadership structure by the end of the first week
  • When the two groups meet for competition, in-group solidarity and cooperation increases and inter-group hostility is strong
  • People tend to overestimate the abilities of their own group members and to minimise the abilities of out-group members
  • Contact between two groups is not enough to reduce hostility
  • When groups needed to work together, exchanged tools, shared responsibilities and agreed how to solve problems, friction was reduced – working towards a superordinate goal once was not sufficient, there needed to be numerous cooperation tasks to achieve this



  • There were controls, such as the careful sampling and the briefing observers so they all followed the same procedures, this meant that cause-and-effect conclusions could be drawn more justifiably than when observing naturally-occurring groups
  • There were several data collection methods and the findings agreed, so validity was claimed – for example, derogatory behaviour and recordings found derogatory remarks against the out-group
  • The group conflict could be seen as prejudice; reduction of friction would be reducing the prejudice, therefore the study has a practical application


  • It was unethical in the sense that there was no informed consent obtained, there was no right to withdraw for the participants (also, the boys’ parents were not allowed to visit – to prevent them feeling homesick – but this meant they could not check on their children’s welfare)
  • It was hard to generalise to other situations because the sample was restricted to boys with a specific background

1.11 reicher and haslam (2006)

Aim: To investigate tyranny at a group level

In 1973, Zimbardo carried out the famous Stanford Prison experiment where one group of people acted as guards and others as prisoners, all of which were participants. The study looked at the psychological effects of becoming a prisoner or a prison guard. The experiment was conducted at Stanford University, where 24 undergraduates were selected to play the roles in a mock prison in the basement of the Stanford Psychological Building. Those chosen were chosen due to their lack of psychological issues, crime history, and medical disabilities, in order to obtain a representative sample.

Roles were assigned based on a coin toss. Prisoners and guards rapidly adapted to their roles, stepping beyond the boundaries of what had been predicted, and leading to dangerous and psychologically-damaging situations. One third of the guards were judged to have genuine sadistic tendencies, while many of the prisoners were emotionally traumatised and two had to be removed early on. The study was meant to last for two weeks, but after Zimbardo’s girlfriend pointed out that he was allowing unethical acts to happen directly under his supervision, he concluded that both prisoners and guards had become too engrossed in their roles and terminated the experiment after only six days for their safety.

The BBC Prison Study

Reicher and Haslam (2002, 2006) wanted to test the idea of social identification and to see how many people come to condone tyranny or become tyrannical themselves, following on from the events of World War II. The study builds on the work of Milgram, Tajfel and Zimbardo. It builds upon the Stanford Prison experiment, but is not an exact replica as Zimbardo’s work was unethical.

Reicher and Haslam called it an experimental case study, as they set up a one-off situation and then studied it to collect in-depth, detailed data using observational studying, video and tape recording, analysis of conversations and psychological and physiological assessments.

The study was discussed with colleagues, a university ethics committee and the British Psychological Society (BPS). Safeguards used within the experiment included:

  • thorough screening of the participants
  • a signed, detailed consent form which told participants that they could be at risk of stress and confinement
  • independent monitoring of the study by two clinical psychologists and an ethics committee
  • security guards, able to intervene if the behaviour ever became dangerous

The BBC recorded the study and organised it into four programmes. They were broadcast in May 2002. The participants knew they would be appearing on national television. A detailed explanation of the study is provided by Reicher and Haslam, in conjunction with the BBC, at


The BBC Prison Study

Reicher and Haslam (2002, 2006) wanted to test the idea of social identification and to see how many people come to condone tyranny or become tyrannical themselves, following on from the events of World War II. The study builds on the work of Milgram, Tajfel and Zimbardo. It builds upon the Stanford Prison experiment, but is not an exact replica as Zimbardo’s work was unethical.

Reicher and Haslam called it an experimental case study, as they set up a one-off situation and then studied it to collect in-depth, detailed data using observational studying, video and tape recording, analysis of conversations and psychological and physiological assessments.

The study was discussed with colleagues, a university ethics committee and the British Psychological Society (BPS). Safeguards used within the experiment included:

· thorough screening of the participants

· a signed, detailed consent form which told participants that they could be at risk of stress and confinement

· independent monitoring of the study by two clinical psychologists and an ethics committee

· security guards, able to intervene if the behaviour ever became dangerous


In general, the authors aimed to:

i           provide detailed data on the developing interactions between two groups of unequal power and privilege

ii         develop practical and ethical procedures for conducting important large-scale studies in social psychological research

iii        analyse the conditions under which people will:
define themselves in terms of, and act on, group identities
accept of challenge group inequalities

In particular, the following predictions were based on the Social Identity approach:

·         dominant group members (the guards) will identify with their group from the start and impose their power

·         subordinate group members will only identify and challenge inequality if relations are seen as impermeable and insecure

Text Box: AIMS In general, the authors aimed to: i	provide detailed data on the developing interactions between two groups of unequal power and privilege ii	develop practical and ethical procedures for conducting important large-scale studies in social psychological research iii	analyse the conditions under which people will:   -    define themselves in terms of, and act on, group identities   -    accept of challenge group inequalities In particular, the following predictions were based on the Social Identity approach: •	dominant group members (the guards) will identify with their group from the start and impose their power •	subordinate group members will only identify and challenge inequality if relations are seen as impermeable and insecureThe BBC recorded the study and organised it into four programmes. They were broadcast in May 2002. The participants knew they would be appearing on national television. A detailed explanation of the study is provided by Reicher and Haslam, in conjunction with the BBC, at

11 Responses to “Social Psychology”

  1. 1 J.Morris1
    October 30, 2009 at 19:45

    This is an EXEMPLAR website that everybody should use and I can see you have really immeshed yourself in this subject. I thoroughly enjoyed reading it in between my P*** bashing schedule and bowling instead of swimming in a pool.

    From J.Morris1 and not just J.Morris and keep working hard-I’m being serious.

  2. 2 Isabelle
    January 21, 2012 at 16:21

    Those notes are really helpful! Thank you so much !

  3. 3 Michelle
    April 13, 2012 at 12:20

    This is amazing ..helped me soooo much. Thank you Thank you !

  4. 4 RIO
    December 8, 2012 at 16:09

    this is the best website i love you thanks!!

  5. 7 J.Crooch
    October 14, 2013 at 12:01


    Love from Mr.Crooch

  6. January 27, 2014 at 12:42

    Really helpful site. If anyone is studying for AQA (A) Psychology my website has a good set of resources for A2.

    You can find it here:

    It has model essay answers for Psya3 and Psya4 but i plan to release some various materials for all the papers to aid in revision soon.

  7. March 19, 2014 at 10:15

    SUCH A useful notes and helps me a lot…thnx:)

  8. 10 will
    April 25, 2014 at 20:10

    This is the single most useful resource for psychology on the internet
    thank you so much !

  9. 11 josh
    January 12, 2015 at 17:56

    the explanations are really good however it would help if evaluations were done by RAMP or GRAVE this would make it a lot better

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Browse the site and read the various units of the course. This is Edexcel Psychology and only contains notes for that syllabus. Feedback is appreciated, feel free to check out the other sites in the series.

Also, please download the AS Revision 101 toolbar for free at which includes links to all the AS revision sites, a chat room widget to discuss revision and links to other associated sites

Copyright Notice

The content of this website belongs to Daniel Holloway
Significant contribution from AS Edexcel Psychology by Christine Brain
If a copyright holder of any diagrams or texts would like them removed from this site, please contact me via the site and that will be done as soon as possible


Get every new post delivered to your Inbox.

Join 56 other followers

%d bloggers like this: