Jane Elliot Experiment 1984

Aim : To demonstrate Social Identity Theory
Method: On the first day she made blue eyes is better than brown eyes and on the second day, brown was better than blue. The brown eyed students (on the first day) were discriminated and were called stupid. On that day, she designated the blue-eyed children as the superior group, giving them extra privileges like second helpings at lunch, access to the new jungle gym and five minutes extra at recess. She said blue-eyed children was linked to intelligence and ability, as the “blue” pigmentation would result in an increase of these qualities. The next day she did vice versa.
Findings: At day 1 brown eyed children took 5 minutes in working with flashcards because they felt bad. At day 2 it took brown eyed children 2.5 minutes with flashcards because they have been promoted to feel more superior.
Evaluation Has high ecological validity, not natural experiment as she manipulated the experiment, independent variable is the color of eyes, confounding variables is difficult to control as students might had bad day which caused to act in certain ways, student have needs to be in group.
Here is a link to watch the experiment: A classroom divided

Nisbett et al (1973)

Male college students were more likely to attribute their own choices of girlfriend and major study to external facts (e.g. she’s a relaxing person, chemistry is high paying).
When attributing their friend’s choices, were more on dispositional factors (e.g. need someone to relax with, want to make a lot of money), without considering their situation.

Milgram’s study (1963) - Study of obedience

Aim: To investigate how far people will go in obey an authority figure based on how he wanted to prove the Germans were different from people that they were able to carry out barbaric acts against Jews and other minority people.

  • Each participant was assigned with a confederate, and true participant believes confederate to be another participant
  • Participant and confederate were told that experiment was on human learning
  • They drew lots and it was made in such a way that the leaner would always be the confederate and the teacher, the participant
  • Participant was shown how the ‘learner’ was being strapped to an impressive-looking machine, the generator. Then he/she was given a test shock of 45 volts to convince them it was real.
  • Participant leaves confederate and enters a separate room where their task was to read out 4 key words and 4 possible pairs, while the confederate had to press the correct button.
  • Each incorrect answer, a 15 volt shock was applied and for every consecutive incorrect answer, 15 volts was increased by another 15 and so on up to 450 volts.
  • Milgram wanted to see how far would the participant go


  • Levels of obedience expected… psychology students and professional colleagues had a mean of 1.2%
  • Level of obedience obtained: 65% continued to max 450 volts and no subjects stopped before 300 volts.
  • Subjects’ realisation of shock effects (How painful to the learner were the last few shocks), mean response was 13.42 out of 14.
  • Subjects were seen to sweat, tremble, stutter, bite lips, groan. There were nervous laughs but they clearly admitted it was not funny at all. There were actual uncontrollable seizures observed in 3 subjects. On one occasion they had to call a halt to the experiment.


  • procedure was well standardised and obedience was accurately operationalised as the amount of voltage given
  • It took part in Yale University, a prestigious university, laboratory looked realistic
  • There was good control and little variation in experience for each participant as up to 75 volts, confederate would only have grunts of discomfort. At 150 volts, the leaner would shout and say he did not want to continue. At 300, he would refuse to give answers.
  • The experimenter would also occasionally give prompts and say “You must continue” or “It is essential that you continue”.
  • It lacks ecological validity
  • Some ethical problems (subjects were decieved, physically hurt and mentally)
  • Subjects were only American and male
Sherman (1980) - Compliance

Experimenters called residents in Indiana (USA) and asked them if hypothetically they were to volunteer to spend 3 hours collecting for the American Cancer Society.
Three days later a second experimenter called the same people and actually requested help for this organization.
Those responding to the earlier request was 31%, this was much higher than the 4% of the group of people who volunteered to help when approached directly.

Moscovici and Lage (1976) - minority

Involved 4 participants and 2 confederates. Minority of two confederates described a blue-green colour as green. Minority was able to influence about 32% of participants to make at least one incorrect judgement about colour of slides shown. In addition, participants continued to give incorrect responses even after 2 confederates had left experiment.

Lee et al. (1977) - Fundamental Attribution Error

Aim: See if student participants would make fundamental attribution error, even though they knew that all actors were simply playing a role.
Method: Participants were randomly assigned to one of three roles: a game show host, contestants on game show or members of the audience. The game show hosts were instructed to design their own questions. They audience then watched the show through the series of questions.
When show was over, observers were asked to rank intelligence of people taken part.
Results: They consistently ranked the game show host as most intelligent. They failed to attribute the role to the person’s situation, instead attributed the person’s performance to dispositional factors.

  • The sample is problematic, they only made use of student participants
  • University students spend their days listening to professors, who are seen as authorities who ask questions and give answers and is not a learned response rather than attribution error.
  • Question… if findings are generalizable.
  • However it can reflect what we see in everyday life… people with social power usually initiate and control conversations. Medical doctors and teachers are often seen as experts and when they publish something outside their field, their work is rarely challenged.
  • Miller discovered that Indian-Hindus took situational factors into account for explaining actions of when someone has done wrong, juxtaposing how FAE applies to Americans.
Leon Festinger et al. (1956) - When Prophecy Fails
  • Most well known covert observation.
  • In Chicago, there was a religious cult that believed the world would end on 21st December. They believed that when natural catastrophes began, they would be rescued by flying saucers as long as they followed the prescribed rituals and read the sacred texts. They were to remain isolated from all non-believers (making psychologists difficult to study them). Festinger and his team decided to become cult members to carry out a participant observation.
  • They remained until that fateful day of 21st December and when nothing happened, Festinger monitored the group members’ doubt, debate and rationalization. The members of the cult, as part of maintaining their self-esteem, decided that God and not destroyed the world because of their prayers.
    Principle: People have a social self and identity. Therefore they have a desire to protect their self-esteem. This is demonstrated in this study as the cult had decided based on their prayers that they have saved the world.
Festinger and Carlsmith (1957) - Cognitive dissonance
Aim: Festinger and Carlsmith (1959) investigated if making people perform a dull task would create cognitive dissonance through forced compliance behaviour.
Method: In their laboratory experiment, they used 71 male students as participants to perform a series of dull tasks (such as turning pegs in a peg board for an hour).
They were then paid either $1 or $20 to tell a waiting participant (a confederate) that the tasks were really interesting. Almost all of the participants agreed to walk into the waiting room and persuade the confederate that the boring experiment would be fun.
Results: When the participants were asked to evaluate the experiment, the participants who were paid only $1 rated the tedious task as more fun and enjoyable than the participants who were paid $20 to lie.
Conclusion: Being paid only $1 is not sufficient incentive for lying and so those who were paid $1 experienced dissonance. They could only overcome that dissonance by coming to believe that the tasks really were interesting and enjoyable. Being paid $20 provides a reason for turning pegs and there is therefore no dissonance.

Tajfel et al. (1971) - Kandinsky vs Klee

Aim: To test the Social Identity Theory

He conducted a study according to various situations in which Bristol schoolboys were assigned at random to one of two meaningless groups. They were assigned based on their preference on either Kandinsky or Klee’s paintings. They were asked to rate the in-group and out-group members on traits such as likeability.

Findings: Tajfel found that the out-group was rated less likeable, but never actually disliked.

Conclusion: There seems to be a preference of the in-group rather than out, however it is not clear that they make social comparisons to enhance either self-esteem.
Later research shows group identity alone appears not to be responsible for intergroup conflict. In the absence of competition, social comparison does not necessarily produce a negative outcome.

Crutchfield (1954)

Without physical presence, he placed subjects in individual cubicles with electronic display boards which supposedly let each subject know what the others had answered. However each were believed to be the last to answer and presented them with uniformly wrong group answers on half the tasks. He tests over 600 subjects using various stimuli.
Asch’s line comparison test - 30% conformity
46% conformity that a picture of a star had a larger surface area than a circle (when it was 1/3 smaller)
37% agreement to the statement ‘I doubt that I would make a good leader’ which none agreed to when asked on their own.
Plus point:
- Gave an insight to conformity
Criticisms of conformity:
Artificiality - Above studies used well controlled and standardised procedures but reflected conformity under laboratory conditions, meaningless stimuli
It is not etic as this type of conformity was true for USA in the 1950s but it may differ over time and for different cultures
Ethics - subjects were deceived

Cialdini et al. (1976)

Demonstrated social comparison with college football supporters.
After a successful football match, supporters were more likely seen wearing college insignia and clothing than after defeats. It is assumed that our need for a positive self-concept will result in bias in these intergroup comparisons. Tajfel calls this “the establishment of positive distinctiveness”.