Intergroup Discrimination


Prejudice and discrimination are two different but closely-linked things.

Prejudice is an attitude towards an individual or group because they are part of a certain group. Discrimination is the behavioural manifestation of prejudice. Both are usually negative.

Some psychologists believe that being part of a certain group makes you automatically prejudiced towards people from other groups. However, Tajfel believed that it goes further than just group membership. He agrees with psychologists who believe prejudice comes from competition between groups, but goes on to say that even a perception of grouping can produce discrimination. This means that although there might be no official groups, if someone believes that they are part of a distinct group then they will discriminate against others who they see as ‘outsiders’. So basically, Tajfel believed that categorisation/grouping was the cause of discrimination.

He decided to use an ingroup and outgroup system. An ingroup is the group a person believes they belong to, an outgroup is the group a person does not belong to.

Henri Tajfel

Henri Tajfel

The first sample consisted of 64 boys (16 in Experiment One, 48 in Experiment Two), all aged 14 to 15 years old. They all knew each other before the experiment because they were from the same comprehensive school in Bristol.

The experiment had two separate sub-experiments. The boys would enter the laboratory in groups of 8.

Part I:
The first part of the experiment was used to establish an intergroup categorisation.

The boys were brought into a lecture room and told that the experimenters were studying visual judgement. Forty clusters of varying numbers of dots were shown on a screen and the boys were asked to write down an estimate of the number of dots in each cluster. This experiment had two conditions:

  1. Over-estimators and under-estimators condition: After the boys made their estimates, they were told that some people consistently overestimate the number of dots while others consistently underestimate the number of dots, but that these tendencies were not related to accuracy.
  2. Better and worse condition: The boys were told that some people are consistently more accurate than others.

Two groups of eight boys were used in each of the above conditions.

After the judgements had been made and checked, the boys were told that they would be grouped according to their visual judgements. In reality, the grouping was completely random. The boys were taken to separate cubicles and informed of the group they were in.

Part II:
The second part of the experiment was to assess the effects of grouping on intergroup behaviour.

The boys were given a book of matrices (I know, I know, maths just seems to pop up everywhere, the pest!). They had to assign rewards and penalties of real money to the other boys. Each boy had his own code number so their identity was kept private. The only information given was the group of the boy (“your group” or “of the other group” was written).

There was three possible options:

  1. Ingroup choices: The top and bottom row of the matrix referred to members of the boy’s own group.
  2. Outgroup choices: The top and bottom row of the matrix referred to members of a different group.
  3. Intergroup choices: One row referred to a boy of the same group, the other row referred to a boy of another group.

The second experiment involved 48 new boys.

The boys were first shown slides of paintings by Paul Klee and Wassily Kandinsky. They were asked to state their preference without being aware of the painter’s identity. They were told they would be grouped according to their preference but, as previously done, the grouping was completely random.

They were also given matrices and asked to give rewards or penalties to other boys. However, these matrices were structured differently and had three variables:

  1. Maximum joint profit: Boys could give the largest possible reward to members of both the ingroup and outgroup.
  2. Largest reward to ingroup: Boys could choose the largest possible reward for the member of their own group regardless of the reward given to a boy in an outgroup.
  3. Maximum difference: Boys could choose the largest possible difference between rewards for both groups, in favour of their own group.

If you’re a maths dunce who faints at the mention of matrices (like me), you might be kinda sleepy-eyed right about now. For your sake, here’s a summary of the results (well, aren’t you special!):

  • When making intergroup choices, most of the subjects gave more money to members of their own group.
  • The ingroup and outgroup choices were pretty much 50/50 so they were conducted quite fairly.
  • The boys chose a maximum difference between giving rewards in Experiment Two. So instead of simply giving themselves a larger reward, they purposely chose the option that maximised the difference between the fellow group member’s reward and the outgroup member’s reward.

Type of research method
This was a straightforward laboratory experiment.

Experimental design
This study used independent groups.

Independent variable
The IV was options they were given in the matrices.

Dependent variable
The DV was the choices made by the boys when giving rewards and penalties.


  1. High level of control: The experiment took place in a laboratory with controlled conditions, therefore there was very high control over extraneous variables.
  2. Replicable: The procedure itself can be replicated (i.e. random grouping and giving matrices with different options). This means that the study can be repeated and the result’s reliability can be checked.


  1. Not generalisable: The study was conducted on a bunch of teenage boys from Bristol. Not very representative of the wider population, therefore the results aren’t applicable to the rest of us (unless you are also a 14 year old boy from Bristol, which would be kinda weird).
  2. Lack of ecological validity: The study was conducted in a laboratory with a completely artificial setting. This means the results may have occurred simply because the environment was not true to real life.
  3. Demand characteristics: Since the setting was artificial and the experimenters were around, the boys may have been acting in a way that was not totally natural.

Ethical issues

  1. Informed consent: Children cannot give their consent to be used in a study. The guardian or parent must give consent if a child is to take part in a study. Tajfel would have had to ask permission from the parents, but he does not mention whether or not he did in his study so that could be a concern.
  2. Deception: The boys were deceived on the nature of the grouping, as well as the reasons behind the experiment.
  3. Confidentiality: The boys’ identities were kept confidential so their privacy was respected.
  4. Emotional or physical harm: There was no harm inflicted on the participants in this experiment.
  5. The right to withdraw: Tajfel did not make it clear during the experiment that the boys had a right to withdraw. This would be against ethical guidelines and has not been properly handled in the study.
  6. Debriefing: To our knowledge, the boys were not debriefed about the real nature of the study in any way. Tajfel didn’t really give much details here.

Reference: Tajfel, H. (1970). Experiments in Intergroup Discrimination. Scientific American. 223: 96-102.

5 thoughts on “Intergroup Discrimination

  1. hey, just to let you know the experiment design is repeated measures as the boys had been in both groups, in- group and out-group.

  2. HI, Can i know why in Henri Tajfel’s study, only at part two of the study where 48 new participants were told to make aesthetic preference, can only determine MJP MIP and MD, but not for the first part of the study where 64 boys asked to estimate 40 clusters of number of varying dots ?

  3. Can I just say that I laughed out loud at the ‘maths dunce’ comment! Oh geez, I really was getting sleepy-eyed but that was an amazing way of snapping me back into attention. Great sense of humour!

Leave your comments below!

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