Evaluate the Classic Study of Sherif et al.
The Robbers Cave experiment that Sherif et al had conducted aimed to find how competition between two
groups lead to prejudice towards outgroup and ingroup favouritism. This is a field experiment that consists of a
sample of 22 middle-class 11-year-old protestant boys from Oklahoma, USA. Furthermore, the independent
variable of the study is cooperation or competition, while one independent variable is the number of friends in
the outgroup. The parents gave consent for the boys to participate in the experiment. Moreover, the experiment
starts with the boys being divided into two groups: Rattlers and Eagles. The first stage of the experiment is the
group formation, a non-competitive task or activities were given to each of the group to create bond such as
building campfires. The second stage is creating friction, the researchers created tournaments with prizes for the
winners, the activities include tug of war. Lastly, the third stage is reducing friction, both groups have more social
contact, they watch movies together and have superordinate goals or one goal that requires intergroup
cooperation such as mending a broken water tank. The experiment went on for three-works using a covert
observation, recordings, and questionnaires that led to both quantitative and qualitative data.
One strength of the experiment is the researchers carefully selected the sample of boys to be considered in the
experiment. For example, they did interviews, and they also tested the potential of the participants that they
allocated in groups based on matching personalities, skills, and interests which helps in ingroup favouritism. This
highlights that the study uses good controls in getting their samples increasing the internal validity of the study.
However, the samples are completely biased only consisting of middle-class and protestant American boys. This
is a weakness of the study since it is lacks representativeness, therefore findings are not generalisable to the
whole population such as for working-class, girls, over 11 years old and other ethnicities.
Furthermore, the findings from Stage 1 shows that leaders established in each of the groups. The groups also
differ in social norms such as Rattlers who are tough and who swore a lot, while Eagles cried more when injured
and were anti-swearing. In Stage 2, the groups developed hostility and are eager to challenge each other in
different activities, they also fight, trash each other's cabin, and name-calling. 6.4% of Rattlers were friends with
the Eagles and 7.5% Eagles are friends with Rattlers. They also show ingroup favouritism by ranking their
members as friendly, brave, and tough while the outgroup is seen sneaky and stinkers. Lastly, Stage 3 shows that
superordinate tasks had little effect in reducing friction in the beginning. The percentages had increased with
36.4% of Rattlers being friends with Eagles, and 23.2% of Eagles are friends with Rattlers.
On the other hand, one weakness of the study is there is evidence of failed replication of findings. For example,
Tyerman and Spencer (1983) conducted the experiment with sea scout troop of 30 boys who belongs to one of
four patrols and knew each other well. Findings show no increase of ingroup solidarity from the two-week long
camp, it even decreased a bit. The troops were not hostile towards each other and interacted as one troop. This
is important as it highlights that if people know each other very well, even if there is competition it could not
lead to prejudice only if they do not know each other.
The study concluded that intergroup competition increases ingroup favouritism and outgroup hostility. They also
concluded that social contact is not enough to reduce prejudice, but a series of superordinate goals will.
Moreover, another strength of the study is its application in reducing prejudice in society. For example, Aronson
and Bridgeman (1979) developed a jigsaw classroom based on superordinate goals, where students work
together increasing likeness and empathy for the outgroup members which also increased academic
performance of black minority students. This is important as it highlights that the experiment had great impact
in increasing educational achievement and performance of ethnic minority students rather than being
discriminated by their own peers.
In conclusion, Sherif et al's study has its strengths for having good internal validity from selecting their sample
and it also has good applications in education of black ethnic minority students. However, there are certain
limitations of the study such as having a non-generalisable sample and failed replication of the study that showed
different findings.