Sherif et al. (1954, 1961) Robber's Cave Experiment (Edexcel A-Level Psychology): Revision Notes

Sherif et al. (1954, 1961) Robber's Cave Experiment

Background

Muzafer Sherif and colleagues conducted a series of experiments between 1949 and 1961 to investigate intergroup conflict and cooperation. The research involved boys attending summer camps who were divided into groups and made to compete in camp activities, allowing researchers to observe how competition influenced group behaviour.

First experiment (1949): Conducted in Connecticut with 24 boys attending an 18-day camp. Boys initially spent time together before being divided into two groups. They participated in separate activities, developed group identities (naming themselves Red Devils and Bull Dogs), and formed in-group friendships. Conflict emerged between the groups, with boys rating members of the other group negatively despite previous friendships.

Second experiment (1953): Held in upstate New York with groups calling themselves Panthers and Pythons. This study was unsuccessful as the boys suspected the researchers were deliberately creating friction between groups, leading to the experiment being stopped.

Third experiment (1954, 1961): The successful study conducted at Robber's Cave State Park in Oklahoma, which forms the focus of this research.

Aim

The researchers aimed to investigate intergroup relations over time when various experimentally induced situations were introduced. Specifically, the study examined:

• How groups form and develop identity
• The effect of competition on intergroup relations
• The conditions under which intergroup conflict could be resolved

Participants

The study involved 22 boys aged 11-12 years old (one boy was 12, the rest were 11) from middle-class Protestant families in Oklahoma City. These boys were described as similar and 'normally adjusted'.

An initial sample of 200 boys was opportunistically selected from schools in Oklahoma, with 22 finally chosen. The researchers ensured boys were not acquainted with each other before the study. Parents were paid US$25 to incentivise participation and not visit during the two-week camp.

Procedure

The study took place at Robber's Cave camp in Oklahoma, an isolated location previously used as a hideout for outlaws. The experiment followed three distinct stages:

Stage one: in-group formation

During the first 5-6 days, the two groups of boys were kept separate from each other. Each group participated in activities designed to encourage in-group formation, such as camp outings and hikes.

Researchers, posing as camp staff, observed the verbal and non-verbal communication within each group and the relationships that emerged. Sociometric data (quantitative data about personal and social relationships) was gathered, measuring how boys rated each other on popularity, initiative and other characteristics.

The groups developed their own identities during this stage. One group named themselves the 'Rattlers' and the other the 'Eagles'. They created group norms, rules, nicknames and symbols. At the end of stage one, groups were made aware of the other group's existence, and an 'us versus them' attitude began to emerge.

Stage two: competition and friction

Over the next 4-6 days, the boys were brought into contact during competitions as part of a camp tournament. Each individual group member had to contribute to win points towards the tournament total. Both groups were subjected to orchestrated frustrating situations that they believed were caused by the other group.

During this stage, stereotypes about each group were recorded, and behaviours and attitudes towards the other group were noted.

Activities included:

• Baseball games
• Tug of war
• Touch football
• Tent pitching
• Cabin inspections
• Skits and songs
• Treasure hunt

Prizes for winning teams included a trophy, medals and a four-bladed knife.

Stage three: superordinate goals

The final 6-7 days focused on bringing about conflict resolution through introducing common goals that required cooperation between groups. Three problem situations were created that could only be resolved if both groups worked cooperatively:

1. Fixing the water tank that provided water to both groups
2. A joint camp-over where group members had to work together for food and sleeping arrangements
3. Starting the broken-down camp bus

Results

Stage one results

Boys in each group formed their own group norms and rules, establishing group identity. They participated in various cooperative activities and had defined a name for their group by the end of the stage. When asked to self-report who their friends were, around 93% selected exclusively from their own in-group.

Stage two results

The boys began showing signs of hostility towards the other group and persisted in demanding competitive activities. Leaders emerged in each group, and boys became territorial. When the tournament was announced, fighting began, name-calling occurred, and the Eagles burned the Rattlers' camp flag.

Strong in-group favouritism and negative out-group bias developed, resulting in derogatory terms being used about the other group (such as 'stinkers', 'braggers', 'sissies'). Skirmishes and camp raids occurred between the groups.

Stage three results

During the final stage, researchers attempted to bring the groups together in various situations. Initially, mere contact alone (such as in the dining hall or watching a movie) was not sufficient to reduce hostility. The boys persisted in name-calling and fighting.

Tasks involving superordinate goals (goals that require cooperation between groups) were introduced. The first cooperative task involved fixing the water tank. Boys were divided up and each had a role in identifying the cause of the water shortage. When the blockage was identified, boys were observed mingling with each other and no longer name-calling. However, harmony did not persist, and the boys soon displayed negative out-group bias during supper, with derogatory terms being used and food being thrown.

The boys were informed they could secure a movie if they collectively paid for it. They worked out a payment strategy, and a noticeable reduction in hostility was observed during supper that evening and turn-taking the following day at breakfast. The boys worked together to pull the camp bus, which had been rigged by researchers not to start.

At the end of the study, researchers reassessed friendship choices. A notable increase in the number of boys whose friendships were now with the out-group was found compared to choices made in stage two.

Conclusions

Sherif and colleagues concluded that:

• Strong in-group identities were formed initially, even without competition
• With the introduction of competition, negative out-group bias quickly emerged
• The introduction of superordinate goals had a cumulative effect in reducing negative out-group bias by removing competition
• The research supported realistic conflict theory, which proposes that prejudice can be brought about through competition for resources

Evaluation

Strengths

High ecological validity: The Robber's Cave study took place within a natural environment for the boys. Unlike laboratory studies where group behaviour is heavily manipulated, the behaviour observed between the boys occurred relatively naturally. The camp setting was a realistic context for studying intergroup conflict.

High level of control: The researchers used careful planning at each stage. Staff were participant observers who were unaware their behaviour was being recorded as part of a psychological investigation. Staff were only permitted to intervene in decision-making and conflict between groups when there was a risk to safety. This control ensured staff did not directly influence the boys' behaviour.

Careful participant selection: To decrease the influence of individual variables on results, Sherif carefully selected the 22 boys based on gender, age, IQ, social class and religion, matching the groups as carefully as possible. This prevented individual differences in character and attributes affecting the behaviour of the boys.

Weaknesses

Ethical concerns: Although parents consented to the study, the boys were unaware they were part of a psychological investigation on intergroup hostility, believing it to be a study about leadership. This means fully informed consent was not obtained from the boys themselves. Many have questioned the ethical nature of the study as it deliberately induced prejudice and placed boys in situations where they could come to harm. This was exacerbated by the fact that Sherif was running out of funding, meaning the Robber's Cave study was perhaps his last attempt to obtain successful outcomes, potentially leading to accusations of deliberately inducing high levels of conflict and hostility.

Researcher interference: Unpublished researcher notes and interviews with the boys as adults revealed that researchers may not have been as independent as instructed. Many were reported to be actively encouraging intergroup hostility and creating opportunities for conflict by breaking down items of a rival group and blaming the other group. They were also reported to encourage physical conflict by not intervening in skirmishes between boys, which fuelled hostility beyond what would normally be experienced.

Sample limitations: As only boys were used in the study (two dropped out during the first stage), the sample may be insufficient and gender-biased, meaning the results may not be generalisable to girls or mixed-gender groups. Sherif selected boys who were reputed to have good athletic ability and were keen on sport, which could explain the degree of conflict, as they were perhaps naturally competitive.

Alternative explanations: It was apparent at the end of stage one that intergroup hostility had already emerged at the mere knowledge of another group, even before competitions formally began. This perhaps offers some support for social identity theory, as competition was not necessary to create prejudice – it merely crystallised later on in stage two.

Limited generalisability: Tyerman and Spencer (1983) asserted that it is not a natural condition for strangers to meet and compete against one another. In real life, group members will be familiar with one another and have a history of social interaction. Using a Scout troop whose separate patrols normally interacted a couple of times a year, they partially replicated the summer camp study. Using four patrols at a two-week camp, patrol leaders assessed behaviour using a range of different measures (cooperation, atmosphere in camp, solidarity). They found that hostility did not emerge between groups, and competition did not inevitably lead to hostility. As the Scout patrols were familiar with one another, this represents a more realistic outcome and has greater generalisability than the Robber's Cave study. However, the competitions employed did not involve direct contact between patrols, nor were trophies offered, which may explain the lack of competitive hostility.