Lesson 8 of 8 · 20 min

Reducing prejudice through intergroup contact

Describe how intergroup contact that is sustained, with superordinate goals, mutual interdependence and equality of status reduces prejudice, with the evidence, and explore other ways to build intercultural understanding.

What you will be able to do

  • Describe the four conditions of intergroup contact that reduce prejudice.
  • Explain the Robbers Cave and jigsaw classroom evidence, including its limitations and ethical issues.
  • Interpret data on contact and prejudice, choosing and interpreting the right statistical test.
  • Explore ways to increase intercultural understanding to reduce conflict.

If prejudice grows from unfamiliarity and competition, bringing groups together should help. Sometimes it does. But history is full of groups that lived side by side for generations and stayed hostile. In 1954 Gordon Allport proposed the contact hypothesis: contact between groups reduces prejudice, but only under the right conditions.

The four conditions of intergroup contact the syllabus names
ConditionWhat it meansWhy it reduces prejudice
SustainedContact is repeated and lasts long enough for real relationships to form, not a single meeting.Time lets people see outgroup members as individuals, so stereotypes break down; friendships form.
Superordinate goalsBoth groups want a goal they cannot reach alone.The groups' interests line up instead of competing; "us and them" becomes "we".
Mutual interdependenceEach group needs the other to succeed; outcomes depend on each other.People must cooperate and rely on each other, which builds trust and liking.
Equality of statusMembers of both groups have equal standing in the situation.Neither group is in charge, so contact does not confirm stereotypes of one group as inferior.
Interactive

Switch on the conditions of contact

Two groups in a setting. Switch each condition on or off and see how the setting changes and what the classic research found.

The Robbers Cave experiment. In 1954, Muzafer Sherif and colleagues ran a summer camp for boys at Robbers Cave State Park in Oklahoma. The boys, who did not know each other, were split into two groups that did not at first know the other existed.

  1. Group formation. Each group bonded and named itself: the Rattlers and the Eagles.
  2. Friction. The groups competed in a tournament with prizes for the winners only. Name-calling, flag burning and cabin raids followed; each group rated the other with unfavourable stereotypes.
  3. Integration. First the researchers tried simple contact: meals and films together. Hostility continued, with food fights. Then they set superordinate goals: the camp's water supply "failed" and both groups had to find the fault; the boys pooled money to rent a film; and both groups pulled together on a rope to start a stalled food truck.
Friendship choices and out-group ratings at Robbers Cave (Sherif et al., 1961)
MeasureEnd of friction (stage 2)End of integration (stage 3)
Rattlers: best friends chosen from the Eagles5 of 78 choices36 of 99 choices
Eagles: best friends chosen from the Rattlers5 of 67 choices (7.5%)about a quarter of choices
Rattlers: ratings of the Eagles that were unfavourable53%4.5%
Eagles: ratings of the Rattlers that were unfavourable76.9%22.6%

Calculate the percentage of the Rattlers' friendship choices that went to Eagles at the end of the integration stage. Give your answer to 1 decimal place.

in %

Round to 1 decimal place.

Key idea

What Robbers Cave shows

Competition created prejudice between groups of boys who were alike in background. Contact alone (eating together) did not reduce it. A series of superordinate goals, requiring interdependence, over days of sustained contact between groups of equal status, did.

Note

Limitations and ethics

  • The boys were chosen to be similar (same age, similar backgrounds), and the "groups" were created for the study, so the results may not generalise to groups with real cultural differences, histories and power differences.
  • It was a field experiment with one camp: no control group and little replication.
  • The boys did not know they were being studied (deception), and the researchers deliberately created hostility between children. Today this would breach the principle of protection from harm and need full informed consent from parents and the children's assent, withdrawal rights and a debriefing.

The jigsaw classroom. In 1971, schools in Austin, Texas, had just been desegregated, and classrooms were tense. Elliot Aronson's team designed the jigsaw technique. Students work in small, mixed groups. Each student learns one piece of the lesson and teaches it to the group; nobody can complete the task without every other member's piece. That builds in mutual interdependence, equal status (every piece matters equally) and a superordinate goal (the group's learning), and it ran several times a week (sustained). Students in jigsaw classes came to like each other more across groups and showed less prejudice than students in ordinary classes (Aronson & Bridgeman, 1979).

Here are illustrative data, made up for practice, from a class that used jigsaw groups for eight weeks. Ten students completed a prejudice scale (higher scores mean more prejudice) before and after.

Interactive

Did prejudice scores fall?

Each student was measured twice, so the paired t-test is the right one. Illustrative data.

For these data, the paired t-test gives p=0.0005p = 0.0005 and the unpaired t-test gives p=0.055p = 0.055. Which conclusion is correct?

Choose one answer

Worked example

Analyse before-and-after data from a jigsaw program

  • Complex unfamiliar
  • Technology active

Illustrative data: ten students completed a prejudice scale (higher scores mean more prejudice) before and after eight weeks of jigsaw groups.

Student12345678910
Before34293831273633303528
After30273330243131293027

(a) Identify the design and the appropriate t-test. (b) Calculate the mean of the differences (before − after) and the t statistic, to 2 decimal places. (c) With p=0.0005p = 0.0005, state a conclusion. (d) Identify one limitation of the design and suggest an improvement.

6 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Match the test to the design (paired for repeated measures or matched participants).

  2. Compute each difference, then dˉ\bar{d}, sds_d, SE=sd/nSE = s_d/\sqrt{n} and t=dˉ/SEt = \bar{d}/SE.

  3. Compare p with 0.05 and state the conclusion in words.

Watch out for

  • Using the unpaired test: it gives p=0.055p = 0.055 here and would wrongly suggest no significant change.

  • Claiming the program caused the fall without a control group.

How strong is the evidence overall? Thomas Pettigrew and Linda Tropp (2006) combined 515 studies with about a quarter of a million participants in 38 nations. The mean correlation between contact and prejudice was r = −.21. In a single class study an r that small (below 0.25) would be described as no correlation, but averaged over 515 studies it is a small, very reliable effect: more contact went with less prejudice in 94% of the studies. The more rigorous experimental studies found a stronger effect (about r = −.33, a weak correlation). Contact under Allport's conditions usually reduced prejudice more, though the conditions helped rather than being essential. Later analysis found that contact works mainly by reducing anxiety about the other group and increasing empathy, and less by increasing knowledge (Pettigrew & Tropp, 2008).

Note

Contact and empathy in the brain (an Australian study)

Yuan Cao and colleagues (2015) at the University of Queensland scanned the brains (fMRI) of Chinese students who had arrived in Australia between six months and five years earlier while they watched videos of people of their own race and of another race receiving painful touch. A brain area linked to empathy, the anterior cingulate cortex, responded more to own-race pain: a racial bias in empathy. But the more contact a student reported with people of the other race, the more strongly it responded to their pain. The design is correlational, so less biased students may simply have sought more contact; but the neural measure is harder to fake than a questionnaire.

Interactive

Contact and prejudice in a survey

Illustrative survey of 12 students. Correlation alone cannot show that contact causes lower prejudice: less prejudiced people may simply choose more mixed activities.

In the survey, r=−0.69r = -0.69 (n=12n = 12, p=0.014p = 0.014). Which statement is the best conclusion?

Choose one answer

Increasing intercultural understanding. Contact is one tool. Others aim at the same processes: less anxiety, more empathy, accurate knowledge, and fair decisions.

  • Cooperative structures in schools, sport and workplaces: mixed teams with shared goals (jigsaw, team sports, buddy programs).
  • Learning about cultures and histories, including the history and languages of the First Nations peoples of the local area, taught with and by those communities. Acknowledgements of Country and Welcomes to Country by Traditional Owners recognise the peoples whose Country a community is on.
  • Perspective-taking and stories that let people imagine another's experience, which builds empathy.
  • Structuring decisions to limit implicit bias: blind review of applications, clear criteria decided in advance.
  • Visible norms and leadership: schools, workplaces and leaders who state and enforce that racism is unacceptable.

Not every program works. A review of hundreds of prejudice-reduction studies found that many interventions are brief and measured only in the short term, so the strongest evidence is still for sustained, structured contact (Paluck et al., 2021).

Worked example

Evaluate a contact program against the four conditions

  • Complex unfamiliar

Two neighbouring high schools, one mostly attended by students from a recently arrived migrant community, have a history of conflict. A council proposes a one-day "cultural fair" at which students from the second school visit stalls run by the first school's students, with a prize for the school whose students sell the most food.

Evaluate the proposal using the conditions for reducing prejudice through intergroup contact, and propose a better program.

5 steps

Try it yourself first, then check the working one step at a time.

Watch out for

  • Praising the fair because it brings groups together: contact alone is not enough, and competition makes it worse.

Common mistake

Contact is not enough on its own

"Put the groups together and prejudice will fall" loses marks. Robbers Cave showed that eating together did not reduce hostility, and competitive contact increased it. Always name which conditions the contact meets.

Exam tip

Describing ways to reduce prejudice

Name each condition in the syllabus's words (sustained, superordinate goals, mutual interdependence, equality of status), say what it looks like in the program described, and explain how it reduces prejudice. Support with evidence: Robbers Cave, the jigsaw classroom, or Pettigrew and Tropp's meta-analysis.

Summary

  • Contact reduces prejudice most when it is sustained, with superordinate goals, mutual interdependence and equality of status.

  • Robbers Cave (run in 1954; Sherif et al., 1961): competition created hostility; contact alone did not end it; superordinate goals did. Out-group friendships among Rattlers rose from 6.4% to 36.4%.

  • The jigsaw classroom builds interdependence, equal status and shared goals into everyday lessons.

  • Across 515 studies, more contact went with less prejudice (mean r = −.21), mainly through less anxiety and more empathy.

  • Repeated measures data need the paired t-test; correlational data cannot show cause.

  • Intercultural understanding also grows through cooperative structures, learning cultures and histories, perspective-taking, fair decision processes and clear norms.

Finished the lesson?

Try the 3 quick checks above and the lesson completes itself, or mark it complete now.

Lesson 8 of 80% read

Save your progress with a free account