Psychology › Section L · Social psychology › Topic L.5

Prejudice and discrimination

Prejudice is not only open hatred: it includes quick assumptions, biased expectations and rules that quietly disadvantage some groups. This topic separates beliefs, feelings and actions, asks where prejudice comes from, and looks at what actually reduces it.

Stereotypes, prejudice and discrimination

These three terms map onto the ABC of attitudes from the topic on attitudes. A stereotype is a belief that members of a group share certain traits, applied to individuals whatever their own characteristics ('older people can't learn new technology'). Prejudice is the feeling: an unjustified, usually negative attitude towards people because of their group membership. Discrimination is the behaviour: treating people unfairly because of their group. The three often travel together but can separate. Someone can hold a prejudice without acting on it, and an organisation's rules can discriminate even when no individual intends harm.

Forms of prejudice

Racism is prejudice and discrimination based on race or ethnicity; sexism, on sex or gender; ageism, on age, which can target young people as well as old; homophobia, on sexual orientation; and transphobia, against transgender and gender-nonconforming people. Field experiments show discrimination in action. Marianne Bertrand and Sendhil Mullainathan (2004) sent matched CVs to employers in Boston and Chicago. CVs with names common among white Americans received about 50% more callbacks than identical CVs with names common among Black Americans. In 2020 the US Supreme Court ruled in Bostock v. Clayton County that firing someone for being gay or transgender is unlawful sex discrimination.

Explicit and implicit attitudes

Explicit attitudes are the ones we consciously hold and report. Implicit attitudes are automatic associations that can operate without awareness. Because open prejudice is socially unacceptable in many settings, explicit measures often show little bias while implicit measures show more. The best-known implicit measure is the Implicit Association Test (Anthony Greenwald and colleagues, 1998), which compares how quickly people pair group labels with good or bad words. Its use is debated: individual scores shift noticeably between sessions, they predict a person's behaviour only weakly, and training that changes scores has not reliably changed behaviour. It is a research tool, not a test for labelling individuals.

Why prejudice forms

Children learn prejudice from parents, peers, media and social media, and conformity to group norms keeps it going. Henri Tajfel's minimal group studies (around 1970) assigned boys to groups on trivial grounds, such as preferring one abstract painter to another, yet they gave more rewards to their own group. Social identity theory (Tajfel and John Turner, 1979) explains this: part of our self-esteem comes from the groups we belong to, our in-groups, so we favour them over out-groups. This in-group bias is often joined by the out-group homogeneity effect, seeing 'them' as all alike, and by scapegoating, blaming an out-group for frustrations.

How stereotypes confirm themselves

A self-fulfilling prophecy is an expectation that changes behaviour in ways that make it come true. Carl Word, Mark Zanna and Joel Cooper (1974) found that white interviewers sat further from Black applicants, made more speech errors and ended interviews sooner. When white applicants were then interviewed in that cold style, they performed worse. Teacher expectations can work in a similar way: Robert Rosenthal and Lenore Jacobson's Pygmalion study (1968) is famous, though later reviews found such effects real but usually small. Confirmation bias, noticing evidence that fits a stereotype and overlooking evidence that does not, then makes the stereotype feel proven.

Stereotype threat and its replication debate

Claude Steele and Joshua Aronson (1995) reported that Black students did worse on a hard verbal test when it was described as measuring ability than when it was not. They called this stereotype threat: worry about confirming a negative stereotype about one's group, which may undermine performance. Hundreds of studies followed, including many on women and maths. The effect is now debated. Meta-analyses suggest that publication bias inflated the early literature, and a large preregistered study of Dutch secondary school pupils (Flore, Mulder and Wicherts, 2018) found no effect on girls' maths scores. Many researchers now think the effect, where it exists, is smaller and more dependent on conditions than once claimed.

Reducing prejudice

In the 1954 Robbers Cave study, Muzafer Sherif's team split boys at an Oklahoma summer camp into two groups. Competitions bred open hostility, which eased when the groups worked on superordinate goals, shared goals neither could reach alone, such as restoring the camp's water supply. Gina Perry's book The Lost Boys (2018) argues that the staff helped stir up the conflict, and an earlier 1953 attempt was abandoned when the boys did not turn on each other as planned. Gordon Allport's contact hypothesis (1954) says contact reduces prejudice under equal status, common goals, cooperation and support from authorities. A meta-analysis of more than 500 studies (Pettigrew and Tropp, 2006) found that contact generally reduces prejudice. Elliot Aronson's jigsaw classroom gives each pupil one essential piece of a shared task.

Worked example: separating the three

A manager reviewing job applications thinks people over 55 cannot keep up with new software, feels irritated when an older applicant's CV arrives, and screens out every CV that lists a graduation year before 1990. 1. The belief about software is a stereotype, the cognitive part. 2. The irritation is prejudice, the affective part. 3. Screening out the CVs is discrimination, the behavioural part, and in the United States it may break the Age Discrimination in Employment Act, which protects workers aged 40 and over. 4. Check: the belief is applied to every older applicant, whatever their actual skills; that is what makes it a stereotype rather than an assessment.

Worked example: weighing a stereotype threat claim

A school reads that reminding girls of a 'gender gap' before a maths test lowers their scores, and plans to change its test instructions. 1. Identify the claim: stereotype threat lowers performance. 2. Look at the evidence: many early experiments were small, and meta-analyses detected signs of publication bias. 3. Look for strong tests: a large preregistered study of more than 2,000 Dutch pupils found no effect. 4. Weigh costs: neutral instructions cost nothing and avoid needless worry, so the change is reasonable even if the effect is small. 5. Check: disputed does not mean disproved; the conclusion should match the strength of the evidence.

Common mistakes to avoid

First, using stereotype, prejudice and discrimination interchangeably: they are a belief, a feeling and an action. Second, assuming prejudice requires conscious hostility: implicit associations and biased rules can produce unfair treatment too. Third, treating an IAT score as proof that a particular person is biased: individual scores are unstable and weak predictors. Fourth, citing stereotype threat as settled fact: it is debated and probably smaller than early studies suggested. Fifth, thinking any contact reduces prejudice equally: it works best with equal status, cooperation and institutional support. Sixth, assuming in-group bias needs real conflict: trivial groups are enough.

Key takeaways

A stereotype is a belief, prejudice a feeling and discrimination an action towards people because of their group. Forms include racism, sexism, ageism, homophobia and transphobia, and CV experiments reveal discrimination in hiring. Implicit attitudes can differ from explicit ones, but the IAT is weak at the individual level. Prejudice is learned, held in place by norms and fed by in-group bias, which even minimal groups produce (Tajfel), and by scapegoating. Self-fulfilling prophecies and confirmation bias keep stereotypes alive. Stereotype threat (Steele and Aronson, 1995) faces a serious replication debate. Superordinate goals, structured contact (Allport, 1954) and jigsaw classrooms reduce prejudice, though Robbers Cave has its own critics.

Flashcards

Flashcard

Stereotype

A belief that members of a group share certain traits

The cognitive part, applied regardless of the individual.

Test your understanding

Question 1 of 5

A shop owner believes teenagers are likely to steal and so follows every teenage customer around the shop. Following them is an example of:

Question 2 of 5

In Tajfel's minimal group studies, boys sorted into groups by their preference for one painter over another:

Question 3 of 5

Which statement best reflects the current status of research on stereotype threat?

Question 4 of 5

According to Allport's contact hypothesis, contact between groups reduces prejudice best when:

Question 5 of 5

An interviewer expects an applicant to be weak and acts coldly; the applicant grows nervous and answers poorly, which seems to confirm the expectation. This chain is:

← Previous topicNext topic →