Keywords

Introduction

Prejudice and discrimination against social out-group members are ubiquitous phenomena in society. Feelings of being threatened by refugees, migrants, and other minorities; the successful election of individuals and political parties promoting more or less overt devaluation and discrimination of specific social groups; disadvantages and social exclusion based on religion or sexual orientation; new and old forms of anti-Semitism; overt or covert violence toward people who look different or have different cultural lifestyles; and new forms of terrorist threat—these are all well-known examples of the social realities to be found in probably every society in the world (see, e.g., Dovidio, Hewstone, Glick, & Esses, 2010).

Currently, these problems have to be faced at a time when there is an increasing need for individuals and social groups to cooperate in dealing with the realities of multicultural and diverse societies and rapidly progressing economic globalization. Basically, these problems call for a promotion of personal competencies along with an understanding of which conditions need to be encouraged to promote nonviolent and tolerant social progress and which conditions are optimal for successfully living within plural societies. The following sections present interventions designed to prevent prejudice and discrimination and to promote tolerance and social relations between different social groups in order to exert a positive influence on the aforementioned problems and bring about significant change in our societies.

Prejudice is a multifaceted construct that includes negative intergroup feelings, attitudes, and behavior (Brown, 2010) that can already be observed in children from preschool age onward (see Aboud, 1988; Raabe & Beelmann, 2011). The emergence of prejudice thereby can be viewed as a result of a complex interplay between societal, social, and individual variables (Levy & Killen, 2008). Accordingly, several social-developmental theories of prejudice have been formulated that address the role in the formation of prejudice during childhood and adolescence of, for example, cognitive and social-cognitive development (Aboud, 1988; Bigler & Liben, 2007), identity development (Nesdale, 2004), moral development (Killen & Rutland, 2011), intergroup contact and friendships (Davies, Tropp, Aron, Pettigrew, & Wright, 2011), social norms (Rutland, 2004), or intergroup threat (Bar-Tal & Teichman, 2005). Hence, this diversity of models is also reflected in a large number of different intergroup interventions, programs, and initiatives aiming to reduce prejudice and discrimination and to promote intergroup relations (Aboud et al., 2012; Beelmann & Heinemann, 2014; Oskamp, 2000; Paluck & Green, 2009; Ponterotto, Utsey, & Petersen, 2006; Stephan & Stephan, 2001; Stephan & Vogt, 2004). These approaches originate from a range of different disciplines such as social psychology and the educational, developmental, political, and media sciences to mention just a few. We limit this chapter mainly to the prevention of ethnic, racial, and national prejudice, although the approaches may generalize to other domains such as gender, disability, and sexual orientation. We categorize the measures of prejudice prevention and reduction along two dimensions addressing the intervention content and the intervention strategy (see Table 1). The content dimension (A) refers to the “what” or the theoretical background where contact interventions (A1), knowledge-based interventions (A2), and individual skills promotion (A3) can be distinguished. The strategic dimension (B) refers to the intervention methods applied and therefore to the “how” of the measures. We differentiate between educational measures (B1), standardized structured learning programs (B2), and media-based interventions and cultural events (B3). It should be noted here that the use of the term prevention in the sense of absolutely avoiding any negative evaluation of social out-groups is—as in other prevention fields (e.g., aggression)—unrealistic, at least on a population basis (in contrast to an individual level of prejudice). For prejudice, however, there is no clear threshold regarding whether a certain level of prejudice is normal, dysfunctional, or a threat to members of the social out-groups exposed to it (in contrast to, e.g., clinical definitions of behavioral disorders). In addition, there are numerous assessment methods for prejudice with unknown normative distributions and no universal definition of prejudice. Therefore, it is difficult or even impossible to make a clear conceptual distinction between the prevention (avoiding any level of prejudice ever occurring) and the reduction/remediation of prejudice (changing people with a more or less severe level of prejudice). However, from a pragmatic perspective, we use the term prevention throughout this article and restrict our presentation to any intervention aiming to reduce prejudice or (positively formulated) promote intergroup relations and do not apply it to interventions that try to remediate people with severe levels of prejudice such as political or religious extremists. Nonetheless, it has to be noted that this distinction is continuous and not categorical with a clear determination point.

Table 1 Classification of anti-prejudice interventions to promote intergroup relations with examples
Full size table

Types of Intergroup Interventions: Concepts and Evaluation

Contact Interventions (A1)

Intervention Concepts

A major group of interventions has either been developed in the context of the contact hypothesis or can be assigned to the theory’s core assumptions (see Allport, 1954; Pettigrew, 1997). These assume that contact experiences between members of different social groups lead not only to reduced levels of prejudice and discrimination against social out-group members but also to improved intergroup relationships in general. Contact interventions are based on the seminal work of Allport (1954) who also emphasized key conditions for intergroup contact: the possibility to establish personal relationships with out-group members (e.g., close friendships); equal status between groups; common goals; cooperation between groups; and support by authorities, law, or norms.

Numerous interventions based on the assumptions drawn from the contact hypothesis have been developed. These range from integrated schools and specific educational learning strategies to youth exchange programs and media campaigns. The first implementations of contact interventions in the 1950s were integrative school systems in the race-segregated United States designed to prevent prejudice and discrimination in childhood and adolescence (Brown v. Board of Education, 1954; Schofield, 1995). Inclusive classes composed of different social group members (typically different ethnic groups but also other social groups such as disabled and nondisabled) aimed to increase the frequency of cross-group interactions, thereby leading to a sustained improvement in intergroup relations.

One variation of contact interventions does not rely on direct but on indirect (or extended) intergroup contact (Wright, Aron, McLaughlin-Volpe, & Ropp, 1997). Indirect contact interventions attempt to extend the contact principle to situations in which a direct personal encounter or the necessary and supportive conditions of direct contact are difficult to establish, because, for example, relevant out-group members are not living in the local context. Mediators of such extended contacts are parents, teachers, peers, or media celebrities. Accordingly, indirect intergroup contact interventions seek, for example, to bring about extended contact experiences via interactions between in-group friends and potential out-group members or via prominent advocates of social groups (e.g., television spots with prominent soccer players from two different ethnic groups). One example of indirect contact was presented and evaluated by Cameron and colleagues (Cameron & Rutland, 2006; Cameron, Rutland, & Brown, 2007; Cameron, Rutland, Brown, & Douch, 2006). They developed comic adventure stories in which a child belonging to the ethnic majority (in this case, English children) makes friends with a refugee child, and both children share as well as solve social problems in their everyday lives. The principle—as in other extended contact interventions—is for participants to experience successful and satisfying social interactions that prevent or decrease negative out-group evaluations between salient groups. At the same time, social interactions with out-group members and the development of cross-group friendships are supposed to be promoted by in-group role models. Another contact intervention using the principle of indirect intergroup contact is the so-called coexistence programs (see, e.g., Stephan, Hertz-Lazarowitz, Zelniker, & Stephan, 2004). This term characterizes a large and diverse group of intervention programs used in the contexts of historically grown and existential conflicts between social groups (e.g., the conflict between Arabs and Jews in Israel or Catholics and Protestants in Northern Ireland). The basic idea behind such programs is to reconcile intergroup conflicts via the intermediation of reciprocal respect and the fundamental acceptance of out-group rights. For this purpose, intervention participants process and reflect on, for example, written descriptions of individual victim biographies (e.g., parents who have lost a child in a military attack). Other programs in this group also use supplementary information about the cultural history of the out-group.

The most recent form of indirect intergroup contact is to be found in the conception of imagined intergroup contact (Crisp & Turner, 2009). Imagined contact describes the mental simulation of a social interaction with one or more out-group members (Crisp, Stathi, Turner, & Husnu, 2009). These mental simulations of positive contact experiences are designed to activate the cognitive concepts associated with successful interactions with out-group members (e.g., meeting a Muslim stranger for the first time at a party or cooperating with a foreign colleague at the workplace). Thus, imagined intergroup contact should have a positive emotional impact on future expectations of positive, relaxed, and comfortable direct contacts, thereby reducing the fear of negative encounters with out-group members as well as negative attitudes toward the social out-group.

Evaluation

Interventions based on the contact hypothesis have a long tradition in social psychology and have been evaluated extensively. For example, Pettigrew and Tropp (2006) synthesized data from more than 500 studies with a total of about 200,000 participants that focused on the effects of intergroup contact on attitudes toward ethnic and other social groups such as the elderly or disabled. Overall, the authors found that intergroup contact had a significant positive intervention effect on prejudice reduction. Effect sizes ranged between d = −0.42 and −0.49, thereby indicating a 20–25% reduction of prejudice via intergroup contact interventions.Footnote 1,Footnote 22 These effects remained stable even after controlling for methodological limitations. Furthermore, Pettigrew and Tropp (2006) found that contact interventions under Allport’s (1954) optimal contact conditions (see above) have a significantly stronger effect on prejudice (d = −0.60) than interventions with unstructured contacts between groups (d = −0.42). Additionally, the authors found different contact effects depending on the salient out-group as well as age effects (contact effects appear to be stronger for children and adolescents), but no moderation by participants’ gender. In a supplementary meta-analysis, Pettigrew and Tropp (2008) examined the processing variables between intergroup contact and attitudes toward social out-groups and found that positive contact effects are mediated via reduced levels of intergroup anxiety and by increased empathy. On the other hand, increments in knowledge about the relevant out-group affected the relation between cross-group contact and prejudice to a significantly lesser extent.

Positive effects of direct intergroup contact have also been confirmed in a meta-analysis on cross-group friendships (Davies et al., 2011). These authors integrated 135 studies with 208 individual samples and found that cross-group friendships improved attitudes toward the social out-group with a medium effect of d = 0.53. This effect appeared to be independent from participants’ age, gender, or social group status. Cross-group friendships elicited smaller effects when friendships were between ethnic or racial groups and stronger effects between groups that differed in sexual orientation or religious affiliation. Furthermore, by analyzing longitudinal studies, Davies et al. (2011) confirmed a medium-sized long-term effect (d = 0.48) and showed that cross-group friendships improve out-group evaluations and attitudes especially via heightened values of time spent and self-disclosure with out-group friends, thereby indicating the significance of such behavioral engagement.

The effects of extended intergroup contact interventions have been summarized in a recent meta-analysis by Zhou, Page-Gould, Aron, Moyer, and Hewstone (2018). These authors covered 20 years of research on the extended contact hypothesis and summarized the results of 115 studies to analyze its effects on intergroup attitudes. Again, the mean effect between extended cross-group contact and intergroup attitudes was medium (d = 0.52), indicating improvements in out-group evaluations via extended intergroup encounters. Furthermore, the authors confirmed the existence of extended contact effects on improved out-group attitudes independent from direct contact experiences. Although the effect sizes decreased after controlling for direct cross-group friendships, there was still a significant extended contact-attitudes relation (d = 0.35). Regarding potential moderation, Zhou et al. (2018) found no effect of the country of study conduct, participants’ mean age, or gender, indicating that the extended contact effect holds for a wide range of applications. Another meta-analysis by Miles and Crisp (2014) focused on imagined intergroup contact in which imagining a positive interaction with an out-group member should reduce prejudice and encourage positive intergroup behavior. This meta-analysis summarized 70 studies and found a small-to-medium overall effect (d = 0.35) on intergroup bias. The interrelation between imagined contact with out-group members and intergroup attitudes was stronger the more that participants had been instructed to elaborate on the situation in which the imagined intergroup interaction was set (e.g., workplace, school, leisure activity). Imagined intergroup contact effects were also stronger for children than for adults, indicating that imagined contact may be particularly able to promote social change in educational settings.

A final meta-analysis by Lemmer and Wagner (2015) integrated studies testing the impact of direct and indirect contact programs on ethnic prejudice in real-world settings. Outcomes were assessed directly after the termination of the intervention (k = 123 comparisons with N = 11,371 participants) and in follow-up tests at least 1 month later (k = 25 with N = 1650 participants). Results indicated that direct contact interventions led to a stronger decrease in ethnic prejudice than indirect contact interventions (effect sizes between d = 0.29 and 0.41 vs. between d = 0.23 and 0.33, respectively). Furthermore, follow-up studies indicated that intergroup contact intervention effects persisted over time (follow-up effect sizes ranged from d = 0.23 to 0.35). In addition, the authors found a slightly higher mean effect size when studies were conducted in countries with severe societal conflicts such as Israel (d = 0.31) compared to other regions (d = 0.27). Finally, Lemmer and Wagner (2015) found evidence that contact interventions work better for majority group members (d = 0.38) than for low-status group members (d = 0.20).

In summary, there is convincing evidence from intensive empirical research and meta-analytical results that interventions based on the assumptions of Allport’s (1954) contact hypothesis clearly reduce prejudice toward social out-group members and improve intergroup relations. Intergroup contact effects appear to operate regardless of participants’ gender, age, social context, social status, or the duration of contact experiences. The latter finding is quite surprising, because contact interventions range between some days up to 12 months (Lemmer & Wagner, 2015). However, the intensity of interventions is at least confounded with the level of directness of contact, with direct contact interventions usually being more intensive than imagined contact (which lasts for only a few minutes; see Miles & Crisp, 2014) and eliciting stronger effects on intergroup attitudes. In addition, positive outcomes also require the realization of Allport’s (1954) contact conditions, and these can probably be implemented better within direct encounter. Independent from intensity and level of directness, intergroup contact interventions reduce prejudice more efficiently among majority group members than among minority groups and show higher effects for children and adolescents than adults. In addition, contact interventions reduce prejudice over time in different social contexts and for different social groups and therefore yield the best prospects for future implementations designed to promote intergroup relations and improve social cohesion in diversified societies.

Knowledge-Based Intergroup Interventions (A2)

Intervention Concepts

A second group of interventions is based on the idea that providing information about social out-groups and imparting positive intergroup norms and values associated with democracy, cultural diversity, tolerance, and human rights will theoretically reduce prejudice and promote intergroup relations. These approaches use social identity theory (Tajfel & Turner, 1979) to describe the impact of social categorization processes on negative intergroup evaluations. Interventions then aim to reduce prejudice by deemphasizing social categories (i.e., group members become individuated) or applying decategorization (i.e., the use of social categories is terminated or at least qualified), cross-categorization (i.e., in-groups and out-groups are formed according to orthogonal combinations of two simple categorizations such as race and nationality that lead to mixed social categorizations), and multiple classification (i.e., simultaneous identification with multiple social categories) or by creating an inclusive, superordinate social category such as a European identity compared to a national identity construction (cf. Brewer, 2000; Brown & Hewstone, 2005; Cameron et al., 2006; Gaertner & Dovidio, 2000; Mummendey & Wenzel, 1999).

Again, this group of interventions relates to a variety of different interventions and programs. One more general strategy is that taken by civic or citizen education programs such as the Active Citizenship through Technology (ACT) program (Bers & Chau, 2010) which facilitates participation in constructing a “Virtual Campus of the Future” together with other students, campus administers, and academic departments. These programs usually aim to promote citizenship experiences or normative political participation and engagement (e.g., voting, joining political parties)—naturally with a clear orientation toward principles of democracy and human rights (Manning & Edwards, 2014). Clearly, such an orientation seems to be incompatible with prejudice and discrimination of ethnic or other social out-groups. Programs are often not just information based (e.g., addressing democracy, citizenship, and other concepts); but these also try to apply democratic principles through behavioral exercises. For example, service learning programs (Celio, Durlak, & Dymnicki, 2011) combine a community service with an academic curriculum, thereby aiming to teach youth to take responsibility for the community and encourage insight into the need for participation in a democratic society.

Other more focused approaches using information-based strategies are diversity or multicultural training programs (see Garcia, 1995; Paluck, 2006; Paluck & Green, 2009). These programs seek to increase an understanding of differentness and to promote tolerance between members of different social, cultural, or religious groups by imparting information on the diversity of human cultures. This greater intercultural knowledge then is hypothesized to lead to a reduction in prejudice, resentments, and discrimination. One example is the “A World of Differences” anti-bias curriculum developed by the A World of Differences Institute of the Anti-Defamation League in New York (see www.adl.org). One version of this curriculum for children from kindergarten to grade 5 contains five units with 25 lessons and addresses issues such as “understanding my strengths, skills and identity,” “understanding and appreciating differences,” and “understanding bias and discrimination.” Other diversity programs focus on communication-related aspects, aiming primarily at persons who have to adapt to another culture for professional reasons (see Kulik & Robertson, 2008).

In contrast to diversity programs that try to establish more tolerance, antiracism programs aim to reduce and avoid severe forms of social devaluation of out-group members. In most cases, these measures provide historical information on serious violations against human rights such as the Holocaust and try to inform about the political and social backgrounds of such crimes against humanity. In addition, this type of program deals particularly with the intermediation of social values and norms such as democracy or general human rights that facilitate an understanding of the significance of a shared social value system and should decrease any motivation toward negative out-group attitudes and discriminatory behavior.

Another group of training programs focuses on unconscious or implicit biases. Implicit biases are learned stereotypes that affect automatic forms of cognitive information processing and are able to influence behavior (Dovidio & Gaertner, 1986; Fazio, 1995; Noon, 2018). Implicit bias trainings confront participants with their own biases, provide knowledge about the social functions of prejudices and discrimination, and discuss tools that can be used to adjust automatic patterns of cognitive processes and eliminate discriminatory behavior. These programs use the implicit association test (IAT; Greenwald, McGhee, & Schwartz, 1998) to assess participants’ baseline implicit bias levels before giving them unconscious bias training tasks to change or control individual implicit biases or presentations on automatic information processing and finally reevaluating their bias levels in long-term posttests (Devine, Forscher, Austin, & Cox, 2012). Training techniques cover counterstereotyping (e.g., imagining powerful women to decrease gender bias), negation trainings that encourage participants to actively reject cognitions that reinforce their biases, perspective-taking trainings, and meditation approaches especially loving-kindness meditation (LKM), which is a “Buddhist technique for cultivating unconditional kindness toward all living things” (Aspy & Proeve, 2017, p. 104).

Evaluation

When evaluating the effectiveness of inter- and multicultural intervention programs, Stephan and Stephan (1984) concluded that the majority have positive effects and reduce prejudice. An updated meta-analysis (Stephan, Renfro, & Stephan, 2004) synthesized 35 studies and found small-to-medium effects on attitudinal measures (e.g., prejudice, negative stereotypes, and sympathy toward out-group members, d = 0.25) as well as on behavioral measures (e.g., behavioral preferences and intentions such as to interact with members of the social out-group, d = 0.38). Interestingly, these effects were more than twice as large at follow-up assessments (8–64 weeks after the intervention) although only a few studies provided such follow-up data. However, increased follow-up effects are not uncommon in prevention research, and it can be assumed that certain contents require a given period of time to deliver sizable effects. Apart from this, the systematic review by Stephan, Renfro, et al. (2004) illustrates a familiar pattern: Programs were more effective when implemented in combination with real direct contacts with social out-group members.

A recent meta-analytical integration of over 40 years of empirical research on diversity trainings by Bezrukova, Spell, Perry, and Jehn (2016) addressed gaps in the previously conducted systematic reviews on the effectiveness of diversity training programs (e.g., Kalinoski et al., 2013; Smith, Constantine, Dunn, Dinehart, & Montoya, 2006). It assessed the effects of diversity training programs on four outcome dimensions over time and across different characteristics of training context, design, and participants. The four outcome dimensions were the following:

  1. (a)

    Cognitive learning (referring to how far participants acquire knowledge about cultural diversity);

  2. (b)

    Behavioral learning (the development of participants’ skills in, e.g., situational judgment or objective behavior);

  3. (c)

    Attitudinal/affective learning (changes in participants’ attitudes toward diversity); and

  4. (d)

    Reactions of participants (feelings toward an instructor or toward the training overall).

The analysis integrated 260 studies of 29,407 participants reported in 236 research articles. The largest effects of diversity trainings were on participants’ reactions toward training instructors (e.g., their competence, credibility, and experience) or the likelihood of content transfer in daily routines (d = 0.61) followed by cognitive learning (d = 0.57), behavioral learning (d = 0.48), and attitudinal/affective learning (d = 0.30). The authors also calculated effect sizes for delayed posttest effects (up to 24 months after diversity training) showing that only cognitive learning outcomes were maintained over time. Furthermore, results revealed that diversity training effects were larger when accompanied by other diversity initiatives (e.g., diversity training within integrative educational contexts) targeting both awareness and skills development as well as when they were conducted over a sustained period of time and training groups contained a greater proportion of women. In sum, diversity trainings contribute to the reduction of prejudice and discrimination against out-group members by providing information (and sometimes skills) with which to adjust to and cope with increasing diversity of societies.

Quite similar evidence has been obtained for antiracism programs (McGregor, 1993). Albeit the scarcity of research in this field, this meta-analysis of seven controlled studies found a moderate effect size (d = 0.48) on the reduction of racist attitudes. Besides the scarcity of evaluations, reviews on antiracism programs fail to demonstrate whether and how such programs can be applied to and influence members of risk groups (e.g., adolescents who already have contacts with extreme right-wing groups).

This criticism also applies to the large number of civic/citizen education and service learning programs, although recently conducted systematic reviews found some evidence of their efficacy. Nonetheless, these reviews focus mainly on effects on civic engagement such as voluntary services in the community and political participation such as voting (Geboers, Geijsel, Admiraal, & ten Dam, 2013; Lin, 2013; Manning & Edwards, 2014) or on attitudes toward the self, school, and learning and on social skills and academic performance (Celio et al., 2011), but not on intergroup attitudes and relations. Accordingly, it is difficult to draw conclusions on the efficacy of such programs in terms of prejudice reduction and decreased levels of discrimination. However, Lösel, King, Bender, and Jugl (2018) confirmed an orientation toward democratic values as a protective factor against political and religious ideologies and violent radicalization. Hence, it can be assumed that civic or citizen education is probably an efficacious measure for reducing prejudice and discrimination as well.

Regarding the effectiveness of implicit bias trainings, Lai et al. (2014) investigated 17 intervention studies on implicit prejudice (total N = 17,021). Interventions used a broad variety of training techniques, such as counterstereotyping, perspective-taking, empathy training, or evaluative conditioning methods. Non-Black US participants were evaluated on their preferences for Whites compared to Blacks via IAT pre-posttests and self-reported racial attitudes. The average effect of implicit bias trainings was d = 0.36, whereas trainings using counterstereotypical methods (d = 0.38), intentional strategies to overcome bias (d = 0.38), or evaluative conditioning (d = 0.27) were especially effective in reducing implicit preferences. Interventions applying perspective-taking methods (d = −0.01), approaches to egalitarian values (d = 0.05), or emotion induction (d = 0.06) tended to be ineffective in altering implicit prejudices. However, these effects account exclusively for change in implicit bias values. No intervention was able to reduce explicit forms of racial prejudice. Furthermore, it is questionable whether IAT-measured implicit prejudice reveals small (Greenwald, Poehlman, Uhlmann, & Banaji, 2009) or even any effects at all (Oswald, Mitchell, Blanton, Jaccard, & Tetlock, 2013) on explicit attitudes or behavior.

Individual Skill Promotion (A3)

Intervention Concepts

A third heterogeneous group of interventions to prevent or reduce prejudice places more emphasis on training and promoting individual competencies such as multiple classification (see above), perspective taking, empathy, conflict resolution, or social competencies in general. Approaches are based on social-cognitive developmental theories on the origins of prejudice and discrimination (Aboud, 1988; Raabe & Beelmann, 2011) as well as on social learning theory (McKown, 2005). These concepts assume that individual deficits in the aforementioned cognitive or social-cognitive competencies encourage prejudice and discriminatory behaviors, whereas tolerant attitudes, in contrast, are strengthened by the promotion of these competencies (Beelmann & Heinemann, 2014).

A series of interventions can be differentiated on the basis of these assumptions (Aboud & Levy, 2000). One fundamental approach refers to Aboud’s (1988) contribution on the development of prejudice that focuses on multiple classification skills. Children are trained by delivering social information (e.g., characteristics of minority group children) in order to learn that individuals have multiple affiliations to different or varying social groups (Aboud & Fenwick, 1999). Other authors have tried to promote empathy and perspective-taking skills as significant correlates of intergroup-related attitudes (Miklikowska, 2018; Stephan & Finlay, 1999). These approaches range from simple perspective-taking and empathy exercises (e.g., describing discriminated persons and imagining their feelings; see Finlay & Stephan, 2000) to simulations of one’s own discrimination experiences. The latter uses, for example, the relatively prominent social psychological “blue eyes-brown eyes” simulation (see, e.g., Steward, LaDuke, Bracht, Sweet, & Gamarel, 2003). This method works with arbitrary group divisions (originally based on eye color—hence the name of the technique—but other attributes can be used instead) and the simulation of group advantages or disadvantages (e.g., the devaluation of out-group members). It allows children, adolescents, and even adults to experience the feeling of being discriminated against and should lead to favorable attitudes and own behavior when engaging in social relations with members of different groups. Nonetheless, from an ethical point of view, the experience of discrimination is not completely unproblematic, because fake discrimination events may have negative psychological consequences in participants (e.g., increased feeling of anxiety about future encounters with out-group members or anger reactions).

A relatively unspecific approach to the prevention of prejudice is the promotion of social competencies, especially by training problem-solving and conflict resolution skills. Contrary to the aforementioned interventions, these programs train general social behavior without a specific focus on the reduction of prejudice and discrimination. However, social skills in problem solving and conflict resolution facilitate nonviolent behavior in conflict situations and enhance the capacity to solve social problems between groups in diversified societies. Training in conflict resolution skills, for example, is used in adult mediation training programs and in elementary schools to train coping with peer conflicts (see, e.g., Johnson, Johnson, Dudley, & Acikgoz, 1994; Johnson, Johnson, Dudley, & Magnuson, 1995; Sandy & Cochran, 2000).

The same mechanism works for a vast number of social training programs (see Durlak, Domitrovich, Weissberg, & Gulotta, 2015) and programs to prevent antisocial behavior (e.g., aggression, violence, delinquency, or crime). However, the focus of these approaches is seldom on preventing group-based aggression (see, for reviews, Beelmann & Raabe, 2009; Farrington, Gaffney, Lösel, & Ttofi, 2017). Current publications deal with hate crimes and crimes in the context of extremism (see Beelmann, 2014; Borum, 2014) and highlight how pejorative attitudes, prejudice, and discrimination influence violence and delinquency. In this area, however, prevention research is scarce (International Center for the Prevention of Crime, 2015).

Evaluation

Measures related to skills acquisition have proven to reduce prejudice and discrimination or foster tolerance toward human diversity. Numerous empirical evaluations reveal the effectiveness of cognitive and social cognitive programs especially among children, adolescents, and adults (Aboud & Levy, 2000). However, most of the effects of these diverse approaches and programs do not address biased out-group attitudes, beliefs, and discrimination tendencies but focus mainly on improving cognitive skills (e.g., Bigler & Liben, 1992; Katz & Zalk, 1978; Schaller, Asp, Rosell, & Heim, 1996); perspective taking and empathy (e.g., Feddes, Mann, & Doosje, 2015; Stephan & Finlay, 1999; Steward et al., 2003; Weiner & Wright, 1973); or moral development, problem-solving, social and conflict resolution skills (see Beelmann & Lösel, 2006; Garrard & Lipsey, 2007; Johnson & Johnson, 1996; Lösel & Beelmann, 2005).

However, one meta-analytical review by Beelmann and Heinemann (2014) summarized the effectiveness of a variety of structured programs designed to promote individual competencies in reducing prejudice and improving intergroup attitudes in children and adolescents. The overall effect size for 45 studies on cognitive and social-cognitive training programs was moderate (d = 0.40). Trainings in perspective taking and empathy as well as in social skills yielded the strongest effects in terms of reduced levels of prejudice or improved attitudes toward out-groups (both d = 0.50), followed by trainings on moral development (d = 0.36), interventions promoting problem-solving skills (d = 0.20), and trainings in classification/social categorization (d = 0.16). Hence, individual training in perspective-taking, empathy, and social skills seems to offer one of the best ways of reducing prejudice and discrimination—at least in childhood and adolescence.

Educational Measures (B1)

Concepts

A number of interventions have been designed for the educational context. Two particularly well-known concepts are integrative schooling (see above) and bilingual education. The latter later should lead to enhanced knowledge of other languages and cultures and finally to more acceptance of the respective social out-groups (see Stephan & Stephan, 2001). However, cooperative learning techniques are used more frequently to promote intergroup attitudes. These techniques are applied when students work together on tasks within heterogeneous groups (e.g., in terms of race), and their work assignment is arranged in such a way that all group members have to cooperate to achieve a high overall performance. This arrangement is hypothesized to promote not only group cohesion (and finally intergroup relations) but also academic performance.

Different forms of cooperative learning techniques have been developed such as the Student Teams-Achievement Divisions (Slavin, 1990), the Learning Together Techniques (Johnson & Johnson, 1994), or the Jigsaw-Technik (Aronson & Patnoe, 1997). These differ in whether, for example, the individual performance of each student can be evaluated afterward or whether these establish a competition between the small groups within the classroom. However, independent from the different methods and from possible implementation problems (e.g., strain on underachievers), these cooperative learning techniques do meet the conditions for successful intergroup contact (i.e., mutual task, same status within the contact situation, support by authorities).

Evaluation

Reviews have reported mostly positive evaluations of the effectiveness of educational measures although simultaneously identifying some critical aspects. For example, Schofield’s (1995) review of the effects of integrative schooling found that interethnic contacts within schools lead to positive intergroup attitudes only when programs succeeded in initiating personal relationships and friendships between group members (Aboud & Levy, 2000). This has led to modifications of concepts of integrative schooling over the years (see Pfeifer, Brown, & Juvonen, 2007; Schofield, 2006; Schofield & Hausmann, 2004) by, for example, combining them with additional methods or advanced training for teachers.

Only a few studies have evaluated the effects of bilingual education systematically and then mostly in the context of ethnically integrated schools. Although these also revealed some positive effects on prejudice reduction (see, e.g., Genesee & Gándara, 1999), it is difficult to see how their effects could be generalized to other intergroup constellations such as the relation between handicapped and nonhandicapped people. In contrast, the outcome of cooperative learning techniques has been subject to intensive evaluation research (Johnson & Johnson, 1989, 2000; Slavin, 1995). For example, Slavin and Cooper (1999) confirmed a significant reduction in prejudice in elementary and secondary school children independent from the type of cooperative learning techniques applied. After summarizing over 180 studies, Johnson and Johnson (1989, 2000) concluded that—compared to individual and competition-oriented techniques—cooperative learning techniques not only reduce prejudice but also promote interpersonal attraction between members of different social groups in the classroom. The effect sizes for cooperative learning techniques were large (e.g., d = 0.66) and especially pronounced for intergroup contexts with handicapped children. Independent from these differential outcomes, cooperative learning techniques had positive effects on academic achievement and the students’ general satisfaction with their school.

Standardized Training Programs (B2)

Concepts

A second intervention strategy encompasses more or less standardized or manualized training programs. These concepts are mostly highly structured, consecutive, and designed to teach and practice concrete cognitive or social competencies. For example, Beelmann, Saur, and Ziegler (2010) developed a 15-session multimodal training program for elementary school children based on developmental risk factors and processes of prejudice and other forms of negative intergroup attitudes (see Beelmann, 2011; Raabe & Beelmann, 2011). The program covers three domains: (a) reading and discussing indirect contact stories as proposed by Cameron et al. (2006), (b) imparting intercultural knowledge (e.g., by taking a fictitious world trip), and (c) promoting those cognitive and social-cognitive competencies that correlate negatively with prejudice and discrimination (e.g., empathy, perspective taking, multiple classification, social-problem solving).

Evaluation

A meta-analytical review conducted by Beelmann and Heinemann (2014) provides some insights into the effectiveness of a variety of standardized training programs for reducing prejudice and improving intergroup attitudes. The authors summarized the results of 81 international research reports with 122 intervention-control comparisons of structured programs designed to reduce prejudice or promote positive out-group evaluations in children and adolescents. They analyzed contact programs (direct, indirect), knowledge acquisition trainings (on out-groups, values, and norms), and—as mentioned above—social-cognitive trainings or combinations of these three types of program. The overall effect on reducing prejudice was small to medium (d = 0.30). In comparison with knowledge acquisition programs and social-cognitive skill programs, intergroup contact interventions yielded the strongest effects in terms of decreasing the level of prejudice (d = 0.43). However, the effect size for social-cognitive trainings was only slightly lower (d = 0.40). In particular, the targeted out-group moderated the effects of trainings on intergroup attitudes, with programs addressing attitudes toward persons with disabilities eliciting the highest effect sizes and proving to be even more effective than interventions addressing attitudes toward ethnic out-group members. No further characteristics of trainings (e.g., duration, intensity rating) or characteristics of participants (e.g., age group, gender, in-group) accounted for further effect-size variability, indicating that the programs are suitable for a broad range of applications. However, in line with Pettigrew and Tropp (2008), the authors (Beelmann & Heinemann, 2014) found that contact interventions showed the strongest effects when combined with social cognitive trainings on, for example, empathy and perspective-taking skills. The advantages of multimodal or combined prevention strategies could be confirmed by follow-up data from the aforementioned program by Beelmann et al. (2010). Follow-up assessments made 1 and almost 5 years after the termination of the program revealed high effects on the children’s intercultural knowledge and medium effects on their prejudice and tolerance toward ethnic out-groups compared to an equivalent control group (Beelmann, 2018; Beelmann & Karing, 2015).

Media-Based Interventions and Cultural Events (B3)

Concepts

A final intervention strategy is based on the media (print and visual) or cultural events. Once again, this covers a broad variety of provisions such as public campaigns or festivals. Such interventions are used widely and seem to be among the most popular practical approaches for lowering prejudice and discrimination. Their popularity is based on the assumption that the reasons for prejudice and discrimination are information deficits or low problem awareness and that such broadband and cost-effective strategies are a good way to tackle these problems. Therefore, the aims of media-based interventions and cultural events are twofold: on the one hand, a broad distribution of information on, for example, discriminative social groups or social or political grievances such as injustice and, on the other hand, films, television series, and spots as well as cultural events to promote responsiveness in target groups and enhance problem awareness within the general public and society. Examples for media-based interventions are public awareness campaigns with celebrities from sport or entertainment as conducted, for example, during the soccer World Cup to counteract racism at sport events. Other examples are films or television series that impart information on social out-groups and apply forms of indirect or extended contact (e.g., contact between children of different ethnicities in Sesame Street; see Cole, Labin, & del Rocio Galarza, 2008). More recently, Bilali and colleagues applied an audio-based intervention (so-called radio drama) in different countries in Africa (see, e.g., Bilali & Vollhardt, 2013; Bilali, Vollhardt, & Rarick, 2016). This intervention contains entertaining stories on fictional interethnic conflicts that try to enhance the understanding about the development of intergroup conflicts and give role models to learn about avoiding prejudice and violence and opportunities for reconciliation.

Evaluation

The relative shortage of systematic evaluations also applies to media-based interventions, public campaigns, and cultural events—despite their relatively high popularity. However, some analyses of public campaigns indicate not only some positive effects but also negative side effects such as sensitivity effects (e.g., participants become worried about a topic that they did not know about before) and increased feelings of threat (see, e.g., Vrij & Smith, 1999). These are important findings, because public campaigns are disseminated broadly by definition, and could therefore also have broad negative effects. For example, it is conceivable that the large-scale dissemination of campaigns with multicultural content may lead to increased feelings of being threatened by foreigners in certain population groups. Therefore, differential and further deliberations are necessary to avoid negative side effects of media and public campaigns—at least in the area of prejudice prevention (Winkel, 1997). Evaluations of television series are also difficult to summarize, although these measures are widely used, at least in the United States (Persson & Musher-Eizenman, 2003). An older narrative review by Graves (1999) did reveal some cautious positive evidence. The main problem was uncertainty about the intensity of the intervention (i.e., who has viewed, for example, a certain television film or series) and whether or not this does indeed lead to more interethnic contact and less prejudice. However, in recent studies, Bilali and colleagues showed that a radio drama intervention in African countries was able to reach a high proportion of the population and had a significantly positive effect on several intergroup attitude measures (Bilali & Vollhardt, 2013; Bilali et al., 2016).

Summary and Conclusions

Various interventions have been designed to reduce prejudice and promote intergroup relations. They take the form of either direct or indirect (extended) contacts between members of different social groups; are based on information about these groups; or aim to promote social, cognitive, or social-cognitive competencies that correlate empirically with intergroup attitudes and behavior. Some have been evaluated extensively; others still lack convincing evidence (see Aboud et al., 2012; Oskamp, 2000; Paluck & Green, 2009; Stephan & Stephan, 2001). Especially positive are the effects of contact interventions in general along with cooperative learning methods and prevention programs that foster empathy and perspective taking. Results on other programs such as multicultural trainings and civic or citizen education or value education are also promising. However, despite these positive conclusions, there are also limitations and methodological caveats. The most challenging is the lack of long-term evaluations showing stable and long-lasting effects on prejudice and intergroup relations. In addition, most evaluations measure assessed effects on intergroup knowledge and attitudes or behavioral intentions but not on intergroup behavior or further outcomes such as intergroup friendships. Further problems concern the need to go beyond pilot projects in order to implement and disseminate programs in routine social settings such as schools or communities (Beelmann, Malti, Noam, & Sommer, 2018; Malti, Noam, Beelmann, & Sommer, 2016). The last 20 years of prevention research confirm impressively that it is not just the content and methods of interventions that lead to the intended outcomes but also the context and implementation conditions (Durlak & DuPre, 2008). Hence, more intensive intervention research is needed to develop programs that are effective in real-world settings. Finally, there are still insufficient links between programs addressing intergroup relations and research in related fields such as crime prevention or the prevention of radicalization and violent extremism (Beelmann, 2014). More integrative research combining results from diverse research fields should lead to more promising concepts and approaches. For example, programs are still insufficiently linked to developmental knowledge and on the risk factors involved in prejudice, discriminative behavior, and more general problems such as antisocial behavior and crimes (Farrington, Gaffney, & Ttofi, 2017; Nivette, Eisner, & Ribeaud, 2017; Rutland & Killen, 2015). Such synergies will be necessary to exploit the full potential of the programs and interventions described here and may significantly reduce the real societal problems reported in the introduction.