What works to reduce prejudice and discrimination? - A review of the evidence

Section Four: Prejudice-Reduction Interventions

Having summarised some of the important theoretical contributions to prejudice-reduction, I will now present a summary of the main types of interventions with evidence on effectiveness, drawing on case studies and suggesting some principles which may be usefully applied elsewhere. Again it is vital to note that the case studies are not intended to be directly applicable to prejudice-reduction in Scotland. For instance, some of them talk about successful interventions to improve intergroup relations in post-conflict societies, which may be dealing with tensions that often spills over into actual (violent) conflict, and we may also assume that these are likely to be more 'reactive' than preventative. However, it may be appropriate to apply some of the 'universal principles' emerging from these to future strategies.

A mixture of lab-based interventions and evaluations of prejudice-reduction initiatives 'in the field' make up the growing literature on 'what works', however the majority of studies are controlled and experimental, have taken place in psychology laboratories, and often with psychology students as participants. Fewer studies take place in 'real-life', in schools or communities for example.

For the purposes of summary and analysis, the interventions that are most frequently studied and that are useful for this report can be roughly divided into three categories:

1. Educational strategies (including but not limited to school-based interventions)
2. Short-term diversity training courses
3. Media campaigns

1. Education and Re-education

Unsurprisingly, education has long been a key area of interest for scholars in all disciplines who have looked at 'what works' to reduce prejudice. Educational prejudice-reduction initiatives build on contact theory through the premise that activities such as cooperative learning; discussion and peer influence; instruction; and multi-cultural curriculum will help to reduce prejudice in a way that contact alone might not be sufficient to. Educational initiatives are concerned with promoting positive relations through challenging stereotypes and 'myths' about out-groups. This may involve groups being in direct contact with each other, for example pupils from different faith schools taking part in shared learning, or peer discussion between different groups on topics that might be said to create divisions and tensions (such as certain historical events).

Some educational initiatives may draw on 'extended contact' principles such as empathy and perspective taking, and might take the form of vicarious/imagined contact. These techniques may be useful for more 'hidden minorities', and situations where direct contact is either impractical (for instance, when dealing with prejudice against transgender people, who make up a very small proportion of the population) or might prove problematic (such as in post-conflict societies).

This section of the report outlines various case studies of these principles being implemented in prejudice-reduction interventions. Although much of the existing research in this area is lab-based, there is value in also highlighting those interventions that were carried out and evaluated 'in the field'. The case studies selected were sampled from a large number of interventions, on the basis that they cover different international contexts, different age groups, and deal with different types of prejudice: as such, they are intended to be merely an indication of the types of studies that exist. The final point to make is that 'diversity training', in the sense of short-term initiatives which often take place in corporate workplaces, is dealt with in a separate section, as the principles discussed in this section tend to focus on mid-long term educational strategies, and tend to be aimed at younger people rather than adults.

Shared education curriculum

A recent intervention with a rigorous longitudinal evaluation is the 'Promoting Reconciliation through a Shared Curriculum Experience Programme' report, published 2013. Undertaken by the Centre for Effective Education at Queens University Belfast, the study was a two-year evaluation of the above programme, which was designed to address the "propensity of teachers to avoid controversial issues relating to sectarianism and the conflicted past in Northern Ireland" (1). The study was a clustered randomised controlled trial involving 27 primary and secondary schools in Northern Ireland, with a total of 840 children taking part. 12 'lessons' were delivered (by teachers) over a 6 month period, and the evaluation included pre and post-test questionnaires, interviews with teachers, focus groups with students, and observations. The programme was carried out in a 'curriculum only' or 'contact & curriculum' basis, to test the 'contact' effect (shared learning) as well as the impact of talking about the issues.

Findings were positive, in terms of children learning about people from other religious backgrounds, and signs of improved intergroup relations. The role of the facilitator (the teacher) was noted as very important. The study seemed to show support for Walsh (1988) argument about critical thinking in education, as participants became more critical of the in-group (perhaps questioning old assumptions). The report claims that the intervention helped to challenge everyday understandings about 'outgroups', particularly in the context of Catholic-Protestant relations. This might support the argument that acknowledging and discussing historical events would be helpful in terms of breaking down existing barriers and challenging the residual prejudice apparently stemming from historical conflict and poor relations.

Some limitations were noted. Firstly there is a question over long-term impact, which is the case with virtually all studies of this type, even if they show encouraging results. Results from this study also suggested that effects were different when lessons were delivered in single group versus contact settings. 'Shared learning' involving contact between groups at times actually counteracted the benefits from the curriculum. This does not suggest that contact in general is not beneficial, however it may be that when confronting history and attempting to challenge prejudices against out-groups, there are advantages of delivering this in single settings. This question is worthy of future research, and where possible a balance should be sought.

Secondly, one concern raised by teachers taking part in the study in Northern Ireland was that by raising sectarianism as an issue, it could in a sense worsen the situation by creating a problem where one does not exist:

"…many children, particularly those from more rural areas were 'unaware' of sectarianism and found the concept difficult to grasp; that the programme appeared to direct children to 'defend their own culture' instead of 'accepting the culture of others'; and in so doing, 'encouraged sectarian identifications".

Some of the feedback included use of symbols that children may not understand, for example paramilitary symbols. There are challenges associated with this type of intervention, as there is danger of essentialising group categories (Bekerman & Zembylas 2011). Again this is something that has to be carefully considered when designing, implementing, and monitoring prejudice-reduction initiatives based on intergroup theories. It would suggest that regular feedback is sought as part of on-going evaluation of projects, and acted upon when necessary in terms of changing content or delivery style. Notwithstanding the risks associated with transferring any policy from one jurisdiction to another, some of the principles raised in this intervention might be useful if applied carefully elsewhere.

Sharing perspectives: conflict resolution

Unsurprisingly, many of the key prejudice-reduction interventions have taken place in areas in which ethnic or other prejudice results in or is exacerbated by overt conflict, or at least has done in recent times. As such, quite a lot of literature on the topic of teaching history / education focuses on post-conflict settings. In relation to contexts with less overly problematic intergroup relations, such as Scotland, we of course have to be wary of what conclusions might be drawn from 'what works' in these settings, however there may be useful strategies that could help to influence prejudice-reduction initiatives more broadly.

With findings published in 2008, 'Enabling Adolescents in Culturally Diverse Environments to Peacefully Resolve Ethnic Group Conflicts' was a project based on the idea that change is best delivered through small groups. The project brought different ethnic groups together at two diverse Midwestern High Schools in the United States^[2]. The programme was designed to maximise the benefits and diffuse the potential risks of contact, based on 'intergroup dialogue programmes' combined with 'conflict mediation'. Over a three-year period and with a total of 178 participants, school students explored dynamics of intergroup relations (in their own school and with another school) by exploring stereotypes, and examining their attitudes towards others and vice versa. The project evaluation was based on pre and post-test surveys, as well as qualitative interviews. Among the key findings was a reduction in prejudiced attitudes and stereotypes, reports of new friendships, and more knowledge. The authors note that crucial to the success of the intervention was careful choice of facilitators; involving those who had previously completed the programme to help run it the following year; careful attention to feedback; and the collaboration of researchers, practitioners (in this case teachers), and participants (in this case students).

In the same volume, Bargal (2008) describes the effects of an intervention with Israeli and Arab youth in Israel, which focused on reducing conflict and negative stereotypes between the two groups. Like the Michigan University project outlined above, the intervention was based on the principles of Lewin's 'reeducation' theory. Youth from both groups were recruited to participate in a three-day conflict management workshop, and participants dealt with issues such as intergroup conflict, stereotypes, prejudice, and discrimination. Central to the intervention was discussion of the complexity of the Arab-Jewish conflict, and the opportunity for both groups to discuss their personal experiences of living with conflict. According to Bargal, group facilitators played a crucial role:

"Facilitators point out the similarities as well as the differences between the two groups. They emphasize the overall need for national group identity and the importance of each group's unique history. (p. 56)"

Clearly, provision of accurate information is important, and a setting which facilitates debate and discussion of what might be considered contentious issues increases the potential efficacy of an intervention. Bargal noted that there had to be high levels of motivation to take part, emphasising that people have to want to at least be open to challenging their own biases. Moreover, it was concluded that potential effects in terms of attitude change could have been encumbered by the fact that the intervention was short term.

Peer-based learning

The final real-life case study is an intervention developed and tested as part of a psycho-educational initiative at various universities across the United States. Souweidane's (2012) 'An Initial Test of an Intervention Designed to Help Youth Question Negative Ethnic Stereotypes' was based on perspective-taking principles and the idea of reducing prejudice by challenging stereotypes. 192 high school students (from two schools with high concentration of Arab American and Jewish American students) took part, and were divided into 'immediate intervention', 'delayed intervention', and 'control group'. Pre and post-test surveys as well as observations were used to test effectiveness of the intervention. Part of the activities included using media to talk about stereotypes - for example, looking at websites such as 'Facing History and Ourselves' to learn more about the history of anti-Semitism.

The results from the study were encouraging in terms of improving tolerance and positive relations, and improvements in negative stereotypes. Findings chime with some of the contact and education/reeducation theories outlined in the previous section, highlighting the value of grounding interventions in theory. The intervention was said to have positive effects on participants' critical thinking (which Walsh 1988 argued is crucial to reduce prejudice), and this was especially the case for peer educators. Particularly promising was the effectives of peer-based learning, which is supported by social learning theory and action research. Young people may play an important role in helping their peers confront and address negative ethnic stereotypes. Involving young people as leaders and educators in interventions targeting young people has numerous merits acknowledged in the literature (cf. Stukas et. al., 2000). This theory was supported by the example outlined here:

"The present study has also provided evidence supporting the positive impact of promoting youth to become engaged in interventions targeting attitude and behavioral changes among their peers. Research on this topic may be enhanced by adopting empowerment theory practices in the development of an intervention. The significant improvement among the peer educator group supports this approach. This study has demonstrated that empowering youth to take on a leadership role, such as a peer educator, positively affects the youth leader. Future research may want to focus on engaging youth in interventions targeting them so that we gain greater understanding of the youth educator role effect and so that improved outcomes may be achieved" (123-4).

Some limitations included that the intervention was limited to four sessions, and the author suggests that more time (for example, a semester-long class) would be more likely to affect change (122).

Lab-based study: 'A prejudice habit-breaking intervention'

Although this report focuses on real-life interventions in order to get a sense of what might be most straightforwardly transferred to other contexts, findings from lab-based studies can also be applied (carefully) elsewhere.

A particularly useful lab-based intervention to include as a case study is Devine et al (2012) 'Long-term reduction in implicit race bias: A prejudice habit-breaking intervention'. The three month longitudinal study aimed to determine whether interventions can have long-term effects in terms of reducing implicit biases. The intervention took place in the University of Wisconsin, USA, with 91 non-Black participants, and the authors claim that their study "is the first to our knowledge to produce long-term change in implicit bias using a randomized, controlled design" (1276). As discussed in section 3 of this report, the existence of implicit biases mean that even people who are consciously non-prejudiced and consider themselves to be supportive of equality can unintentionally act in prejudiced ways towards others. As Devine et al (2012) comment, "this process leads people to be unwittingly complicit in the perpetuation of discrimination" (1267).

In this study, the authors seek to build on promising but limited results from what they term 'easy-to implement strategies' such as perspective taking and imagining counter-stereotypic examples. They note that reductions in prejudice from such short-term interventions are "likely to be highly contextual and short-lived" (1268). Participants were therefore engaged in a long-term process, with intentional efforts to overcome biased responses. The intervention was multifaceted, with a bias education and training programme complementing the use of the Black-White Implicit Associated Test (IAT). The results of the IAT were as anticipated: a high proportion of participants (in this case 90%) implicitly favoured White people over Black people. Being confronted with this evidence was thought to increase awareness of bias, and this awareness was developed through the other aspects of the intervention, such as the training section which provided participants with strategies to overcome these in everyday settings (for example through perspective taking and stereotype replacement). Importantly, the study noted improvement of attitudes over time, perhaps as people became increasingly self-aware and used the strategies taught to overcome instances of prejudice.

Of course it is important to note the limitations of the study. Participants were all psychology students, common in lab-based studies but controlled for as best as possible. It is also unclear how easily this type of intervention would be administered in real-life - perhaps in school settings it might be more feasible than other settings. Another key issue is to think about who the interventions are targeting and who they are likely to be missing. Most people do not consider themselves to be prejudiced so whether they would commit to a goal of 'breaking the prejudice habit' is questionable. However, it may be the case that such strategies are still useful in that they address the problem of people wanting to be tolerant and free of prejudice, but still holding implicit bias. Given that we have already established the difference between many people's intentions regarding equality and their attitudes to the implementation of measures which aim to actually tackle inequality, it is likely that such interventions would address an important discrepancy.

2. Short-term 'diversity training' courses

In many ways, the interventions discussed in the previous section could be termed 'diversity training' because the objective is to help people value diversity, as opposed to fearing difference - a key cause of prejudice. This section, however, looks at more short-term and isolated diversity training programmes, rather than focused and longer term interventions targeted at certain populations. These often take place in corporate workplaces, and with adults as opposed to children and adolescents, though some do focus on younger people. This type of training comes in many forms, with some 'instructional' in nature such as showing movies or delivering lectures, and others encouraging interactive activities such as role plays and discussions. Diversity training may involve group discussions about 'difference', based on the same values which are at the heart of educational initiatives: overcoming ignorance; expressing hidden assumptions; and feeling empathy for other groups or individuals (Paluck 2006: 581). Diversity training is an industry with huge levels of investment, yet as Abrams (2010) comments, there is "almost no adequate evaluative research" (74). Following a discussion of the general theoretical concerns with short-term diversity training, this section will draw on two case studies of applied prejudice-reduction interventions to explore the strengths and weaknesses of this type of approach. The first empirical example took place in Australian workplaces with adults, and the second in the UK with children and adolescents in school and community settings.

'Backlash' and other limitations

A central criticism regarding diversity training programmes is that they are rarely "guided by the theoretical models of learning or prejudice reduction" (Paluck and Green 2009: 354). Pendry et al (2007) further highlight the separation between theory and practice as they comment that despite diversity trainers and social psychologists having similar objectives (i.e. improvement in intergroup relations and reduction in prejudice), "they currently operate in a fairly separate fashion with limited dialogue (28)." This disconnect is likely to result in more 'piecemeal' initiatives compared to educational programmes which may be grounded more thoroughly in theory.

Moreover, diversity training programmes are often considered to have potential 'backlash' effects, perhaps as a result of the 'blanket' designs often applied, the short-term nature of most of these initiatives, and delivery not always being sensitive to its environment. There is a strong suggestion that programmes can reinforce inequalities/discrimination felt by minority participants by drawing attention to difference. As discussed in the previous section, discussing group difference can be positive in terms of improving attitudes towards out-groups, however it is important that these discussions are handled carefully. Paluck (2006) suggests that diversity training courses might reinforce stereotypes, and actually 'backfire' by increasing, renewing or even fostering new sensitivities. Plaut et al (2011) suggest that majority participants may also in some cases feel excluded, for example if the emphasis is put on the celebration of minority cultures.

Yet a 'colour-blind' approach which suggests that everyone is equal is similarly problematic. As Abrams (2010: 72) comments, we know that everyone is not equal; there remain huge inequalities in all societies. Therefore, initiatives that 'pretend' everyone is equal and do not highlight difference and inequality might be seen to lack credibility and sophistication. Pendry et al (2007) point out that diversity training "differs from the superordinate concept of diversity management in that it does not necessarily imply any background change in system-level structure, decision making or organization ethos" (28). This is important: an organisation with management dominated by middle-class white men compelling its staff to attend 'diversity training' may appear insincere if a commitment to diversity is not shown in the institution as a whole.

Some general limitations of diversity training courses which are similar to those highlighted in the educational initiatives section are also worth mentioning. Firstly, diversity training programmes are often not evaluated at all, or are evaluated by participants directly after sessions, making it impossible to track any long-term effect on attitudes or behaviours. It is also important to reiterate the point that real change is only possible if people are motivated to change:

"Unfortunately, field research on prejudice reduction does not have much to say about influencing those who do not sign up for anti-prejudice interventions (Paluck and Green 2009: 352).

This report so far has emphasised the point that people have to want to overcome prejudice, and that meaningful change will generally only occur over time. It is questionable whether compulsory attendance at a workplace 'diversity' training course, for example, which may be one-day in length, and often shorter, would satisfy this criteria.

Tackling racism in Australia

One of the few academically-evaluated applied prejudice reduction programmes was published in 2001, the culmination of research in Australian workplaces in the 1990s: 'Stereotype Change and Prejudice Reduction: Short- and Long-term Evaluation of a Cross-cultural Awareness Programme' by Hill and Augoustinos. The Cross-Cultural Awareness programme was an anti-racist educational course used in South Australia in various institutions including some government agencies. Staff attended a three-day training programme on either a compulsory or voluntary basis, depending on the type of role. The programme's objective was to reduce prejudice towards Aboriginal Australians, a group frequently stereotyped, stigmatised, and discriminated against, and to promote knowledge and appreciation of indigenous culture.

It is important to point out the methodological limitations of the study. As well as the small sample (62 participants), there was no control group and the study was non-random due to location (workplace). However, it was evaluated using a social-psychological approach, and given the oft-cited issue of interventions failing to be grounded in theory, it is worthy of consideration. Moreover, it included a 3 month follow-up, addressing another key limitation of such interventions, in that long-term attitude or behaviour change is rarely captured. The training course involved group discussion, role-play and videos, and was facilitated by Aboriginal employees (the target outgroup). Participants were encouraged to reflect on their own beliefs and stereotypes, and to think more broadly about prejudice and discrimination.

The results of the intervention were relatively positive. There was a significant improvement in knowledge, and a reduction in negative stereotyping and 'old-fashioned prejudice' (p. 258). However, there were limitations. Firstly, effects seemed to reduce after the 3 month period. This could reflect a deficiency with the intervention, however it is likely to be an indication of the fact that negative stereotyping is a difficult habit to break. Furthermore, the decrease in 'old-fashioned' racism was not matched by a decline in 'modern' racism - for example, the belief that Aboriginal Australians have too much influence as a result of Government initiatives to promote equality. This chimes with the earlier discussion about the discrepancy between people's broad attitudes to equality and their attitudes towards specific measures to work towards this. Finally, the authors emphasise the importance of such programmes being a part of - not an alternative to - broader systematic attacks on prejudice at all levels: 'the individual, the intergroup, and the institutional/structural levels'. They note that:

"The piecemeal use of such programmes 'here and there' in the community, is unlikely to be effective if there are no serious challenges to the social realities that shape and govern intergroup and structural relations" (260).

Genuine institutional and cultural change is undoubtedly more difficult to achieve, but this evaluated case study stresses the importance of bearing in mind that prejudice is not simply a 'personal pathology', and that interventions should look at the structural arrangements of society as a whole. This echoes Pendry et al (2007) argument that diversity management is crucial.

Tackling racism in the UK

Another relatively rare example of evaluated short-term diversity training initiatives are outlined in a 2005 report entitled 'The Search for Tolerance: Challenging and changing racist attitudes and behaviour among young people', produced for the Joseph Rowntree Foundation (JRF). These studies reinforce the complexity of the 'contact theory' as it plays out in real-life, in the form of diversity training programmes. The report is based on five separate case studies of small projects in the United Kingdom. The initiatives focused on tackling racism and improving intergroup relations, and are described below:

"Two are educational and delivered as part of citizenship education in schools, 'Show Racism the Red Card' in Stafford - run by a police officer - and 'You, Me and Us' in Peterborough - run by the local authority's youth service. Tower Hamlets Summer University is a voluntary sector, informal education project. The 'Diversity Awareness Programme' for convicted racially motivated offenders is run by probation officers. The Jubilee Football Tournament was run by two housing associations and could be described as a community cohesion project" (1).

Six hundred young people, mostly 11 and 12 years old, took part in the five projects in total. A mixture of quantitative (survey) and qualitative methods were used. Some of the key points on 'what works' and 'what didn't work' that emerged from the report included the need to have a clear structure; a range of activities; sufficient time given to interventions; and consideration of the potential for backlash.

In practice diversity training programmes are often 'instructional' in form, as a result of lack of planning, resources, and time. The studies described in the JRF report were a mix of interactive and instructional, and the findings highlight the limitations of instructional approaches:

"Activities that encourage young people to reflect on their own experiences and debate local events and concerns are more likely to have a lasting impact than presenting general information about racism, which seems distant and superficial and therefore of little relevance" (57).

This echoes findings in the previous section and supports the theoretical arguments that 'learning through doing' is more effective than simply being told that certain attitudes and behaviours are 'wrong'. Creative methods are more likely to attract and engage participants.

For instance, the 'Tower Hamlets Summer University' initiative was criticised for a lack of interactivity and variety of activities. In contrast, the 'You, Me, and Us' programme in Peterborough, which was a series of workshops within schools involving drama, poetry, storytelling, music and art, proved more popular and because of this potentially more effective. Notwithstanding the limitations associated with self-reporting through questionnaires, participation in the programme appeared to have positive effects, with a significant proportion claiming to have "a better understanding of the complexities and subtleties of racism and cultural difference" (28).

Of course it is very important to be careful not to draw conclusions about the effectiveness of creative methods alone in terms of changing attitudes. Another attempt to use creative methods to promote equality and improve intergroup relations was the 'Jubilee Football Tournament' in Rochdale, an area characterised by divide between white and South Asian communities. However, a closer examination of the football tournament raises questions about the long-term effects of such initiatives. Feedback through discussions with participants suggested positive short-term effects, as for the duration of the tournament young people from different communities were brought closer together. However, these effects were short-lived, and it is suggested that this was a consequence of a lack of a clear anti-racist purpose, and failure to build on initial signs of potential improvement in relations. The two communities remained quite separate after the intervention. This highlights the importance of a clear strategy for all prejudice-reduction interventions, and careful monitoring during and afterwards. It also reiterates that creative methods may well improve the popularity of programmes, but are not necessarily any more effective in terms of changing attitudes or reducing prejudice, especially in the long term. The most important intervention elements remain contact and education which encourages a self-critical approach.

Some of the studies in the JRF report also highlighted the need to be extremely careful when designing and managing discussions about sensitive topics such as racism. For example, the authors note that the in 'You Me and Us' programme:

"Some of the responses suggested that not all the messages had been understood as intended. For example, the day began with a discussion about football hats and scarves, with an implicit message about not making decisions about other people based on stereotypes. One young person, however, seems to have come to a slightly different conclusion, saying he had learnt: that you are racist just by booing some ones hat. ('British', male)" (28).

As noted earlier in the discussion of what actually constitutes prejudice, there is a complexity that is perhaps not always recognised when designing prejudice-reduction interventions. Crucially, the reception of messages and interventions will be influenced by the existing predispositions of participants, so effects will vary. This is particularly important when considering interventions for young children. As Aboud et al (2012) note, "messages need to be tailored to the cognitive and emotional maturity of the children who in most cases already have well-formed opinions of themselves and others" (333). The 'Show Racism the Red Card' initiative in Stafford was said to be successful in reinforcing the point that racism is wrong, but did not really focus attention on sources of prejudice or specific issues raised by the children (e.g. their concerns about the situation in Iraq). This suggests that short-term diversity training programmes may risk being seen as superficial, inevitably lacking the depth that more long-term interventions can have.

Conclusion

Despite the aforementioned limitations and complexities, it is not suggested that diversity training has no value. Rather, that the key message is that it is important to be careful about how these activities are approached and to consider that some types may work better than others. The JRF report recommends empathy-inducing interventions as most valuable in terms of challenging prejudiced attitudes. Its findings also resonate with some of the key principles of the education/reeducation theory, including experiential learning and the value of learning about history and political events. Short-term activities may be useful in conjunction with other interventions (for example, on-going education, increase in contact). And as the Australian study showed, these should take place in a broader context of commitment to diversity and anti-racism, with institutional and cultural change. The most successful educational initiatives discussed in the previous section were designed on the basis of the existing social psychology research, and as the studies outlined here emphasise, such considerations are important for short-term interventions too.

3. Media-based interventions

Introduction

The media is regularly used as an approach to tackle prejudice, through TV, radio, and the internet. The media can provide an informational or a normative function, and initiatives may include poster campaigns, advertising, storylines on television programmes, and plays. Anti-prejudice campaigns in the media tend to fall into three categories: general awareness-raising; encouraging of reporting discrimination/abuse; and campaigns targeting certain groups/particular settings (e.g. 'Show Racism the Red Card' - tackling racism in the context of football). Although 'media' is a separate section from education/reeducation and diversity training for the purposes of categorisation for this study, media campaigns have educational purposes as well as providing a moral challenge, and the educational initiatives like those discussed in the previous sections use different types of media sources in their interventions, such as books and film clips. This section will cover these briefly, as they complement findings in previous sections and are worth reiterating, and will then move on to look at media campaigns.

Use of media in educational interventions

Emphasising the links between the different prejudice-reduction techniques, Aboud et al (2012) suggest that media could represent a useful alternative or addition to contact, for instance in cases in which direct contact between different groups is not feasible:

"Media is a particularly convenient way of providing children with an indirect or vicarious form of contact, especially children with little or no opportunity for direct contact" (331).

Aboud et al carried out a systematic review of 32 studies published from 1980 - 2010 of various types of interventions to reduce prejudice in early childhood. Interventions took place in various different countries, and all were delivered to young children under the age of eight. The authors found that media/instruction forms of intervention had a 47% success rate in terms of improvement in attitudes (and to a lesser extent on peer behaviours). Of course such figures should be treated with caution - the different studies will each have had different evaluation methods, and the recurring problem of short-term vs. longer term attitude-change will be pertinent here too. Nonetheless, it is a positive indication that media can be used successfully as part of educational strategies to reduce prejudice.

Their evidence also strongly suggests that the type of media content very much matters. For example, in relation to using media as part of educational curriculum, "scenes and stories of intergroup contact among peers" fared far better than 'multicultural education (331). Using media as a form of indirect contact, as opposed to focusing on the culture of a minority group, may therefore be a more effective approach.

Media campaigns

Media campaigns - for example, by campaign organisations, the Government, or criminal justice agencies - are also extremely popular, and frequently used with the intention of promoting change through raising awareness and challenging attitudes and stereotypes. Some academic research has looked into the effectiveness of such campaigns, and the results are mixed. Sutton et al (2007) suggest that despite the frequency of such campaigns, there is little evidence of their effectiveness. To date there has been very little research, and the authors point out that we cannot easily evidence effectiveness in terms of changing attitudes.

Abrams suggests that 'informational' media messages might not be the most efficient way of influencing people and that normative pressure can be much more successful. The normative communication functions of media can be considered more controversial to libertarians, and again the issue of to what extent Government has the right to intervene in this way is contentious. There is also the issue of the 'credibility of the messenger'. Abrams (2010) points out that it matters where the attempt to persuade comes from:

"Groups become more persuasive if we identify with them and less persuasive if we see them as out-groups" (70).

We are more likely to be persuaded to change attitudes if there is a general consensus amongst our own group. Findings in the educational initiatives discussed in the previous section strongly supported peer engagement, suggesting that change is best affected from within peer groups where possible. Similar lessons could perhaps be learned for media-based interventions.

'One Scotland Many Cultures'

The 'One Scotland Many Cultures' campaign by the Scottish Government was launched in 2002 and involved advertising (through TV, radio etc.), poster campaigns, a website, and other related awareness-raising activities in the media. The campaign's objectives were to celebrate multiculturalism, create empathy for victims of racism, and state a moral appeal for equality and tolerance. However, a retrospective evaluation carried out by Sutton et al (2007) found that the campaign was not based on the key theories and evidence from the social-psychological literature and empirical studies. One significant consequence of this was that the campaign conformed to minority ethnic stereotypes, such as 'Asian shopkeepers and doctors and Black footballers' (47). Theory tells us that prejudice-reduction interventions can backfire if they are regarded as 'favouring' certain groups, or if they reinforce stereotypes, yet this does not appear to have been fully appreciated. This is a common problem with media interventions in general. The authors also discuss some of the media campaigns on racism in football, noting that:

"If a campaign depicts racial discrimination at football matches as coming from far-right neo-fascists, rather than by more everyday supporters, it will not ring true, and so have less impact on prejudice and discrimination at matches" (29).

Moreover, when designing interventions the pre-existing opinions of the audience should be given careful consideration. Maio et al (2001) carried out an experimental study on how people respond to anti-racism messages in the media, and found that results are greatly dependent on existing opinions. Crucially, existing opinions or attitudes could result in messages backfiring. Reinforcing stereotypes and failing to properly consider the target audience and what messages the campaign wishes to get across are problems frequently raised in studies on diversity training and educational initiatives too. Such oversights risk alienating audiences, so this highlights the importance of utilising the available academic evidence when planning interventions.

There is also a tendency to lean towards 'hard-hitting' messages, provoking anger, fear, or guilt, with the premise that triggering powerful emotions such as these will capture people's attention. This may be appropriate in certain contexts, such as health promotion or crime awareness. It perhaps falls into the what 'should' work category when talking about reducing prejudice, but the theoretical research encourages us to be careful in this respect. Abrams (2010) warns that making people feel guiltier about inequality seems unlikely to be a useful solution - people are prone to reacting defensively (similar to the findings noted in the section on short-term diversity programmes). As noted earlier, the social psychological theories state that inducing empathy and compassion are the most effective ways of challenging attitudes, so when designing media interventions it is important to bear this in mind. Hard-hitting messages are also in some cases based on exaggerated interpretations of an issue, and as such are not necessarily accurate. If it appears that facts are being distorted and what is being depicted is not a true reflection of reality, there may be a risk of alienating the intended audience. An example may be plays or films about a particular type of prejudice. We have established that prejudice often exists in subtle, everyday manifestations. However in order to maximise appeal, particularly dramatic interpretations of a problem may be deployed. Initiatives that use such methods should be aware of these risks.

As well as making best use of the available evidence and social theory when designing interventions, Sutton et al (2007) note that evaluation is often not properly considered:

"Evaluation of these initiatives has also tended to be done as a quick afterthought with a consequent lack of the rigour required to identify good practice" (20).

This lack of evaluation echoes problems raised in previous sections. The 'One Scotland Many Cultures' project was criticised for poorly-designed surveys which made evaluation even more difficult. Sutton et al propose that initiatives should be tested with target audiences in pilot projects before launching, then monitored throughout. One recommendation of this report would be academic evaluations of prejudice-reduction initiatives, which may well have a more thorough approach. Finally, the literature on media interventions suggests that repetition is an important point - repeating an argument continually may have a greater effect than one-off campaigns, similar to the finding regarding educational interventions discussed in the previous section, in that short-term one-offs are less effective than on-going programmes.

Conclusion

It is reasonable to suggest that, at best, media campaigns might be deemed effective in relatively 'vague' ways. However it is possible to draw from the available evidence some suggestions of what is most likely to have positive effects and least likely to potentially 'backfire'. Given the popularity of such campaigns, effectiveness might be increased by taking into account some of the lessons discussed in this section.