Preventing Domestic Abuse for Children and Young People (PEACH): A Mixed Knowledge Scoping Review

Aims

1. To identify and synthesise the evidence on effectiveness, cost-effectiveness and acceptability of preventative interventions addressing domestic abuse for children and young people under 18 years of age in the general population.

2. To produce advice on what form future research might take in the context of England and Wales.

Objectives

1. To locate and describe the existing body of evidence relating to preventative interventions addressing domestic abuse for children and young people under 18 years of age in the general population.

2. To identify the range of short-, medium- and long-term outcomes achieved by preventative interventions for children and young people under 18 years of age to date.

3. To distinguish between different preventative interventions including educational programmes, media and community campaigns and other initiatives in terms of effectiveness, cost and cost-effectiveness.

These aims and objectives were used to generate a series of specific research questions.

Research questions

1. What is the nature of preventative interventions in domestic abuse for unselected children and young people under 18 years of age, and what theories underpin the chosen intervention strategies?

2. What outcomes are assessed in studies of preventative interventions in domestic abuse for unselected children and young people?

3. Which elements of the described programmes or interventions have proved to be effective, for which groups of children and young people, and in which contexts?

4. What is the cost of preventative interventions in domestic abuse for unselected children and young people under 18 years of age, and which elements of programmes or interventions have been the principal cost drivers?

5. What are the experiences and views of children and young people about interventions aimed at preventing domestic abuse, and are these influenced by gender?

6. Which of the successful intervention programmes are most likely to be acceptable to stakeholders, and cost-effective in the context of services and developments to date in the UK?

Some of these questions have proved more likely to yield answers than others and this is primarily due to the uneven nature of the evidence. For instance, the systematic review provided few data in relation to cost-effectiveness. These gaps are identified throughout the report.

Children and young people

This term includes all school-age children, that is those aged between 5 years and 18 years.

Preventative interventions

While the original research brief for the study called for a focus on interventions to prevent children and young people becoming victims/perpetrators of domestic abuse in later life, it was evident from an early stage, both from the UK mapping survey and from the literature retrieved, that much activity in this field was aimed at assisting children and young people to manage domestic abuse as they might currently or soon experience it in their own intimate relationships as well as in their parents’ relationships. Prevention in this review is, therefore, not confined to the prevention of domestic abuse in adulthood.

Preventative programmes are delivered in schools and other settings (such as young people’s centres) to children and young people under 18 years of age in the general population. They aim to prevent domestic abuse through raising awareness and changing attitudes and behaviour. They address domestic abuse in young people’s own interpersonal relationships and often also address any experience of domestic abuse in their parents’ relationships.

Preventative interventions also include media and community campaigns and initiatives aimed at preventing domestic abuse that address children and young people in the general population.

We have tried to distinguish clearly between different forms of initiatives throughout this report. However, in places, particularly in Chapter 3 where we report the findings of the mapping study, the term ‘programme’ is used as a shorthand to indicate a range of initiatives that includes both manualised programmes and one-off lessons or events such as a school assembly.

Domestic abuse

In line with the government definition, this study has adopted a broad definition that includes coercive and controlling behaviour in addition to physical, sexual, threatening, emotional/psychological or financial abuse of those who are or have been an intimate partner, regardless of gender or sexuality. 17 The definition adopted includes ‘honour-based’ violence and forced marriage but not female genital mutilation (FGM) (in line with definition adopted by NICE’s Public Health Guidance18 on domestic abuse).

Primary research question for the literature review

The research questions (see Chapter 1) were reframed into a single overarching question for the purpose of the systematic review of published literature. This was:

What is the nature of, underlying theory for, and evidence of effect of interventions designed to help children and young people avoid and/or deal with domestic abuse, and what interventions work to trigger effective mechanisms for change in specific groups and individuals in which specific contexts?

Search strategy

The search strategy was designed to be broad and to ensure that a wide range of relevant literature was included. Included studies were those published in any language between 1990 and 2012 (updated in February 2014). The year 1990 was chosen as the start date for inclusion of material for review, as prevention programmes emerged in North America in the mid-1980s28 and evaluations and research into such programmes did not appear until the 1990s.

Study types

Each of the detailed research questions in the protocol was mapped against the knowledge matrix of Petticrew and Roberts29 (modified for this specific research topic) to establish the kinds of data that were most likely to answer the question posed (Table 1). This indicated that the search strategy should not be limited to studies with randomised designs. The strategy therefore allowed for inclusion of studies using a wide range of methods. Because we were aware that the existing reviews in this area either were narrowly focused on randomised controlled trials (RCTs) or quasi-RCTs,13,14 or included data that were at least 12 years old,15 we did not include systematic reviews, although we did ensure that our search strategy located all of the relevant studies included in these prior reviews.

Databases searched and other means of locating literature

A wide range of databases were searched, as follows: Allied and Complementary Medicine Database (AMED); Applied Social Sciences Index and Abstracts (ASSIA); Cumulative Index to Nursing and Allied Health Literature (CINAHL); EMBASE; Education Resources Information Centre (ERIC); MEDLINE; PsycARTICLES; PsycINFO; Social Policy and Practice; Social Work Abstracts; Sociological Abstracts; Studies on Women and Gender Abstracts; Australian Education Index; British Education Index; and the Centre for Reviews and Dissemination NHS Economic Evaluation Database (NHS EED). Searches were undertaken for the period 1990 to February 2014. The initial date of searching was 3 July 2013 and the searches were updated in February 2014.

The protocol planned for the establishment of a Zetoc Alert list, but in the event two members of the team (Jane Ellis and Nicky Stanley) were in receipt of a wide range of relevant journals regularly. They are also experts in the field, with extensive networks in this area. Along with a search of the Controlled Trials Register, the National Institute for Health Research portfolio of ongoing studies, and formal contact with leading authors in the field, this was deemed to be adequate to scope the ongoing and new studies in the field. The following journals were, therefore, reviewed regularly across the period of the study: Journal of School Health; Children and Society; Journal of Adolescent Health; Violence Against Women; Violence and Victims; Prevention Science; Journal of Interpersonal Violence; Journal of Primary Prevention; Scandinavian Journal of Public Health; and Child Abuse Review.

Search terms

The search terms were chosen to generate a wide range of hits in the first instance. The example terms described in the protocol were amended as the study progressed in the light of the data and insights emerging from the consultation groups and from the initial testing of the search for parsimony and comprehensive inclusiveness. They were structured using a version of the PICO framework (population, intervention, context and outcome, which was divided into general and intermediate outcomes, such as attitudes and knowledge, and specific types of violence and/or perpetration outcomes) relevant for a realist review of studies employing a range of headings. Table 2 sets out a summary version of the search terms that were used. Appendix 2 gives an example of the full search in EBSCOhost.

The reference lists of all included studies were searched to check for any frequently cited studies not identified by the primary search (‘back-chaining’).

Inclusion and exclusion criteria

Table 3 details the inclusion and exclusion criteria for included studies.

Quality appraisal

In realist review methods, the quality assessment of the included literature is less important than the information it generates about programme theory and design, the contexts it has been used in and the mechanisms (both intended and unintended) that might have been triggered as a consequence: criteria that have been labelled relevance and rigour. 19 In the case of the current review, however, we wanted to focus particularly on programmatic and mechanism effects for studies where the programme under review seemed to have had an effect either on the hypothesised mediating variables, or on any component of relationship/dating violence. To establish if any such effect was real and likely to be generalisable beyond the original setting for the study, we needed to undertake a basic assessment of quality.

In the original protocol, it was planned that all studies meeting the inclusion criteria would be subject to quality assessment using a specific tool relevant to the methods used, such as Consolidated Standards of Reporting Trials (CONSORT; for any RCTs that are identified)30, Strengthening the Reporting of Observational Studies in Epidemiology (STROBE; cohort, case–control or cross-sectional designs)31 or Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PRISMA; systematic reviews). 32 As it had been agreed that all identified studies regardless of quality would be included in the literature review, the initial quality screening was undertaken using relevant Critical Appraisal Skills Programme (CASP) tools,33 as an overall guide to the quality of the included studies. In the event, this process provided a fair guide to the quality of the included studies, and, indeed, the CASP tools include the key elements of all of the specific assessment tools we originally planned to use. We therefore did not undertake more detailed quality assessments of the quantitative studies, especially as there were no plans to meta-analyse the findings from the included data set. For the qualitative data, the search produced so few papers that the decision was made not to exclude studies on quality grounds, and that, in these circumstances, the CASP tool provided an adequate assessment of general quality. 34 The interpretation of the quality scores for quantitative studies is given in Table 4.

Analysis

Six intersecting narrative analyses were planned: a description of study characteristics; an examination of the theoretical basis of the programmes; a summary of the kinds of interventions used; a description of the outcomes assessed; an overview of the effectiveness of the included interventions and programmes, both overall and for specific groups (including analysis by gender, ethnicity and prior risk status) and contexts; and a synthesis of the views and experiences of participants and staff involved in interventions in this area. Costs and cost-effectiveness data derived from the review data are reported separately in Chapter 7.

Quantitative findings were reported by study and summarised narratively across the data set. Qualitative data were synthesised using a meta-synthesis approach that included reciprocal and refutational narrative analysis, and the production of a line-of-argument synthesis. The summary of the findings entailed an interpretation of the findings of the review against prior theoretical constructs for what might work, for whom and in what contexts.

Detailed analytic process

The studies were grouped and described in a range of different ways to allow for insights to emerge. The structure of these descriptive processes varied somewhat from those proposed in the protocol in order to reflect the concerns arising from the consultation groups, and the nature of the emerging data from each phase of the study:

1. The characteristics of each study were logged on a pre-designed data extraction form by study type (controlled trials; cohort/caseload studies; qualitative data). This included a description of the intervention components of, participants in, context of and outcomes assessed in each study.

2. The outcomes measured used across the studies were then described to assess the commonality and differences between them and the number of studies that included each specific outcome. Outcomes were summarised under four headings: measures of knowledge, attitudes, behaviours (such as help-seeking), and incidences of victimisation or abuse related to relationships.

3. Studies assessing similar intervention components and/or which described elements of the same programme were then grouped together (regardless of methodology or outcomes examined at this stage) to establish the type of interventions tested in the included studies, how these fitted with the theoretical principles identified and what the range of resource requirements might be for each programme.

4. Following the first set of consultation meetings, the researchers developed a matrix of theories that might have been used to frame programmes. A wide range of potential theories was identified, ranging from those addressing causation (such as various feminist and social norms theories) to learning and education theories and theories of change and adoption of innovation. The team then assessed the underlying programme theory in each paper included in the search (where this was evident or could be deduced) to determine whether or not any of the theories identified in the initial theory matrix were present in the reviewed programmes and interventions.

5. The findings of each study were assessed. Outcomes noted to be significant at the p = 0.05 level (or p = 0.01 where it was evident that more than 100 separate analyses were carried out without correction for multiple testing) were logged, by time point (within 1 month of the end of the intervention; between 2 and 5 months; 6 months to under 4 years; and 4 years or more). The same organisational structure was used for the outcomes as in point 2, above.

6. For each included study/programme, data showing post-intervention changes that were statistically significant (using the criteria in point 5 above), and that were identifiably linked to gender, grade, age and history of perpetration/victimisation at baseline, were separately analysed by subgrouping by the relevant data and then summarising them narratively.

7. Where authors provided logic models for their programmes, these were examined for similarities and differences.

8. Given the lack of strong evidence of effect in the included studies, formal context–mechanism–outcome (CMO) models were not constructed.

Review of the grey literature

As this was a scoping review, the included studies were not limited to formal published research papers. Much of the literature on UK interventions is available only as ‘grey literature’, and so this aspect of the review was fundamental to building a fuller picture of the nature and impact achieved by prevention work in the UK context. This element of the review addressed UK-based grey literature, including local independent evaluations, national reports, technical reports and theses; in-house evaluations were excluded. The time parameters used were the same as those for the systematic review.

The systematic review search yielded three grey literature publications and an online search of relevant websites (listed in Appendix 3) yielded a further nine. Back-chaining produced an additional five items and the remainder consisted of publications identified through requests to consultation group members and in response to a request circulated by Women’s Aid as well as through the research team’s own knowledge of the field.

Forty-six documents were identified by these means and data were extracted and recorded on a pre-designed form using the same categories as were employed for the peer-reviewed papers. A narrative approach was utilised for this aspect of the review, with analysis identifying the main findings and themes in independent programme evaluations undertaken in the UK.

Mining the mapping survey data

The mapping survey administered in 18 local authorities (see Table 6 for the local authorities surveyed) consisted of 46 questions about each programme reported. A subset of the 23 questions listed in Table 5 was identified as being relevant to assessing the costs and benefits of the programme; responses to questions marked with an asterisk in Table 5 were deemed necessary to be able to draw useful conclusions. Therefore, those programmes reported in response to the survey, which provided complete data in these 13 areas, were included in this aspect of the study.

For the included programmes, responses to the questions listed in Table 5 were extracted and tabulated. These data were described narratively and common features were identified. Resource inputs were identified as far as possible using the information reported, with particular attention given to who delivered the programmes, the pattern of delivery and who funded the programmes. Outcomes were identified from the responses provided to the question about achievements (question 44) and evidence of any relationship between resources and outcomes was noted.

Extracting economic data from the literature review

All papers included in the literature review were scrutinised to identify any data on resources required to run a programme and/or the cost of a programme as described in the paper.

The following data, relevant to cost and cost-effectiveness analysis, were extracted, where present, from all papers included in the literature review:

• resources required to set up the programme, for example training

• resources required to run the programme

• pattern of delivery, for example number and frequency of sessions

• cost information

• funder of programme development and implementation or, if appropriate, the research

• outcomes.

Information about resource use was reviewed and each programme assigned a broad category of high, medium or low resource use. This classification was judged on the information provided in the literature, and was carried out by two members of the team (Soo Downe and Sandra Hollinghurst). The criteria used to assign the categories were the intensity and scale of resources reported.

The resource requirements were described narratively and common features identified. Resource inputs were identified using the information reported, and particular attention was paid to who delivers the programme, the pattern of delivery and the source of funding. Outcomes, as reported in the papers, were noted, as was the reporting of any relationship between resources and outcomes. The general pattern of resources and outcomes was reviewed, and conclusions about any relationship between them were described narratively.

Extracting economic data from the grey literature and personal contact with experts

The UK grey literature identified in the main study was examined for any evidence about resource use and/or cost. In cases where some information was available, we contacted authors of papers and reports directly for clarification or more detail.

During the process of identifying important unpublished reports and conducting the consultation groups with experts, the researchers followed up any mention of resource use and/or cost that was judged to be potentially useful. The research team contacted all experts whose work was brought to our attention with the aim of obtaining precise details of the cost of programmes and understanding how decisions about cost-effectiveness were handled. The information collected was recorded narratively by programme. In respect of analysis, the individual and diverse nature of these data prevented any systematic synthesis and analysis was, therefore, restricted to a descriptive framework.

Extracting economic data from the expert consultation groups

Forms of violence addressed in programmes

Respondents were asked to select from a list of seven options what forms of violence or abuse the 98 programmes reported addressed (here, programme is used to mean any initiative, project, one-off event, media campaign, school assembly, lesson plan, scheme of work or external resource). The majority of programmes (62%) focused on domestic abuse/violence in young people’s intimate relationships, with a substantial number of programmes (46%) addressing the broader issues of peer violence/bullying and domestic abuse/violence in adult relationships (Figure 2).

FIGURE 2.

Forms of violence and abuse addressed in reported programmes (n = 98).

Respondents were asked to indicate any ‘other’ forms of violence or abuse addressed by programmes in an open-text box. Other examples given included raising awareness, healthy relationships, conflict management, FGM and honour-based violence, staying safe, feelings, gender inequality and positive role models. Every programme reported addressed more than one form of violence; the most common combination was domestic abuse in adult relationships with domestic abuse in young people’s relationships (n = 42), with the latter combined with child abuse being the second most frequently reported combination (n = 22) along with child abuse and peer violence (bullying).

Topics covered

The survey collected information on all the topics covered by programmes. Participants were given a list of 23 options and were asked to tick all those that were addressed; they were also given the opportunity to add information on other topics covered in an open-text box. Figure 3 shows that most common topics included in the 98 programmes reported were types of abuse (51%), recognising domestic abuse when it is happening (49%), personal safety (48%) and definitions of domestic violence (46%). The least common topics covered were domestic violence and issues for children with disabilities and/or learning difficulties (12%) and FGM (10%). Other topics indicated by respondents included aspirations, positive male role models, self-esteem, gender inequality and emotions. While there appeared to be an emphasis on awareness and recognition of violence/abuse, a significant number of programmes, over one-third in all cases, also addressed issues related to keeping safe and help-seeking, including topics on safety strategies (44%), information on support services (40%), disclosure and safeguarding (37%) and help-seeking (37%). Content that aimed to address the needs of minority groups was less frequently reported.

FIGURE 3.

Topics covered in reported programmes (n = 98). BAMER, black, Asian, minority ethnic and refugee; LGBT, lesbian, gay, bisexual and transgender.

Programme aims

An open question asked respondents to describe the aims of programmes; information was received for 51 programmes. The most frequently mentioned aims were to raise awareness (n = 21); to increase knowledge and understanding (n = 14); and to increase knowledge and understanding of healthy/unhealthy relationships (n = 19). Thirteen sought either to provide support directly to children and young people living with domestic abuse or to provide information on support services. In 10 cases, safety for children was an identified aim; this was mostly in relation to programmes delivered in primary schools. The focus in almost all programmes was the reduction or prevention of victimisation rather than perpetration, and only three reported a specific focus on promoting gender equality. A comprehensive long-term strategic aim was reported in one case:

Develop consistent preventative work on domestic violence and abuse accessed by all young people across their childhood, with recognisable themes from nursery to college. Each element has a different approach and set of aims depending on age and setting of delivery.

Venues

Respondents were asked to indicate all venues where a programme was delivered; these are shown in Figure 4. Programmes were mostly delivered in mainstream secondary schools (n = 49), with mainstream primary schools (n = 28) being the second most frequent setting. Twenty-three programmes were delivered in one venue only; of these, 10 were in secondary mainstream schools only and 11 in primary mainstream schools. Nine school-based programmes were delivered in a range of types of schools. Five programmes were delivered in more than six venues and all included work with young people in colleges, the secure estate and/or young people’s centres. Programmes involving a media and/or online aspect (n = 6) were delivered alongside work in schools. Of the community-based programmes, one programme was delivered in a health centre; other community venues included domestic violence advice centres and a local conference centre. A peer education project had undertaken street-based work with young people as well as using other venues. Eleven programmes, two schools and four other organisations (five were unknown) reported that they used outreach as an approach to delivering the programme. (Outreach is the delivery of services to people where they spend time or in their homes; it is common practice in youth work.)

FIGURE 4.

Venues where programmes were delivered. a, Secure estate refers to a young offender institution, a secure children’s home or a training centre. PRU, pupil referral unit.

Facilitators

There are ongoing debates in the literature regarding which groups of professionals are best equipped and placed to deliver these interventions (e.g. Hale et al. 38). The survey asked for information on who delivered the programmes. Of those who responded to this question (n = 65), the vast majority (89%) reported that external staff were involved in some way in delivering programmes, with only 11% of programmes being delivered solely by teachers. However, teachers were involved in a further 31% of programmes working jointly with one or more other staff, meaning that, overall, teachers were involved in 42% of programmes (Figure 5). How co-working between teachers and external staff was organised varied and its adoption could be linked to a number of factors. One domestic abuse organisation reported that teachers were merely present in the classroom while specialist staff from the domestic abuse organisation delivered the programme.

FIGURE 5.

Staffing arrangements for programmes.

A range of teachers was involved in programme delivery, including class teachers in primary schools, form teachers in secondary/high schools, specialist PSHE teachers and ‘other’ teachers. As Figure 6 shows, school counsellors/guidance teachers and school nurses were involved in a small number of programmes (n = 7 and n = 4, respectively).

FIGURE 6.

Practitioner groups involved in delivering programmes. PSE, personal and social education; SPHE, social, personal and health education.

The practitioners most widely involved in programme delivery were those from specialist domestic abuse organisations who were involved in 63% (n = 41) of the programmes for which staffing data was provided; in almost half of these cases (31%) they were the only practitioners involved. A further 14% of programmes were delivered by other professionals from outside schools; these included sexual health workers, youth workers and staff from voluntary organisations (other than domestic abuse or children’s voluntary organisations).

In addition to the staff listed in Figure 6, 12 programmes involved ‘other’ people delivering the intervention; these included a community safety officer, a safeguarding team, a trainee social worker, theatre company staff, nurses (not school nurses), a domestic violence co-ordinator (local authority) and youth offending service practitioners.

Facilitator training

The survey asked respondents to indicate whether or not those delivering the programme had received specific training to enable them to do so. The question was unanswered for 45% of the programmes reported. Figure 7 shows that, for those programmes where information was available, almost two-thirds (61%) reported that all of the staff had undertaken specific training to deliver the programme, with only 9% reporting that staff had not had any such training. A small number (15%) reported that ‘some but not all’ of the staff had undertaken training.

FIGURE 7.

Staff who had undertaken specific training to deliver the programme.

Location of programmes in the curriculum

Where programmes were delivered in schools, respondents were asked to indicate in which lessons programmes were delivered. As expected, school programmes were most likely to be delivered in PSHE classes, with a few programmes being delivered in citizenship or drama lessons. None of the reported programmes were taught in science lessons. Two respondents reported that programmes were delivered in English lessons; one respondent who described ‘Equate: A Whole School Approach’ programme reported that one aspect of the programme – ‘personal space’ – was delivered in physical education lessons and another element – ‘global gender inequalities’ – was taught in geography classes. Five respondents reported that programmes were delivered in separate sessions outside mainstream lessons either to individuals or in small groups; this was likely to occur in cases where young people had disclosed abuse.

Methods of delivery

All respondents who answered a question on methods used to deliver programmes (n = 56) reported using multiple methods. Figure 8 shows that the most commonly used method of delivering programmes/interventions was small-group discussion (71% of those programmes reported), with heavy use also made of whole-group discussions (64%) and digital versatile discs (DVDs) (52%). The DVD material was likely to have been drawn from national and regional campaigns and one respondent explicitly identified the media campaigns as an influence on the genesis and content of their local programme. Theatre in education (7%), community events (7%) and adverts (13%) were less frequently mentioned, and this is likely to reflect the costs attached to such approaches.

FIGURE 8.

Methods of programme delivery.

The survey also asked respondents whether or not reported programmes were delivered in conjunction with programmes for other non-school groups, for example parents or carers. The majority of respondents answering this question did not know whether or not this was the case. Where this question was answered, 11 programmes were described as delivered in conjunction with programmes for professionals working in the local community, six were described as delivered in conjunction with programmes for parents/carers (n = 6) and two were delivered alongside programmes for other adults in the community (n = 2). One programme was also delivered in conjunction with a programme for teachers as part of in-service training.

Target audiences by age

The survey classified children and young people receiving programmes into four age ranges: 0–4 years (early years), 5–10 years (primary), 11–15 years (secondary) and 16–18 years (further education). As the focus of the study was interventions for children in the general population, programmes targeted at those over 18 years of age or at specific groups, for example pregnant women, were excluded. Where programmes were reported as aimed at a wide range of age groups, for example at children aged 5–16 years, these were described as targeting combined audiences. Table 9 shows that one-third of the programmes for which this information was received (n = 49) were targeted at secondary school-aged children, nearly one-fifth were delivered to primary school age and 14% were aimed at both primary and secondary school ages. Only a small number of programmes were targeted at young people in secondary and further education but the survey was not sent to further education colleges. None of the programmes reported were targeted at children under 5 years of age.

The survey asked respondents whether or not different components of the programme were delivered to more than one age group. Where this information was available (n = 56), 54% (n = 30) said that programme content was adjusted for different age groups, 29% (n = 16) answered ‘no’ and 18% (n = 10) did not know. Respondents described delivering differing content to different age groups in respect of material addressing their children’s or young people’s own intimate relationships. Topics such as violent crime or drugs were delivered to older children, while younger children were more likely to be offered material addressing friendship or internet safety. Different materials such as books and DVDs were selected for primary- and secondary-aged children, and the depth of the discussion varied according to age, with games and role-play seen as more appropriate for younger children. While most of this adaptation and selection was described as being undertaken in-house by the programme facilitators, one programme, ‘Equate: A Whole School Approach’, was described as designed to include age-appropriate components, with a built-in ‘shift from general domestic abuse awareness to more specific issues such as sexual consent or teenage romantic relationships as children move up through the school’. Three programmes designed specifically for primary school-age children were cited: the ‘Helping Hands’ programme (this programme is discussed in detail in Chapter 7); ‘Miss Dorothy’ (also known as ‘Miss Dot’ and ‘Miss Dorothy Watch Over Me’) and ‘Themwifies’.

Programme participants

Thirty-six respondents provided information on the number of children and young people participating in programmes in the previous 12 months (Figure 9). Half of the programmes for which this information was provided had between 30 and 499 participants; the largest number reported was 2300 participants. Four newly established programmes had no participants to report as yet.

FIGURE 9.

Number of children and young people participating in programmes in the previous 12 months.

Programmes delivered by gender

Figure 10 shows that most respondents reported that programmes/interventions were delivered to both mixed-sex groups and single-sex groupings. Only a small proportion of those answering this question described programmes that were targeted specifically at girls only and only one reported a programme targeted at boys only. Three of the girls-only programmes were delivered in single-sex schools.

FIGURE 10.

Target audiences by gender grouping.

Among the reasons given for programmes being delivered in mixed-sex groups were that it was relevant, suitable and important for both boys and girls; that it ‘mirrors society’; that it was ‘in line with the agency ethos’; and to ‘get balance’ and a range of views. Other responses focused on practical arrangements such as ‘it reflected class composition’ and to ‘achieve maximum coverage’. The most commonly stated reasons given by those advocating single-sex groups were that young people talked more freely in this context; it enabled discussion on gender-specific experiences; and that girls could talk ‘without offending anyone, i.e. potentially boyfriends can be in the group’. Those employing both mixed- and single-sex groupings reported that flexibility enabled the programmes to be more responsive to the needs of particular groups and brought the advantages of both approaches to learning by affording opportunities to discuss issues in different settings. One respondent described such an approach in more detail:

Most [name of organisation] elements are delivered in class groups that are mixed sex, but certain elements such as Personal Space include issues around sexual consent, rights and responsibilities. This part is delivered to single sex groups as it needs to be targeted specifically to meet the needs of either young men or women.

Programme participants by race and disability

Respondents were asked to identify the proportion of black, Asian, minority ethnic and refugee (BAMER) programme participants and those who had a disability or special needs. Forty-two per cent of respondents (n = 41) provided information relating to ethnicity and a similar number did so in relation to disability and special needs.

Respondents’ reports of numbers of BAMER participants on programmes ranged from none (n = 4) to 90% (n = 1), and in 15 cases (37%) this information was not known. As we lack the information required to identify respondents’ locations in a number of cases, we are unable to report on the extent to which this represents the make-up of the local population in these areas. However, although the number of respondents answering this question was low, Figure 11 shows that the majority of those responding described programmes that were delivered to audiences of which 10% or more were made up of BAMER children and young people. This can be contrasted with the finding from the analysis of programme content, reported above, that only 17% of programmes were explicitly addressing the issue of domestic abuse in BAMER families.

FIGURE 11.

Proportions of BAMER participants on programmes in last 12 months.

Thirty-eight per cent of respondents (n = 42) did not know the proportion of participants with disabilities or special needs on programmes. Where this information was provided, the numbers of children and young people with disabilities or special needs on programmes in the last year ranged from none to 25% (Figure 12). For comparison purposes, the number of pupils with special educational needs in England, recorded by the Department for Education, in 2012–13 was 18.7%,39 while in Wales it was 22.3%40 and in Northern Ireland it was 21.1%. 41 In Scotland, the term ‘additional support needs’ is used to encompass a wider range of children with support needs (e.g. bereavement is included). The recorded figure in 2013 was 19.5%. 42

FIGURE 12.

Proportions of programme participants with disabilities or special needs in the previous year.

Funding

The survey asked respondents to specify the main funders of programmes. Twenty-seven respondents did not answer this question (28%), and of those who did, 10 respondents (14%) did not know who funded programmes. The most common source of funding identified for the 71 reported programmes was community safety partnerships (17%) followed by local domestic violence organisations (10%), with schools being the main funder in 9% of cases. Other funders included trust/foundations, the police, local authority children’s education services or social care. The NHS (3%) and local authority youth services (1%) were the least frequently cited sources of funding for these programmes (Figure 13). These patterns were consistent across high and low responding low authorities. Additional information provided in respect of other funding sources identified a wide range of national and local funders such as the Big Lottery, Lancashire Children’s Trust, a local solicitor’s office, Preston Children’s Trust, Comic Relief, Coalfields Regeneration Fund, Scottish Government and Public Health.

FIGURE 13.

Main sources of programme funding.

Influence on programme design

The survey collected information on whether or not programmes/interventions were built on an existing programme. Sixty-four programme accounts provided information on this issue; the most frequent response (42%) was ‘no’, that the programme was designed locally, while 39% reported that the programme was modelled on another intervention and in 19% of cases the original was not known. Those who answered ‘yes’ were asked to describe what influenced or informed the design and content of the programme. Various influences were described including feedback from young people themselves, personal knowledge and research as well as replication of other established programmes. The range of responses is outlined below:

Feedback from the young people

Inventor’s own personal involvement with subject and also media campaign

Successful outcomes from similar work in the US. MVP [Mentors in Violence Prevention] Scotland is based on the MVP model developed by Dr. Jackson Katz in the mid1990s. Content driven by current issues within society around abuse and exploitation.

Respondents were also asked whether or not children and young people had been involved in designing the programme. Where this question was answered, there was a fairly even spread between responses; 37% (n = 23) answered ‘yes’, 27% (n = 17) answered ‘no’ and in 36% (n = 22) of cases this was not known.

Programme adjustments

Respondents who had been involved in setting up programmes were asked if they thought that the programme had changed over time. The 39 respondents who answered ‘yes’ (18 reported no change over time) described programme adaptations which had addressed programme content, methods of delivery and the context in which they were delivered. Five reported having expanded the content to embrace newly emerging topics such as online abuse, sexting and sexual exploitation. One reported that the programme now had ‘more focus on homosexuality and BAMER’; another respondent had widened the programme’s scope to address issues for lesbian, gay, bisexual and transgender (LGBT) and male victims. Four had introduced new methods including ‘more interaction, more role-play’, all in response to feedback from young people. Overall, such responses reflected an ambition to keep programmes up to date and flexible and so increase responsivity to different audiences. Four responses focused on how external changes had impacted on the delivery of programmes. These mainly related to reductions in funding; however, one respondent described how the privatisation of a local authority education service had made it more difficult for schools to access the programme.

Extent to which programmes were sustained

Stanley et al. 28 have argued that short-term funding, together with the failure to embed these interventions in the national curriculum, has limited the sustainability of these programmes. Respondents were asked when programmes began, whether or not they were still running, and if not, when they ended. The length of individual programmes was then calculated. This information was not given for 37 (35%) of the programmes reported. In 21 cases, this information was ‘not known’, either because the start or the end date of a programme was not known or because programmes were still currently running but respondents did not know when they began. For programmes (n = 42) where start and end dates were known, 22 had been in operation for less than 2 years, 11 had lasted between 2 and 3 years and nine had been running for over 3 years, indicating a lack of sustainability (Figure 14). Just over half (54%) of all 98 programmes reported were still operational between October 2013 and January 2014 when the survey was completed.

FIGURE 14.

Period over which programmes sustained.

Reasons for programmes ending were identified in six cases: in three cases this was because funding had been discontinued and in another three cases programmes had been intended to be time limited.

Evidence for impact

Respondents were asked whether or not they had discovered if the programme had had any impact on its participants. Nearly two-thirds (62%) of those answering this question (n = 52) claimed to have collected this information; 11 (21%) had not done so. Respondents were also asked to describe how this process was carried out. Information for 58 programmes was reported on. Of these, 24 programmes had undertaken this as part of a more formal evaluation process, with eight specifically stating that they had used pre and post questionnaires. Three reported having used specific measures, for example the Strengths and Difficulties Questionnaire. Qualitative measures cited included user satisfaction forms and focus groups; one reported using a ‘survey of learning journey’ as well as ongoing monitoring from session to session.

Evaluation of programmes

The survey collected information on whether or not the programmes had been evaluated. The majority of those answering this question (n = 33; 62%) reported that the programme they were describing had been evaluated; most of these were in-house evaluations. Ten respondents described the evaluation as independent but no reports from these evaluations were provided.

Programme achievements

Respondents were asked for their views on what programmes had achieved and comments were made in relation to half of the programmes reported. The most common achievement reported (n = 24) was ‘raised awareness’ or increased knowledge (of domestic abuse/abusive relationships), and ‘raised self confidence or self-esteem’ for children and young people was often associated with this:

It has helped young people look at their relationships, gained self confidence.

Raising confidence amongst young people to challenge abuse, raising awareness of issues and support available, based on feedback from schools.

Seven respondents identified changes in behaviour, although some of these were anecdotal:

. . . changed behaviour in schools, improved relationships, children able to express emotions and has addressed bullying.

Qualitative work has identified many other pluses including positive school climate, knowledge of the issue, ability to intervene and support friends.

Three described disclosures of experiences of abusive relationships as a consequence of delivering programmes:

The project has achieved: changes in knowledge; changes in behaviour; changes in attitude and increased disclosures of domestic abuse.

A small number of respondents described children and young people enjoying or valuing the programme:

The programme is still very new but it has already generated a great deal of interest in the students it has been offered to. The vast majority, in excess of 80%, report no previous education regarding domestic abuse.

Programme challenges

Thirty-nine responses described challenges encountered in developing and delivering programmes. Two-thirds of the comments received related to either accessing or getting buy-in from schools (n = 13) or freeing up time in the curriculum (n = 6):

Schools have such differing opinions on whether this is a valid or useful topic for their students.

Being able to deliver this – local schools and colleges who are working to the curriculum seem to not have any spare sessions to fit this in!

Similarly, it had been difficult to establish priority for preventative work in one area where:

The low level of reported teenage relationship abuse locally does not put the subject very high on the agenda of children’s services. If you can’t see the problem then it doesn’t exist.

Other problems encountered included difficulties in accessing the funding required for staffing or to deliver the programme to a greater number of children and young people (n = 7):

Funding did not materialise from the C&YPT [Children and Young People’s Team] so the school underwrote the programme.

Other comments related to tensions in multiagency working:

The various agencies within [local authority area] do amazing work, and although they seem on face value to work side by side and sing from the same hymn sheet . . . there are still a lot of boundaries between organisations and in some instances rivalry. This makes it more difficult, if one agency won’t work with you because others will.

One respondent reported that it was difficult to find

. . . appropriate male facilitators to help develop the project content and co-facilitate the sessions.

Included studies

As Figure 15 shows, 28 quantitative papers and six qualitative papers were included in the review; there was some overlap, with three of these papers containing both quantitative and qualitative data, and so in total there were 31 papers included in the systematic review. These covered 23 separate programmes, as some programmes were described in more than one paper. Details of the characteristics of the 28 quantitative studies are given in Appendix 7 [see Table 34 (controlled trials) and Table 35 (cohort and case–control studies)]. The findings from the review of the six qualitative studies are discussed as a group later in this chapter.

FIGURE 15.

Flow chart of included studies.

Characteristics and quality of included studies

Context of included studies

Two studies were undertaken in the UK,57,58 and one in India. 66 None of these were randomised trials. The remaining 25 studies (17 programmes) were undertaken in the USA (n = 18 papers44–49,51–55,59,61–63,65,68,69) or Canada (n = 7 papers43,50,56,60,64,67,70). The Indian study was an extension of the USA study that assessed the ‘Coaching Boys into Men’ programme. 66

Nature, validity and reliability, and comparability of outcome measures in all studies

Tables 11 and 12 give the outcome measures reported on in the randomised trials and cohort/case–control studies, respectively. The only tool used in more than one study was the Conflict Tactics Scale. 43–45,60,61,70 A range of other existing, modified and new (developed for the specific project) data collection tools were reported across all of the studies. This made comparison of effectiveness between studies extremely difficult. However, the data arising from most of the studies could be categorised into four distinct groups. These were knowledge, attitudes, behaviours (such as help-seeking) and incidence of perpetration/victimisation.

In most of the randomised trials, as a minimum, the internal consistency of the measures was cited for the study population (usually based on Cronbach’s alpha). Reliability (test–retest or split-half analysis type data) was not usually reported. The only exception was Wolfe et al. ,50 who did not report any validity or reliability data for any of the tools used, though these may be available in associated study reports.

In terms of the cohort and case–control studies, testing was less often cited, and five studies did not mention any internal consistency/validity/reliability checks on the tools they used. 44,57,63,64,68 One of these five, Miller et al. ,66 used the same tools as reported in Miller et al. ,44 and these were subject to testing.

Nature of the interventions for all studies/programmes

Tables 13 and 14 list the interventions used in the studies by study type, and programme.

All included programmes, except Families for Safe Dates,53 were school based. The programme components varied widely. The minimum contact time was three 35-minute sessions58 and the maximum was 21 classroom sessions, totalling 28 hours (Fourth R56). In the latter case, the programme also included material on substance abuse and sexual health. Most used a mixture of techniques, ranging from taught sessions and/or discussion groups to role-play. A number explicitly stated that they used audio-visual aids as part of their programme (see Table 14), but others did not specify the detailed components of their programme. In two, theatre was the main vehicle for the programme delivery. 57,59 In the case of the Healthy Relationships programme evaluated by Bell and Stanley,57 the theatre performance was developed and delivered by a professional theatre company, after extensive engagement with teaching and health staff from the school and local youth and health services. The programme evaluated by Belknap et al. 59 was explicitly based on the theory and underpinning philosophy of the Theatre of the Oppressed, and the play was informed by extensive local engagement with young people living in the local Latino/a community in prior qualitative research. The play was developed with and performed by undergraduate students and a professional theatre director. Foshee et al. 54 noted that their programme included a theatre production performed by peers, but did not provide any further details of this.

Three programmes made explicit reference to bystander theory:44,62,66 in the case of the ‘Coaching Boys into Men’ programme,44,66 this was translated as sessions delivered by coaches to the athletes they were training which aimed to provide examples of respected authority figures advocating positive behaviours, including bystander interventions. ‘Mentors in Violence Prevention’62 was delivered through trained peer mentors. In the only non-school intervention, ‘Families for Safe Dates’,53 the programme was delivered by regular mailings of information and activities to families, along with reinforcement telephone calls. Details of the diversity and nature of the programme interventions, where these were supplied, are given in Tables 13 and 14.

Of the programmes where the professional delivering the programme was mentioned, seven were delivered by teachers. The ‘Coaching Boys into Men’ programme involved school sports coaches in the USA44 and India. 66 Three programmes explicitly involved young people in designing and/or delivering the programme. 45,54,62 A range of external volunteers and professionals, including counsellors, lawyers and police officers, were involved in delivering other programmes (see Tables 13 and 14).

Training for staff involved ranged from 60 minutes44 to 12 days over 4 months. 66 These extremes were both associated with the same programme, ‘Coaching Boys into Men’. In the former case, the programme was developed in the USA. However, the authors found that when they tried to translate the programme to India, they first had to deal with the negative attitudes of the coaches they wanted to use as role models. The 9-day training and education sessions held over 4 months to train the coaches in implementing the programme followed an initial 3-day workshop that was designed to address gender norms and attitudes. This illustrates the problems of assuming that those teaching or running programmes to reduce interpersonal violence hold the same views as those designing and implementing the programme.

Black et al. noted that those running their programme met regularly with the programme designers to discuss progress and possible blocks and barriers to fidelity. 61,63,68

Nine of the 20 included programmes identified various kinds of school- or community-level activities aimed at supporting the core programme. 45,50–52,54–58,62,64,65,69 These ranged from providing a counselling service for those affected by the programme to a wide range of school and community activities.

Theories underpinning the included studies

Table 15 sets out the explicit/implicit theory(ies) underpinning the reported intervention/programme design, based on the reports of the intervention/programme that was delivered. We were unable to obtain permission to reproduce the logic models developed for various interventions in this report, with the exception of that of Foshee et al. ,54 which is included in Appendix 9. This section discusses the nature of the logic models of the three programmes that included such models in their papers. 44,49,51,54,66

Generally, programme theories included some version of behavioural change, with the intention being to change knowledge and/or attitudes towards specific social norms (usually, but not always, gender norms) with the assumption that this would change behavioural intention, and, eventually, actual behaviour, with a resulting effect on the incidence of perpetration and/or victimisation. As noted above, many of the programmes used a mix of methods, including didactic teaching, theatre, videos and discussion groups. The programmes that used more explicitly affective techniques, such as role-play and theatre, seemed to regard the mechanism of retained knowledge, improved attitude and consequent behavioural change as being more likely to be enacted through an increase in young people’s emotional and empathic engagement with the material and/or through increasing self-esteem and relationships skill building than through the simple transfer of knowledge.

The authors of three programmes (four papers) provided logic models to explain the hypothesised causal relationships in their programme design. 44,48,51,54 Both Foshee et al. ’s and Miller et al. ’s models altered between their earlier and later papers. In the case of Foshee et al. ’s model, two hypothesised mechanisms that might underpin the intended outcome of reduced perpetration and/or victimisation appear in both models: belief in the need for help, and awareness of services. Interestingly, however, the arguably more direct potential mechanism of active help-seeking found in the 1998 version (see Appendix 9) was removed in the 2005 paper. The other element that disappeared was additional community activities (a form of whole-system approach to solving the wider problem of the local environment in which abuse occurs). In both cases, the fundamental mechanisms of change hypothesised are changes in gender norms and in stereotyping attitudes and increases in conflict management skills.

The models in the two Miller et al. papers44,66 are fundamentally the same, and both suggest that the mechanism that might improve outcomes is male athletes’ responses to role modelling by respected coaches. This mechanism of change is hypothesised to influence uptake of knowledge and changes in awareness and gender norms, and to alter bystander behaviour. The nature of the assumptions about triggers that create the mechanisms of change embodied in the coaches is made more transparent by the additional elements added to the logic models for the extension of the ‘Coaching Boys into Men’ programme from the USA to India. 66 This required an additional component to explain how the coaches needed to be trained/reorientated towards gender-positive approaches, and how this also necessitated environmental and organisational support. The recognition of the need for a wider community-level intervention contrasts with the removal of this component from Foshee et al. ’s logic model over time. The new precursor component in the later Miller et al. model66 is framed around a notion of ‘positive deviancy’ as an explanatory framework for what was required from the coaches if they were to become agents of change. This provides a useful insight into the epistemological and ontological basis of the Miller model that is not so apparent in the other logic models or programme theories in this review.

Taylor et al. , in 201048 and 2013,47 hypothesised (and tested) the effectiveness of programmes based on two different logic models. One was an approach based on affective change, through interactive encounters and the building of interpersonal skills. It was theorised that this would trigger mechanisms in students that would change beliefs and attitudes (presumably making students less likely to stereotype and belittle those who they have come to understand and empathise with through the programme) and, thus, decrease the will to abusive behaviour for these students. The other arm of the logic model hypothesised that exposure to facts about the law and justice would increase knowledge (both about those who might be subject to violence, and about the legal consequences of perpetrating such violence) and, therefore, create intellectual disincentives to perpetration.

Outcomes in the included studies

The findings in this chapter demonstrate that a wide range of programmes and interventions to reduce relationship violence have been developed and tested for young people over the age of 10 years in the general population. We did not locate any formal research examining programmes for children under the age of 10 years.

The quality of the studies using randomised designs was generally higher than those using other designs. The conclusions drawn from these studies were also more conservative. However, even the randomised studies demonstrated a range of design and analysis issues that served to limit the generalisability of their results. These limitations led to authors making claims for findings that were statistically significant in a formal sense, but, in many cases, these findings were based on very small effect sizes, and/or resulted from multiple testing with a lack of control for the risk of type 1 errors arising from this.

Study findings that do not reach a chosen level of statistical significance may be interesting if there are sufficiently large differences between intervention and control groups in important variables to justify investment in larger studies to test the finding more definitively. In this situation, the findings become hypotheses generating for future research, rather than guides for roll-out to practice. Equally, findings of statistical significance do not necessarily mean that the outcomes are large enough or relevant enough to be meaningful in everyday practice. For this review we wanted to identify findings that might justify investment in roll-out of one or more of the programmes that had been tested. Owing to very large resource implications of such a roll-out, and the large number of analyses carried out in most papers (totalling many hundreds across the whole set of included studies), only findings that reach p ≤ 0.01 have been reported in this review.

As can be seen in Table 16, only a handful of analyses reached this level: 27 of those assessed immediately after the intervention, 15 that were assessed between 1 and 6 months, and around 18 at 4 years or more after the intervention. Most effect sizes were small or very small and, as noted in discussions above about the baseline characteristics of the participants included in the reviewed studies, five papers reported large SDs around the means or large CIs, suggesting that a small proportion of the whole population was skewing the findings.

This finding is important, as it could be interpreted as suggesting that programmes targeted at whole populations might not be the most effective use of resources. Targeted programmes might achieve more change at lower cost. However, balanced against this is the need to identify individuals at highest risk and to ensure that they take part in interventions and are not stigmatised for doing so. The evidence from the included studies that those who drop out of general population programmes are, in general, those who are likely to be less prosocial suggests that these programmes might be good at screening for such at-risk individuals, but may not be the most appropriate approach for those who are most in need of effective interventions.

Most effects in the included studies were in elements hypothesised by the behavioural change model as being intervening variables between the programme interventions and the adverse outcome of victimisation and perpetration. The biggest effect sizes were found in levels of knowledge. In most cases, the knowledge differences between intervention and control groups, or between baseline and follow-up, were most evident immediately after the intervention and up to 6 months later. Knowledge is important in itself, as long as it is sustained over time, but the general lack of strong association between knowledge, attitudes, behaviour and incidence of perpetration/victimisation in the included studies raises a question about the efficacy of the behavioural change model that underpins the programme philosophies in many of the programmes and interventions. The only study that specifically looked for causative process variables51 found that conflict management skills and belief in a need for help were associated with at least some of the perpetration and victimisation variables assessed. However, the associations did not always reach significance, and there are problems with the data analysis approach in this paper (and particularly with how the large number of missing data were handled), as noted above. Indeed, in the original analysis of the Safe Dates data,54 knowledge of services and help-seeking showed large increases immediately post intervention, but while knowledge about services persisted up to 6 months, help-seeking did not. 55

Help-seeking was also a measured outcome for the programme run by lawyers. 45 Perhaps unsurprisingly, perceived helpfulness of lawyers and help-seeking from them were both increased up to the final point of analysis (1 year). For the subgroup of those with lower levels of English proficiency in this study, help-seeking from doctors and nurses was also increased, although the size of the increase was only moderate.

For some programmes, increases in adverse outcomes are reported, suggesting that the use of one-tailed tests in some of the other included studies may have missed the opportunity to identify unintended consequences secondary to other programmes. In Taylor et al. ’s later evaluation,47 although the building-only intervention improved some outcomes, the prevalence of general sexual harassment increased at the 6-month post-intervention point. In the earlier Taylor et al. study,48 there was an increase in reported perpetration of all violence at the 1-month point. Jaffe et al. 64 showed negative changes in attitude scores in 6 of 48 attitude items by 6 months post intervention.

Studies showing a moderate or greater positive influence on perpetration and/or victimisation

Three programmes reported an impact of their intervention on incidence measures. Foshee et al. 52 found that, overall, those in the intervention group did better than the controls. However, the intervention subgroup who also got the booster did worse than the intervention subgroup who did not get the booster. The differences were evident in reductions in physical perpetration, serious physical perpetration, sexual violence, victimisation for serious physical violence, victim sexual violence and any acts ever perpetrated/experienced. The reanalysis of the data from their programme51 found moderate reductions in a range of perpetration and victimisation variables but, as noted above, the treatment of missing data and the use of multiple analyses might mean that these findings are confounded.

As well as finding increases in general sexual harassment, Taylor et al. 47 also found decreases in the prevalence of dating violence victimisation for the building-only intervention in their programme. This conflicting finding is difficult to explain, and the meaning of the finding of a 50% difference is hard to interpret, as the basic descriptive data were not given.

The only study to find a moderate difference in the incidence of perpetration of violence based on apparently robust analysis beyond the initial few months after the intervention was Wolfe et al. ’s evaluation of the ‘Fourth R’ programme50 at 2.5 years. The difference was, however, small (2.4%), and the 95% CI ranged from 1.00 to 6.02, which includes the possibility of no effect. Subanalysis by those who had dated in the last 12 months did not reveal a difference. The effects were higher for boys (see Gender, below) but the sample was unusual, as the incidence of reported perpetration of violence by girls was much higher than that of boys, a finding that ran completely counter to those of all of the other studies included in this review.

Do programmes work for specific groups?

Twenty-two papers examined gender effects. Subanalyses were also carried out in six papers (four programmes) for groups that were at increased risk of victimisation or perpetration at baseline. 46,52,54,55,67,70 Only one looked for age differences by junior or senior students64 and one subanalysed by ‘advanced’ level. 65 Two looked for differences in ethnicity52,62 and one in levels of English proficiency. 45 The results of these subanalyses are given below.

Results

Programme aims and theoretical basis

The majority of programmes targeting young people in secondary schools or in young people’s services (n = 1579–85,88,94) broadly aimed to raise awareness, increase knowledge and change attitudes to domestic abuse in order to equip participants to learn how to conduct non-abusive intimate relationships. In addition, information on help-seeking and services was provided so that those experiencing domestic abuse either in their own or in their parents’ relationships could seek appropriate support. UK interventions differ from the predominantly North American programmes included in the systematic literature review which focused on ‘dating’ violence and therefore tended not to address children and young people’s experience of domestic abuse in their parents’ relationships. Programmes for primary school children in the UK had broader aims focusing on children’s safety and friendship, although raising awareness of domestic abuse was introduced to older primary school children (see Programme content). All but one of the four programmes aimed solely at primary aged children were based on ‘Protective Behaviours’, a personal safety programme which in essence is ungendered. 95,96 No reference was made to domestic abuse in the evaluation of ‘Miss Dorothy’,78 a programme specifically designed for children of primary school age that focuses on personal safety through encouraging self-empowerment and raising self-esteem and which aims to prevent experiences of victimisation. However, it was retained in the review as it was the second most commonly reported programme in the mapping survey (see Chapter 3).

The programmes ‘Challenging Violence, Changing Lives’,91,92 ‘Practical Prevention’93 and ‘Tender’,79 while having similar aims to those secondary school programmes discussed above, were also aiming to implement prevention through a ‘whole-school approach’. Along with curriculum work, a whole-school approach involves action at institutional and policy levels, the engagement of all members of the school community, awareness raising and training for staff and work outside the classroom with students. A whole-school approach has been widely adopted in the UK to help children and young people learn about many issues including health,97 bullying98 and, more broadly, values and affective education. Some developers of prevention programme stress the importance of a whole-school approach in order to change school culture and to create an ethos where violence and abuse, including gender-based violence, is not tolerated. 99,100 In three cases, theories explaining domestic abuse which informed programmes were made explicit. In some cases, the theoretical basis of the programme was implicit rather explicit. For instance, it was evident that in 11 programmes a feminist understanding of domestic abuse either explicitly or implicitly informed the rationale given for the work or had shaped the programme content. Three programmes – ‘Helping Hands’,86 ‘Practical Prevention’93 and ‘Tender’s Healthy Relationships’79 – incorporated a theory of change, or logic model, in which the goals, indicators of success and the actions to achieve goals were articulated. Other programmes did not articulate how raising awareness about domestic abuse or healthy relationships, a main aim in many programmes, linked with changes in behaviour or social action.

Programme content

A complex set of topics and themes ran through the reported programmes and these varied with the age of the audience. None of the reports included extensive detail of lesson content although most (n = 19) provided a list of topics. Table 23 provides a summary of the most common topics across programmes. As noted above, for those young people in secondary schools and young people’s services the topic of domestic abuse was directly addressed, although over time there appears to have been a move to shift the focus to abuse in young people’s own relationships90,92,93 rather than domestic abuse in their parents’ relationships. 81,88,94 This shift may reflect the impact of Barter et al. ’s study,4 which provided evidence on the prevalence of domestic abuse in young people’s relationships in the UK. Work in primary schools was focused less on domestic abuse and more on wider relationship issues such as friendship, respect and children’s safety. Where domestic abuse was tackled, it was with children aged 8 years and over. In those cases where programmes were designed for both primary and secondary schools, two complimentary programmes were created,38,81,88 which notionally offered opportunities for progression and continuity for learners.

Many of the programmes (n = 18)38,79,81–84,87,92–94 dealing with domestic abuse addressed topics relating to gender such as gender stereotypes and roles, gender (in)equality and discussions of power in relationships. However, it was unclear whether other programmes, where coverage of such topics was not identified, used gender-neutral definitions or explained abuse as an outcome of interpersonal conflict rather than locating it in the broader social context of gendered power relations. Changing attitudes to gender violence was expressly stated as an aim in a small number of programmes (n = 3)79,83,88 with aspects of values education, including topics such as equality, acceptance of difference, respect for self and others and responsibility included in 19 programmes. 38,79–85,88,91–94 The inclusion of work on children’s, girls’ and/or women’s rights was explicitly reported in three programmes,81,91,92 although this does not mean, necessarily, that these topics were not included in others.

Affective education, in the form of learning to identify, name and express emotions in non-violent ways, was included, and this was particularly evident in programmes for primary school children. Lang et al. define affective education as that part of ‘the educational process which is concerned with the, feelings, beliefs, attitudes and emotions of students, with their interpersonal relationships and social skills’ (p. 4). 101 In the ‘Protective Behaviours’ programme where ‘keeping safe’ was the central tenet, the focus was on children learning to identify, trust and act on their feelings about being (un)safe. Often linked with this were attempts to raise children’s self-esteem with an underlying assumption that this would help to avoid victimisation. The teaching of prosocial skills to equip children and young people to conduct non-abusive relationships was evident in 10 programmes;38,79–85,88,94 this variously included conflict resolution/management, anger management, communication, problem-solving and assertiveness.

Topics aimed at supporting children/young people who had lived with, or who were living with, domestic abuse were evident. Coverage of help-seeking and information on support services was reported in 11 programmes. 38,78,79,81–84,88,93,94 In one case, specific mention was made of an initial activity about respecting boundaries within groups. 81 This was undertaken to establish a safe and respectful environment but was also linked to confidentiality and safeguarding in the presumption that the work might elicit disclosures of maltreatment from some children and young people. Although a single example, this does not preclude the possibility that other programmes addressed such topics. The grey literature reviewed yielded one other example of work on boundaries38 comparable with Taylor et al. ’s47 New York programme, discussed in the previous chapter. The importance of creating a safe environment for these interventions was noted by Manship and Perry, who concluded that:

A safe and confidential environment in which participants can reflect on their own experiences, their hopes and aspirations for the future and talk openly was felt to enable programmes to run more effectively and to encourage high levels of engagement.

p. 3182

Responding to disclosures

An argument for undertaking prevention work is the number of children and young people who have experienced domestic abuse either in their families or in their own relationships so that one broad aim of such programmes is to support those currently living with domestic abuse. 37 Despite this, very few evaluations explicitly reported on provision for those who disclosed maltreatment as a result of taking part in programmes. In general, those programmes delivered by specialist domestic abuse organisations would have had direct links to support services within their own organisations and the assumption is likely to have been made that schools would deal with disclosures through existing safeguarding protocols. The ‘Zero Tolerance’ programme had a helpline available, although no report of its use was provided. Forty-four per cent of the staff interviewed for this evaluation reported being aware of children/young people who had required support during the project and ‘issues had arisen in all but two of the settings’88 (support for staff had also been made available and two who had experienced domestic abuse had sought help). At least three of the seven projects reported on by Manship and Perry had integrated one-to-one or group therapeutic work and this was ‘felt to be a vitally important part of the programmes’ (p. 32). 82 The ‘Domestic Violence Awareness Raising’ programme provided open surgeries each week in all but one school where the programme was delivered in which participants could talk about issues that had arisen from the programme. These were popular with children in primary schools, especially girls, but there was no take-up in secondary schools. 81 This was the only evaluation where the number of disclosures that occurred during the period of the programme was reported even though anxieties about prevention work in schools eliciting disclosures which then have to be managed are often reported81,102 (see Chapter 6). In this case, ‘11 children/young people disclosed child abuse or domestic violence from a cohort of 532 (a ratio of approximately 1 : 48) (p. 181). 81

Diversity

Very few programmes appeared to take account of or to address the specific issues faced by BAMER, LGBT or children and young people with learning difficulties or disabilities so that little attention was paid to addressing the complexities and issues for marginalised groups of girls and women. However, two programmes, one developed and delivered by Southall Black Sisters and the other by Respond, were specifically designed to work with and address, respectively, issues faced by BAMER young people and young people with learning difficulties who attended special schools. 93 Manship and Perry82 reported that the content of the seven programmes they evaluated in Kent and Medway did not address LGBT or male victims and lacked cultural sensitivity for BAMER students:

It is through running prevention programmes that further needs could be identified . . . Gaps in the content of programmes could . . . be established, including support for Lesbian, Gay, Bisexual and Transgender young people and the need to provide appropriate information around male victims of domestic abuse (since many of the programmes discussed in the Findings are aimed at women and young females). Finally, the issue of cultural awareness is an important factor for consideration.

pp. 35–682

Recognition of equality and diversity had been built into the Northern Irish ‘Helping Hands’ programme in the ‘design and resources at all levels’ (p. 20),86 including making the materials accessible to Irish speakers. However, some teachers suggested amendments be made to some aspects of the resources ‘for pupils with complex learning needs . . . to enhance accessibility, diversity and equality best practice’ (p. 21). 86

Staffing and training

The majority of programmes reported in the evaluations reviewed were delivered by external staff (n = 16);38,79–83,88,90,94 of these, 12 were staff from specialist domestic abuse or violence against women and girls organisations. 38,79,81–83,94 The remaining four were delivered by police officers,90 police community safety officers,82 arts professionals79,83 and professional actors from the National Youth Theatre who worked as peer educators. 80 Despite the increasing use of peer educators in the USA,63 only one other example of this approach was reported in the evaluations reviewed. In the 2-year intervention by Tender,79 much of year 1 had been focused on developing a peer education model. The intention, as part of a whole-school approach, was that the young people who had previously trained in peer support would act as ambassadors or advocates for healthy relationships and influence the peer culture within a school. There was also an intention that they would be trained to deliver sessions in addition to those led by teachers.

Little information was reported about the qualifications or training of the staff from specialist domestic abuse organisations involved in programme delivery, although three programmes were delivered by staff identified as children or young people’s workers. Hale et al. 38 reported that the ‘Relationships without Fear’ programme was ‘delivered in schools by specialist practitioners who work for Arch’ (p. 92); however, ‘specialist’ in what respect was not made clear. Manship and Perry reported that the staff who delivered programmes in Kent and Medway had ‘Relevant postgraduate and undergraduate degrees (including teaching and youth work)’ (p. 26). 82 Staff external to schools who are involved in programme delivery had also undertaken, among other things, ‘specific training courses including working with young people affected by domestic abuse and working effectively with schools (provided by Respect, AVA [Against Violence and Abuse] and Tender)’ (p. 26). 103

Where practitioners from external organisations delivered the programmes, school staff were often offered training. This focused on increasing school staff’s knowledge of domestic abuse and information on the programme content. There was often an expectation or insistence that teachers would ‘sit in’ on the lessons. This was usually to help manage children and young people’s behaviour, to deal with any disclosures or to support children if they became upset and needed to leave the session (see for example Ellis,81 Manship and Perry82 and South Essex Rape and Incest Crisis Centre84). Again, there was little information about the content of the training staff had been offered, although data on the take-up of the training were given in two of the evaluations. Five of the eight schools involved in the ‘Respect’ programme84 had sent ‘the key member of the staff’ to the in-service training session. Uptake of such training was inconsistent in the nine participating schools in the ‘Domestic Violence Awareness Raising’ programme; in three schools no staff took part in the training and yet in one primary school all most of all of the staff (n = 35) had undertaken it. 81

Ten programmes were delivered by teachers in partnership with staff from a specialist domestic abuse organisation and two in partnership with staff from children’s organisations; the latter were aimed at primary school children. 78,87 In all of these instances, the teachers received training prior to the intervention from the staff in the partner organisation or from colleagues who had undertaken training in order to train others to deliver the programme. 79,85 In all of the evaluations reviewed, staff who received training mostly reported finding it positive and helpful, but in almost all cases there were suggestions for changes. For example, Reid Howie Associates88 reported that 97% of staff thought the training was ‘OK’ or ‘better than OK’, although about half said that their understanding of the issues had not improved and some thought that the teaching material needed better explanation. For most staff, the training was crucial to their being able to deliver or support a programme. However, ongoing support was also important, as ‘Teachers who did not feel supported were likely to feel under-confident in using the materials and dealing with the issues’ (p. 25). 94 However, in the case of the 10 programmes delivered by teachers themselves, ongoing support throughout the intervention was usually provided (see for example Reid Howie Associates88).

The gender of staff delivering programmes was explicitly reported and discussed in two evaluations. Manship and Perry noted that five of the Kent and Medway initiatives were facilitated by women, one by a man and two by a mixed team. One of the male facilitators regarded his gender as advantageous as it ‘gave a different perspective, especially for the young male participants’ (p. 25). 82 Good practice guidance developed in the UK for adult perpetrator programmes suggests that men’s groups should be facilitated by at least two staff, one woman and one man,103 and one mixed pair endorsed this approach and saw it as ‘providing role models’ to the young people (p. 25). 82 The significance of the gender of facilitators (both women in their twenties) was highlighted by a male teacher in Ellis’s evaluation. In reporting on the negative responses of some of the boys, the teacher noted that:

It almost got their backs up a little bit – to have two women coming in to say to them, you know, men are bad or men are the ones who do domestic violence. So they were sort of being defensive of their sex then, their gender.

p. 15781

No references were made in any of the evaluations reviewed regarding the ethnicity of facilitators or the impact this might have had on group dynamics and learning.

Delivery methods

A wide variety of methods was employed to deliver the work, including whole- and small-group discussion, direct instruction, role-play, creative and performing arts, quizzes, worksheets, real-life stories, identifying true/false statements, DVDs and clips, writing exercises, graffiti walls and art. In primary schools, circle-time was used along with jigsaws, poetry, music, stories and movement activities86 [Women’s Aid Federation Northern Ireland (WAFNI) and Department of Education Northern Ireland] with less use of role-play and whole-group discussion. 37 Circle-time aims to raise children’s self-esteem and is based on a belief that the most effective way of promoting positive behaviour and respectful relationships is through working on moral development. It is used widely in PSHE in primary schools to explore a range of issues. In practice, it involves all participants sitting in a circle with everyone taking equal responsibility for solving the problems and issues that are brought to the circle. The group agree ground rules but principally participants must take turns to raise their issues or ideas, speak and listen.

A considerable number of programmes had been designed to include a breadth of methods that might engage children/young people. Six evaluations reported the use of drama or theatre, which are widely thought to be an effective way to deliver PSHE because they are interactive and participatory. 104,105 Some specifically employed drama that was created and performed to peers by young people themselves,79,84,89 and in one case the performance was delivered by students from a local college. 94 ‘Cheshire Schools Project’ employed a professional theatre company to perform a play to present information and act as a stimulus for discussion. In the ‘Westminster Programme’,85 ‘forum drama’ was used: this involved a theatre group performing with young people invited to guide the characters to act differently, and aimed to explore power and control in gender relations. 85 Forum theatre is derived from Boal’s106 ‘Theatre of the Oppressed’, which aims to use theatre to achieve social change (see discussion of Belknap et al. ’s study59 in Chapter 4). Drama and theatre are less dependent on literacy skills than many other methods and may be more inclusive, especially with children/young people for whom English is an additional language or for those with learning difficulties.

Maxwell et al. ’s92 account of the development of a whole-school approach reported the use of peer research as a way of raising gender abuse as an issue in school and determining the content of taught programmes. This approach was deployed alongside the use of peer mentors, the establishment of a subgroup of a school council and campaigns by young people within school.

Audience

The majority of programmes selected for this review (n = 25, 86%) were targeted at young people in secondary schools, usually those aged 11–14 years, although some were aimed at older students too (Figure 16). Six programmes worked with both primary and secondary students, with only three working with young people outside schools in non-formal settings. Targeting young people was explained as a particularly important period for intervention because it was conceptualised as a key transition point and a time when gender abuse manifested itself. As Mahony and Shaugnessy state, the programme ‘is situated in secondary schools because we recognise that this is a crucial time when young women and men are developing their sexual and social identities’ (p. 1). 91

FIGURE 16.

Programme venues.

All of the programmes worked with both boys and girls in mixed-gender groupings, with three examples of single-gender delivery which resulted because the schools or groups involved with the programme were already single-sex organisations or groups; for example, this approach was adopted in a girls’ school, a boys’ school and a girls’ group in a young people’s centre. Maxwell et al. ’s92 report was the only evaluation describing an intervention focused on gender equality that was intentionally delivered to a girls’ group. This does not necessarily mean that this was not done elsewhere.

The numbers of children and young people involved in programmes ranged from 20 to 8000 (see Figure 17 and Table 22). Of the 20 reports of participant numbers, the majority (n = 15) had had fewer than 1000 taking part, with eight programmes having between 101 and 500. While these figures show differing levels of provision, they can be misleading in that the time period over which they had been delivered varied greatly, as did the overall length of time programmes had been operating and levels of programme funding.

FIGURE 17.

Numbers of participants in programmes (note: data are available for only 20 programmes).

Only two evaluations provided description of the geographic area where the programme was delivered, locating the work in a social and economic context. 38,92 Two other studies reported the characteristics of those included in the study, including age, ethnicity and self-reported special needs. 81,83 Worthy of note, however, is Maxwell et al. ’s finding that ‘Despite the five schools being quite different demographically, similar issues of gender inequality and experiences of sexual bullying and abuse were found in all of them’ (p. 17). 92

Children’s and young people’s views of programmes

Children’s and young people’s responses to a programme are an important measure of the degree to which they engage with it, which in turn impacts on their learning. 107 Ten of the evaluations reviewed reported qualitative data about a range of topics on aspects children and young people liked, did not like or would change about programmes. The comments can be grouped around overall views, programme content and methods of delivery. Children’s and young people’s views of their learning are reported in Outcomes of school-based programmes, below.

The evaluations reported that, overall, the vast majority of children and young people were interested in programme topics, took it seriously, and saw it as worthwhile and positive; as Hester and Westmarland noted, ‘Children and young people across the projects indicated they wanted and valued lessons on relationships and on abuse’ (p. 21). 94 Similar comments were reported by, for example, Reid Howie Associates,88 Ellis,81 Maxwell et al. 92 and Hale et al. 38 As one primary school girl said, ‘We should have had [the programme] from the start of the year until the end of the year, and that would actually be more fun (Focus group 1, girl 3)’ (p. 4). 38 Hester and Westmarland94 reported that pupils in one project (Thurrock), who had fewer sessions, had been less positive. DMSS noted that ‘the size of the groups, age and gender balance also made a difference to how they [young people] engaged’ (p. 49). 79

In many cases, the ways in which the programme was delivered emerged as an important factor in student engagement and enjoyment. Participative and interactive methods were valued by children and young people, including creative and visual activities such as drama and role-play.

I thought they were excellent . . . because like science and stuff you have to actually write quite a lot of stuff out. But like the citizenship lessons that we actually did the domestic violence on, they were kind of like hands on and we could actually do things ourselves and like act out things and choose like how we would actually deal with the situation. Instead of just writing things down on a piece of paper or whatever.

Boy, p. 2485

Games that involved activities such as ‘[True/false card sorting . . .] were quite good because you got involved in them (Focus group 5, girl 3, secondary)’ (p. 45). 38 DVDs, discussion and group work were all reported as enjoyable by children and young people: ‘Young people preferred teaching methods such as group discussions, role-plays, student-led tasks and using visual material to deliver the lessons’ (p. 69). 92 However, some children and young people found interactive approaches challenging; these approaches required a level of group work skills and group management skills on the part of facilitators that some did not possess, often leading to confused and noisy lessons/sessions. This was a common theme in many of the evaluation reports reviewed and was identified as a key aspect that children and young people would like to change:

. . . ‘everyone talking at once’ or ‘couldn’t hear cos too much talking’

p. 2180

I didn’t like when everyone was shouting and then you didn’t get attention

Girl, 10, p. 17381

The sessions could be more structured and the Workshop Leaders needed more control over the class.

p. 2779

Overall, the evaluations reported that most children and young people thought that the content was relevant and interesting, although the gendered approach to domestic violence adopted by most programmes was sometimes resisted by some boys who described the work as ‘anti-men’ or ‘sexist’ (as did some adults). A defensive approach from some boys was explicitly reported in five evaluations. As one participant stated, ‘That’s not fair when it’s always men beating women – it’s sexist against men (Focus group 2, Boy 1)’ (p. 46). 38 Little detail was given as way of explanation in the evaluations; however, in some cases this might be a consequence of programme delivery being overcondensed and rushed,81 and there may be a need to consider who these ‘resistant’ boys are and to work with these responses.

Where children and young people were specifically asked what they would change about programmes, two themes were commonly reported (in addition to improved group dynamics noted above): ‘nothing’ and the length of the programme. Several evaluations (see e.g. Reid Howie Associates88 and Ellis81) reported that a proportion of children and young people said they would change nothing – this could be viewed as either a level of satisfaction or a reluctance or lack of interest in critically commenting. The length of the intervention was commented on in at least five cases. DMSS reported that:

Several young people wanted more time – either because they were enjoying it so much: Make the project longer – it was so much fun! Or because they thought the final product would have been better with more time: It would have been better if we spent longer on the final performance so it was slicker. Then we could have put on more nights and invited bigger audiences.

p. 2779

Similarly, Walton reported young people stating ‘there was not enough time’ (p. 21);80 this was noted particularly where there was a single session which limited the extent to which young people could fully engage with the complexity of the issues. However, some young people who had received multiple sessions also reported wanting more: ‘I would have liked more time and longer lessons because they were interesting and it brought the subject to your attention (Girl, 13)’ (p. 187). 81

Adults’ views of programmes

The evaluations of 12 programmes reported, to differing extents, the views of adults;78–83,85,86,88,91,92,94 these were either facilitators or teachers who supported external staff who had delivered programmes. Overall, it appeared that adults thought programmes were ‘successful’, although the criteria against which this was judged were rarely made explicit. Certainly, most adults thought programmes were on the whole well received by children and young people; this may have formed part of adults’ assessment of ‘success’. There was also general agreement that programmes raised awareness of domestic violence where this was the aim and that this enabled young people to think in a more informed way about forming non-abusive relationships:

It makes the students look at relationships in different ways. They were totally engaging and the students were interested.

A teacher discussing the Swale Action To End Domestic Violence Healthy Relationships Project (SATEDA), p. 3082

In a small number of evaluations, some teachers, both those delivering material themselves and those supporting external staff, were uncomfortable with the topics. Ellis81 found that a small number of teachers thought the topic inappropriate for primary school children and secondary school staff expressed anxiety over how parents might respond to the topics, particularly where sexual assault was discussed. Hester and Westmarland reported that ‘a small percentage of teachers were uncomfortable with the focus on gender’ (p. 21);94 perhaps this was more clearly articulated by Mahony and Shaughnessy, who stated that:

Whilst teachers’ perceptions of the programme was generally positive, they often lacked confidence and felt ill-equipped to handle the perceived ‘political nature’ of the content.

p. 491

A level of adult resistance to a gendered approach was noted in five evaluations, which led in some cases, where the work was delivered by teachers, to ‘some [teachers] subverted the content and instead approached the topic in ways they considered more sensitive and less confrontational’ (p. 21). 94 This raises the ongoing debate about advantages and disadvantages of external versus internal staff delivering programmes (see Chapter 6 and Fox et al. 75). On the whole, support for programmes was positive from teachers; however, comments from both young people and staff identified ways in which teachers could undermine programme messages through inappropriate remarks. Walton reported that ‘One observer noted that in the workshop she saw, staff members said things which were either unhelpful or factually incorrect’ (p. 22). 80

Outcomes of school-based programmes

Evaluations of 13 programmes reported findings that suggested that children and young people gained increased knowledge and understanding of the nature and extent of domestic abuse after participating in programmes. In addition, where addressed, learning was reported about help-seeking, rights in relationships and gender equality. Some attitude change was reported in five studies. In only one programme was behavioural change reported as an outcome – the ‘Educational Domestic Abuse Project’ (‘Project Salus’) – although no detail was presented in the report. 82 The main outcomes for each programme are shown in Table 24. In all cases, however, it is difficult to know if such change was sustained as only one study included a follow-up period, of 3 months for the intervention group. 38 The lack of evidence for longer-term outcomes and for behavioural outcomes is partly owing to limitations of many of the evaluations which were small and local. It is also worth noting that many of the programmes aimed to increase knowledge and understanding and change attitudes rather than behaviour.

In a number of the evaluations, children, young people and adults were asked about the impact on participants and their learning. It was reported that most teachers regarded the impact and outcomes as positive for children and young people. In the ‘Cheshire Schools Project’, teachers thought that primary school pupils had understood the messages (97%), gained knowledge (81%) and gained in empathy and understanding of responsibilities in relationships (86%). 94 Similarly, Thiara and Ellis85 noted that teachers delivering the work:

thought [it] was an enriching experience for children and young people, allowing them to build skills (speaking and listening) which had also enabled them as teachers to relate differently to young people (listening to them and getting to know them).

p .2785

Positive changes in the behaviour or attitudes of young people were reported in response to the Tender programme, with teachers stating:

A couple of girls were getting detentions for poor behaviour and this has reduced dramatically if not completely ceased. I have seen a real boost in their self-esteem through being attached to the project.

One boy has been having many issues around the school, but he has attended every lesson and has been really active.

p. 3379

Short interventions, usually referring to a one-off session, were regarded by teachers and external facilitators as having short-term impacts94 and as even counter-productive. 81 DMSS reported that young people who received a ‘high dosage’ input were more able to recall the content of sessions and were more reflective than those who were exposed to the programme for less time and that they ‘tended to comment on their personal development as well as their learning’ (p. 31);79 for instance:

I learnt a lot of skills as well, leadership skills, because it was hard going up there in front of people. It was good to learn those skills, because I am a lot more confident now.

p. 3179

Children and young people themselves reported on things they had learned:

It was a subject I had never discussed before and I found out things I never knew.

p. 684

I have learnt what the early warning signs are and that people are willing to help me.

p. 2579

I [went] home after learning about abusive relationships and it made me feel like encouraged that I have got a nice home to go to and I am a bit more like . . . my attitude has lifted . . . I didn’t stomp up the stairs when asked to clear my bedroom – I just got on with it.

p. 4792

I have learnt being horrible to others is not nice.

p. 19381

It is also noteworthy that, irrespective of the length of a programme, unanticipated outcomes were often reported, suggesting that it is not possible to precisely predict learning as children and young people interpret and make sense of the content through their existing knowledge and experience; such outcomes could be both positive and negative. One girl, aged 13 years, stated that the programme had reinforced her view that it was possible for women to ‘live your own independent life’ (p. 202)81 and yet women’s rights had not been explicitly mentioned or discussed. Hale et al. , however, reported that ‘some children negative unintended messages away from the programme – especially in relation to stranger danger and the culpability of victims’ (p. 15). 38

Media campaigns

The Home Office led and funded the ‘This is Abuse’ campaign, which began in February 2010 and was delivered in six waves up to April 2014. Table 25 shows details of the campaign waves, their focus and the dates delivered. Unpublished reports produced for the Home Office and shared with the researchers are the basis for information included in this section. The reports provided data on the six waves of the campaign listed in Table 25.

A considerable amount of formative research was undertaken in the planning of the initial campaign, which was targeted at 13- to 18-year-old boys and girls, with a slight bias towards those in social classes/grades C2, D and E, parents and carers and a range of agencies working in partnership with the Home Office on the campaign. The principal aims of each wave of the campaign were to challenge attitudes, mobilise communities and provide information on relevant services. The campaigns were multiplatform and included a dedicated website, television, advertisements in a range of locations including online and on mobile devices, the production of materials for partner agencies to distribute and use, and material for parents/carers. The campaign had developed iteratively, with data from each wave being used to inform the development of subsequent ones. Latterly, the focus had shifted towards the conduct of boys as potential instigators of abuse rather than on girls as victims.

The findings reported here are based on material supplied in January 2014 and do not take account of information that may have become available subsequently. The evaluation of waves 2 and 3 of the campaign showed that partner agencies experienced increased levels of help-seeking and the use of ‘Need Help’ section on the dedicated website also provided evidence of the response to these waves of the campaign. There were over 540,000 visits to the website during 2011–12, with 2500 comments being made on the discussion forums. A high proportion of comments (48% on the rape prevention campaign) were from victims of abuse. Longer-term outcomes from the campaign have not been captured; however, such measures are difficult to obtain. 108,109 The comments left by some young people on the campaign website’s discussion forums show that the adverts helped them to understand what abusive behaviours were and that they had been raped:

I used to convince myself that what happened to me wasn’t rape and that it was my fault, but after seeing this advertised made realise that what happened to me wasn’t my fault and I feel so much better about myself that he was an animal and I was just the victim. Thank you for this video as I was only 15 and am now 16, has helped a lot.:)

this website is so good and helpful! I never spoke to anyone about my experience because i was worried no one would understand but on here people who actually know what you have been through can reply to you, it’s helped me so much and now i am seeking the right help to get my life back to normal and to deal with my feelings. I’m so relieved i found this website, I’m now having councilling but i could have never done it without getting advice from this website, i hope everyone else on this website who has been hurt can now feel more confident and now feel as though they aren’t alone :) xxx

Aims of preventative interventions

Group members identified raising awareness of domestic abuse so that young people could recognise it in their own relationships as a legitimate and important aim of any preventative intervention. They argued that young people often see domestic abuse as something that happens between adults and that successful interventions were those that would make young people recognise that this something that was happening in their own lives. This point was made in more than one group, but one group member summed up the message that needed to be delivered to young people as: ‘Even if you don’t realise it, it’s abuse; it is abuse, sometimes you don’t know what abuse is’ (young people’s consultation group 1). Another participant pointed out that developing young people’s consciousness of the prevalence of domestic abuse could facilitate disclosure of personal experience:

. . . if you’re being abused and . . . you might not realise it’s abuse and you might not think it’s that important, but if you see all these statistics on the TV and anywhere else where it’s being advertised . . . you might feel more comfortable with it, thinking that it’s something which you can have a say in because it’s something that matters and it’s not something which is just happening to you . . . but is something you can speak about openly . . .

Young people’s consultation group 1

However, some young people made the point that awareness was necessary for change but was insufficient on its own. They argued that young people needed to know that, once they acknowledged that they were in an abusive relationship, services would be available to help them: ‘it makes people aware but then they need the help afterwards’ (young people’s consultation group 1). Group members commented that young people also needed to know the consequences of seeking help:

I think that they should bring to light what actually happens after you call the people . . .

Young people’s consultation group 1

It was noted that there was a fear that disclosure of domestic abuse in their parents’ relationship might lead to children being taken into care.

A particularly sophisticated critique of the Home Office’s ‘This Is Abuse’ 2012–13 media campaign (see Methods of delivery for further views on this campaign, which was generally positively rated) was delivered by one group participant, who argued that the campaign focused too heavily on individual responsibility for changing behaviour and that more emphasis should be placed on the role of wider social support:

It [abuse] kind of seems to be something you take upon yourself to stop doing . . . it’s only the two people in the relationship’s problem and not anybody else’s . . . it’s kind of just reminding them of the situation they’re in, rather than what to do.

Young people’s consultation group 1

Methods of delivery

Most of the young people participating in the consultation meetings had encountered the Home Office’s ‘This Is Abuse’ campaign (discussed in Chapter 5) through the television. In the main, they considered this an effective, hard-hitting campaign and they noted that television campaigns had a wide reach, did not require the audience to make an active decision to view them and could be watched without incurring stigma (young people’s consultation group 1). It was also noted that television or film had the capacity to reach those who were less able or ready to use written material:

. . . having moving images, videos, is more effective than text. (Male)

Yeah, you’ve got to read text and if there’s a lot of text then a lot of people don’t want to read it. (Female)

Young people’s consultation group 2

Group members were less enthusiastic about the use of advertising or pop-ups on social media to communicate preventative campaigns. There was a sense that such messages were intruding on their space and attempting to distract them from their own activities:

. . . people will always put adverts on Facebook and on Twitter and on YouTube thinking that young people are going to watch them but I don’t think they ever work really, because when you’re on Facebook and YouTube and Twitter . . . I don’t think you want to be coming across these kind of things. . . (Female)

. . . it’s a bit like cold calling really . . . (Male)

Young people’s consultation group 2

However, it was noted that when such information was communicated via social media by someone who was well known, its credibility and authority were higher and it was more likely to be viewed.

Active participation in prevention initiatives was considered to enhance learning – ‘people learn better by doing’ – (young people’s consultation group 1) and group members were positive about their involvement in school peer mentoring schemes which had been established with the aim of older students supporting younger students with a wide range of problems and queries. Raising awareness through participation in creative arts was described by one group member who had been part of a ‘flashmob’ dance performed globally on the International Day for the Elimination of Violence against Women. Other group members reported reading about it in the local newspaper or watching the performance in school. The participative element together with the public recognition that the event achieved at the levels of the school, the community and nationally were felt to be important for achieving momentum towards change:

. . . doing it on a regional level but I think when you do something and it’s on more of a national level . . . you like feel like you’re a part of something, like obviously the dancing, there are people doing it all over the world, so you felt like it’s a worldwide movement.

Young people’s consultation group 1

However, it was noted that not all young people had the confidence to participate in drama or dance. Non-participative theatre could also be an effective medium for engaging an audience:

. . . it’s quite eye-opening and people are sat there watching it . . . you’ve not really got much of a choice but to watch it.

Young people’s consultation group 1

Gender and sexuality

It was argued that some preventative programmes were ‘biased’ in their focus on women as victims of domestic abuse and that this perception could provoke resistance. One young person noted in response to reading materials designed for discussion in PSHE classes:

. . . it all seems so biased . . . it just seems like it’s only men that do it. I think that it would cause a lot of tensions and arguments . . . if I was a teacher . . . it’d be a lot of effort to plan something that would probably end up causing people to just start moaning or arguing . . .

Young people’s consultation group 1

Both young women and young men participating in the consultation groups felt that it was important that interventions also identified boys as vulnerable to abuse in their relationships, and they cited a storyline in Coronation Street (a popular soap drama) which depicted a man as the victim of domestic abuse:

I don’t know if anyone watched Coronation Street, but Tyrone, the storyline of Tyrone getting abused by his wife, rather girlfriend, and he never fessed up about it . . . and it’s just situations like that actually do happen in real life and it’s because there’s no awareness about it.

Young people’s consultation group 1

At the second meeting of the consultation group, group members reported that they had been inspired, in part by the experience of participating in the preventing domestic abuse for children and young people (PEACH) study, to write three short drama pieces on the theme of domestic abuse. By the end of the study, these had been performed and produced on DVD by a local theatre group and distributed to schools for use in PSHE classes. The three scenarios involved:

A male being violent to a female, female being violent to a male and then a same-sex couple . . . we wanted kind of the same thing to happen in each one, but it was more to see the reactions to . . . the changing of who kind of carried out the violence to see if that made an effect on people . . .

Young people’s consultation group 2

An emphasis on approaches that emphasised equality between men and women was reiterated across the groups in different ways. This young woman argued for programmes that identified commonalities rather than gender difference:

I think to an extent, what they need to get out of these classes, is they’re [boys and girls] not that different. If maybe you’re a guy or a girl, you can exude, I don’t know if that’s the right word, but the same kind of violence in a relationship. And I think the one thing that needs to be emphasised is equality because there is a lot of double standards I think.

Young people’s consultation group 3

Young people also identified the potential for gender bias in the delivery of programmes. Young women in the consultation groups argued that teachers often focused their teaching on boys in the classroom and they suggested that a perception of boys as abusive might feed into this tendency to focus on boys’ needs and could contribute to reports of behavioural changes following exposure to the ‘Fourth R’ programme in boys but not in girls:

More people think guys are more angrier . . . which is why when these lessons or things like that are being delivered, I think sometimes there is a possibility that the concentration goes on guys more than girls.

Young people’s consultation group 3

There was uncertainty whether programmes should be delivered to single-sex or mixed groups, with those who went to single-sex schools saying that it had felt ‘more comfortable and natural’ (young people’s consultation group 1) to address such issues in single-sex groups. Mixed-sex groups were felt to contain potential for gender-based conflict:

Because if you have a mixed class . . . if you showed the advert that’s on [This Is Abuse], some girls will probably say: ‘Oh, I bet he’s like that or whatever’ . . . And then the guy would be like: . . . ‘that girl’s probably frigid’. . .

Young people’s consultation group 3

In the third consultation group, where the young people discussed the possibility of programmes having differential effects on boys and girls, it was suggested that programmes could be delivered in part to single-sex groups and, after some initial sessions, to mixed groups.

Delivering interventions in schools

Although those participating in the consultation meetings had patchy and limited experience of school programmes that had addressed domestic abuse prior to entering the sixth form, young people were clear that schools had a role to play in delivering preventative interventions:

The point of school is to educate so . . . it is really good to do it at school as in terms of raising awareness.

Young people’s consultation group 1

Group participants thought that programmes should be delivered in schools to small groups if possible and on a regular basis:

. . . if they’re coming once a year, that’s not going to make an impact . . . if it’s more regular, that makes more of an impact.

Young people’s consultation group 1

However, there were concerns expressed about whether or not schools could manage the confidentiality required for disclosure and this reflects young people’s insistence, discussed above, that interventions need to take account of the process of disclosure and its possible consequences for young people.

Moreover, some schools were described as lacking the confidence to address the issue:

I think with like some schools . . . they are a bit nervous about bringing in . . .

Touchy subjects.

Because some schools are scared that if they bring it up . . . people in the class are going to tell other people outside the classroom and then it’s just going to get spread . . .

Young people’s consultation group 1

This theme of schools’ concern about protecting their reputations was reiterated by the expert education consultation group.

Another point expressed by both the expert education consultation group and by the young people’s group was the way in which the demands of the curriculum could make it difficult to allocate space to teaching on issues such as domestic abuse. Participants in the young people’s group thought that this was particularly the case for the over-16s, who were preoccupied with exams.

Young people described teachers as often lacking the necessary expertise to deliver programmes on domestic abuse:

. . . you can tell when someone’s like bluffing it . . . especially like teachers, especially when they’ve been given briefs that they don’t know anything about and they’ve just been asked to do a presentation, you can tell they’re practically just reading the slides on their notes, so that’s really bad, so obviously whoever’s doing it got to have the knowledge . . .

Young people’s consultation group 1

Knowledge needed to be reinforced by experience or expertise, and teachers often failed to convey this:

If it’s like just a teacher delivering it and they’ve got no experience and it’s almost like, well why are you telling me? You don’t know anything about it.

Young people’s consultation group 1

Similarly, participants in the third consultation group who discussed the ‘whole-school approach’ (see Chapter 5), were dubious that staff across the school would have the confidence or skills to deliver preventative messages on domestic abuse:

Those kind of teachers where they’re so stuck in the past, where they wouldn’t want to do areas like this. Where they’re just focused clearly on educating their subject and don’t want to let anything else get in the way . . . they might be scared of bringing it up . . .

Young people’s consultation group 3

However, another group member described an English teacher who had succeeded in delivering preventative messages about domestic abuse in the context of studying Of Mice and Men.

Target audiences

In response to hearing that some programmes seemed to benefit some groups of young people more than others, one young person suggested that programmes should be targeted to those who are vulnerable:

. . . those people who are, I’m going to use that word, vulnerable, from either being victims of it and the perpetrators . . . I think people with low, what is it, self-confidence in themselves.

However, this point provoked disagreement:

. . . I just feel very strongly that it should be for everybody because I don’t think you can pick out the vulnerable from the non-vulnerable . . . You also isolate people if you take them out and say, well you’re most vulnerable and you’re the most likely to perpetrate.

Young people’s consultation group 3

The second consultation group noted that school populations could differ considerably but it was thought that all schools should deliver preventative teaching as those from more protected backgrounds should know that domestic abuse ‘actually happens’ (young people’s consultation group 2). Young people thought that it was important that schools addressed domestic abuse at an early stage but they were uncertain if programmes that explicitly addressed domestic abuse could be delivered before the age of 13 or 14 as ‘any younger . . . it wouldn’t be appropriate to bring it up’ (young people’s consultation group 1).

Authenticity: making it real

A key theme running through all the consultation groups with young people was the importance of authenticity or making ‘it real’ in investing a message with potency and achieving impact.

Authenticity as described by the young people seemed to have different components. In the first instance, it involved material that was recognisable and meaningful to young people such as the Home Office’s ‘This is Abuse’ campaign, which featured young people close in age to the members of the consultation group:

. . . because of our like age group, we could relate to it a bit more, it seems more real.

Young people’s consultation group 2

It also entailed expertise and/or relevant knowledge and experience in those delivering the message. As noted above, teachers were often felt to lack this authentic expertise and were described as just delivering material on healthy relationships so they ‘can tick it off the list . . . once a year’ (young people’s consultation group 2). Similarly, young people distinguished between celebrities such as Jahméne Douglas, who promoted Women’s Aid through his appearance on television talent show The X Factor, who, because they had relevant experience of the issue, were judged to be ‘genuine’, and those who were perceived to be using a campaign to promote their career.

Authenticity also involved a message with an emotional charge:

We had a firefighter come in school once and talk to us about fire safety . . . and he was talking and suddenly . . . he’s seen so many horrific things that he started like properly crying and everything in front of us and he was very embarrassed about it . . . and, yeah, that changed my opinion . . . seeing real emotion.

Young people’s consultation group 2

This emotional charge could derive from the use of real life accounts, narrative or drama. Drama was described as making ‘it real’ (young people’s consultation group 3) and as ensuring that preventative messages were relevant to young people’s own experience:

It’s like in front of you and then you realise, actually, it doesn’t happen miles away, you know, it happens here. And it’s so close to home and it happens to people that you might know and, you know, it can easily happen to anyone. And so I think drama kind of conveys that a bit more.

Young people’s consultation group 3

Young people in the third consultation group discussed the merits of drama that they themselves developed and performed versus a professional performance. They thought that both could be valuable and described a piece of theatre performed at school by a theatre in education company that had involved questions and after-the-show discussion with students as well as a professional performance:

They kept addressing us like, so say if there was a problem or something, they’d turn to us, like the actual audience and then . . . you would feel involved in the problem. So it felt like it more applies to you . . . So that’s why it was a lot more hard hitting because it wasn’t just watching, it was you’re involved in it, you’re watching, you grow attached to the characters . . .

Young people’s consultation group 3

Campaign aims

Group members identified a range of aims for campaigns for children and young people. In common with discussions in the young people’s consultation group, raising awareness in the individual was considered to be a primary goal:

[E]ffectiveness would be . . . they’ve now thought of what’s going on in their relationship.

Media consultation group 2

There was also an expectation that campaigns would provide a ‘step on to action’ or help-seeking (media consultation group 1) and it was argued that once ‘you raise awareness you’ve actually started an intervention’ (media consultation group 1). In common with the young people’s consultation group, group members thought that there was a responsibility to ensure that services were available to follow up the consequences of campaigns:

. . . what we don’t want is a really cohesive prevention programme, whereby young people, perhaps internalise the message, actually I’ve been or I’m being abused or actually, I’m worried about my behaviour towards other people. And thinking, ‘great, I now know that this has happened, now where do I go?’

Media consultation group 3

However, at the broader societal level, it was argued that campaigns had a role in facilitating conversations about topics that were previously taboo so that:

Domestic violence is something that can actually just be something you talk about without . . . being worried about who’s in the room or who’s listening or are you offending anyone in particular.

Media consultation group 1

Group members also reiterated the point made by the young people’s group who had argued that interventions needed to be ‘real’ and suggested that campaigns had a role in making the content of taught programmes more relevant to the ‘real’ world outside school:

A big role is reinforcing those taught programmes because a lot of the time it can feel like . . . something in school is what applies in school and then school is not necessarily applicable to real life . . . the point of campaigns . . . is to reinforce that some of the stuff you learn in school applies and seems relevant.

Media consultation group 1

Mechanisms of change

A number of mechanisms of change were identified by this group. The first of these drew on behavioural theories and addressed changing social norms:

. . . reinforcement is key to changing social norms. And that’s what we want to do, is we want to change the social norm. And by making it, by putting that pressure, basically, in a lot of different places, that’s how you change the norm.

Media consultation group 3

Destigmatising the topic of domestic abuse and making it an issue that could be discussed openly ‘around the water cooler’ (media consultation group 3) was seen as part of this process.

The other type of theory employed to explain change was psychodynamic theory, which was implicitly invoked in that campaigns were seen to evoke empathy for the victim’s perspective with behaviour change envisaged as resulting from this position of increased empathy. However, group members argued that this form of change was more likely to be found where abusive behaviour was not deeply embedded in a young person’s family and social environment and where it was not premeditated: ‘raising awareness can help that group because it brings the empathy’ (media consultation group 1).

Emotional identification was also seen as the mechanism that made for the effectiveness of drama or narrative:

It works because of the emotional engagement, it’s not the fact that it’s a story, it’s the fact that it engages in emotion . . . Whatever engages people with the heart as well as the head is probably going to be effective.

Media consultation group 1

In this sense, emotion was seen as an aid to learning: impact was achieved through the visceral charge that accompanied a message.

Audiences

There was discussion about the need to ‘granulate’ campaign material to target different subgroups within the general population of children and young people. Group members felt that, in the first instance, campaigns should emphasise commonalities and target whole populations – ‘find the things they have in common first before the differences’ (media consultation group 1) – with subsequent campaign elements, such as a particular page on a website, targeted at particular subgroups. The need to target subgroups was felt to vary according to the particular subgroup. It was stressed that disabled children and young people could resent messages that singled them out.

It was argued that preventative interventions were too often focused solely on female victims and that campaigns often failed to reach boys, who were the main perpetrators of abuse:

All the awareness campaigns that I’ve really had anything to do with have . . . been focused on raising awareness for victims, but not about raising awareness for young people who might be using [that] behaviour . . .

Media consultation group 1

Similarly, it was highlighted that preventative campaigns and programmes ‘tend to be very heterosexual focused’ and that interventions should recognise and speak to LGBT young people and promote talking ‘about the different relationships that we can have’ (media consultation group 1).

However, group members noted that it could be a struggle to identify ways of addressing diversity in a short media advertisement lasting less than a minute without risking associating domestic abuse with a particular subgroup. It was felt to be easier to reach subgroups through developing associated materials that targeted particular subgroups or could be used in particular settings such as youth groups, disabled groups, sports groups or young people in uniform such as the Girl Guides, Scouts or Army Cadets. Gadd et al. ’s111 research was cited as evidence that many young men with high levels of abusive behaviour could be reached through the criminal justice system, particularly the youth offending service.

It was argued that appropriately designed and formatted materials were required for younger children who needed to be alerted to the issue of abusive relationships early and that age-appropriate material was already available and was being delivered to the 7–11 age group in schools.

Some immigrant and refugee groups were considered to have brought their own experiences and patterns of interpersonal violence with them to the UK. Group members suggested that those designing preventative interventions in the UK should work with professionals in countries in Eastern Europe and elsewhere to understand how best to work with culturally specific dynamics and attitudes.

Delivery

Group members involved in commissioning and designing media campaigns cited the wide range of media that could be used for delivering campaign material. Mainstream television was only rarely used for advertising because of the costs involved but campaigning organisations had had considerable success with delivering preventative messages through long-standing television programmes which attracted large and faithful audiences. The BBC programme Newsround was seen as a useful means of directing messages towards younger children, and group members described how the Home Office’s ‘This Is Abuse’ campaign had moved to using the channels that young people viewed as well as popular soaps to deliver messages on domestic abuse:

. . . now there’s a lot more different media channels being used than there were three years ago . . . using Hollyoaks for instance this time round and MTV, and I think they’re trying to move to a place . . . where young people are already rather than dragging young people to where they want them to be.

Media consultation group 2

Delivering messages through the narrative medium of a soap drama was considered to be a highly effective way of raising awareness, especially when:

You already know the character, if it’s someone that’s in their house every day that they recognise . . . that has . . . the integrity of a celebrity that [you] genuinely believe, believe more, the character that you sort of know and love.

Media consultation group 1

Where on-screen advertising was used, it was more likely to be through YouTube™ (www.youtube.com) ‘so your film’s coming up before Beyoncé’s new video’ or ‘on-demand channels which reach boys quite well on mobile Xbox, PS3 [PlayStation 3™]’ (media consultation group 1). It was also noted that boys were more likely to view material through YouTube. However, this approach conflicted with the views of the young people’s consultation group, who described such advertisements as intrusive. It was noted that where local campaigning groups were posting material on YouTube, the quality needed to be high in order to compete with other material.

The groups discussed the use of online games to deliver interactive preventative messages to young people noting that ‘that’s where the kids are socialising these days’ (media consultation group 2) and cited examples of such games from both the USA and the UK. Group members argued that apps could offer those playing the opportunity to engage at a number of different levels and that they needed to be ‘multiple branched and strongly narrative’ (media consultation group 2). There were concerns that the quality of such games was not always sufficiently high to compete with commercial products and that, while a gaming format might be a ‘pull factor’ for children and young people, the message might become lost in the game format. It was noted that the effectiveness of such approaches had not been robustly evaluated.

While social media were considered to be particularly accessible to young people, more traditional media, including magazines, radio, billboards, and posters on buses and toilet doors, were still felt to have impact. It was noted that free newspapers had retained large readerships and posters and that leaflets could be effective if they had ‘a coherent visual identity’ (media consultation group 1). Memorability and repetition were identified as essential features of campaigns that had impact.

Group members argued that increasingly, a successful campaign was one that was multiplatform and could be delivered at the local as well as the national level by diverse organisations:

. . . it has to permeate different areas, so a media campaign, then [a] teacher talks about what’s happening about it in school . . . And then youth centres are doing the same thing . . . so it’s more cohesive.

Media consultation group 1

Such an approach would offer a child experiencing domestic abuse ‘a whole range of ways . . . to be able to buy in to the solution’ (media consultation group 2). National and international organisations, such as the Girl Guides, were seen to provide structured networks which could be used to disseminate material and messages to a large audience. However, it was pointed out that competitive tendering processes risked generating competition rather than collaboration between organisations.

It was also suggested that different types of campaigns for children and young people, such as those that targeted bullying and those addressing domestic abuse, could be delivering congruent messages that overlapped so that ‘one campaign knocks on to another’ (media consultation group 1).

There was considerable interest in identifying ways in which football could be used to deliver preventative messages to young people, both young men and young women. It was noted that UK premiership football clubs had large educational programmes and sizeable budgets for community work. Football clubs also had the benefit of good reach into their local communities. One representative of a domestic violence organisation described plans for a ‘Kick Domestic Violence out of Football’ campaign to be delivered during the 2014 World Cup and it was agreed that football clubs could exert ‘massive influence’ (media consultation group 2). However, group participants also noted that football clubs sometimes failed to distance themselves from the behaviour of players who were convicted for abusive behaviour and that clubs needed to take on campaign messages in a way that was genuine so that partnerships between campaigning groups and clubs were real; football needed to:

. . . understand the messages that we want to put through . . . there’s a massive space we have to get through.

Media consultation group 2

There was discussion about the use of celebrities to spearhead campaigns, and some concern was expressed about the potential for celebrities to deliver conflicting messages. In common with the young people’s consultation group, participants stressed the need for celebrities to have authenticity and cited the example of Jahméne Douglas who had grown up with domestic abuse and, on his own initiative, had used his appearance on The X Factor to campaign and raise awareness about domestic abuse:

He tweeted about Women’s Aid, he asked everyone who, who voted for him to give money to Women’s Aid, he did an incredible amount, but none of it was prompted by us . . . he’s like . . . I’m going to use music to raise money for . . . abused children and women . . . people have connected with him because they connected with his story . . . he’s interested and genuine, he’s always available because he actually genuinely cares about it . . .

Media consultation group 1

It was argued that young survivors of domestic abuse who were known and recognisable to young people provided powerful role models and had credence for young people. Male survivors such as Jahméne Douglas were particularly valuable in that they were able to address young men’s perspectives.

Campaign content

Group members argued that campaigns that employed positive messages and images and avoided ‘hitting someone over the head’ with their message were more likely to be successful, particularly with young men:

. . . because then it’s saying, this is positive images of how to be masculine in society . . .

It’s that solution focused bit, isn’t it?

Media consultation group 1

This is consistent with Stanley et al. ’s study,112 which found that men in the general public would resist messages that they perceived as too negative towards or blaming of men. At the same time, however, successful messages would need to have ‘shock value’ (media consultation group 1) if they were to change thinking and raise awareness.

As noted above in relation to television soaps and apps, a strong narrative was identified as a central feature of a successful campaign. A story could personalise an advertisement and evoke emotional engagement from the audience, which was felt to be key to a campaign achieving an impact. Successful advertisements were solution focused and worked by:

. . . telling the story, saying, you know, this is a problem you might encounter, this is how you solve it.

Media consultation group 2

The group discussed successful commercial advertisements that used a story to convey the message that ‘you may be a loser like this kid, but look how you can transform yourself into a winner’ (media consultation group 2) and argued that such narratives held particular appeal for boys. It was suggested that boys were especially interested in messages that promised to increase their ability to attract a girlfriend; one representative of a domestic violence organisation reported:

We just did a big piece of research on that and that just seems to be what boys want. They want to know how to be successful with girls, how to be a good boyfriend.

Media consultation group 3

Group members also highlighted that multiple messages would be required in order to successfully target different groups of young people. While raising awareness might be a sufficient outcome in some groups of young people, with other groups the task was to change abusive behaviour that was established and reinforced in their families and local communities:

There is no one type of instigator . . . raising the empathy is enough for some young people . . . for others, particularly with the more disadvantaged, it’s so ingrained, their lives are so much more complex, they haven’t got that ability just to change their lives like that.

Media consultation group 2

Group members argued that campaigns had ‘to really target what we think is behind the behaviour’ (media consultation group 2).

One of the main challenges identified for campaigns was competing with conflicting messages aimed at young people:

The competing . . . messages that are . . . contradictory. But actually, that comes from family and neighbourhoods as well as . . .

Well it also comes from exactly the same environment that we’ve just been discussing. What you can get on a music video or what you can get on extreme pornography . . .

Media consultation group 3

Group members suggested that, in addition to encouraging young people to approach the media in a discriminating manner, such conflicting messages could be countered by showing positive images of people changing their behaviour and achieving positive relationships:

I did a lot of work with Oxfam, in terms of poverty. And . . . they were saying, people get desensitised . . . And . . . their take on it is to show the after-effect and really innovative work that’s being done to kind of tackle this. And I think that’s definitely a way that could be, you know, going forward, in terms of showing people actually what support is available . . .

Yes, actually we need to see some stuff about . . . what a healthy relationship looks like.

Media consultation group 3

Involving young people in campaigns

Involving young people in developing and delivering campaigns was identified as a means of ‘making it relevant’, and group members stressed the value of interactive approaches. Young people were keen to be involved with initiatives involving social and other media – ‘because we’re using media, it resonates with young people, it chimes with them’ (media consultation group 2) – and delivering messages that young people helped to devise or voice made it more likely that the language was appropriate and that the message was closely targeted. Involving young people in the production of campaigns had benefits that extended beyond the life of the campaign:

There’s some really good examples of co-design work that’s youth led and that, that has really good impact, because you effectively create evangelists, you create disciples for your issue . . .

Media consultation group 2

Involving young people in a proactive way also served to ‘give them a voice’ and recognise their expertise, and was considered to invoke a children’s rights perspective:

. . . co-creation across the whole cycle, we’re not just using them as research subjects they’re there to actually create the campaign, so . . . they’re no longer passive recipients . . . they become sort of active agents.

Media consultation group 2

. . . they have a right to have a say in what’s being delivered, how it’s being delivered and to participate in that.

Media consultation group 1

Delivering messages to young people through their peers was argued to be an effective means of communicating, as it replicated their natural tendency to look to their peer group for support. Some campaigning organisations reported using youth ambassadors to reach children and young people more effectively:

We have theme ambassadors for our seven themes . . . and then we have a nationwide network of young people’s groups and we consult them on all of our marketing campaigns . . . the theme ambassador might speak at an event specifically but we would consult everybody on what the event topic would be . . .

Media consultation group 1

Group members also described a recent initiative funded by the Department of Health where young people who gave regular advice to their subscribers through online video blogs on YouTube produced a series of videos called ‘Awkward Conversations’ that addressed issues such as talking to parents about sex. This involved ceding control of the initiative to young people themselves, which could be challenging, but it was noted that new technology made a shift towards the cocreation of such campaigns more likely:

Technologies democratise communications to the extent that young people . . . instead of just consuming content themselves, they are now architects of that, they are making it, and we have to accept reality that we can’t retain all control, we have to allow conversations to happen . . .

Media consultation group 2

Group members pointed out that a high proportion of young people did have positive attitudes towards relationships and that those positive attitudes should be harnessed through peer pressure to achieve change. Bystander approaches (see Chapter 4), which aimed to empower young people to challenge abuse in their peer group, were mentioned in respect of such an approach (media consultation group 1). In the same vein, it was argued that creating a positive climate in which young people could express their own views was more likely to achieve change than delivering prescriptive messages:

. . . creating those conversations as opposed to just giving what those answers are . . . I think sometimes when young people are able to defend and, and have a dialogue with each other about that, it’s better than us saying this is the answer.

Media consultation group 2

Peer mentoring was seen to be another means by which young people could deliver prevention messages to other, usually younger, children and young people. The importance of providing peer mentors with appropriate training and support was emphasised, and it was suggested that this provision could be built into further and higher education programmes.

It was agreed that, in order to work with young people in a participatory way, organisations developing campaigns needed to evoke the trust of children and young people. For example, it was noted that young people might resist messages delivered by the criminal justice system or social services, although group members noted that the probation service or police sometimes partnered up with other organisations to deliver services or messages to specific groups of young people. The importance of trust between young people and the information source was identified in a number of meetings of this expert group, and it was noted that the format in which a message was delivered could impact on that trust. For example, young people were often sceptical about advertising, which could affect how a campaign was viewed:

. . . we can’t be seen to be advertising . . . That’s the problem. We’re using, a lot of the time, the same mechanisms, but if we’re seen as advertising you immediately lose trust and genuineness.

Media consultation group 3

Measuring effectiveness

Participants in the media consultation group agreed that there was a need for more evidence on what made campaigns effective. They expressed concern that current campaigns aimed at preventing domestic abuse were being developed from a slim evidence base consisting mainly of the previous experience of those designing campaigns, and that there was a considerable amount of reinventing the wheel with local organisations developing their own campaigns ‘from scratch’ (media consultation group 1). Underfunding and short-term funding were identified as basic problems resulting in a lack of robust evaluations and pilot projects which were not sustained: ‘just as it takes off, the money disappears’ (media consultation group 1).

Measures of campaign impact identified by the group included increased help-seeking or service use, an increased readiness to discuss domestic abuse openly at a social level and increased empathy for others from perpetrators. It was suggested that evaluations of preventative interventions should include evidence concerning subsequent disclosures. Health outcomes were also mentioned and it was anticipated that a successful campaign would result in an increase in reported incidents of domestic abuse, followed in the longer term by a reduction in reporting.

There was also interest expressed in commissioning research that examined the targeting of preventative interventions, especially the degree to which media campaigns should be gender specific:

That’s, for me, one of the stand out [pieces of] information that would be extremely useful to organisations who will say, ‘am I going to do a gender specific programme? Am I going to make sure that I, when I do evaluate my programme, if I’m delivering it to both girls and boys, am I going to look at boys and girls differently in a way, and how’s that going to affect anything later on that I do?’

Media consultation group 3

Programme aims

There was some resistance within this group to conceptualising material on domestic abuse delivered in school as a preventative intervention that aimed to change behaviour:

Are you trying to change behaviour or are you trying to change expectations of what behaviour is acceptable and what is not, which is a different thing I think.

Yes, I think we’re talking about attitudes.

Education consultation group 1

This was in part because changing behaviour was considered too ambitious an aim for an intervention delivered in school:

. . . you’re expecting an awful lot of a PSHE class, if you get one, to actually change behaviour . . . if you take the teenage pregnancy thing . . . SRE [sex and relationships education] was one aspect of eight different things to reduce the pregnancy rate . . .

Education consultation group 1

Group members, particularly those located within education, emphasised that children were exposed to a wide range of competing influences outside school; these group members were more likely to identify the aims of such programmes as contributing to well-being or as an aspect of care guidance which was considered a more congruent task for education:

Schools now are definitely showing an interest about how they carry out their duty of care to children who they know are identified as suffering domestic violence, so . . . it’s within the education system.

Education consultation group 1

It was highlighted that care guidance or pastoral work was now recognised as making an important contribution to the goal of academic achievement and group members argued that, from a child’s perspective, the two aims were integrated:

You’re never going to raise standards if your relationships aren’t working . . . schools would say ‘well there’s a pastoral curriculum but then there’s the academic curriculum’, no there’s not, there’s the experience that children have in school, they don’t separate it like that.

Education consultation group 1

Group members also noted that schools had a statutory responsibility ‘to prepare children and young people for the roles and responsibilities for adult life’ (education consultation group 1).

However, those group members involved in designing and developing programmes for domestic abuse organisations to deliver in schools were more likely to conceptualise programmes as aiming to reduce domestic abuse, and wanted to make that objective explicit:

It’s helping teachers, whole school communities and pupils recognise that there is a link between bullying behaviour generally, poor relationships generally and the potential to become a future perpetrator or victim survivor . . . And if you don’t use that language then it’s almost like the domestic violence does, does become invisible.

Education consultation group 1

Theories and mechanisms of change

Unsurprisingly for a group consisting of those working in or close to education, much of the discussion around mechanisms of change drew on cognitive theories:

You want young people to be more knowledgeable . . . to have a chance to reflect on their values and clarify them.

Education consultation group 2

You have the choice to behave like this . . . you have a choice to behave like that, and the consequences of that and that or this and this . . .

Education consultation group 2

In common with the arguments made by the media consultation group (see above), group members argued that preventative interventions needed to promote understanding of what constituted positive behaviour as well as assist children to identify abusive behaviour:

. . . you need to understand what a healthy relationship is in order to be able to understand an unhealthy one.

Education consultation group 2

However, group members also argued that children learnt behaviours through modelling the behaviour that was observed around them:

If we’re looking at this type of learning so much of it is absorbed unconsciously through that day-to-day interaction, how we do things here.

Education consultation group 1

What happens in the corridors and the playgrounds . . . is actually more important . . . than what’s delivered in 40 minutes or an hour and 10 minutes, once a week.

Education consultation group 1

As in the other consultation groups, there were also references to the affective component of a programme, and group members argued that the relationship formed with whoever delivered the programme was crucial:

. . . it’s actually the relationship . . . It’s about a young person saying I met somebody from the outreach service, they were really nice . . . That’s the critical learning, not what they talked to you about, so when you have a problem in your own life, you’ll think I’ll phone them, they were nice, and that can be more powerful than actually the content.

Education consultation group 1

There were identifiable differences between group members regarding their understandings of the nature of the change that might be achieved. Group members representing domestic violence organisations were likely to consider that programmes were raising awareness about gender equality, while those from within the education system were more likely to focus on imbuing positive caring values or developing understanding of healthy relationships. Some group members from the domestic violence sector who were involved in producing programmes for schools to use noted that the language and concepts used in schools differed according to the school they were working with:

I know the schools where I can say ‘hey there’s another resource for violence against women and girls’ . . . And at another school I’d have to say ‘ooh, I’ve got some more staying safe stuff’.

Education consultation group 3

Distinguishing audiences

Personal, social and health education

It was generally agreed that schools were an appropriate venue to deliver preventative programmes addressing domestic abuse, as a school:

. . . is a community where children are for a large number of hours and therefore in that sense alone it’s a very good place to site any work that you’re going to do because you’re touching most children.

Education consultation group 1

Personal, social and health education was a suitable place to locate these programmes, although it was noted that academy schools might not include PSHE in the curriculum. In these schools, material on healthy relationships or domestic abuse might be delivered via form periods and form tutors. There were concerns that academy schools were prioritising academic learning and that PSHE lacked a statutory place in the curriculum and was, consequently, low in status and under-resourced:

It’s probably the most patchy theory or subject on the curriculum, in fact in some schools it’s not even a subject . . . we’ve got non-statutory PSHE . . . we’ve got no training for PSHE, worth mentioning in initial teacher training . . . well-intentioned teachers being transferred into PSHE.

Education consultation group 1

An Ofsted report113 that identified that only 39% of secondary schools were teaching PSHE was cited. The low status of the subject and the lack of PSHE training at the qualifying level meant that although some teachers had accessed specialist PSHE training post qualification, some of those teachers delivering material on healthy relationships were not trained PSHE teachers and lacked the relevant skills and confidence:

. . . we’re being asked to evaluate Concorde. But we’ll not take into account the fact that some of the people flying it will be trained pilots and some people won’t know what, how to pull the wheels up.

Education consultation group 1

Whole-school approaches

There was general support across the three meetings of this group for whole-school approaches, which were seen to increase children’s exposure to the key messages:

. . . the intervention is being reinforced from lots of different sources, so it’s a sort of multimodal approach . . . you’ve got the actual curriculum; you’ve got teaching; changing teacher attitudes and so changing teacher behaviour; you’ve got the young people . . . with buy-in. So . . . basically doubling, tripling, quadrupling the amount of influence the intervention’s got because it’s coming from so many different sources.

Education consultation group 1

However, while material addressing domestic abuse could be delivered in a range of lessons including drama, physical education, information communication technology (ICT), English and history, it was argued that PSHE was the place where these messages were drawn together:

You still need somewhere to say to young people, remember we did that work on assertiveness in drama? Do you remember we did that work on, on communication in English? . . . let’s bring that together . . . let’s really get into it now, and draw on that learning.

Education consultation group 2

The National Healthy Schools Programme114 was cited as an example of an integrated approach ‘with a planned PSHE programme . . . supportive curriculum wrapped round good pastoral care’ (education consultation group 3) but there was concern expressed that ‘after 10 years of Healthy Schools and all that work, suddenly it’s all collapsing’ (education consultation group 3).

Involving children and young people in planning and delivery

As in the media consultation group, there was considerable support for involving children and young people in the design and delivery of programmes. Group members from organisations external to schools were particularly keen for young people to ‘have ownership of the issue being explored’ (education consultation group 1). It was argued that:

All the . . . emotional health and well-being programmes, the ones that were more likely to show impact were those where the young people were involved from the very beginning.

Education consultation group 1

However, it could be challenging to ensure that ‘you really do get sort of a representative group of young people rather than the more skilful ones, the more articulate ones, the ones who are always volunteering for things’ (education consultation group 1), and year heads could assist in identifying those young people who would benefit from involvement. In common with the media consultation group, members of this group identified the potential for creating lifetime ambassadors among young people.

School councils were also identified as a useful mechanism for securing young people’s involvement in the design and delivery of programmes but it was noted that this was not consistently implemented. There was also enthusiasm for peer mentoring and it was noted that peer mentors ‘hear the problems’ (education consultation group 1). However, group members emphasised the need for peer mentors to be appropriately trained and supported to respond to queries about domestic abuse at home or abuse in young people’s relationships, as young people did not necessarily give one another positive advice.

Parental and community involvement

Parental involvement was conceptualised as an additional feature of the whole-school approach:

You’ve got to have that whole school approach but then take it even further and the parents have got to be informed, the parents have got to be supporting the aims . . .

Education consultation group 2

However, this could be hard to achieve and it was pointed out that young people ‘don’t necessarily want parents involved with that, do they?’ (education consultation group 1). Moreover, it was questioned whether or not parents always ‘have the skills and force to be able to then support their children . . . Because there are lots of parents who are absolutely freaked out to talk about [it]’ (education consultation group 3).

It was also argued that parents from faith communities could be resistant to the early introduction of healthy relationships material:

In faith schools, that would equally be the barrier with the parents, and what parents think is acceptable to be taught in schools and what’s not. I think it’s in the teacher’s head a lot of the time.

Education consultation group 1

However, a group member described his organisation’s success in working with parents in faith groups outside schools:

Work we’re doing in youth centres and out in the community, inviting them to performances and events . . . we had 250 black parents in this church . . . led by young mums . . .

Education consultation group 1

There was less discussion about how schools worked with other community organisations on this issue, but it was felt that schools were now better linked with other community agencies, such as the police and health and safeguarding services, than they had been in the past.

Programme content

It was noted that national media campaigns had facilitated the delivery of teaching or programmes in schools. Programme material needed to be interactive, fun and engaging. Interestingly, given that some of the trialled programmes such as ‘Fourth R’ (reviewed in Chapter 4) delivered material on dating violence alongside material on sexual health and substance abuse, there was a case made for creating links between teaching material aimed at preventing domestic abuse and other PSHE issues:

One boy said to me, ‘my PSHE is like Buffy the Vampire Slayer, every week a different monster. Last week we did gonorrhoea . . . week before that we did HIV [human immunodeficiency virus] . . .’ And I think there’s a real danger at the moment that we’re compartmentalising all of these issues and . . . that there may be a core set of values and principles that we need to be exploring to underpin all of those.

Education consultation group 3

Gender equity was a strong theme in the young people’s consultation group and members of this group noted the need for programme material to acknowledge that boys and men could be victims of interpersonal abuse. Group members commented on the potential for evoking resistance from boys if they were consistently depicted as perpetrators of abuse:

. . . the boys will say: ‘oh . . . we’re always labelled the abusers’. And young people are very sensitive to that so I think you have to be very careful . . .

Education consultation group 3

It was felt to be important that young people ‘perceive you to be fair in your analysis’ and it was noted that relevant statistics could be deployed to convey the point that domestic abuse was a ‘gendered crime’ (education consultation group 3).

Group members also advocated taking a positive approach in work with boys and focusing on how a positive approach to relationships could help them to acquire a girlfriend:

There’s a lot of sex educators are saying that boys want to be good at sex . . . they want to know what good behaviour is . . .

Education consultation group 3

Drama and theatre

In common with the media consultation group, group members noted that drama could be a valuable interactive programme component which offered children and young people a creative means of engaging with the material. Moreover, drama could be effective for children who ‘in their own individual personal and social and emotional development won’t actually be ready to pick up the stuff that they need at a cognitive level’ (education consultation group 1).

However, the use of externally provided theatre in education could be expensive, and group members also suggested that theatre productions could present domestic abuse in its more extreme and dramatic manifestations rather than identifying ‘that very low-level aggressive behaviour that young people are more likely to experience in the early stages’ (education consultation group 1).

Externally delivered theatre could also be an excuse for the teacher to sit back and let the professionals take over, and it could bring to the surface difficult emotions and attitudes which would need to be dealt with safely.

Who delivers programmes?

There were differences of opinion in this group regarding who should deliver these initiatives in schools. Some group members argued forcefully that this was a task for teachers, who had an ongoing relationship with their class:

I believe really strongly, that teachers manage the learning in their classroom . . . a class belongs to that teacher, they are the professional, it doesn’t mean they can’t be supported by outside providers who bring in a different set of skills, different set of knowledge, but I do feel really strongly that the class is managed by their teacher, and I think it’s always dangerous where teachers somehow step away from that responsibility.

And I think you’re right about the teacher and the trust and when you ship people in from outside . . . they don’t build that trust necessarily with those people.

Education consultation group 1

However, some group members argued that, while specialist PSHE teachers might be equipped to deliver these programmes, class teachers often lacked the skills and confidence to do so:

. . . certainly not with domestic abuse, they’re not skilled . . . this is out of their comfort zone.

Education consultation group 1

There were also concerns that some teachers did not present as positive models for handling conflict:

There are teachers who would use very . . . abusive behaviours . . . In terms of behaviour management.

Education consultation group 1

‘Upskilling the teaching force’ (education consultation group 1) was considered to be essential if teachers were to take sole responsibility for delivering these programmes.

One of the arguments for using external facilitators based in a service that provided support for those experiencing domestic abuse was that the contact made through the programme could encourage young people to access that service in the future:

If people go in and are genuine and authentic with young people, they carry the brand of their organisation and it may encourage young people to recognise that when they’re in crisis that is an organisation that I know about and I feel good about them, I’m OK to ring them up.

Education consultation group 3

However, a reliance on external facilitators could contribute to a lack of sustainability:

Programmes don’t survive because . . . part of their short-termism is that there’s somebody who will go in and do it and will negotiate all those pitfalls . . . and then if their funding goes or whatever and it’s just left to the schools to deliver it then that . . . understanding, the more subtle understanding goes.

Education consultation group 1

The strength of the relationship between the school and the external provider of a programme was considered to be essential for its successful implementation. Similarly, the use of external facilitators could be effective ‘providing they work closely with the school and they’re part of the community’ (education consultation group 1).

Disclosure

There was a considerable amount of discussion about disclosure in the three meetings of this group and the potential for teaching on domestic abuse ‘to open the floodgates to disclosure’ (education consultation group 1) was identified as a disincentive for some schools to deliver these interventions. However, there was substantial variation in the extent to which group members had found that delivering these interventions in schools prompted disclosures of domestic abuse:

When we did domestic abuse awareness in schools, I mean one lesson for an hour, we got nine disclosures from pupils from that.

Education consultation group 2

One of the things that’s been quite remarkable about the programme over the years is that there’s been very little disclosure . . .

Education consultation group 2

It was acknowledged that disclosures of domestic abuse could be difficult for schools to handle appropriately and that staff needed support to do so. Group members described using learning mentors both to offer confidential opportunities for children to disclose and to liaise with other agencies that might need to be involved in responding to a disclosure. While children valued teachers as a trusted source of information, they might struggle to disclose to a teacher, as they would fear losing control of what happened to the information they supplied. Learning mentors were seen to be accessible and non-threatening repositories for disclosures and it was noted that sometimes it was easier for children to disclose to school staff other than the form teacher, including playground assistants and art teachers.

Other group members described disclosures being handled by the school’s safeguarding team or officer but difficulties in passing safeguarding referrals onto children’s social services were identified and attributed to ‘the huge barrier between education and social care [and] misunderstanding of what each sector does’ (education consultation group 1).

It was agreed that school staff could be prepared for disclosures by being equipped with information about helplines that children could call and contact details for relevant local agencies. Group members argued that the teacher’s role was primarily to signpost young people to appropriate sources of support rather than to act as a counsellor.

It was pointed out that domestic abuse could be an issue for school staff as well as for pupils, and that this should be addressed as part of any intervention:

I mean all the stuff on well-being, you know, there’s a big movement now to say we cannot do anything on the well-being unless you’re addressing the well-being of staff and, you know, sort of simply how can you look at one half of a community without looking at the other half?

Education consultation group 1

Support and funding for programmes

A common theme reiterated across these groups was the need for senior management commitment and leadership to underpin programmes in schools. Involvement from the governing body was also felt to be crucial to the success of a programme and the need to inform governors about domestic abuse programmes was emphasised:

Governors . . . they’re very powerful in this . . . if you did governor training, or it was part of governor training to be aware of these kind of programmes, that would be really useful.

Education consultation group 3

Outside schools, some of the support and resources required for these initiatives were missing and this was attributed in part to the loss of the support and monitoring role formerly undertaken by local authorities. There were also concerns expressed about the reduction in funding for the independent domestic violence sector where children’s support workers were the posts most likely to be affected by restrictions on local authority spending. Group members stressed the need for central government to support preventative initiatives and several group members suggested that Ofsted could also play a role in promoting and monitoring programmes:

It’s about the government’s initiative that drives things and if they’re not driving this . . .

It’s going to drop off the end.

Education consultation group 1

Group members felt that the relationship between school and external provider was key to a programme’s sustainability. However, it was also argued that, given the ephemeral nature of funding for the domestic violence sector, programmes needed to be embedded in schools if they were to be sustained:

Schools have got to have something that they can roll out year after year and not be dependent on external providers all the time.

Education consultation group 1

It was suggested that a directory of available programmes would be useful; this would allow schools to know what was available and exercise choice as to which programme they delivered. Currently, schools selected programmes in an opportunistic manner when funding became available:

Somebody bowled up and said they’d do it . . . The communication comes at the right time and there’s a bit of money in the kitty or the budget and we’ll go for it . . . Somebody strong is leading it and then they feel passionate about it . . . there are exceptions where it’s actually part of a planned school priority or a target . . . But, on the whole, it tends to be like that. And also, it tends to be, you know, the cheaper, the freer . . .

Education consultation group 3

Presenting the mapping survey findings (see Chapter 3) on sources of programme funding to the group generated considerable discussion on this topic. Group members from the independent domestic abuse sector thought that, in future, they would be increasingly likely to charge schools for programmes. There was disagreement over whether or not this would be a major disincentive for schools or whether schools were increasingly able or willing to find the funding for these initiatives. There was a suggestion that, in some areas of the country, Behaviour and Attendance Partnerships could provide such funding.

Measuring outcomes

It was noted that, if schools themselves were to fund programmes, they would need to be able to justify the expenditure in the face of competing priorities:

Schools do have budgets that they can draw on for these sorts of interventions, but they need to be able to justify their expenditure. And increasingly, they’re asked to justify everything, in terms of outcomes.

Education consultation group 3

In common with the media consultation group, group members acknowledged that many local programme evaluations ‘just skim the surface really’ (education consultation group 1).

The group discussed what outcomes could be used to measure the impact of school programmes. Group members reported anecdotal evidence of programmes boosting students’ GCSE (General Certificate of Secondary Education) drama grades or acting to improve attendance for children whose attendance had previously been poor (education consultation group 2). It was suggested that schools could use the data they already collected to measure outcomes:

. . . the sort of data they already collect and they might look to see an intervention of this sort reflected in that data, are teenage pregnancies, academic retention, pupil achievement, the progress and achievement of particularly vulnerable groups and health related behaviours.

Education consultation group 3

However, there was also concern expressed that setting up attainment as an outcome was ‘a red herring’ (education consultation group 2) and that those types of impact likely to result from domestic abuse prevention programmes were not formally recorded. Outcome measures related to help-seeking were felt to be more relevant; these included disclosures of domestic abuse, the use of a school helpline or an increase in referrals. It was suggested that students’ ‘actual knowledge of where to go’ (education consultation group 1) for help would be a meaningful outcome measure.

It was pointed out that defining the goals of a programme tightly made it more likely that demonstrable outcomes would be produced:

. . . the more targeted the programme, the more likely you are to get an impact, the broader the goals of the programme, the more difficult it is to demonstrate a measurable outcome . . .

Education consultation group 1

This could make it particularly difficult to demonstrate impact from some of the broad outcomes such as ‘changing attitudes, well-being’ (education consultation group 1) or ‘more mature behaviour’ (education consultation group 3) that might be anticipated from preventative programmes on domestic abuse.

Context

Delivery

Mechanisms of change

The individual interviews with experts offered the opportunity for the question of mechanisms of change to be pursued in more depth. The mechanism of change most frequently identified was that of changing social norms:

. . . a notion of kind of influencing social norms as opposed to influencing outcomes for specific groups

USA 4

. . . trying to change or transform social norms or lift up more pro social norms and discourage antisocial social norms around these issues

USA 3

This was described by some as ‘starting a conversation’ (USA 5) and was seen as a process that took place at a number of levels: within the peer group, in a school, in a community and in the wider society.

The ‘whole-school approach’ represented the fullest expression of social norm theory where ‘organisational change in the environments where young people live and work and learn and play’ (Australia 1) was conceived as a means of changing values and attitudes. This approach, in common with the bystander approach, drew on the peer group as a mechanism of change:

. . . bring about a healthier school so that the school climate is improving . . . in any classroom of 25 kids, five of those kids might be at risk, five or even 10 of them might be at risk of an abusive relationship. The other 15 are there to keep that from happening . . . the other kids know what to say, the other kids they now have the language, so that peer component is critical.

Canada 2

In contrast, other approaches focused on providing young people with the confidence or skills to negotiate healthy relationships:

I see it as the route to a kind of empowerment, to say well get out of my face or get your hands off of me, or, you know, to initiate, you know, to be mutual, consensual, so it’s this notion of boundaries.

USA 4

. . . aiming to enable young people to use ethical frameworks . . . it’s trying to give people the ways of thinking I suppose to make decisions in complex situations where there’s social pressures, gender pressures and sometimes alcohol and other things involved.

Australia 1

Some experts identified a shift from an earlier focus on victimhood to addressing perpetrators. Rather than equip girls to be more assertive, the aim was now to change boys’ behaviour:

If you are aiming these programmes that are trying to somehow help girls be victimised less then it’s tough because really it’s totally up to whoever might victimise them to change their behaviour . . . Primarily, you want to target potential perpetrators . . .

USA 1

One interviewee involved in developing campaigns argued that the focus had switched to providing boys with knowledge about what girls wanted in a relationship, as what motivated boys was the possibility of being successful in attracting a girlfriend.

Deconstructing gender norms could be part of this approach:

You’re actually pulling some of those gender norms out kicking and screaming into the centre of the room and you’re talking about whether or not they’re helpful for people, . . . it gives young men the space to talk about . . . how pressuring it feels to be the one who’s always expected to know what to do and the way he’s always expected to lead it, and if they get it wrong in front of their friends then everybody knows.

New Zealand 1

Similarly, interventions for younger children that enabled them to recognise and distinguish safe relationships were described as drawing on the same mechanisms that underpinned work with adult victims of domestic abuse:

The purpose is to increase children’s understanding of feelings, feeling safe and explore and promote behaviours which will contribute to a safe environment . . . it’s what we would have been doing with women for years about how do you keep yourself safe, how do you recognise your fear, even that level of what risk you are at . . . it’s the same message.

UK 3

Evaluation

It was generally agreed that, currently, there was limited robust evidence of programme effectiveness and it was noted that ‘there have only been a few programmes that have really been able to do these kind of thorough evaluations that are really, really convincing’ (USA 1). In common with the views expressed by the expert consultation groups, interviewees considered that a lack of strong evidence meant that untested interventions were being selected for implementation in a manner that was often opportunistic rather than evidence-informed.

Interviewees argued that there was a need for more longitudinal studies and for studies that measured behavioural as well as attitudinal change:

. . . a big challenge is behaviour, to be able to measure behaviour rather than to measure attitudes or knowledge. It seems like many, many programmes look at attitude change because it’s more accessible, kind of, you know, the whole idea of sort of measuring something not happening is a little bit of a challenge.

USA 1

Researchers interviewed commented on the difficulties of getting access to schools to evaluate programmes, challenges in securing comparison sites, the potential for losing participants to follow-up and the difficulty in controlling for other intervening factors:

. . . it’s hard to figure out if it’s really just your game that’s making a difference or if it’s other things that are being done in the school or in the home . . . we are trying to measure them pre test, post test and follow-up. To find kids, get them actually to do the evaluation is not easy.

USA 2

The problem of confounding influences or ‘noise’ was a particular challenge for those seeking to evaluate media campaigns:

You can’t necessarily say with 100% confidence all the time that it’s your campaign that is moving the needle either way, right? Because there’s a lot of other noises out there.

USA 5

It was also suggested that the scale of change envisaged required a longer time frame than most evaluations afforded:

We can’t possibly just be evaluating our programmes on the basis of the absence of violence, we’re not necessarily able to achieve that because of the time frame we have to deliver programmes and also we know that the genuine absence of violence can take years if not generations to achieve because of the level of cultural change required.

Australia 1

Some minimal criteria for evaluation were proposed:

Measures of attitudinal and behavioural change, long-term follow-up at least 6 months after, control of comparison groups . . . randomisation if possible or at least, you know, control.

Australia 3

It was noted that the reliance on self-report was a limitation of many studies in this field. While some experts considered that existing measures of change were valid and useful, others were seeking new approaches:

We need to start developing ways of thinking about behaviour change that are shown by indicators relevant to the changes we’d like to see. So if we want young men to understand sexual consent and also be able to negotiate it and communicate it without applying pressure or power in their personal relationships, particularly with young women, then we would need to look at well what are the indicators of that behaviour happening?

Australia 1

There were also differences found between those who were seeking more specific indicators of change and those who wanted to adopt broader indicators:

If it’s about well-being then there’s a whole suite of measures around well-being really. I think having little specialist indicators is probably death for this really, it has to be integrated into what are the big things . . .

UK 2

Those who were involved in evaluating media campaigns were more likely to use proxy measures such as calls to helplines to evaluate impact. In general, while researchers from the USA and Canada expressed confidence in existing outcome measures, those from Australia, New Zealand and the UK were more interested in developing new approaches to measuring change:

We haven’t really developed comprehensive ways of measuring changes in behavioural intent . . . or actual behaviour.

New Zealand 1

Analysis

Analysis

‘This is Abuse’: the Home Office

The government of England and Wales has been running a series of media campaigns since March 2010, directed at preventing relationship abuse and helping teenagers to understand what ‘consent’ means within their relationships. The umbrella name of the campaign is ‘This is Abuse’, and the constituent parts are ‘Teenage Relationship Abuse’, which ran on three separate occasions (February to March 2010, September to December 2011 and February to April 2013); ‘Teenage Rape Prevention’, which ran twice (March to April 2012 and December 2012 to February 2013); and ‘This is Abuse’, which combines the key messages from both previous campaigns and which ran most recently (December 2013 to April 2014).

The campaigns were directed at girls and boys aged 13–18 years and the stated aim was to prevent teenagers from becoming perpetrators and victims of abusive relationships. The original campaign utilised television, cinema, radio, press and poster advertising which directed teenagers to the campaign website for further advice and support (http://thisisabuse.direct.gov.uk). Further developments for the later waves included the use of social media, collaboration with makers of television and radio programmes aimed at a teenage audience, and widespread dissemination of support material.

The findings reported here are based on limited data on two waves of the campaign following a meeting with the Home Office in January 2014 and this does not take account of information that may have subsequently become available.

Wave 2 of the Teenage Relationship Abuse campaign, which ran for 4 months, cost £1.5M in total and wave 1 of the Teenage Rape Prevention programme, which ran for 2 months, cost £1.1M. In both cases, the major cost elements were expenditure on television and cinema advertisements, outdoor publicity such as sheets in shopping malls and panels in changing rooms and college washrooms, and online display and search. Website costs were relatively modest.

The average cost of a case of domestic violence to the taxpayer was estimated to be £10,000 (derived from Walby116). This implies that for the monetary benefits of wave 2 of the Teenage Relationship Abuse campaign to equal the cost, 150 cases of domestic abuse need to be prevented. Over the period of the campaign there were over 9000 visits to the ‘need help’ section on the website, implying that if 1.7% of these visits led to a case of domestic violence being prevented, the programme would be cost neutral. There is, however, no means of confirming who the website hits were from and whether or not any have led to a change in behaviour.

The cost of a rape victim to the UK taxpayer was estimated to be £96,000 (derived from Walby112). This implies that for the monetary benefits of wave 1 of the Teenage Rape Prevention campaign to equal the cost, 12 cases of rape would need to be prevented. During the campaign 8394 people accessed the ‘need help’ section of the website, implying that if 0.14% of the visits resulted in a prevented case of rape, the programme would be cost neutral. However, there is again no indication as to who the website hits were from and whether or not any have led to a change in behaviour. However, comments left on the discussion forum on the ‘This is Abuse’ website show that the adverts are helping people to understand what abusive behaviours are, seek help and, in some instances, leave the relationship.

Just finished with my boyfriend of two years after seeing these videos and I can relate to each one, the word ‘your pathetic’ ‘little tart’ and ‘you don’t do anything without my say so’ just shock me back to reality I know I deserve better than what I was getting put through

This is a great campaign and really needed. If I had seen this advert a year ago I would have realised what I was going through much sooner and left. Thanks x

I’ve just broken up with my boyfriend of nearly 6 months. After watching this video, the exact words of ‘you’re pathetic, are you gonna go cry to your friends?’ echoed in my mind.

He had said those exact words to me on many occasions. He put me down, stopped me from talking to my friends who were boys, checked my inbox, my texts, physically pushed me around . . . I don’t see why I didn’t see this before!

Helping Hands: Women’s Aid Federation Northern Ireland and Department of Education Northern Ireland

As discussed in Chapter 5, ‘Helping Hands’ is an activity-based programme used in primary schools in Northern Ireland, which aims to portray keeping-safe messages for children and young people using games, stories and activity booklets. The programme is delivered by teachers who receive 2 days’ training, which is delivered by peers who have been trained to train. Information about the cost and benefits of the programme was obtained from an internal evaluation document and an interview with the training manager of WAFNI.

The costs of the programme fall into two main components: training the teachers and providing the materials. Teachers are trained in clusters of about 16 and the reported cost of training and equipping one cluster and their classes was given as £6068. The breakdown of these costs is given in Table 29.

This programme was supported by an annual grant of £60,000 from the Department of Education Northern Ireland; 10 training sessions were carried out per cohort, which produced 160 newly trained teachers each year. The cost of providing substitute teachers during training was covered by a child protection training budget provided by the Department of Education and/or the Area Library Boards.

Ongoing development of the programme included amending and updating the training materials and running marketing days to support local groups with enrolling schools. This, together with the training of local group staff to deliver the training to teachers, was met by WAFNI.

The evaluation document (see Chapter 5 for a fuller discussion of this report) was largely positive in its assessment of outcomes. It identified:

Increased understanding of the context and impact of domestic abuse, particularly on a child in a school setting and an increased level of ‘comfort’ expressed by teachers if domestic abuse is raised in a classroom setting.

p .5586

It also noted:

. . . increasing . . . confidence and competence of teachers to address sensitive and sometimes challenging issues

pp. 56–786

The evaluation document made no reference to cost or value for money; however, the training manager reported that she ‘does not see this as an expensive process which once in place enables the school to continue to deliver the Helping Hands year after year’ (Regional Finance and Training Manager of WAFNI, 4 March 2014, e-mail to the authors).

Drama workshops

One UK-based charity, which uses drama and arts to work with young people informally, shared information about the cost of their programme. This organisation worked in schools, youth centres, PRUs, offices and health-care settings ‘creating tailored projects which offer support and meet specific needs’. The workshops were delivered by trained facilitators to a small group, typically a class of 30, and the programme culminated in a production presented to a larger group comprising the students’ peers. Two models were available for schools to purchase: a 10-week programme of 1 hour per week, which at the time cost £4500, and a model where the same material was condensed into 2 full days and cost £2000. In each case, the ratio of direct costs of delivering the workshops to the indirect costs such as management, overheads, training, and monitoring and evaluation was roughly 50 : 50. The organisation pointed out that the cost per student was £150 for the 10-week programme and £67 for the 2-day programme, although if all students who benefited were taken into account, including those watching the productions, these costs were £21.40 and £9.50, respectively.

When the charity was asked about the advantages and disadvantages associated with each model, the response was that the organisation promoted the 10-week programme as their ideal because this allowed time for the students to absorb learning between sessions, but our informant acknowledged that the swiftness and lower expense of the 2-day programme suited some schools better. To the researchers’ knowledge there has been no formal comparative evaluation of the two models. However, informal internal evaluation suggested that the two models had a similar impact on the students participating but the performance/presentation that was created and then shared with a wider peer audience had less of an impact in the 2-day model. An explanation offered for this was that the condensed model allowed less time to delve into the issues and unpick them to a point where the group was ready to share essential learning with their peers. The purpose of the presentation was described as:

To share learning with a wider group of young people but it is also about consolidating learning and having a tangible end point for the process so we currently keep that element in [the 2-day model], even though it might have less impact.

Chief executive, 2 September 2014, personal communication

Voices Against Violence

This programme was developed jointly by the World Association of Girl Guides and Girl Scouts (WAGGGS) and United Nations (UN) Women. It is a non-formal coeducational initiative designed to be delivered by peer educators and youth workers, and is intended to be non-location specific and international so that members of the Guiding Association can utilise it across the world.

No information on costs was obtainable but conversation with individuals involved in the development and dissemination of the programme provided a description of resources.

The training programme was intense, typically lasting a week, with 25–50 trainers being trained by six experts, and the model was designed so that the training was cascaded out within the community: those trained then trained another group in the same way down to level of individual Guide leader. The handbook for leaders is downloadable from the website (www.unwomen.org/ru/digital-library/publications/2013/10/voices-against-violence-curriculum).

Four age-related booklets were available for use with groups of children and young people for ages 5–7 years, 8–11 years, 12–16 years and 17–25 years, and were available to organisations once they had signed an agreement to adhere to the principles of the curriculum and deliver it safely.

Development of the material took 2 years of partnership between the two organisations and involved several drafts and iterations. It was piloted in 20 countries during the process to ensure that it was acceptable internationally and covered all relevant topics.

To our knowledge no formal evaluation of the programme has been carried out.

Sortir Ensemble et Se Respecter: Swiss (French) adaptation of ‘Safe Dates’

The ‘Safe Dates’ programme developed by Foshee et al. 54 in the USA was adapted for a Swiss (French) population. 76 Conversation with those involved in the adaptation revealed some relevant information about resources and costs.

Programme facilitators received 2 full days’ training, which cost CHF520 (£340), including the provision of the programme handbook. Ideally, the programme would be delivered by two facilitators.

The rights to ‘Safe Dates’ have been bought by Hazelden Publishing (Foundation),117 and a manual and compact disc are available for purchase. Any translation/adaptation of ‘Safe Dates’ requires a licensing agreement and payment of licensing fees to Hazelden.

Costs and benefits

These extracts show that representatives from the education and media organisations generally found it difficult to talk about cost and cost-effectiveness. Even when asked directly, they frequently sidestepped the question because they did not have the information, did not want to disclose it or had not previously considered it, and a lack of formal evaluations was reported. For example:

I don’t think there’s been enough evaluation . . . there’s no formal evaluation that goes into the cost of the campaigns that we’re working on . . .

Media consultation group 2

Benefits were easier to talk about and ‘effectiveness’ was seen as important, appearing to drive decisions about funding or implementation decisions, but there was little consensus as to what constituted ‘effectiveness’, which was not clearly defined even for similar programmes. A tension between desirable outcomes and measurable benefits emerged, indicating that greater importance was sometimes placed on a process or an outcome just because it could be measured rather than because these were identified as the outcomes that mattered but which were difficult to quantify (see Chapter 6). This tension is conveyed by these quotes:

Cost per whatever is the way we do everything, cost per click on Facebook adverts, cost per view on YouTube videos . . .

Media consultation group 1

. . . the benefits are the process and the outcome . . . and the engagement . . . of young people.

Media consultation group 2

I think it’s a real problem to measure the effectiveness of a campaign with prevention.

Media consultation group 1

Media representatives were enthusiastic about involving young people in the development of programmes:

And because our commitment is co-creation across the whole cycle, we’re not just using them as research subjects they’re there to actually create the campaign, so . . . they’re no longer passive recipients of care, they become sort of active agents for themselves, they become . . . the movement.

Media consultation group 2

They were also specific about the time required to create, for example, a short film. This involvement was seen as being cost-effective in the sense that young people were often paid expenses only, or a minimal amount, and there were process benefits to be gained from cocreation. It was accepted that young people benefited from the involvement in the development phase as well as the implementation.

Relationship between resources and outcomes

The concept of a relationship between cost and outcomes was unfamiliar to many group members and treated with some scepticism by others because of the difficulty in measuring (and valuing) benefits and the context-specific nature of most evaluations:

Well there’s a feeling that people use the resources they’ve got to do the best they can with it, and then you look at the results . . . but do you then learn anything about cost-effectiveness? I’m not sure because people are using different ways of doing things, aren’t they?

And the results you get are slightly different because you’re working in a different context and again you might have these different ways of seeing the issue, so it’s really difficult.

Education consultation group 2

There was an admission that the relative scale of costs and benefits was not often considered – explicitly or even implicitly. An expensive programme was not necessarily expected to reap large returns and yet a small-scale project might benefit many:

There may be a kind of implicit relationship about . . . a big budget, hopefully it’ll have a big impact and a small budget, small impact . . . but it’s not even as far as that . . .

No, I don’t think it . . . it doesn’t work like that.

No.

You can spend so much on advertising and unless you’ve got the proper objectives behind it, the messaging, you’ve got the right audience, it won’t work.

Media consultation group 2

This view was particularly true for the media group, who viewed websites and social media as being potentially far-reaching methods of communicating a message for low cost:

[Video games] are fairly low budget and there’s places online . . . who have millions of hits to their content every day. Young People’s YouTube trenders, young people who have their own channels, who get millions and millions of hits.

Media consultation group 2

On the other hand, television advertising, although far-reaching, was considered to be ‘expensive’ by several contributors to the media group, and there was an indication that, increasingly, there was some doubt about its reach:

How many people watch TV live now? Or how many watch it . . . later . . . and fast-forward the ads?

Media consultation group 1

We don’t advertise on television any more.

Media consultation group 2

However, in contrast to this, members of the young people’s group argued that television advertising was effective because it targets ‘everyone . . . who watches the programme’ (young people’s consultation group 1), adding that television is a safe environment compared with the internet, where:

your partner might see [you] . . . or family members . . . typing ‘domestic abuse’.

Young people consultation group 1

After prompting, some members of the education group agreed that more resources might lead to greater benefits, expressing the view that a long-term programme carried out in schools was more likely to be effective in the long term (although no evidence was offered to support this view). There was a feeling that a programme lasting at least a term would have a better chance of changing attitudes, deepening relationships and creating a cultural shift than a short, albeit intense, programme:

There was time to develop relationships with senior management and head teacher and to, to open up channels of communication in a way that there hadn’t been time to do previously.

Education consultation group 2

A positive point was made about the use of existing networks. These were seen as an efficient way of cascading a message out, and the Guiding and Scouting movements were cited as examples of this process by members of the group. Within these networks, a large number of local leaders could be trained to deliver, at modest cost, a programme that could have the potential to reach many young people.

And then you’ve got something like the Girl Guides . . . you’ve got all these leaders and I trained only 20 leaders who were then supposed to roll it out to their regions . . . there’s something in there about the networking and how you’re distributing it. It’s very efficient.

Media consultation group 2

Budgets and funding

For most organisations, budgets were an important limiting factor in what they were able to achieve; thus, affordability appeared to take precedence over cost-effectiveness:

If we have no budget then . . . there’s things you can’t do. That’s a top, that’s the first thing. In a lot of cases that’s the only thing. If we have no budget then . . . there’s things you can’t do.

Media consultation group 3

Several comments were made about the limitation of budgets, their inflexibility and, in some cases, their unpredictability. Frustration was evident in that sustained programmes of work could not be planned because of uncertainty around long-term funding:

They come up with a resource, and it gets . . . sent out to schools . . . But it’s . . . a drop in the ocean. There’s no join up with any other things. And then . . . suddenly the budget goes . . . it’s gone. There’s no kind of sustainability.

Education consultation group 3

Although the consultation groups were asked about the source of funding for preventative projects, very little information was provided, and it was unclear in many cases from where funding was derived. The limited discussion centred on the police as a potential source, as they were identified as having funded a particular project, and mention was made of a ‘pot of money’. Media consultation group 3 suggested that the police might be the most likely source of future funding.

Resources and cost

The majority of educational programmes identified by the literature review and the survey were described in sufficient detail to be able to understand what resources were required to deliver the programmes. However, these descriptions were largely provided in order to understand the different components of the programmes and their purpose rather than the resources themselves and are therefore not reliable for comparative cost purposes. Indeed, as these reports mostly focused on the delivery of the material and, in some cases, pre-delivery training, there were likely to be important resources that were not included. These might be, for example, resources invested in developing the programme, the cost of teachers’ time to attend training and wider implications (positive or negative) on other sectors such as health and criminal justice.

Information about some aspect of cost was obtained for five programmes: ‘Fourth R’, ‘This is Abuse’, ‘Helping Hands’, the drama workshops, and the Swiss version of ‘Safe Dates’. The last of these provided only limited information about the overall cost, while ‘Fourth R’, ‘Helping Hands’ and the drama workshops expressed cost in terms of cost per child/student involved. Only in the case of the Home Office’s ‘This is Abuse’ campaign was any attempt made to evaluate the programme by comparing cost with a benefit. In this case, a relationship was made between the cost of the campaign and potential benefits in terms of money saved through preventing cases of abuse or rape. This technique of return on investment (ROI), where benefits are valued in monetary terms and compared with the investment required to produce these benefits, is a form of cost–benefit analysis. A positive ROI indicates that the intervention is cost saving. In the case of ‘This is Abuse’, the number of website hits was used as a proxy for prevention.

Benefits

Benefits recorded in the published evaluations of school-based interventions tended to be some combination of knowledge, attitude and, in a few cases, behaviour. These were measured in a variety of ways using questionnaires and scales and so it could be argued that an evaluation of cost per student benefiting (e.g. improving their knowledge, attitude or behaviour by a specified amount) could have been carried out. However, none of the studies included in the review did this. Benefits of media programmes emerged as less well defined and harder to quantify. Members of the consultation groups talked in terms of ‘reach’ and ‘number of hits’, but translating this into tangible benefits was fraught with difficulty. Indeed, the professionals in the media consultation group regarded television advertising as ‘expensive’ (implying that it was not cost-effective or good value for money), but the young people’s consultation group favoured the anonymity of this method, and the fact that it could be accessed without intention and without the fear of incurring stigma. These benefits, beyond those that are immediately obvious, may be reflected in the large proportion of the total expenditure on ‘This is Abuse’ taken up by television advertising.

The difficulty in identifying, measuring and valuing the benefits of a preventative programme was discussed by members of the consultation groups in the context of considering cost-effectiveness or value for money of the programmes they knew about. In a world driven by short-term and unreliable budgets, rigorous examination of wider societal benefits, including those impacting on other sectors or likely to take place in the future, was not a priority. This is not surprising. The interventions we have reviewed are complex and multisectoral and an evaluation taking account of all societal benefits would be a considerable challenge.

Guidance on costing

Some guidance exists about methods of costing programmes such as those identified in this report, which could be a useful step forward. A detailed microcosting exercise was carried out by Edwards et al. in evaluating a parenting programme,118 and one output of this exercise was a step-by-step guide to the microcosting of a complex intervention aimed at the general population. 119 A standardised approach to such an intervention may not always be possible and there may be numerous drawbacks and pitfalls, but the general principles advocated are worth consideration. The guidance advocates a multiagency perspective, taking account of all costs to all stakeholders and separating out non-recurring set-up costs, such as training, from the recurring costs associated with delivery of the intervention. The authors recommend that diaries are used by all those involved in the intervention in which they can record all resources used on a daily or weekly basis, as they suggest that by using this method the number and scope of data are more likely to be complete, and a reasonably accurate cost will be arrived at. Transparency is all-important and although suitable estimates can be used when microcosts are not available, these need to be well documented. It does not seem that any of the evaluations considered in this review used such rigorous data collection methods. A further consideration is the cost of adapting a programme to changing circumstances either over time or in a different context. This factor was not considered in any of the literature or discussions included here. The ‘Voices Against Violence’ programme developed by the Guiding Association and UN Women was introduced with the aim of using it across 20 countries and, although extensive revising and editing formed part of the development of the material, no mention was made of adapting the content as conditions and environments changed.

Combining cost and benefits

Combining costs and benefits to draw conclusions about the cost-effectiveness of a preventative intervention is a challenge that was not comprehensively tackled by any of the evaluations in the literature reviewed and it is a concept that was not considered explicitly by experts we contacted or members of our consultation groups. The technique of ROI can be useful, although from a public sector or societal perspective the wider social ROI (SROI) is preferred. This technique has gained credence as a method of evaluating complex interventions implemented in the public and third sectors since the publication of the Cabinet Office’s ‘Guide to Social Return on Investment’120 and it has been critiqued by Arvidson et al. 121 SROI is akin to cost–benefit analysis conducted from a broad societal perspective, though Arvidson et al. 121 suggest that, with SROI, the process of carrying out the analysis creates its own benefits, and these need to be taken into account in addition to the benefits of the programme itself. It also appears to be a labour intensive and expensive method of analysis.

Despite the increasing attention being paid to different ways of evaluating public health interventions, no standardised method exists. In practice, a range of types of outcomes is used depending on the research question and the rigour of the evaluation depends on its purpose. There appears to be a genuine tension between directing funds at preventative programmes themselves and conducting an evaluation that will provide evidence to inform future development.

Scale

The question of whether more resources result in more or better outcomes was something we attempted to discern from the literature and our consultations. Broadly, evidence from Wolfe et al. ’s evaluation of ‘Fourth R’ programme50 and the drama workshops suggests that longer-term education-based interventions might be more successful; this was backed up by others in the consultation groups who suggested that longer, more intense programmes were more likely to result in a permanent shift in attitudes, which could lead to behavioural change in the future. However, the evidence is scant. In contrast to these views, the systematic literature review (see Chapter 4) suggested that length on its own did not predict impact and those involved in developing media campaigns felt that there was very little relationship between scale and outcomes and that ‘getting it right’ was more important.

Funding

The UK-based education programmes and media campaigns identified in this study were funded from a range of sources including local authorities, the police service, education, the probation service, the health service and third-sector organisations. There was no clear sense that this was ‘joined up’ (as one member of the consultation group put it) and this resulted in uncertainty and lack of sustainability. At the same time, the potential benefits of a programme or campaign that successfully reduces domestic violence are widespread. Walby116 estimates that within the public sector, 45% of the cost of domestic violence is borne by health care and 33% by the criminal justice system, with the remaining 22% falling on social services, housing and refuges, and civil legal services. However, public services are themselves only one-quarter of the wider societal cost, as lost economic output represents a further 12% and human and emotional costs account for the majority 63%. This finding indicates the importance of a societal response and a need for a co-ordinated effort to develop and implement multisectoral initiatives.

Programme content

The systematic review showed that the focus of the work in North America was dating violence rather than young people’s experience of domestic abuse in their parents’ relationships and, as noted earlier in this chapter, a shift in this direction has occurred in the UK. Nonetheless, all programmes included in this review addressed a number of themes and topics that related to their aims and varied depending on target audiences. All interventions aimed at young people of secondary school age incorporated raising awareness of domestic abuse, usually through imparting knowledge about it: what it is, its prevalence, and how it impacts on victims and, in the case of adult victims, their children. The literature reviews and mapping survey also revealed that information on the services available to help survivors was commonly included with the intention of enabling children and young people to seek help and support safely. Where young children were the target audience, the grey literature and mapping survey showed that domestic abuse was never addressed directly but that the focus was on keeping safe and issues such as friendship, bullying and respect. Some programmes also incorporated teaching of skills such as conflict resolution and communication. Similarly, aspects of values education were frequently incorporated into programmes where values such as equality, acceptance of difference and respect were promoted with the aim of developing prosocial attitudes. Affective education, in the form of learning to identify and express emotions through non-violent means, was included in the content of some programmes, and with younger children this was focused on raising self-esteem so that children would be confident in protecting themselves.

The inclusion of topics on gender equality and gender stereotyping (masculinities and femininities) was common in programmes for older children and young people. Evidence from the literature reviews and consultation suggested that the ways in which these topics and the gendered nature of domestic abuse were approached were vital as some boys resisted what was sometimes described as the ‘sexist’ approach of such messages. The consultation emphasised the need for messages, particularly those targeting boys, to be positive to counter any such resistance. Wolfe et al. ’s evaluation of the ‘Fourth R’ programme50 showed that resistance from boys is not inevitable. Crooks et al. 122 argue that, in the absence of a gendered understanding of interpersonal violence, young people require a ‘gender-strategic’ approach, and they describe how the Fourth R programme builds gender awareness through a range of activities and by including opportunities for single-sex group discussions:

. . . both boys and girls will be hypersensitive to messages that they hear as ‘boy bashing’ (Tutty et al., 2002). The challenge is to understand this reality, yet increase awareness of adolescents’ understanding of gender and societal constructs of gender. 122

Programme structure

The findings showed that the length and structure of school-based interventions varied considerably from one-off sessions to manualised programmes of up to 10 or more sessions, such as, the ‘Fourth R’ programme in North America. In the UK, where a whole-school approach was adopted, interventions were of indeterminate length as they had often been developed or adjusted in response to particular groups and settings. 92 The survey and review of grey literature found that the most common structure for most school-based programmes in the UK was six 1-hour sessions delivered over 6 weeks, although no clear rationale for this arrangement was reported; however, it is roughly equivalent to one half-term period in schools. Much of the work was then relatively short-term and, therefore, less likely to be embedded into schools. Although the systematic review did not provide conclusive evidence of this, some evidence from the analysis of costs and benefits and the review of the grey literature suggested that, in the case of taught interventions, longer programmes that used more resources had most impact, but this would need to be tested further. Most interventions were targeted at specific age groups and did not form part of a comprehensive programme which linked together opportunities for children and young people to develop their learning on this issue throughout childhood and the teen years. Educationalists in the consultation group identified the need for a ‘spiral curriculum’ whereby basic ideas are repeatedly revisited, building children and young people’s learning across their school career.

Methods of delivery

The literature reviews and mapping survey data revealed that in the main a range of methods were used to deliver the content of interventions. Most commonly this included didactic approaches, group discussion, role-play, quizzes and visual stimulus such as DVDs, with fewer using creative arts such as drama/theatre. Both the young people and the experts who were consulted argued for the use of drama/theatre and narrative, as these approaches had the capacity to deliver an emotional charge which contributed to authenticity as well as having benefits for children with lower levels of literacy. While the cost of professional theatre productions could be prohibitive, both the systematic review and the grey literature showed that performances developed with young people themselves could be used to engage young people in delivery and infuse messages with authenticity. However, the extent to which drama and theatre are effective in changing behaviour requires further research.

Authenticity was also achieved through the use of messages and material that were recognisable and meaningful to young people and which made ‘it real’. Young people consulted for this study considered that authenticity was enhanced when interventions were delivered by those with relevant expertise or experience, and these genuine messages were contrasted with those that lacked conviction or plausibility. Media campaigns that used narrative and were developed with assistance from young people and/or which featured young people, such as the Home Office’s recent ‘This is Abuse’ campaign, were also perceived as authentic by the young people’s consultation group. As noted above, media campaign materials are used within school-based programmes and are increasingly likely to influence their content. The key message from the qualitative data reviewed was that children and young people find participative and interactive methods most engaging.

Who delivers?

Data from all elements of the study showed that interventions were delivered by a range of professionals, including teachers and other school staff such as school nurses, with external staff from specialist domestic abuse services, young people functioning as peer educators and other educationalists such as youth workers extensively involved in facilitating programmes both in the UK and elsewhere. There was considerable discussion in the consultation groups and expert interviews about who should deliver these interventions and the reviews of the literature revealed no consensus on this matter. External staff from specialist services have knowledge and expertise on domestic abuse and yet, their capacity to reach all children and young people, and thus to make provision universal and comprehensive, is limited. Programmes are then highly dependent on external funding, much of which has been shown to be short term, which limits the sustainability of interventions. In addition, while staff from outside schools are able to flag up the presence of community services for young people, they are less likely to have an impact on school culture or to provide continuity and progression for learners, making long-term change more difficult. On the other hand, teachers possess expertise in working with, and have ongoing relationships with, children and young people, which allows them to recognise and work with students’ responses to these interventions. The grey literature described some school staff resisting teaching on domestic abuse as they lacked confidence and competence, particularly in respect to dealing with any disclosures that the work might elicit. Examples of training for teachers to equip them to deliver programmes which had been provided by domestic abuse specialists were found across the data included in this study. The mapping survey found that nearly half of the interventions identified were delivered by multiagency teams consisting of teachers working in partnership with specialist services, and partnership models were advocated by those participating in the consultation groups and interviews.

The UK grey literature suggests that the majority of staff delivering taught programmes are women; however, there is evidence from the reviews of both the qualitative and the grey literature of an emerging debate concerning the gender of facilitators. Although there was no consensus about the importance of the gender of facilitators, some boys appeared to value a man facilitating the work. 57,82 Where work was cofacilitated by a woman and a man, there were opportunities for staff to embody the message of gender equity. This approach mirrors recommendations for good practice in work with perpetrators of domestic abuse. 103