This article was written by professor Rachel Jewkes, director for DfID’s development program, What Works to Prevent Violence Against Women and Girls.
Over half of all children in the world — 1 billion children, aged 2–17 years — are estimated to experience violence annually.
Forms of violence overlap and reinforce each other — at school, in the community, and at home — and range from peer violence and bullying, to corporal punishment by family members and teachers, sexual abuse and harassment, coercion, rape and emotional abuse and neglect.
Girls are more likely to experience sexual and dating violence, while boys are more likely to experience physical violence and corporal punishment.
DFID’s What Works to Prevent Violence against Women and Girls Global Programme (What Works) has conducted three evaluations in very different settings to see whether schools could provide acceptable and effective entry points for interventions to reduce violence in Pakistan, Afghanistan and Kenya.
Our research has confirmed the very high prevalence of violence experienced by children in their early teens, the major health, social and educational impact, as well as the potential for the right interventions preventing violence.
Violence breeds violence
Violence experienced by school children in their early teens is widespread in all three countries.
We have seen that in Pakistan 93% of boys and 79% of girls in our What Works study had experienced peer violence in the past month and 91% of boys and 61% of girls had experienced corporal punishment at school in the past month, despite this being illegal.
In the Kenyan study, 7% of girls (mean age 12 years) had been raped in the past year alone. Children with disabilities and children from the poorest families are also particularly vulnerable to violence. In our Afghanistan and Pakistan studies, children with a disability were more than twice as likely to be victims of peer violence as those without a disability.
Other research has shown us how different forms of violence are connected: In South Africa and the United States experience of bullying is closely related to dating violence perpetration (for boys) and experience (for girls).
Children learn about dominance and control within the family through observation and experience of the behaviour of parents, older siblings or significant others, and through socialisation among peers.
Thus, we hope that by impacting on violence among children at one point in their life trajectory, we will be able to alter other, and even later, use and experiences of violence. This is critical for breaking the cycle of violence.
The interconnectedness across generations was shown by our finding from Pakistan that 1 in 7 of the children who had perpetrated peer violence in the past month had witnessed their mother being abused in this period, compared to 1 in 20 of those who had been victims of peer violence, and less than 1 in 100 of those with no experience of violence at all.
Violence undermines children’s confidence, increases depression and leads to absenteeism and drop-out.
In our Pakistan study, peer violence was associated with depression, poorer school attendance and performance. We also learned that one of the reasons that girls are withdrawn from school early is to prevent them from being exposed to sexual harassment travelling to and from school.
In Kenya we found that teenage girls who had been raped were at much higher risk of dropping out from school, especially if this had occurred more than once. On a more positive note, however, our evaluations have shown us that with the right intervention, schools can provide positive opportunities for preventing violence in school and at home.
The right place and the right time
An important message from What Works is that the right intervention deployed in the right setting through schools can result in reductions in violence.
Our ground-breaking work has demonstrated that this is possible even in some of the world’s most fragile and conflict affected environments. A play-based life skills education programme in Hyderabad, Pakistan implemented by the international NGO Right to Play was subject to rigorous assessment through a large randomised controlled trial.
The students were in grade 6 (mostly 11-12 years) when the intervention commenced. The intervention is a structured play-based life skills education program, delivered by coaches and junior leaders, selected and trained from among the scholars, in 120 sessions over two years.
Each activity or game was followed by a process of critical reflection: reflect, connect and apply, and through this a range of attitudes were challenged and confidence and skills developed.
This study showed that the proportion of boys and girls in schools who had experienced corporal punishment at school in the last four weeks dropped significantly in the intervention compared to control schools over 24 months.
The study also found the intervention was associated with less experience of physical punishment at home and at school in the last four weeks, witnessing of fighting between the child’s father and another man, and witnessing the child’s mother being beaten at home.
The reduction in experience of violence was associated with improvements in the children’s mental health, shown through reductions in depression, and changes in gender attitudes, notably less patriarchal gender attitudes.
Helping Afghanistan’s children
What Works also undertook proof of concept research in a schools-based peace and community education programme in Jawzjan Province, Afghanistan implemented by Help the Afghan Children (HTAC).
The area was conflict-affected during the study, with particular security problems in one of the three districts in which the intervention was set. Students were in grade 7 and 8 (13-15 years) when the intervention commenced.
The interventions that were effective allowed sufficient time for change and were context- and developmentally appropriate
This was evaluated through a modified interrupted time series evaluation conducted over 18 months during a two-year intervention. Over two years, 99 sessions of peace education (about 50 hours) were provided by trained and supported teachers. The lessons sought to build resilience and self-confidence, provide an understanding of the causes of conflict and conflict resolution skills, communication skills, and critical reflection skills.
The intervention also included conflict resolution and peace building training for other adults in the community including parents, young people (aged 18 to 25 years), and community and religious leaders, with discussions on how Islam supports the rights and protections of women and girls, and how the involvement of women in community affairs facilitates safer, more secure and prosperous communities.
There is also training provided for women participating in CSOs and working in governmental departments and use of community radio programming.
Learning from failure
Over the period of the evaluation we saw peer violence reduce by over 50%.
The proportion of boys experiencing corporal punishment at school in the past month dropped from 43.7% to 27.2% and for girls dropped from 35% to 14.2%. Boys also reported that they witnessed their relatives abuse their mother less often, and girls reported seeing their father abuse their mother less often.
Again, there were accompanying improvements in depression scores and gender attitudes.
We learned much from our third evaluation, although here the intervention was not successful in preventing violence.
We conducted a rigorous and independent evaluation of a sexual violence prevention intervention in Kenya delivered by the NGO Ujamaa. The intervention focused on teaching self-defence in schools in Nairobi’s informal settlements and also ran courses with boys to encourage them to challenge ideas of masculinity and the use of violence.
Girls and boys in grade 6 (mostly 11-16 years at baseline) were given a 12 hour intervention. The conclusions that we can draw from this evaluation are limited, as the data is still being analysed.
The girls’ intervention was partly mental and verbal self-defence and 6 of the hours were spent on physical skills. The boys’ intervention focused on manhood, puberty and resisting bullying and aggression. Neither of these interventions were effective in reducing violence.
The girls’ intervention did not prevent their being raped and the boys did not stop their use of violence. The notable difference between the interventions in Afghanistan, Pakistan and Kenya was the time given for the intervention to work.
The What Works evaluations are important as to date there have been significant gaps in the evidence base, with only a few rigorously evaluated programmes demonstrating promise in preventing violence outside of North America.
A notable exception is the “Good Schools Toolkit” intervention in Uganda, which has been demonstrated to be effective at reducing various forms of violence (emotional, physical and sexual) from school staff to students, as well as between peers.
What Works has proven that even in challenging settings where violence is commonplace, it is possible to achieve significant reductions in violence through interventions in schools
We have shown that even in challenging settings, it is possible for interventions in schools can achieve significant reductions in violence — up to 50% — but also, that interventions are not always effective.
The interventions that were effective allowed sufficient time for change (18-24 months) and were context- and developmentally appropriate. This work takes time, especially with young teenagers. The intervention in Kenya was much shorter and less intensive (6 sessions of 2 hours) than those in Pakistan (60 sessions of 35 minutes per year over a two-year period) and Afghanistan (33 lessons of 35 minutes per year over two years) that showed a reduction in violence.
Key elements of success that characterised our more successful interventions were that they were multi-component and had a strong theory of change.
They worked with children but also involved parents and/or the wider community including local leaders to challenge wider community norms around gender and violence. They used effective learning methods including those that developed critical thinking skills, communication, empathy, leadership and non-violence, and included activities to promote critical reflection on gender norms, roles and identities and power.
Ending violence and lifting depression
The intervention in Kenya had a quite different pedagogy.
Further we were concerned from qualitative research that the self-defence intervention might have given girls confidence (a sense of self-efficacy) disproportionate to their physical abilities and they may have neglected more established safety behaviours, such as seeking safety in numbers.
Our interventions were delivered by highly trained teachers, NGO staff or coaches — they had purposefully selected, well trained, experienced and well supported intervention facilitators. The teachers were trained in gender norms, roles and identities as well as in the content of the programme and teaching or facilitating skills.
The interventions in Afghanistan and Pakistan that were successful in reducing children’s experience of violence also reduced children’s depression.
Neither of these used psychotherapeutic methods to achieve this impact. This suggests that by removing an important risk factor for mental ill-health, violence reduction can effectively reduce depressive symptoms, which is likely to have far reaching impacts on children’s wellbeing and education.
By employing these nuanced and culturally sensitive techniques and working with communities, What Works has proven that even in challenging settings where violence is commonplace, it is possible to achieve significant reductions in violence through interventions in schools. — Rachel Jewkes
Read the full evidence brief Violence Against Women and Girls and Education by Erica Fraser and Rachel Jewkes here.
(Picture credit: Death to the stock photo)
*This article has been updated after publication to more accurately reflect the results of the Kenyan violence prevention program.
Remember to mark this article as complete using the button below