Violence by teachers almost halved in primary schools

The Good School Toolkit is a behavioural intervention designed by Ugandan NGO Raising Voices, which aims to foster change in operational culture at the school level. Its materials include t-shirts, books, booklets, posters and guides for around 60 different activities. [1] To evaluate the Toolkit's effectiveness, researchers from the London School of Hygiene & Tropical Medicine, in partnership with Raising Voices, conducted a randomised controlled trial (RCT) in the Luwero District of Uganda. The trial took place over 18 months with 21 schools in the intervention arm and 21 in the control arm.

They found that the Toolkit produced a large reduction in physical violence from school staff -- 42% -- as reported by primary school students. Although it was highly effective for both sexes, results suggested that the Toolkit was more effective for male students. Students in intervention schools reported improved feelings of well-being and safety at school, suggesting that the Toolkit succeeded in changing the school environment.

Study lead author, Dr Karen Devries from the London School of Hygiene & Tropical Medicine, said: "What's notable about these results is that we found a very large reduction. We don't normally see a shift of such magnitude. That said, levels of school violence remain high. We need to do more.

"Given the prevalence of violence observed in the East African region, the Toolkit or similar programmes may have a major effect on the burden of child maltreatment in countries where violence from school staff is common. Further analyses are underway to explore the effect of the intervention on other forms of violence, including violence from peers. A process evaluation, a qualitative study and an economic evaluation are also underway, with results expected in 2015."

In East Africa, in-school violence is more prevalent than violence by parents. Exposure to physical violence in childhood is associated with, among other negative effects, increased risk of depression and suicide, poor educational results and increased risk of partner violence later on. In previous research carried out by the team in one Ugandan district, more than 90% of 11 to 14-year-olds reported receiving physical violence from school staff including caning, with 8% reporting extremely severe acts such as choking, burning, stabbing and being severely beaten up. Despite the health burden associated with violence against children, few programmes exist to tackle in-school violence and even fewer have been rigorously evaluated. The Good Schools study is therefore one of the first of its kind.

Despite large reductions seen in the study, levels of physical violence in the intervention schools remained high, with 30% and 60% of students reporting violence in the past week and past term respectively. Further research is needed to examine: if the Toolkit can further reduce levels of violence if implemented over longer time periods; whether the effects of the Toolkit are sustainable without ongoing support from Raising Voices; what effect the intervention would have at scale, and how best to scale it up; and if it has any effects on violence occurring outside of schools.

The research was funded by the UK Medical Research Council (MRC), UKaid through the Department for International Development (DfID), the Wellcome Trust via the Joint Global Health Trials Scheme, and the Hewlett Foundation.

The authors note that, as with violence research in general, the study findings are based on violence that was reported rather than observed. While school staff (unlike students) might be expected to emphasise an improvement, their reports show very similar effect sizes and direction to those of the students' reports, lending weight to the results. Students in Ugandan primary schools are slightly older than those in higher income countries, a factor to consider when generalising the results to primary school populations with different age profiles. The results should, however, be generalisable to most African settings.