Table 2 The good school study

Study Aim

To evaluate the Good School Toolkit – a whole school intervention designed by Raising Voices (www.raisingvoices.org) which seeks to reduce violence in schools.

Methodology

Multi-component study (randomised controlled trial, qualitative evaluation, process evaluation, economic evaluation) conducted between 2012 and 2014. Interviewed students, teachers, school staff, school administration, parents/carers.

Study Measures

Quantitatively violence was measured trough self-reports using the International Society for the Prevention of Child Abuse and Neglect (ISPCAN) Child Abuse Screening Tool – Child Institutional (ICAST-CI). Disability was measured using the Washington Group Short Set Questions. Qualitatively in-depth interviews using a semi-structured tool were used to explore children’s subjective experiences and understanding of violence and how it impacted on their lives. Interviews with children with disabilities further explored how their disability affected their experience of school and their ability to participate in the intervention.

Ethical Approval

Obtained from the London School of Hygiene and Tropical Medicine and the Uganda National Council for Science and Technology

Permission

Obtained from appropriate national and district government authorities

Consenting Procedures

3-tiered consent process:

1. Consent sought from head teachers for school participation and to approach parents and students

2. Parents given opportunity to opt their children out of participation

3. Sampled children asked to given written consent to participate after emphasising that participation was voluntary, that there existed an obligation to report in case of any disclosure of violence, and that referral services were available. Specific attention was given to ensuring that children with disabilities understood the information with which they were presented and consented to take part by ensuring that information was provided in a manner that was accessible to them for example slowly and clearly for children with auditory impairments.

Interviewer recruitment and training

All interviewers had previous experience of working with children and provided written references. They received 3 weeks of full-time training on violence against children, child rights, disability, strategies to maintain privacy and confidentiality, consenting procedures, interview techniques and referral protocols. Considerable time was also made available for role-playing and practicing to ensure interviewers were as prepared as possible e.g. to respond appropriately to disclosures of violence. Interviewers were also trained on specific techniques for interviewing children with disabilities for example ensuring that they talked slowly and clearly whilst also facing children with auditory impairments. These techniques were practiced through role playing. An excess of interviewers were trained with only the best invited to participate in the data collection.

Referral procedures

The study employed a counsellor. A comprehensive protocol to handle disclosures of violence was developed in consultation with local child protection experts. It specified specific pathways of action depending on the severity and timeframe of what the child disclosed. Decisions on disclosures that would necessitate a referral and to where they would be referred were made considering the legal requirements in Uganda and the local child protection systems.

Interviewer welfare

The study measured signs of vicarious trauma amongst interviewers and provided space for critical reflection through team debriefs and space to decompress through group social activities. Interviewers were also able to speak with the study counsellor.

Challenges and learning

At follow-up, training on disabilities, specifically epilepsy (which is almost always considered a disability in this context), was strengthened given that at baseline, some interviewers were found to have a poor understanding of this disability. Children with the most severe disabilities were unlikely to be at school meaning that children with disabilities who were included in the study are likely to have less severe impairments.