Any VAWG programme which involves working with women and girls and collecting data about their experiences of violence must be done in a safe and ethical manner by taking a Do No Harm approach. The main ethical concern when collecting data on VAWG is the risk of harm to respondents, such as potential retaliation from a partner if their participation in the study is discovered. When collecting data, participants should have the right to decide if they want to disclose occurrences of violence and to determine whether, how, where and when specific information is shared. This includes victims and survivors, their families and supporters, communities, organisations working with survivors, and those involved in data collection itself. To ensure informed consent, it is important to always explain the nature of the study to participants including the purpose, risks and benefits, and provide the option to withdraw at any time. All data collection activities should be rooted in a survivor-centred approach and designed based on the GBV Guiding Principles:

1. Respect privacy and ensure consent – Data collection should prioritize the dignity of respondents and respect their privacy. Participation should be voluntary in nature, with informed consent. The participant should feel free to cancel or reschedule the session if they have any concerns.   

2. Safety – Data collection should take place in locations/spaces and at times that are safe and accessible for respondents. The safety and security of all those involved in data collection and analysis should be continuously monitored.  

3. Confidentiality – Participants should always remain anonymous and protected. Any information must be stored securely with electronic data/soft copies password-protected and encrypted (such as on GBVIMS and Primero/GBVIMS+ incident monitoring and case management information systems) or hard copies in locked locations. With the rise of AI and machine-learning tools and technologies such as mobile phones or web-based platforms facilitating remote GBV data collection, transcribing interviews and analysing textual and big data, researchers must detail in consent forms and explain to participants how their data will be used, analysed and their privacy will be protected. 

4. Non-discrimination – Adhering to the Leave No One Behind principle, survivors should receive equal and fair treatment regardless of their age, disability, gender identity, religion, nationality, ethnicity, sexual orientation, or any other characteristic.  

Participants’ data should always remain anonymous and protected – an important issue given the sensitive nature of information relating to violence and abuse. Secure and ethical data storage involves implementing robust encryption methods, access controls and regular audits to prevent unauthorised access and breaches. Confidential information must be stored securely with electronic data/soft copies password-protected and encrypted (such as on GBVIMS and Primero/GBVIMS+ incident monitoring and case management information systems) or hard copies in locked locations. With the rise of AI and machine-learning tools and technologies such as mobile phones or web-based platforms facilitating remote GBV data collection, transcribing interviews and analysing textual and big data, researchers should clearly explain to participants how their data will be stored and their privacy will be protected.

Any potential risks (physical, emotional and social) to participants, practitioners, and the wider community should be identified and mitigated at the outset to limit possible harm and maximise possible benefits. When collecting data as part of VAWG interventions, this includes minimising the effects of power hierarchies. For example, in many situations, healthcare providers, teachers or traditional leaders are often viewed with great respect and authority, so it is important to put in place measures to mitigate potential risks, such as safe and private interview settings, or anonymous and confidential surveys administered by vetted researchers.

Respondents should be made fully aware of and understand the risks and benefits of participating, have the right to decide and opt out of the study, and have opportunities to ask questions. A risk assessment should be conducted to understand any threats to physical safety, such as holding interviews in unsafe areas and removing these sites from the sampling frame, as well as putting appropriate safety measures in place.

The WHO guidelines state that researchers have an ethical responsibility to ensure scientific soundness and appropriateness to the context and population.

• In the case of research to measure the prevalence, causes and consequences of a specific type of VAWG, this will involve selecting an appropriate sample size and, usually, using complementary qualitative data collection to help interpret results.  
• In terms of intervention research and evaluation, the appropriate data collection methods will depend on the maturity of the intervention (e.g. is it a pilot or proven intervention) and the resources and time available.

In all cases, before selecting a methodology, it’s important to review available secondary data, including any other studies on the prevalence of that type of violence, any relevant evaluation evidence, and practice-based knowledge from similar programmes in the local context and/or globally. This will enable the identification of gaps that you can address through further primary data collection as needed. In the cases of impact evaluations with control arms (experimental or quasi-experimental designs), ethical practice means that you should ensure that the intervention is delivered to the control group once the evaluation is complete.

Ethical approval is not required for routine data collection activities such as gender analysis or collecting programme monitoring data that does not involve personal information. However, research that asks survivors or at-risk populations about their experiences of violence or other sensitive information - through focus groups, key informant interviews, or even anonymised quantitative surveys that gather data on GBV experiences and attitudes - will typically need to obtain ethical clearance from the appropriate Institutional Review Boards (IRBs).

IRBs review the risks associated with the proposed research study and ensure adherence to ethical principles in human research, such as respect, Do No Harm, and equitable treatment of research participants. The necessity for ethical approval can vary depending on the country and sensitivity of the research. Local research partners, universities or government ministries can support with understanding the local context for IRB processes to ensure proper precautions are taken.

When conducting research about VAWG, there is always a risk that a research participant might suffer distress when discussing their or others’ experiences of violence. Participants may also ask for further support. It is, therefore, the researchers’ ethical obligation to provide information about services available to all research participants regardless of whether they report experiencing violence. Before undertaking data collection, research teams need to map locally available referral services providing basic care (physical and psychosocial), support and information (formal and/or informal) for survivors/victims and those at risk.

Researchers can create information products such as pamphlets or small cards with listed resources or develop a directory that professionals can use to make referrals to share with women – when they share these resources with women, they should check it is safe for them to receive these materials (and that their discovery does not risk further provoking a violent partner). Where services such as psychosocial support are not available, research teams can consider hiring a counsellor to accompany the research team and provide support to those who need it. Research should not be undertaken if support services are unavailable in the community or cannot be made available by the research team.

Research on individual GBV experiences should only be conducted by trained research staff or partners who specialise in GBV to ensure the wellbeing of participants especially as women and girls may feel vulnerable while disclosing difficult and painful experiences. All data collection teams should be carefully selected, include female members, and receive relevant and sufficient specialised training. Training on research methods and tools, ethical principles, privacy, confidentiality, informed consent, safeguarding measures, and direct support to respond to disclosures made by participants should be carried out by lead researchers to all of those involved at every stage of data collection, processing and dissemination.

Asking questions and handling information about painful and traumatic events requires interviewers to be trained in how to sensitively approach these topics and recognise signs of distress during the interview. They must be prepared to take appropriate action to support the respondent and/or stop the interview if needed. Although usually exempt, researchers should be aware of mandatory reporting requirements for suspected cases of violence to local authorities or social services.

All team members, including field supervisors, interviewers, transcribers, and data personnel, may be affected by hearing the traumatic stories and experiences that are shared. They should also receive regular psychosocial support and debriefs with professionals such as counsellors or psychologists. This not only helps them process their emotions, but also improves their ability to gather quality data.

Safeguarding processes must be established for any participating children, young girls or adolescents under 18 years old to protect them from any potential harm, including sexual exploitation, abuse, or sexual harassment (SEAH) during the data collection process. In many cases, this may include obtaining consent from a parent/carer/guardian/responsible adult (verbal or written agreement for the child to participate in research or M&E) on their behalf – although there may be circumstances where this is not appropriate or safe. Children must also be given the opportunity to provide age-appropriate, voluntary, informed assent to participate, or refuse to participate. Both children and parents/guardians must be informed of the possible limitations of confidentiality, for instance, when there are legal requirements to report child abuse. Before any research activities are undertaken, children’s safety must be prioritized and risks of harm minimized, by consulting with local stakeholders to identify and outline all potential risks to children in research protocols.