You will need to design a range of indicators to measure change at different levels of your programme in line with your theory of change: for example, indicators to track whether activities and outputs are delivered in the short-term in line with your programme design, and indicators to measure whether the the expected outcomes and longer-term impacts are achieved. The table below gives examples:

When designing indicators, it can also be useful to refer to existing VAWG indicators and monitoring and evaluation frameworks in similar contexts. This is because key VAWG documents typically feature recommended indicators informed by global guidelines and standards. They also provide lessons learned and guidance on how to consider contextual factors. Key sources include: the table of indicators in the UN's RESPECT Monitoring and Evaluation Guidance and in UNICEF's Tip Sheet: List of GBV Risk Mitigation Indicators. 

Methods of data collection and measuring change should be both quantitative and qualitative to provide a more comprehensive understanding of progress towards reducing VAWG, addressing risk factors for VAWG, and providing high quality services to improve survivor wellbeing.

In terms of primary data collection, quantitative methods of information-gathering and measurement typically include surveys, questionnaires and statistics. Qualitative methods include interviews, focus group discussions and safety audits or observations. Qualitative methods can provide contextual information on risks faced by women and girls, perpetration of different types of violence, harmful consequences for survivors, and shifts in social and gender norms.

Secondary can also be included from other sources, for instance, demographic and social and economic indicators from national surveys, data from legal and judicial frameworks, academic and other reports.

A good indicator should be clear and concise. It should focus on a single issue that provides relevant information on a situation; particularly information that provides the strategic insight required for effective planning and sound decision-making. Each indicator should include a description of what it measures, the tools needed to gather the data, and the calculations involved in producing the measurement. Ensuring indicators are SMART is a helpful way to guide this process:

• Specific: Indicators should be specific and clearly defined, with a clear meaning and scope.
• Measurable: Indicators should be quantifiable and measurable, so that progress towards the goal can be tracked over time.
• Achievable: Indicators should be achievable and realistic, meaning that they can be realistically measured given the available resources and data.
• Relevant: Indicators should be relevant to the goals and objectives of the programme or project.
• Time-bound: Indicators should be time-bound, with a clear timeframe for measurement.

While individual indicators can be used independently, in most cases, groups of indicators should be used together to provide coherent overall information to programme managers or policymakers to track change and judge the progress of a programme.

It’s considered best practice to disaggregate indicators where appropriate to account for the range of experiences of different population groups based on their intersectional identities (i.e. by sex, age, race, disability, socio-economic status). Where disaggregation can support with precision, contextualisation, transparency and accountability, it can also benefit equity and inclusivity efforts, which is especially relevant where the language of indicators has been generalised.

While indicators make it possible to measure a range of changes and programmatic impacts, only those that are needed should be included in monitoring, evaluation and learning (MEL) frameworks. This is because each additional indicator requires additional data collection and this can place a burden on populations. So there needs to be enough indicators to meaningfully and robustly measure change, but not an excessive number. This is in line with ethical guidelines for data collection and supports a survivor-centred approach across all areas of work.

Different data collection methods and sources should also be considered critically when designing indicators, and any data collection should be undertaken in a way that respects confidentiality and women and girls’ rights without jeopardising their safety or provoking unintended consequences. This list is not exhaustive, as risks can vary between contexts, but some challenges which can impact the effectiveness or likelihood of measuring certain indicators include:

• Potential harm to beneficiaries, including in creating safety risks for survivors and other women and girls.
• Shortage of qualified, female enumerators/data collectors.
• Stigma faced by survivors who report VAWG.
• Insecurity, including the risk of retaliation by perpetrators and/or the community.
• Lack of harmonised VAWG-related data collection tools and data collection methods.
• Lack of or weak data-protection mechanisms to ensure the safety, security, confidentiality and anonymity of case information.
• Lack of service infrastructure.
• Lack of effective and quality case management services for VAWG survivors.
• Limitations on the mobility of typically marginalised segments of the female population (e.g. older women and adolescent girls or women and girls with disabilities).
• Restricted access to the affected population, especially women and girls.
• Limited time to establish trust and rapport with affected populations.
• Difficulty in establishing adequate interview settings that ensure basic privacy.

Global guidelines and ethical standards on programme design and data collection can help inform the design process, such as:

• WHO (2016) Guidelines for Ethical and Safety Recommendations for Intervention Research on Violence Against Women
• WHO (2007) Ethical and Safety Recommendations for Researching, Documenting and Monitoring Sexual Violence in Emergencies
• Inter-Agency Minimum Standards for GBViE Programming (2019)

Where indicators and measurements of change are being designed for specific populations, it’s important to engage them in the design process, and validate the indicators based on their inputs. Working with women-led organisations or diverse civil society organisations helps promote safe engagement of marginalised groups and women and girls in data collection, and aligns with feminist-informed approaches. Before collecting data, the local availability of care and support services for survivors/victims should also be mapped; if services are not available in the community or cannot be made available by the research team then data collection should not be undertaken.

In addition to selecting indicators, it can be helpful to consider data quality, and how to maximise reliability for Monitoring, Evaluation and Learning. Some data quality criteria include:

• Truthfulness: Responses collected are true, with consideration of how to minimise false reporting and bias
• Completeness: Data collected is as complete as possible, with no missing values
• Coherence: Data is coherent throughout and across a dataset
• Accuracy: Level to which data is precise and matches any means of verification
• Actuality: Most recent available data is used and referenced, as relevant to the timeframe of the assessment
• Safety: Data is secured, anonymised and kept confidential, with additional security such as encryption if any personal information is gathered
• Comprehensiveness: Data is representative of a situation, population, geographic coverage
• Uniformity: Same measures and metrics are used across the dataset
• Integrity: Quality of data for the whole life cycle of the dataset