Understanding GBV in emergencies

1Defining gender-based violence

The Inter-Agency Standing Committee (IASC) Guidelines define GBV as any harmful act that is perpetrated against a person's will and that is based on socially ascribed differences between males and females (gender). It includes acts that inflict physical, sexual or mental harm or suffering, threats of such acts, coercion and other deprivations of liberty, whether in public or in private life.

GBV affects everyone, but it disproportionately affects women and girls, because it is rooted in gender power inequality. Men and boys can also be survivors, particularly of sexual violence in detention or conflict, as can LGBTQI+ people and persons with disabilities, who are often more exposed and less heard.

The term "survivor" is preferred in psychosocial services because it emphasises the person's agency; the term "victim" is still used in the legal and medical sectors.

2The forms of GBV

The GBV Information Management System (GBVIMS) uses six core types, which provide a common classification across organisations:

• Rape: non-consensual penetration, however slight, of the vagina, anus or mouth with a penis or other body part, or with an object.
• Sexual assault: non-consensual sexual contact without penetration (unwanted touching, forced kissing, sexual mutilation).
• Physical assault: hitting, slapping, burning, choking and any non-sexual physical violence.
• Forced marriage, including child marriage: under 18, a child cannot give free and informed consent.
• Denial of resources, opportunities or services: confiscation of income or documents, being prevented from going to school, receiving care or collecting assistance.
• Psychological or emotional abuse: threats, humiliation, forced isolation, harassment.

Other terms describe contexts: intimate partner violence, by far the most common form, including in crises; sexual exploitation (sex exchanged for aid, shelter or money); female genital mutilation; trafficking.

3Causes, risk factors and consequences

The root cause of GBV is gender inequality: social norms that tolerate violence, male control over resources and decisions, impunity. Crises do not invent GBV, they exacerbate it through contributing factors:

• breakdown of family, community and rule-of-law protection;
• displacement, overcrowding, shelters without lockable doors, distant and unlit latrines;
• loss of livelihoods, which pushes towards survival sex and early marriage as an economic strategy;
• presence of armed men, collection of firewood or water in isolated areas;
• poorly designed aid distribution, controlled by intermediaries who abuse their power.

The consequences are serious and lasting: injuries, HIV and other STIs, unwanted pregnancies, unsafe abortions, fistula, psychological distress, suicidal thoughts, stigma, family rejection, loss of income, and sometimes death (femicide, suicide, obstetric complications).

4The principle: assume GBV is happening, and do no harm

The IASC Guidelines set out a central principle: all humanitarian personnel should assume GBV is taking place and treat it as a serious and life-threatening problem, regardless of the presence or absence of concrete evidence or figures. The absence of reported cases never means the absence of violence: shame, fear of reprisals and the lack of services mean that the vast majority of survivors do not speak out.

The practical consequence: no prevalence survey is carried out at the start of an emergency to "prove" the problem. Such surveys expose respondents to risk and add nothing to the immediate response. The WHO ethical and safety recommendations (2007) require in particular that the benefits to respondents outweigh the risks and that services be available before any data collection.

The do no harm principle applies to every action: a badly worded question, a poorly stored file, a photo, a name in a report or a focus group discussion about individual cases can expose a woman to deadly reprisals.

5Analysing the GBV situation in the first weeks: a step-by-step method

The principle of "assume GBV is happening" does not exempt anyone from analysis: it changes the nature of the analysis. The aim is not to prove that violence is taking place or to count it; the aim is to understand which risks affect which people, which services exist and actually function, and which barriers prevent people from reaching them. This analysis feeds the Humanitarian Needs Overview (HNO), the Humanitarian Response Plan (HRP), donor proposals and service design. The GBV Area of Responsibility's Handbook for Coordinating Gender-Based Violence Interventions in Emergencies (GBV AoR, 2019) and the assessment standard of the Inter-Agency Minimum Standards provide the framework.

Step 1: secondary data review (days 1 to 5)

Before any data collection, a member of the GBV team, ideally supported by the sub-cluster's information management officer, gathers what already exists: the country's Demographic and Health Surveys (DHS), which often include a domestic violence module; data from the ministry in charge of women's affairs; reports by local women's NGOs; the Secretary-General's reports on conflict-related sexual violence; previous multi-sector assessments; national legislation (legal age of marriage, legal status of abortion, mandatory reporting obligations, criminalisation of homosexuality, recognition of marital rape). This work fits into a five-page note and must be completed in under a week. It answers a simple question: what do we already know about gender norms, the forms of violence prevalent before the crisis and the legal framework?

Step 2: service mapping (days 3 to 10)

This is the operational priority, because everything else depends on it: without services, no data collection with women is ethical. For each facility, the team checks on site or by direct call, not on the basis of an old list:

• the actual availability of clinical management of rape: post-exposure prophylaxis (PEP) kits in stock, emergency contraception, trained staff, presence of a female health worker, real opening hours, services free of charge;
• the existence of GBV case management, psychosocial support, safe shelter and legal aid;
• access conditions: distance, transport costs, requirement of a police report or identity document, language, accessibility for persons with disabilities.

Step 3: limited and ethical primary data collection (days 7 to 21)

Data collection covers risks, perceptions and access, never individual experiences. It combines:

The eight WHO ethical and safety recommendations (2007) for researching, documenting and monitoring sexual violence in emergencies apply to all data collection: benefits must outweigh risks, methodology must minimise risks, referral services must be available before data collection, safety of everyone involved, confidentiality, informed consent, careful selection and specific training of the team, additional safeguards for children.

Step 4: analysis and feedback (days 21 to 30)

The analysis cross-references three questions: which risks, for whom (sex, age, disability, legal status, sexual orientation if the data allow it without risk), and which existing capacities. The output is a GBV situation analysis note of 8 to 12 pages, presented to the sub-cluster, shared with the other clusters and used for the HNO. It contains specific sectoral recommendations (WASH, shelter, CCCM, food security, health, education) and includes no identifiable account.

Who does what

• GBV sub-cluster coordinator: launches the exercise, avoids duplicate assessments, validates the tools, organises the feedback session.
• Member organisations: provide trained teams (exclusively women for women's groups) and carry out the mapping of their area.
• Health cluster and UNFPA: report on the availability of clinical management and reproductive health kits.
• Other clusters: integrate questions on GBV risks into their own assessments (MIRA, sectoral assessments), in the form of questions about places and perceptions, not about cases.

6The normative and legal framework every practitioner must master

A GBV professional must be able to cite precisely the instruments on which their work is based, because they are used for advocacy with the authorities, negotiation with donors and argument within Humanitarian Country Teams. The table below distinguishes binding international law, Security Council resolutions and humanitarian standards.

National law: the decisive question in the field

International law sets the direction; yet it is national law that determines what a survivor risks and what the team can do. Before opening services, the team must answer eight questions in writing, with the support of a national lawyer:

1. Is marital rape recognised as an offence?
2. What is the legal age of marriage for girls and boys, and are there exceptions (parental consent, customary or religious law)?
3. Is a police report or certificate required to access care after rape? (If so, this is a major barrier that warrants immediate advocacy.)
4. Under what conditions is termination of pregnancy legal (rape, incest, danger to life)?
5. What mandatory reporting obligations apply to professionals (violence against minors, all violence, certain professions only)?
6. Are same-sex relationships or sex work criminalised, which would expose some survivors to prosecution if they come forward?
7. Is a medical certificate admissible if issued by a midwife or nurse, or only by a doctor?
8. Which customary justice mechanisms deal with cases of violence in practice (mediation, compensation, marriage to the perpetrator)?

The answers to these questions are included in the standard operating procedures (SOPs) and explained to survivors from the first contact, particularly mandatory reporting obligations: a survivor can only give informed consent if she knows what will or may be passed on.

7Data and figures: what we know, what we can say, what we must never do

GBV data are among the most misused in humanitarian action. A professional must be able to distinguish three types of data, cite reference figures correctly and reject false interpretations, even when they would suit an advocacy message or a funding proposal.

Three families of data

Reference figures to cite correctly

• World Health Organization, 2021 estimates (2000-2018 data): about one in three women (30%) worldwide has experienced physical or sexual intimate partner violence or non-partner sexual violence in her lifetime; about 26% of ever-partnered women aged 15 and over have experienced violence from an intimate partner; about 6% report sexual violence by someone other than a partner, a figure that is heavily under-reported.
• Child marriage (UNICEF): about one in five young women worldwide was married before the age of 18, with much higher rates in several countries of West and Central Africa and South Asia, and documented increases during crises.
• Female genital mutilation (UNICEF, 2024): more than 230 million girls and women alive today have undergone it.
• Intimate partner violence in crises: studies conducted in conflict and displacement settings show that it remains the most common form, often more common than sexual violence by combatants, contrary to the media image.

Always cite the source, the year and the population concerned. Do not write "70% of women experience violence in crises" unless you can name the study, the country and the method.

Misinterpretations to reject

• Presenting service data as prevalence: "312 GBV cases in the district" does not mean "312 women experienced violence". The correct wording is: "312 incidents reported by people who accessed the services of organisations participating in the information-sharing protocol".
• Concluding that violence has increased or decreased on the basis of reports: an increase often reflects the opening of a new service, an information campaign or greater trust; a decrease may reflect the closure of a service or insecurity that prevents movement.
• Comparing areas that do not have the same service coverage: the area that "has the most cases" is often the one with the most services.
• Demanding figures in order to fund: the response is justified by the presence of risk factors, the affected population and the lack of services, not by a number of cases.

How to justify a need without counting cases

For a proposal section or an HNO chapter, the argument rests on four blocks: (1) pre-crisis national prevalence data, cited with their source; (2) risk factors observed since the crisis (displacement, overcrowding, loss of income, armed presence, service disruptions); (3) service gaps documented by the mapping (for example, "2 out of 7 facilities have PEP kits"); (4) concerns expressed by women and girls in focus group discussions. The number of people in need of GBV services is estimated using the common methodology set each year by the GBV AoR for the humanitarian programme cycle, not on the basis of reports.

Rule of caution: any GBV data presented publicly states the source, the period, the contributing organisations and the sentence "these data do not reflect the prevalence of GBV".

8Groups at heightened risk and an intersectional approach

GBV disproportionately affects women and girls, but not all of them are exposed in the same way. The IASC Guidelines (2015) and the Minimum Standards (2019) call for identifying at-risk groups, not to create rigid categories, but to adapt services and remove specific barriers. The intersectional approach consists of analysing how gender combines with age, disability, legal status, ethnicity, sexual orientation and gender identity to produce particular risks and forms of exclusion.

Emerging forms to include in the analysis

• Technology-facilitated GBV: sharing of intimate images, phone harassment, control of location or mobile money accounts. These also affect humanitarian settings, particularly urban settings and digital cash transfer programmes.
• Economic violence linked to assistance: confiscation of ration cards or withdrawal codes, pressure to sell assistance, sex in exchange for registration. These fall under both denial of resources and sexual exploitation.
• Trafficking: promises of employment or marriage abroad made to adolescent girls and young women in displacement sites, often by intermediaries known to the family.

Including male survivors is a requirement of non-discrimination, but the expertise and resources of GBV programmes remain mainly focused on women and girls, because gender inequality is the root cause. The two requirements are not in conflict: health and mental health services are trained to receive all survivors, while spaces reserved for women and girls are maintained.

9Full example: the initial GBV risk analysis in Mabéla district

Fictional context. For the past five weeks, Mabéla district has been hosting 46,000 internally displaced persons fleeing clashes between the army and a non-state armed group. About 60% live in three spontaneous sites (Kabo, 19,000; Lirou, 11,000; Sanza, 7,500), the rest with host families in the town of Mabéla. The demographic breakdown from the initial registration is as follows: 52% women and girls, 21% children under 5, 14% adolescent girls and boys aged 10 to 17, and 31% female-headed households. An international non-governmental organisation that provides GBV case management in the neighbouring district is mandated by the sub-cluster to coordinate the analysis.

1. Secondary data review

• The latest national Demographic and Health Survey indicates that 34% of ever-partnered women aged 15 to 49 report physical or sexual violence by their partner, and that 41% of women aged 20 to 24 were married before the age of 18 in the region concerned.
• The penal code recognises rape but not marital rape; the legal age of marriage is 18 for girls, with a possible exception down to 15 with parental consent; reporting violence against minors is mandatory for health and education staff; termination of pregnancy is permitted in cases of rape upon medical decision.
• The Secretary-General's annual report on conflict-related sexual violence mentions violence committed by both parties in the region over the past two years.

2. Service mapping (verified by phone call and visit)

3. Primary data collection

Over two weeks, the team (four women and two men, all trained, plus two supervisors) conducts 14 key informant interviews, 18 focus group discussions (6 with adult women, 4 with adolescent girls, 4 with adult men, 4 with adolescent boys), totalling 152 participants, and 6 exploratory day and night walks. Each enumerator is given a referral card; two spontaneous disclosures occur during the groups: the facilitators politely interrupt the account in the group, offer a one-to-one conversation after the session, provide information about the mobile clinic and the hospital, and nothing is recorded in the notes.

4. Analysis: the main risks

5. Reasoning and priorities

The team has no case figures, and does not seek any. The absolute priority is access to care within 72 hours, as this is the only intervention whose effectiveness depends on short time frames. Next come low-cost, quick-impact risk mitigation measures (locks, lighting, latrine segregation, verification of lists), followed by the set-up of case management and a safe space in Kabo, the most populated site. The local women's association is identified as a partner to be strengthened (training, secure storage, supervision), rather than replaced by an international team.

6. The final product

The 10-page analysis note is presented to the sub-cluster in week 5, then to the inter-cluster coordination group. It includes: a two-page summary, methodology and limitations, legal context, service mapping, six priority risks with recommendations by sector, an estimate of the population in need according to the GBV AoR methodology, and an annex of standard questions on GBV risks that other clusters can integrate into their assessments. It contains no precise location of assaults, no personal accounts and no case figures.

10Common mistakes, contextual variations, interview questions and references

Common mistakes

What donors and evaluators check

• A gender and GBV risk analysis prior to the project, including for non-specialised projects (for example through the ECHO Gender-Age Marker or the IASC Gender with Age Marker).
• Direct consultation of women and adolescent girls, documented (number of groups, composition), without identifying data.
• The existence of a service mapping and referral pathway before any data collection.
• No case figures used as prevalence in the project justification.
• National legislation taken into account in the design (mandatory reporting, age of marriage).

Variations by context

• Sudden-onset emergency (natural disaster, mass influx): analysis reduced to 10 days, priority given to mapping clinical management and to risk mitigation measures in collective shelters.
• Protracted crisis: pre-crisis prevalence data and in-depth qualitative studies on social norms become possible and useful; the analysis is updated every year for the programme cycle.
• Camp: concentrated and visible risks (latrines, distributions, lighting), identifiable management structures; watch out for male-dominated committees.
• Urban settings or host families: dispersed population, housing-related risks (exploitation by landlords, overcrowding), less visible domestic violence; rely on public services, neighbourhood associations and helplines.
• Restricted access and remote management: analysis based on trained local partners, phone interviews with female key informants on non-sensitive questions, no identifying data via messaging apps.
• High insecurity: reduce the number of visits, favour neutral locations (health centres), assess the risk for participants who might be seen with an organisation.

Interview questions

Key references

• IASC, Guidelines for Integrating Gender-Based Violence Interventions in Humanitarian Action (2015).
• UNFPA and GBV AoR, Inter-Agency Minimum Standards for Gender-Based Violence in Emergencies Programming (2019).
• GBV AoR, Handbook for Coordinating Gender-Based Violence Interventions in Emergencies (2019).
• WHO, Ethical and Safety Recommendations for Researching, Documenting and Monitoring Sexual Violence in Emergencies (2007).
• WHO, Violence Against Women Prevalence Estimates (2021).
• United Nations, annual reports of the Secretary-General on conflict-related sexual violence.
• IASC, Guidelines on the Inclusion of Persons with Disabilities in Humanitarian Action (2019).
• UNHCR, Working with Men and Boy Survivors of Sexual and Gender-Based Violence in Forced Displacement (2012) and Working with Lesbian, Gay, Bisexual, Transgender and Intersex Persons in Forced Displacement (2011).
• IASC, Gender Handbook for Humanitarian Action (2017).