Methodology

Data on the number of children living in conflict zones is conducted by the Peace Research Institute Oslo (PRIO). The core dataset used to map conflict patterns in this report is the Uppsala Conflict Data Program’s Georeferenced Event Data set (UCDP GED). The UCDP dataset provides the geographical location, timing and intensity of each conflict event globally, covering the years 1989–2022. A conflict event is defined as a lethal incident, either a violent clash between two armed groups or an attack on civilians by a group/groups. To estimate the number of children living in conflict areas, and populations more generally, PRIO cross-referenced the conflict data with population data from Gridded Population of the World (GPW) and from the UN World Population Prospects. It is very unlikely that there are fewer. The data on battle deaths does not distinguish between civilians and military personnel, or between adults and children. 

These numbers are based on reported and verified cases and incidents of the six grave violations against children, as presented in the United Nations annual reports of the Secretary-General on children and armed conflict between 2005 and 2022. Unlike the annual UN reports on children and conflict, we have included verified incidents of military use of hospitals and schools when we add up the grave violations in each conflict setting. The reporting only includes cases and incidents where the perpetrators are armed forces and groups.

Save the Children’s analysis and determination is based on nine indicators. Six indicators show the prevalence of verified cases or incidents of each of the six grave violations (as set out in the annual UN reports on children and armed conflict). Three further indicators draw on PRIO’s research to show conflict intensity measured by battle deaths, total child population living in conflict affected areas, and the share of children living in conflict zones relative to the child population of the country (see more info under Children in Conflict zones) The indicators are separately rated, then the rating is summarized. All indicators are weighted equally.

The data here builds on a series of reports developed by Save the Children and partners: Too little too late (2011), Unprotected (2019), Still Unprotected (2020) and Unprotected Special Edition (2023). Links to the reports can be found on the bottom of the main page.  The methodology builds on humanitarian funding as reported to the Financial Tracking Service (FTS), a global and real-time database created in 1992 and managed by UN OCHA which tracks international humanitarian aid flows. All humanitarian funding flows reported on the FTS are considered in the studies, it includes funding from Humanitarian Response Plans and appeals, the Central Emergency Response Fund, Country-based Pooled Funds, and other funds reported by the European Emergency Disaster Response Information System, government donors, UN agencies, NGOs and private donors.

For refugee settings, the main data source was the Refugee Funding Tracker, which includes funding and budgets for refugee-related appeals and plans such as country and regional Refugee Response Plans since 2012. However, the RFT does not provide sufficient sector-specific data for a situational analysis of Child Protection. Therefore additional sources with sufficient data granularity on Child Protection have been introduced to complement the analysis for refugee settings: (1) the Syria 3RP funding data for Child Protection tracked by No Lost Generation/Syria 3RP Child Protection Working Group, and Bangladesh JRP funding data collected from UN OCHA FTS following the methodology for categorising Child Protection detailed below; (2) UNHCR refugee programme funds on Child Protection, made available and analysed in comparison to the overall inter-agency funding in refugee settings.

Data limitations: Some limitations should be highlighted to put the findings of the studies in perspective.  

– The FTS database relies on voluntary reporting from donors and recipient organisations. It therefore does not capture exhaustively all humanitarian funding. It is, however, the most comprehensive public data source on humanitarian funding currently available. Centrality of Protection in Humanitarian action and Child Protection mainstreaming

– Protection is the central outcome and purpose of humanitarian response and all sectoral responses contribute to protection. As mentioned in the Minimum Standards for Child Protection in Humanitarian Action “All humanitarian actors have the obligation to engage in multisectoral child protection activities.” The studies acknowledges that other sectoral activities contribute to answer the protection needs of children but looks at funding for specialised Child Protection interventions and activities. Child Protection integration across sectors

 – The studies aim to identify funding for specialized Child Protection (CP) interventions. Funding for CP integrated programming is taken into consideration when identified under the Protection and Child Protection sectors of the FTS, but it is extremely difficult to account in an exhaustive manner for all integrated CP activities across sectors due to reporting requirements. For instance, some education or gender-based violence (GBV) interventions, reported under the education and GBV sectors, might include specialised CP interventions, but the studies focused on funding reported under Protection and Child Protection. In Humanitarian Response Plans, CP is often included in the Global Protection sector with no breakdown of data for the areas of responsibility: Funding requirements are still often formulated for the whole Protection sector and not specifically for CP, which led to limitations in tracking CP funding and funding coverage.

Keyword search – The studies include all funding reported on the FTS for the Child Protection sector, but it also delves into the Protection sector to identify funding that may qualify as CP funding. The studies therefore proceeded to a keyword search where a number of keywords were searched through the descriptions provided on the FTS to flag funding flows with a focus on Child Protection (see list of keywords below). Each flagged flow was then controlled individually. This process includes a certain degree of subjectivity in the choice of keywords and the categorisation of funding flows. The quality of this process is also constrained by the details provided in the funding description. For a full list of key words applied, please see the Still Unprotected Report on page 53.  In addition, even when funding for Regional Refugee Response Plans is tracked on the FTS, funding is not tracked by traditional sectors but overall reported under ‘multi-sector’, which hinders the identification of CP funding for these plans.

This data here draws on the following sources: First, bespoke data was produced by the Peace Research Institute Oslo (PRIO) that maps children and armed conflict trends from 1990 to 2020. The core dataset used is the Uppsala Conflict Data Program’s Georeferenced Event Dataset (UCDP GED) and the United Nations (UN) 2019 World Population Prospects.  Living in or close to a conflict zone is defined by the UCDP GED dataset as living 50 kilometres or less from one or more conflict events in a given year. A conflict event or incident is defined as a lethal incident, either a violent clash between two armed groups or an attack on civilians by a group or groups, at a given time and place. Conflicts usually consist of multiple conflict events. Further, PRIO classified countries by conflict intensity, with countries experiencing more than 1,000 battle-related deaths in a calendar year considered high intensity.  Second, PRIO researched new subnational data, estimating the number of children living in conflict zones who were at risk of being recruited into armed groups during the period 1990–2020.4 Children are defined as at risk of recruitment if they live within 50km of one or more lethal conflict events where at least one conflict actor who was reported to have recruited children in a given year was active. PRIO linked data from the UCDP GED with UN population estimates. Additionally, data on conflict actors reported to have recruited boys and girls in a given year were updated with data collected by the Dallaire Institute for Children, Peace and Security in Canada and new data collected by the University of Leiden in the Netherlands and PRIO. More can be found in the report Stop the War on Children – A crisis of recruitment

The data here builds on the report Weapon of War: Sexual Violence against Children in Conflict  (2021) The report seeks to present a compelling new angle on the issue by estimating how many children are at risk of sexual violence committed by conflict actors. In cooperation with the Peace Research Institute Oslo (PRIO) we have calculated how many children live 50km or closer to armed conflicts where at least one conflict actor has been reported to perpetrate sexual violence against children in a given year. For the first time, we approach this from a global perspective, and estimate the number of children at risk in all armed conflicts in the world.

This research relies on the Sexual Violence in Armed Conflict (SVAC) Dataset, which is based on reports by Amnesty International, Human Rights Watch and the US State Department, as well as subnational data on armed conflict events from the Uppsala Conflict Data Program Georeferenced Event Dataset (UCDPGED), and population estimates from the UN Population Prospects. The data gathered covers 30 years, since 1990 (see methodology above under Children in Conflict zones). In the datasets used, the term sexual violence is defined as rape, sexual slavery, forced prostitution, forced pregnancy, forced sterilisation, forced abortion, sexual mutilation, sexual abuse and sexual torture.  It covers violence perpetrated by armed conflict actors, including armed forces, law enforcement and non-state armed groups, but does not account for domestic abuse or sexual violence committed by criminal gangs or peacekeepers.