About GReVD
The Consortium
Resources
As a result of consistency issues between sources and datasets, the current methodologies for monitoring violent deaths can consistently tell us trends within individual data sets, but we cannot say with precision how many people die each year around the world due to violence. The existing approaches to count violent deaths are admittedly imperfect and sometimes only proxy measures exist. This means there is no reliable global picture of violent deaths, which makes it harder to assess overall progress in reducing violence, the first target of SDG 16. We must invest now in the reporting and methodology that will be necessary for monitoring progress toward the goal through 2030. This report was funded by UKAid.
Artificial Intelligence (AI) can help us to better process large amounts of information and identify pandemic relevant events from massive feeds of unstructured data. The Cline Center for Advanced Social Research will apply semantic knowledge extraction methods to two extreme-scale global news sources (one proprietary, the other provided by an external vendor) to identify pandemic-related social disruptions, socio-economic effects, and policy responses. This effort will be supported by collaboration with experts across the GReVD consortium.
This paper aims at describing the present state and methodology of the Small Arms Survey’s Global Violent Deaths database (GVD), currently the broadest database on violent deaths, and how it relates to the vision for a Global Registry of Violent Deaths (GReVD), which ultimately aims to produce a count of actual violent deaths, coded by location and time. After assessing the various data sources and considering gaps and limitations of current data processes, the paper highlights the added value of a complementary registry approach.
This report, written by SIPRI staff for UNDP in 2015, summarized the findings and recommendations of a technical virtual working group convened around indicators and measurement of SDG 16. The reference document includes guidance on indicator selection as well as a 50 page annex with hundreds of indicators, variants, alternative specifications and sources. It is a valuable sourceboook for measurement of issues related to peace, justice and inclusive institutions (SDG 16).
When a violent death occurs, it must be coded through a channel and over multiple stages of coding without error to ultimately be included in a Registry. The 5 x 9 Framework, developed in the Gaps Report, introduces five coding channels, nine stages of coding and a number of errors that can occur in coding, resulting in under- and overcounting.
Violent death has a different meaning to different stakeholders. Conflict-related deaths and intentional homicides are certainly part of violent deaths, but do not wholly capture the concept. Because of these varying definitions, the consortium approach involves identifying upper bound and lower bound estimates to violent deaths and improving precision (reducing the band between bounds) over time. This brief introduces basic definitions and concepts related to violent death and asks the reader where they would draw the line.
The Global Registry of Violent Deaths (GReVD) will be a single catalog of every violent death coded by time and location. The initiative will resolve coding issues across sources and ensure that violent deaths are not left behind through the unified efforts of a wide and growing consortium. This brief describes the initiative. It introduces basic concepts including estimates using an upper bound and lower bound approach, immediate and long-term challenges and milestones and the partners in the GReVD consortium.