Counting and documenting conflict-related sexual violence: Data and methods

This paper examines the importance of data in understanding, documenting and mitigating conflict-related sexual violence, with a specific focus on quantitative data collection methods. It provides a high-level review of approaches to documenting conflict-related sexual violence, highlighting prominent datasets such as the Sexual Violence in Armed Conflict (SVAC) dataset and the sexual violence data subset in the Armed Conflict Location & Event Data (ACLED) dataset. These datasets are explored in the context of their methodologies, strengths and limitations in tracking conflict-related sexual violence over time and across regions. The paper also discusses the crucial role of data in identifying patterns, trends and changes in the scale of conflict-related sexual violence, emphasizing the need for careful consideration of factors such as source diversity, language considerations, sample size and temporal dynamics. Furthermore, the paper critically assesses the biases inherent in data-collection methods, particularly in automated systems like artificial intelligence, which may lead to false positives or omissions. While biases are not necessarily malicious, they can shape research outcomes and policy decisions and must therefore be carefully evaluated. In conclusion, the paper asserts that a nuanced understanding of data-related challenges is essential for effectively preventing and mitigating conflict-related sexual violence, urging researchers and policymakers to approach data with caution, rigor and an awareness of their limitations.