Over the last decades, the number of publicly available databases and datasets on terrorism (DDTs) has increased exponentially, facilitated by access to the internet, advancements in data collection capabilities, and the variety of supporting software. Currently, hundreds of sources provide data and information about terrorism-related issues, such as incidents, perpetrators, victims, and legislation. However, despite their growing prominence, limited attention has been paid to how these DDTs are constructed and how their underlying design choices affect research outcomes. This article presents a metadata analysis of 55 DDTs, focusing on how they document and make publicly available their definitional choices, raw-data inputs, and methodological procedures, as well as the implications for research and analysis. = show substantial variation across DDTs. Key issues identified in the analysis include the lack of retrievable definitions of terrorism, which requires users to infer inclusion criteria; the absence of clear identification of original data sources, which limits traceability to the underlying information; and the lack of publicly available coding and methodological documentation, which constrains transparency, interpretability and reproducibility of DDTs. Overall, while some databases demonstrate promising practices, the field remains a work in progress, characterised by persistent gaps in definitional clarity, the raw-data management, and methodological transparency and replicability.
Photo credit: David Pupăză on Unsplash
