Who said we were terrorists?

Abstract

This article examines the reliability of "terrorism" classifications within the Global Terrorism Database (GTD) by looking at its inclusion criteria as well as filtering mechanisms for ensuring quality case inclusion. Using several descriptive analyses, this article examines how various measures within the GTD affect researcher's ability to adequately analyze various patterns or trends of offending. The underlying limitations of the data, namely data inclusion, the defining of terrorism, and inconsistency in labeling events are examined. It is concluded from the analyses that what is being defined asterrorismmatters downstream when examining the data, given that it is used to make inferences about groups, movements, or the efficacy of governmental policy. Since scholars often lack a proper "error" framework for assessing the quality of big data derived from open sources on terrorism, it makes it difficult for scholars to assess the quality of the data itself. As a result, researchers are encouraged to include an error framework within the GTD for academics to assess the quality of the data they are plugging into their models. Results, limitations, and recommendations are further expanded upon within.