The introduction of the Global Terrorism Database (GTD) nearly a decade ago sparked a revolution in terrorism studies. However, one major flaw in the database continues to plague GTD users. Data lost prior to digitalization, along with unsuccessful data recollection efforts, have left GTD without data on events that took place during the year 1993. The missing data prevents researchers from using the entirety of GTD’s annual range (1970-2014) to conduct reliable time-series analyses. Additionally, it has likely contributed to the formation of theories and claims on faulty empirical ground. To remedy the problem, we have collected data on 4,206 unique terror-attack incidents, with the aim of documenting the universe of 1993 terrorism events. This article showcases our 1993 dataset and illustrates the importance of terrorism events in 1993 for the development of conflicts in Israel, Afghanistan, Colombia, and India.
About the Dataset
Over the last 15 years, scholarship on terrorism and political violence has skyrocketed. Many studies, and numerous large-n analyses in particular, now rely on data from the National Consortium for the Study of Terrorism and Responses to Terrorism’s (START) Global Terrorism Database (GTD). Far more extensive than competing datasets that document individual attacks, GTD has facilitated the analysis of terrorism globally, by country, by organization, by campaign, and in various other important ways. One major flaw, however, prevents scholars from utilizing GTD to run reliable time-series analyses across the entirety of the Database’s annual range (1970-2014), and has likely contributed to the formation of theories and claims on faulty empirical ground. This problem stems from the loss of data for the year 1993. The data loss apparently occurred before the original data collection team at Pinkerton Global Intelligence Services (PGIS) transferred its hard copies to the START headquarters at the University of Maryland at College Park. In order to facilitate the analysis of lengthier and uninterrupted time-series data and in a broad effort to shed light on the events of the largely dark year of 1993, we have set out to identify the universe of attacks that occurred in 1993. Currently, we have amassed 4,206 attack events. Though our data collection enterprise remains ongoing, we view that our dataset as it stands is essential to facilitating reliable time-series analyses at the event and organizational levels, as well as preventing the formation of faulty theories supported by incomplete empirical analyses.
Kristen Ramos, University of Maryland, College Park
For more information, contact Kristen Ramos at: email@example.com