Assessing the role of social media and digital technology in violence reporting

Abstract

The use of social media and digital technologies has radically changed the way that information about violence is captured, reported, analysed and acted upon. People’s use of social media played a significant role in the Egyptian revolution, post-election violence in Kenya, and drug-cartel violence in Mexico. Social media can be used to provide humanitarian agencies, policy makers and academics seeking to understand and respond to violent crises with data unavailable from other sources. After an initial period of uncritical optimism regarding the potential of social media and digital technologies there is now however a growing recognition that they come with new practical, ethical and methodological limitations. Indeed social media content is often the target of conscious distortions, manipulations, or censorship by a range of actors. Bias of several kinds can significantly distort social media data and reduce its representativeness. This paper assesses the role of social media and digital technology (SMDT) in the reporting of violent events, and evaluates its relative strengths and weaknesses as compared to other means available. It seeks to understand how SMDT data is collated, how reliable the data is, and what practical and ethical issues are associated with its collection and use. We start by situating the application of these technologies within the wider discussion of the use of Information and Communication Technologies for Development (ICT4D) and the sub-field of real-time data for development (RTD4D) before presenting the different types of SMDT data collection instruments and methods. We then assess the reliability of SMDT for the reporting of violent events, identifying potential factors of biases in the data, such as geographical coverage, demographic and socio-economic factors, or biases resulting from the nature and configuration of violent events. We also look at whether such data can accurately capture underlying dynamics of violent events. Finally, we look at the practical and ethical challenges associated with the collection of SMDT data on violent events.