Coding One-sided Violence from Media Reports

Please be invited to a Brownbag with Sabine Otto Friday 8th of June from 12:00- 13:00 in the War Room.

Please note: This page refers to an event that has already taken place.

Place: War Room

Event data are apowerful tool to analyse as detailed as possible the violent behaviour andinteractions of armed actors within and across wars. The inclusion of civilianfatalities into the most current created event datasets provides theopportunity to investigate systematically civilian victimizations in civilwars. However, these datasets are often criticized due to its heavy reliance onmedia data, using the argument that gathering information from newspaperscauses various types of biases. Despite these common critics I argue thatspecific challenges show up when coding one-sided violence - the purposeful useof violence against civilians by armed actors - from media reports.

In general,definitions of one-sided violence are complex. On the one side they comprisetheoretically demanding and empirically challenging concepts. On the other sidea certain set of information is required from the consulted news source inorder to code accurately this type of violence.

I will presentresults from my own coding work. Due to the substantial comparison of twodifferent types of event datasets certain challenges have become manifested.First, while the extent of selection bias is difficult to investigate it seemsthat individual cases of civilian victimization are much less reported thanevents in which a higher number of civilians were killed. Second, thedescription of events from media reports often does not provide sufficientinformation in order to distinguish intentional civilian targeting fromcivilians killed as a “by-product” of military interactions between at leasttwo armed actors. This context uncertainty, especially in combination withimprecise coding rules, results in biased estimation of civilian victims incivil wars.