Violent Deaths in Kosovo: Validation of Counts and a Dirty-War-Index Analysis

We examine three separate accounts of fatalities in the 1998-1999 Kosovo war that were produced using three very different methodologies.  The first is the sample survey estimate of Spiegel and Salama (2000).  The second applies capture-recapture theory to multiple lists of victims (Ball et al. 2002).  The last is a long project that attempts to exhaustively list every single victim (the Kosovo Memory Book).  The existence of three estimates using three completely different methods offers a rare opportunity to compare and attempt to cross-validate the a variety of methodologies for estimating and recording war fatalities.  We will compare the numbers both at the national level as well as broken down by time period, region, gender and age.  After completing the validation work we will then analyze the Kosovo Memory Book data using the Dirty War Index, a data-driven public health tool based on international humanitarian law that identifies rates of particularly undesirable or prohibited, i.e., "dirty", war outcomes inflicted on populations during armed conflict (Hicks and Spagat, 2008).