Capture-Recapture Estimation of War Deaths: Foundations, Assumptions, and Challenges

• Activity Number: 499
• Type: Topic Contributed
• Date/Time: Wednesday, August 3, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Nicholas P. Jewell*+ and Michael Spagat and Britta Jewell
• Companies: University of California at Berkeley and Royal Holloway College and Oxford University
• Keywords: Capture-recapture ; Multiple systems estimation ; log-linear models ; counting casualties

Abstract

Capture-recapture estimation has been used to count elusive wildlife and human populations since the late nineteenth century. The ideas were subsequently adapted for epidemiological and social science applications, often being used to count elusive human communities. The specific application of capture-recapture estimation to casualty counting in conflict settings is a more recent development that has now been employed in sufficient examples for the methods to be taken seriously and critiqued. Here we briefly consider some questions underlying this kind of application including: When, and under what conditions, can capture-recapture estimation provide accurate and useful counts of the number of conflict casualties? What is the minimum level of data validity that is required for the methods to be applicable? How do we best choose a model for capture-recapture estimation? How do we best choose a model for capture-recapture estimation? How do we best use multiple lists? How are the results and assumptions of a particular capture-recapture analysis best translated for political consumption and the media without undermining the credibility of the approach?