Estimating Mortality in War-Time Iraq: A Controversial Survey with Important Lessons for Students

In teaching introductory quantitative methods in sociology, I have used a controversial survey of mortality in Iraq before and after the 2003 invasion to highlight to students the power of simple questionnaires, the role of ambiguity in statistics and the place of politics in the framing of statistical results. This brief report summarizes Roberts et al.'s (2004) estimate that the invasion of Iraq resulted in 98,000 (95% CI = 8000 - 194,000) deaths, as well as the intriguing reaction that the survey received in the press. Statistics teachers should find the Roberts et al. study to be an effective way to introduce students to more controversial - and political - aspects of statistical research.