NCJ Number
215544
Journal
SA Crime Quarterly Issue: 14 Dated: December 2005 Pages: 29-36
Date Published
December 2005
Length
8 pages
Annotation
This article examines the reliability of two sources of data on murder in South Africa: the National Injury Mortality Surveillance System (NIMSS) and the Medical Research Council's (MRC's) Burden of Disease estimates.
Abstract
With inconsistencies found in various data the author recommends continuing to rely on police data on murders as a more reliable source. The MRC's estimate of the number of murders in South Africa in 2000/01 came from three sources. One source was an estimate of the number of all deaths in the country, which was derived from the Actuarial Society's model, ASSA2000, with some modifications. This produced an estimate of approximately 557,000 deaths from all causes. To calculate the number of deaths due to non-natural causes, the MRC estimated the proportion of all such deaths from a separate study. This study considered a sample of 12 percent of all death certificates issued between 1997 and 2001. Coincidentally, the study found that 12 percent of these deaths were from non-natural causes. Extrapolated to all deaths, the MRC estimated that 557,000 of these deaths were due to non-natural causes. The MRC then calculated the number of homicide deaths on the basis of NIMSS data, which are compiled annually from a survey of all bodies that come to approximately 35 mortuaries across the country. These data address the time, place, and cause of death as well as various demographic data. The problem with NIMSS data is that they are heavily biased toward urban areas. The result was to produce an estimate of murders that was much higher than police data showed for the same period. 3 figures and 10 notes