What the crime statistics are – and aren’t
Every year in September, the SAPS releases statistics on all the crimes reported to them or
revealed by their actions in the preceding reporting year running from 1 April to 31 March.
They provide the number of recorded incidents of each of about 30 crime types for each of
the country’s about 1130 police stations, as well as the sum for each province and for the
country as a whole, and in the past has also provided rates per 100,000 population
provincially and nationally. These stats affect the hundreds of thousands of people who
work in the Justice, Crime Prevention and Security cluster, the functioning of numerous
other government departments, NGOs, and the life and liberty of potentially every person in
the country or wishing to enter it. They are public property. It is important that they be
reliable, useful, and intelligible.
Unfortunately, the apparent precision of crime figures masks a messier reality. Many factors
get in the way between an incident of crime and its reflection in official stats, such that it
can be difficult to determine whether an observed pattern in the stats reflects the real
pattern in crime or is a feature of other social, political or institutional mechanisms. The
stats should be thought of not as an accurate map of reality but instead as one among the
tools for understanding what is happening with crime. They are often supplemented by
crime victimisation studies (which can reveal the extent and some reasons for
underreporting) and other kinds of research.