Information on the nature of victimisation, the environment in which it occurs and profiles of both victims and perpetrators are critical to the provision of a complete picture for developing policy responses to curb crime. The Johannesburg victim survey provides a relatively accurate crime picture of the extent and, to a lesser degree, the nature of crime in the city. But the victim survey has clear limitations. Primarily, it tells analysts little about the causes of crime, where it occurs (although this could be eliminated through questionnaire development) and the circumstances of individual crimes. Furthermore, victim surveys by definition, provide only clues of the nature of offenders. Moreover, the shortcomings are more serious where accurate police data to supplement survey findings are limited (as occurs in many developing countries).

Despite some of the drawbacks outlined above, victim surveys can usefully determine appropriate crime prevention strategies. The most important policy impact of the victim survey in this regard, is its ability to allow prioritisation of particular categories of crime for policy interventions. With limited resources and capacity among local government and police agencies, prioritisation is critical.

The Greater Johannesburg: Safer Cities strategy has prioritised certain crime categories for intervention as a result of the survey. These include: burglary, mugging, theft of motor vehicles and violence against women. In addition, and by way of conclusion, the survey suggests that interventions are required in relation to four broader issues which cross-cut crime types. These are as follows:

1 Targeting the real victims of crime

The Johannesburg victim survey, as well as other studies,50 indicates that more affluent citizens are at greater risk of property crimes. The poorer sectors of society (in Johannesburg represented by African victims and those living largely in townships and informal settlements), however, are not only at risk of these crimes, but are also especially vulnerable to violent crime. These risks are likely to be more extreme when considering that, according to victims, police protection, service delivery and crime control capacity is lower and of poorer quality in township, informal settlement and inner city areas. Although they have a greater need to protect themselves as a result, people living in these parts of Johannesburg can rarely afford security measures other than fences, locks and burglar guards. Many have no physical protection measures to speak of. Against this profile of vulnerability to victimisation, unequal protection by the police, as well as widespread poverty, it is unlikely that those who are most at risk of being victimised by crime are able to protect themselves. Programmes therefore need to be initiated to address the vulnerability of these potential victims.

2 Improving the police service

Of those victims who reported their crimes to the police, the vast majority were dissatisfied with the police's response. These perceptions need to be addressed: not only do they mitigate against crime being reported, but they reduce the likelihood of victims and witnesses participating in investigations and thus any chance of securing a conviction. The survey results suggest that most victims' perceptions of the police could be dramatically improved by relatively small changes in the levels of local police services to the victims of crime. While these measures could relatively easily improve public perceptions of the police, confidence will ultimately depend on the ability of the police to solve serious (and particularly) violent crimes.

3 Confronting the fear of crime

Formulating an appropriate crime prevention strategy for the city requires an understanding of public perceptions around crime (and the fear of it). This is important given that perceptions (and fears) of crime often differ from the chances of being victimised. For example, Africans in Johannesburg are much more likely to be victims of car hijacking, yet public perception is that white middle class suburban residents are the primary target of this crime. Policies to reduce hijacking need not only to reduce the risk for Africans, but to address the fears of those living in the suburbs. Generally, certain crimes which involve violence or some form of violation of the person or the space (s)he occupies, cause heightened fears of crime, thus seriously affecting the quality of life. The key to successful crime prevention is therefore not only to bring down levels of crime per se, but also to address public perceptions.

4 Targeting repeat victims

Certain people (and places which they may frequent) suffer repeated incidents of crime. Fear of crime is especially high among such victims. Understanding repeat victimisation "directs attention to the victim, the time and possible perpetrator of a likely future crime, and helps target scarce resources cost effectively."51 Thus, planning for crime prevention often relies on the identification of a category of people who have repeatedly been the victims of certain crimes. In most cities, this category is comparatively small and so lends itself to targeted crime prevention. In Johannesburg, levels of repeat victimisation were high across all crime categories. This suggests that crime prevention based on repeat victimisation data may be less successful, given that the numbers and diversity of people involved are too large. However, this does not mean that repeat victimisation analyses cannot be used. Instead, it suggests that the data needs to be disaggregated and locally specific definitions formulated based on an understanding of crime patterns. For example, while there is a high level of repeat victimisation for mugging in the centre of Johannesburg, analysis of the data suggests that a large proportion takes place within specific precincts and that people over the age of 60 who live within the inner city are disproportionately victimised.

Strategies in response to repeat victimisation should adopt this kind of approach in Johannesburg. This, in turn, requires that the police monitor and record levels of repeat victimisation and, where possible, the details about those victims who are affected.

ENDNOTES

*Louw, Shaw and Camerer are researchers at the Institute for Security Studies' Safety and Governance Programme. Robertshaw is the co-ordinator of Greater Johannesburg Safer Cities.

Data for figures 13 to 17 is derived from South African Police Service, Crime Information Management Centre Quarterly Report 4/97, Pretoria, 1997.

J J M van Dijk, Criminal Victimisation and Victim Empowerment in an International Perspective, keynote address, opening session, Ninth International Symposium on Victimology, Amsterdam, The Netherlands, 25-29 August 1997.



GJMC, Greater Johannesburg Research Population Sheet, Greater Johannesburg Metropolitan Council, Johannesburg, 1997.



M Shaw, Towards Safer Cities: The South African Debate on Options for Urban Safety, ISS Monograph Series, 11, Institute for Security Studies, Halfway House, April 1997.



N Tilley & J Webb, Burglary Reduction: Findings from Safer Cities Schemes, Police Research Group, Crime Prevention Unit Series, Paper 51, HMSO, London, 1994.



UNISA, Rape in Johannesburg: The Results of an Ongoing Surveillance Project, press release compiled by the UNISA Health Psychology Unit and Centre for Peace Action, Johannesburg, April 1997. (Statistics updated in November 1997.)



Van Dijk, op. cit.



Ibid.



L Glanz, Crime in South Africa: Perceptions, Fear and Victimisation, Human Sciences Research Council, Pretoria, 1994.



Nedcor, The Nedcor Project on Crime, Violence and Investment, main report, Nedcor, Johannesburg, June 1996.



C M B Naude, et. al., Johannesburg (South Africa), in U Zvekic & A Alvazzi del Frate (eds.), Criminal Victimisation in the Developing World, Publication 55, UNICRI, Rome, 1995.



Zvekic & Alvazzi del Frate, ibid.; P Mayhew, Some Methodological Issues in Victimisation Surveys, in Crime Victims Surveys in Australia, conference proceedings, Criminal Justice Commission, Brisbane, 1995.



Ibid.



Van Dijk, op. cit.



To the extent that the results of a street survey can be compared with those of a household survey, this represents a slight decline in the five year victimisation level from the 68 per cent recorded in 1995 by the IC(V)S survey in Johannesburg; see C M B Naude, et. al., The Second International Crime (Victim) Survey in Johannesburg, UNISA, Pretoria, June 1996.



Van Dijk, op. cit.



Zvekic & Alvazzi del Frate, op. cit.



Van Dijk, op. cit.



Ibid.



CMIC, The Incidence of Serious Crime: January to September 1997, South African Police Service Crime Information Management Centre Quarterly Report 4/97, SAPS, Pretoria, 1997.



Naude, et. al., 1996, op. cit.



C Bridgeman & L Hobbs, Preventing Repeat Victimisation: The Police Officers' Guide, Home Office Police Research Group, London, 1997.



Workshop with Johannesburg SAPS officers, Johannesburg Central Police Station, December 1997.



Van Dijk, op. cit.



UNISA, op. cit.



Ibid.



S Mokwena, The Era of the Jackrollers: Contextualising the Rise of Youth Gangs in Soweto, Seminar Proceedings, Centre for the Study of Violence and Reconciliation, Johannesburg, 1991.



C Campbell, Social Identity and Violence in the Domestic and Political Spheres: A Gendered Common Denominator?, in L Glanz & A Spiegel (eds.), Violence and Family Life in Contemporary South Africa: Research and Policy Issues, Human Sciences Research Council, Pretoria, 1996.



Van Dijk, op. cit.; Zvekic & Alvazzi del Frate, op. cit.



Van Dijk, ibid.



L Rocha-Silva & I Stahmer, Research Relating to the Nature, Extent and Development of Alcohol/Drug-related Crime, unpublished paper, Human Sciences Research Council, Pretoria, 1996; A Louw & M Shaw, Stolen Opportunities: The Impact of Crime on South Africa's Poor, ISS Monograph Series, 14, Institute for Security Studies, Halfway House, July 1997.



These and other police statistics used below are drawn from CIMC, op. cit.



SAPS police areas do not match local authority boundaries. The Johannesburg police area is therefore not synonymous with the Greater Johannesburg Metropolitan Council (GJMC), which is the area covered by the crime victim survey. Soweto, for example, is part of the GJMC, but is a distinct police area, separate from the Johannesburg police area. Also, up to eight police stations which fall outside both the Johannesburg and Soweto police areas are part of the GJMC's jurisdiction.



S Oppler & A Louw, Declining Convictions: An Emerging Crisis, Nedcor/ISS Crime Index, Institute for Security Studies, Halfway House, October 1997.



Van Dijk, op. cit.



Zvekic & Alvazzi del Frate, op. cit.



Ibid.



P Mayhew & J J M van Dijk, Criminal Victimisation in Eleven Industrialised Countries, Wetenschappelijk Onderzoek-en Documentasiecentrum, Amsterdam, 1997.



Van Dijk, op. cit.



Ibid.



Workshop with Johannesburg SAPS, op. cit.



Bridgeman & Hobbs, op. cit.



Van Dijk, op. cit.



Mayhew & Van Dijk, op. cit.



Van Dijk, op. cit.



This finding may relate to the questionnaire design. It was not always clear whether respondents were aware of such services or whether they had actually made use of them. In addition, respondents were not always clear on the types of service which these agencies provide. These shortcomings in the questionnaire are being addressed in future surveys.



Van Dijk, op. cit.



A Alvazzi del Frate, Preventing Crime: Citizens' Experience Across the World, UNICRI, Rome, 1997.



Ibid.



Ibid.



Louw & Shaw, op. cit.



Bridgeman & Hobbs, op. cit.