The Institute for Security Studies has conducted crime victim surveys in four of South Africa’s metropolitan areas, namely Johannesburg (July 1997), Durban (December 1997), Cape Town (February 1998) and Pretoria (April 1998). By using information derived from the victims of crime themselves, these surveys were able to present a view of crime independent of and often more detailed than that provided by aggregated police statistics. These surveys were useful, therefore, as supplements to highlight some of the shadowy areas of the picture painted by the official statistics on crime in these cities.

The same information was sought about crime and policing in the rural areas. Like the urban surveys, a rural victim survey would enable new information to be gathered that could provide detail on:

the extent of crime;

the nature of certain types of crime;

the risk profile of inhabitants of rural areas;

the levels of fear and insecurity;

public perceptions of police service delivery; and

what people in rural areas thought would be appropriate interventions to reduce crime.1

Accessing such information in a single research process is the strength of victim surveys conducted the world over. But, the methodology is limited by a number of factors, chief among which are biases in the data resulting from:

the sensitivity of respondents towards discussing sometimes traumatic incidents;



the ability of respondents to recall and articulate details of their victimisation accurately; and



varying interpretations of what constitutes a crime — respondents may not realise that a particular incident constitutes a ‘real crime’, that is, one that is covered in the survey. This is sometimes the case with spousal abuse that the victim, living with the offender, does not perceive as assault. However, it is more likely in terms of incidents deemed trivial by the respondent, like petty theft or minor vandalism.

This survey, of course, has also been limited by these factors. In addition, as parental consent and specialised and expensive interview techniques are required to interview minors, those aged under 18 years were excluded from the sample.

Although care was taken to ensure that the sample survey was as representative of the African rural adult population as possible, cost constraints meant that comparatively few respondents — 756 — could be interviewed. Therefore, the results of the survey should not be read as definitive, but should rather be seen as broadly indicative of crime and policing in the rural areas.

To ensure the survey’s balance and more detail on policing issues, a range of police officers at provincial, area and station level management in the Eastern Cape, KwaZulu-Natal, Northern Province and the North-West were interviewed between May and July 1998 and in October 1999.

The rural victims of crime survey

The survey of victims of crime in South Africa’s rural areas was conducted in June and July 1998.

The survey was specifically and deliberately focused on African settlements in six of South Africa’s predominantly rural provinces — the Eastern Cape, KwaZulu-Natal, Mpumalanga, Northern Province, North-West and the Free State — as these areas are inhabited by the majority of South Africa’s rural population.

A total of 756 adults — randomly selected after an initial segmentation, which spread the sample over 40 rural magisterial districts in the six provinces and categorised it in terms of gender, age and household status — were sampled by means of face-to-face interviews designed to facilitate completion of a standardised questionnaire.

A total of 430 individuals (56.9% of the sample) indicated that they had been the victim of at least one crime in the period between January 1993 and June 1998 (see table 2). Demographic details of the sample of 756 individuals are presented in tables 3 to 7.

As was expected, the realised sample population indicated that a relatively high number of adults (30%) were not economically active and that a very limited number of economically active people were able to generate an income. High rates of unemployment and meant that only 25.7% of the sample had access to income-generating employment (see table 7).

Just more than half of those employed (51%) were fortunate enough to have full-time formal employment, while 27% were employed on a part-time basis and 22% in the informal sector, implying more sporadic, if not lower income. Therefore, given the profile outlined in table 1 (see p 8), it could not be expected that many of the employed would earn more than R500 per month.

Many households would therefore be dependent on R550 per month paid to the pensioners by the state’s old-age pension fund, and by cash or in-kind support provided by migrant members of the household or other kinship networks, and subsistence farming.

Nevertheless, the size of households was generally large: 211 (27.9%) of the respondents indicated that they lived in households with four or less occupants, 437 (57.8%) said they lived in households with between five and eight occupants and the remaining 108 (14.2%) indicated that their households consisted of nine or more people.

As was expected, migrant labour played a large part in the lives of the sample population. A total of 448 (59.3%) respondents indicated that their households consisted of one or more migrant workers. Most of these households (78%) had one or two migrant workers living there, while 22% had between three and six migrant workers.

Further, a comparatively high number of respondents indicated that they either owned or had access to land that could be used for subsistence farming. In this respect, 376 respondents (49.7%) indicated that they owned their own land, rented land or had access to common land. This reflects a higher proportion than that noted by May who observed that "over one-third of rural households" engage in agricultural production, "making it the third most important livelihood tactic used in the rural areas after remittances and wages from low-skilled jobs."2

In summary, the sample for the victim survey was drawn from the deep rural areas of South Africa and in terms of its defining characteristics, broadly matches that of the greater African rural adult population in the country. As indicated above, the realised sample were generally poor, undereducated and underemployed, lived in large households and were largely dependent for survival on access to arable land and income from migrant labour, remittances, kinship networks and sporadic employment.

Thus, even though the sample size was small, the experience of the respondents, and especially their victimisation, may be viewed as typical of the poorest rural areas of South Africa.

Notes

For more detail on these issues, see A Louw, Crime in Pretoria: Results of a city victim survey, Institute for Security Studies, Halfway House, and Idasa, Pretoria, 1998.



J May, Poverty and inequality in South Africa, report prepared for the office of the executive deputy president and the inter-ministerial committee for poverty and inequality, 13 May 1998.