INTRODUCTION

---

Published in Crime in Pretoria
Results of a City Victim Survey

 

Victim surveys have been conducted across the world for the past thirty years as a means of supplementing police statistics on crime. Information about crime, given a range of factors, is not always accessible or accurate. Compiling official statistics depends on both the public to report crime and the police to record the details once they do. But the nature of criminal events themselves often mitigates against reporting. Some incidents are not regarded as significant enough to warrant the effort; others are too sensitive to disclose details to strangers behind a desk in a police charge office.

Since reducing crime requires an understanding of the extent and nature of the problem, accurate information is critical. Victim surveys are the most important source of information for filling the gaps left by official crime statistics. They present an independent source of information about crime based on questions asked of a representative sample of the population in a specific geographic area. As such, these surveys record the experience of crime from the unique perspective of the victim: not only are crimes which are not recorded by the police captured in the survey, but so too are opinions of victims themselves. Information of this sort — which is critical for crime prevention — is limited in official records since the criminal justice process requires the collation of the offender’s rather than the victim’s details.

The inadequacies of police crime information are a reality in most societies. In South Africa, particular problems — relating to poor police-community relations, as well as more technical issues of data integrity and accessibility — exacerbate the situation,¹ raising the importance of victim survey information. The Pretoria victim survey is the fourth city survey conducted by the Institute for Security Studies. Comparable surveys have been carried out in metropolitan areas of Johannesburg (July 1997), Durban (December 1997) and Cape Town (February 1998). In the case of both the Johannesburg and Pretoria surveys, the results have been channelled directly into initiatives of the respective metropolitan councils to develop local crime prevention strategies.²

This report was prepared for the Greater Pretoria Metropolitan Council, as part of the research process which will culminate in the development of a crime prevention strategy for the metropolitan area. In order to assist the Institute for a Democratic South Africa’s (Idasa) Community Safety Unit in Pretoria to develop the strategy, this report outlines the extent and nature of crime in Pretoria, who is most at risk of particular crimes, levels of anxiety about crime, and the perceptions of victims and of the public more generally, about the police and other initiatives to support crime victims. As such, the data provide a critical balance between the reality of victimisation and public perception, which is essential for formulating targeted and successful local crime prevention strategies.

Victim surveys provide several types of information which are key to the understanding of crime and to developing localised responses to crime reduction:

Determining the extent of crime: Survey data reflect those crimes that are not recorded by the police. Conducted at regular intervals over an extended period of time, this enables an assessment of the extent to which changes in crime levels — as recorded by the police — are real or a function of changing reporting tendencies.



Identifying who is most at risk of particular crimes: Because victim surveys gather information from both victims and non-victims from a representative sample of the population in any area, the data can be used to determine whether particular people are more at risk of victimisation than others. This information (which cannot be obtained from police statistics) is essential to enable the prioritisation of particular crime categories and vulnerable groups for attention.



Understanding the nature of particular crime types, especially those that are poorly recorded in official crime statistics: Victim surveys provide useful details (which are difficult to glean from the South African Police Service (SAPS) databases) on where and when crimes are most likely to occur, and circumstances which characterise certain crimes. Examples include the relationship between victim and offender, weapons used and degree of violence and injury sustained, association with drugs and alcohol, and what the victim was doing when the crime occurred. This information is particularly relevant for crimes such as mugging and assault which are seldom reported to the police.



Measuring levels of anxiety about crime: Feelings of insecurity have social, economic and political consequences for society. By recording the perceptions of both victims and non-victims, and asking specific questions about fear, the surveys illustrate the extent and nature of fear of crime.



Determining public perceptions of police effectiveness and service delivery: Victim surveys provide useful mechanisms for recording the opinions of the public, and more importantly, those people that have had contact with the police, about their performance.



Establishing the opinions of victims and others about appropriate interventions: Crime prevention and victim support are relatively new fields in South Africa. Accurate information about what victims would prefer in this regard, is thus particularly relevant.

These strengths have been recognised for some time abroad, and the Pretoria survey is based on a questionnaire similar to that used by the International Crime Victim Survey (ICVS) in more than fifty countries. The city surveys conducted by the ISS, however, differ from this model in their application as street surveys rather than household surveys.

While the city surveys provide invaluable information about the crime situation in particular localities, certain limitations which relate to the street survey methodology, as well as to victim surveys in general, should be acknowledged.

Studies of methodological limitations related to memory generally conclude that the biases in the data result in an undercount of crime (rather than an overcount as is often suspected).⁶ Trivial crimes (such as minor thefts and vandalism) are most likely to be forgotten in an interview, while more serious crimes are usually well remembered and may even be overcounted, as more important events tend to be "pulled forward in time."⁷ This is relevant to the ISS city victim surveys, which sought information on a selection of serious crimes, and is probably one of the reasons for the high murder rates recorded by the city surveys. This is one example of why comparisons between victim surveys and police statistics should be approached with caution. Others are discussed next.

The data on crime provided by victim surveys should be seen as complementary to those recorded by the police. While the two sets of information measure various aspects of crime which are not always directly comparable, both provide important perspectives on the nature of crime.

Comparing police and victim survey data is complicated by:

Different boundaries: Local government boundaries generally do not coincide with police station boundaries in South Africa, which makes comparisons across specific geographical areas difficult.



Varying definitions of crime types: While definitions of incidents vary, interviewers applied broad police classifications for the following crime types covered in the city victim survey: burglary, robbery, vehicle theft, car hijacking, assault and murder. In the case of ‘sexual assault’ and ‘sexual harassment’, definitions varied considerably from those used by the police.

Definitional issues are less of a problem than they may seem at the outset. In the UK, where a national victim survey has been carried out six times since 1982 (covering nearly 16 500 people in the 1996 survey), analyses have shown that "for many offences police and public definitions coincide." Disagreement is most likely to occur in respect of less serious incidents, such as those whose definition depends on moral judgements (such as some sexual incidents), and those for which there is most discretion for taking legal action.⁸ Victim surveys have generally been found to count a broader set of incidents than police statistics, although this is less likely in the Pretoria survey, which was limited to eight types of serious crime.



Unreported and unrecorded crime: Not all crimes are reported by the public to the police. In Pretoria, for example, 27 per cent of all crimes recorded by the victim survey were not reported. This proportion, however, varies across crime types: the police are notified of almost all car hijackings, for example, but few muggings. Because victim surveys capture incidents which may not have been reported to the authorities, crime levels for some categories of crime are likely to be higher than in police statistics.

Bearing in mind the difficulties of victim surveys generally and the need for gathering information which complements police crime statistics, the methodology of the city surveys required careful consideration. The Pretoria victim survey was initiated to gain an understanding of the levels of crime and violence in different communities within the Greater Pretoria Metropolitan Council (GPMC). This entailed:

the collection, collation and synthesis of all existing and available research and information on generic types of communities (including demographic statistics) and residential areas located within the Pretoria metropolitan area;



the development of broad profiles of the defined constituents, including some select demographic statistics such as age, gender, vocation and place of work and study;



the identification of the ratio of victims to non-victims of crime by race and residential areas; and



the development of the necessary sampling formulae to accommodate this.

The administration of interviews in an uncontrolled environment, such as on the street, at nodal interchanges and in other public spaces such as places of welfare, is unique to the street survey methodology. This is a significant departure from typical crime surveys that rely either on face-to-face interviews in a controlled environment, or on postal surveys. Such surveys have advantages, but are both expensive and time consuming to administer. Given the financial constraints of the project, it was decided to conduct a survey based on a street sample instead. This methodology had been applied in similar studies conducted by the ISS and DRA Development in Johannesburg, Durban and Cape Town.⁹ This resulted in the calculated application of a more refined research methodology as the surveys proceeded in each metropolis. Careful planning was required to establish the sample mix and sample points. This was done in a number of stages.

From the outset, it was decided that the survey results should be analysed categorically by race, gender, age and residential settlement type. A departure from the Durban and Johannesburg surveys included the separation of the grid questionnaire (which provided a profile of both victims and non-victims in the metropolis by age, gender, race, residential area, crime type, victimisation level, as well as repeat victimisation and multiple victimisation) for determining victim ratios (no sexual crimes were included in the grid).

In this scan survey, 2 548 people were approached in forty different areas across the Pretoria metropolitan area. In each sampling node, approximately 64 respondents were approached in four age categories, half of which were male and half female. This provided the minimum required subsample in each age and gender category. Prior to approaching the respondents, no distinction was made between victims and non-victims. The only screening criterion was age.

In the more detailed experience and perception of crime survey, 2 064 people were approached. In this survey, 45 sampling nodes were selected. Variables for analysis were applied similar to those in the grid survey, namely race, gender, age and type of residential settlement.

Stage 2

It was decided that, although the research could not be representative of individual substructures within the GPMC, it should represent the various area types within the metropolis. Since there are very few areas in Pretoria that do not have a residential population, all geographic localities were included in the sample framework.

To target the correct types of sample areas, at least one sample point was selected to represent each residential generic typology. In order to ensure that the selected sample points covered the entire metropolis, and that all residential generic typologies were represented, a total of forty primary sample points were identified for the grid survey. Another 45 primary sample points were identified for the detailed survey. Primary sample points were suburbs, townships, the inner city, or informal settlements.

Within each primary sample point, a number of secondary points were selected where interviews would be conducted. At least five secondary points were identified in each primary sample point. Secondary points included:

shopping and recreation centres (shopping malls, flea markets, corner cafés and stores, plazas, spazas, shebeens, sports grounds, public parks and major streets);



transport nodes (taxi ranks, bus stops, railway stations and parking lots);



education centres (schools, universities and technikons);



health and welfare centres (pension pay-out points, civic buildings, hospitals); and



residential areas (private homes, old-age centres, apartment blocks and informal settlements).

Within each secondary sample point, an interview referral point had to be identified. This achieved two objectives:

to allow the identification and selection of respondents; and


toprovide an appropriate interview environment.

Finally, in order to limit any selection bias, the number of interviews conducted at each secondary sample point was restricted by both time-delay and locality factors.

The questionnaire was refined through the experiences and results of the Johannesburg, Durban and Cape Town city victim surveys. The pilot study thus functioned more as a training tool for refining interview techniques, measuring the length of the questionnaire and establishing a respondent sampling technique. During the pilot survey, forty questionnaires were administered. The length of the Pretoria survey was marginally longer at twenty minutes than those in the other cities, due to the addition of gang, alcohol and drug-related questions. Respondents clearly wanted to speak about their experiences, and the longer than standard questionnaire did not reduce the response rate. On the contrary, fieldworkers often had difficulty in terminating interviews.

As with previous findings, the Pretoria survey confirmed that male enumerators could only interview men, while women could be used to interview both men and women. Of interest was the fact that cross-race interviews were undertaken with relative ease. This suggested that the issue of crime and violence transcends racial inhibitions. Despite this finding, however, and as a precaution, the race of the interviewers in the final survey matched that of respondents.

The successive city studies found that the success rate, both in terms of selecting respondents and the time taken to execute each subsample, was much higher among the enumerators dressed in an identifiable ‘uniform’. The field team therefore wore a ‘t-shirt’ and cap and carried a bag — all bearing the logo’s of DRA Development, the survey company, and the Institute for Security Studies.

An important component of the pilot survey was the construction of a respondent selection technique that would allow the enumerators to obtain their necessary quotas, while ensuring that the selection process was both random and rigorously implemented. During the actual fieldwork stage, dummy respondents were used to ensure that these standards were being adhered to. As a further check, enumerators had to calculate the rate of flow of respondents (who roughly fell in the sample unit that was being targeted) at that specific referral point.

Interviews were conducted at specific times throughout the day to ensure that the views of a broad cross-section of people, including those coming from (and going to) work, and students were captured. In some cases where the sample target of older people (over 60 years) was not realised on the street, interviews were conducted indoors.

Fieldworkers were selected from a variety of sources, although all were Pretoria residents. A half-day training course was followed by a number of training interviews in a controlled environment and in the field. In addition, the field teams were subjected to a one-day workshop on how to empathise with respondents who were victims, and cope with the potential stress of being involved in such a study. A similar exercise was undertaken during the debriefing session that followed the research process.

In the scan survey, a total of 2 547 interviews were undertaken between 27 April and 17 May 1998 (Table 1).

	Sample	Victims	Non-Victims
Gender
Male Female	1 431 1 116	801 591	630 525
Race
African White Asian Coloured	1 532 871 57 87	893 432 33 34	639 439 24 53
Age
15-24 25-39 40-59 50+	460 1 032 758 297	251 594 411 36	209 438 347 261
TOTAL	2 547	1 392	1 155