Victims of Crime Survey 2017-2018
South Africa, 2017 - 2018
Get MicrodataIdentification
ZAF_2017_VCS_v01_M
Victims of Crime Survey 2017-2018
| Name | Country code |
|---|---|
| South Africa | ZAF |
The primary aim of the VCS is to establish the prevalence of particular kinds of crime within a certain population. The objectives of the survey are to:
• Provide information about the dynamics of crime from the perspective of households and the victims of crime
• Explore public perceptions of the activities of the police, prosecutors, courts and correctional services in the prevention of crime and victimization
• Provide complimentary data on the level of crime within South Africa in addition to the statistics published annually by the South African Police Service.
Sample survey data [ssd]
Households and individuals
Version
v1: Edited, anonymised dataset for public access
2019
Version 1 of the dataset was downloaded from Statistics South Africa's website on the 20th October 2019
Scope
The VCS covers experiences of crime from the victim's perspective. The questionnaire includes questions at the household and individual level.
Household questions cover household information, demographics, education, economic activity and income.
Individual questions cover perceptions of crime, government and the justice system, experiences of all types of crime, perceptions of the community, and other similar crime-related information.
There is also a separate questionnaire called section 21 to 28. These questions are answered by a randomly selected household member.
Coverage
National coverage
The lowest level of geographic aggregation available is the metro, otherwise provinces are the lowest level.
Victims of crime 2017-2018 statistics are population estimates of the level of crime in South Africa. The target population of the survey consists of all private households in all nine provinces of South Africa, as well as residents in workers' hostels. The survey does not cover other collective living quarters such as students' hostels, old-age homes, hospitals, prisons and military barracks, and is therefore only representative of non-institutionalized and non-military persons or households in South Africa.
Producers and sponsors
| Name | Affiliation |
|---|---|
| Statistics South Africa | Government of South Africa |
Sampling
The sample design for the VOCS 2017/18 was based on a Master Sample (MS) originally designed for the Quarterly Labour Force Survey (QLFS) as a sampling frame. The MS is based on information collected during the 2001 Population Census conducted by Stats SA. The MS has been developed as a general-purpose household survey frame that can be used by all household-based surveys, irrespective of the sample size requirement of the survey. The VOCS 2017/18, like all other household-based surveys, uses an MS of primary sampling units (PSUs) which comprises census enumeration areas (EAs) that are drawn from across the country.
The sample for the VOCS 2017/18 used a stratified two-stage design with probability proportional to size (PPS) sampling of PSUs in the first stage, and sampling of dwelling units (DUs) with systematic sampling in the second stage. The sample was designed to be representative at provincial level. A self-weighting design at provincial level was used and MS stratification was divided into two levels. Primary stratification was defined by metropolitan and non-metropolitan geographic area type. During secondary stratification, the Census 2001 data was summarized at PSU level. The following variables were used for secondary stratification: household size, education, occupancy status, gender, industry and income.
A randomized probability proportional to size (RPPS) systematic sample of PSUs was drawn in each stratum, with the measure of size being the number of households in the PSU. A sample size of 3 080 PSUs was selected. In each selected PSU, a systematic sample of dwelling units was drawn. The number of DUs selected per PSU varies from PSU to PSU and depends on the inverse sampling ratios (ISR) of each PSU. The sample size for the VOCS 2017/18 is approximately 30 000 dwelling units.
The sampling weights for the data collected from the sampled households are constructed in such a manner that the responses could be properly expanded to represent the entire civilian population of South Africa.
The design weights, which are the inverse sampling rate (ISR) for the province, are assigned to each of the households in a province. The design weights for the VOCS 2017/18 sample were obtained by applying three adjustments to the base-weights. The first adjustment was applied to account for informal and/or growth PSUs. The second adjustment was applied to account for the EAs with less than 25 households, and the third was the non-response adjustment. In addition, there were two types of non-response adjustments: PSU non-response adjustment and household non-response adjustment. The PSU non-response adjustment was applied at the stratum level, whereas the household non-response adjustment was applied at the PSU level.
The final survey weights were constructed by calibrating the adjusted non-response design weights to the known population estimates as control totals using the integrated household weighting method. The lower bound for the calibrated weights was set equal to 50 when computing the calibrated weights with the StatMx software (Statistics Canada software). The VOCS 2017/18 sample will be weighted using the Population Estimates of mid-May 2015. The final weights were benchmarked to the known population estimates of 5-year age groups by population groups by gender at national level, and broad age groups at provincial level. The 5-year age groups are: 0-4, 5-9, 10-14, 15-19, 20-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-54, 55-59, 60-64, 65-69,70-74, and 75 and older. The provincial level age groups are 0-14, 15-34, 35-64, and 65 years and older. The calibrated weights are constructed in such a way that all persons in a household would have the same final weight.
Survey instrument
The VOCS 2017/18 questionnaire was developed based on the questionnaires used in the International Crime Victim Survey (ICVS) and previous VOCSs (both conducted by ISS and Stats SA), with modifications in some instances. The Stats SA questionnaire design standard for household surveys was also used as a normative reference. In order to minimize fieldworker and capturing errors, the questionnaire was mainly pre-coded. During the process of questionnaire design, workshops were held. Participants included representatives from various government departments, especially from the safety and security cluster, non-governmental organizations, researchers, and other stakeholders. The questionnaire was tested behind the glass and within selected households in Gauteng. Before finalization, the draft version of the questionnaire was presented nationally and internationally to gain input. There is a household and an individual questionnaire, as well a subset of questions called sections 21 to 28.
Sections 21-28 required that an individual be randomly selected from the household to respond to questions classified as individual crimes. The methodology used was to select a person 16 years or older, whose birthday was the first to follow the survey date.
Data collection
| Start | End |
|---|---|
| 2017-04 | 2018-03 |
| Name | Affiliation |
|---|---|
| Statistics South Africa | Government of South Africa |
Prior to 2017/18, VOCS respondents were asked about their crime-related experiences in the previous calendar year, but the VOCS changed to a Continuous Data Collection (CDC) method. In this data collection method, respondents were interviewed on a rolling basis over the course of a year and asked about crime experienced in the 12 months prior to the interview. As a result of this, the victimization experiences reported by respondents interviewed in a period of 12 months relate to a broader span of 23 months.
This present VCS, 2017/2018, is the last before VCS became incorporated into a broader survey called the GPSJS. The change to the surveys will likely cause some comparability issues going forward beyond 2018.
Data processing
In general, editing (i.e. invalid or inconsistent responses) and imputation (i.e. blanks within the questionnaire) were used for item non-response. The eligible households in the sampled dwellings can be divided into two response categories: respondents and non-respondents. Weight adjustment is applied to account for the non-respondent households (e.g. refusal, non-contact).
Data appraisal
Comparability over the series:
Prior to 2014/2015, VOCS respondents were asked about their crime-related experiences in the previous calendar year, but since 2014/15 VCS changed to a Continuous Data Collection (CDC) method. In this data collection method, respondents were interviewed on a rolling basis over the course of a year and asked about crime experienced in the 12 months prior to the interview. As a result of this, the victimization experiences reported by respondents interviewed in a period of 12 months relate to a broader span of 23 months.
Prior to 2017/18, VOCS respondents were asked about their crime-related experiences in the previous calendar year, but the VOCS changed to a Continuous Data Collection (CDC) method. In this data collection method, respondents were interviewed on a rolling basis over the course of a year and asked about crime experienced in the 12 months prior to the interview. As a result of this, the victimization experiences reported by respondents interviewed in a period of 12 months relate to a broader span of 23 months.
This present VCS, 2017/2018, is the last before VCS became incorporated into a broader survey called the GPSJS. The change to the surveys will likely cause some comparability issues going forward beyond 2018.
Metadata:
There is an error in the SSA published metadata, which incorrectly states that the survey was designed with 3080 PSUs. The survey was designed with 3324 PSUs.
Data Access
| Name | Affiliation | URL | |
|---|---|---|---|
| DataFirst | University of Cape Town | http://www.datafirst.uct.ac.za | info@data1st.org |
Public use files, available to all
Statistics South Africa. Victims of crime survey 2017-2018[dataset]. Version 1. Pretoria: Statistics South Africa [producer], 2018. Cape Town: DataFirst [distributor], 2019. DOI: https://doi.org/10.25828/8nb3-bg33
Contacts
| Name | Affiliation | URL | |
|---|---|---|---|
| DataFirst Helpdesk | University of Cape Town | support@data1st.org | http://support.data1st.org/ |
Metadata production
DDI_ZAF_2017_VCS_v01_M
| Name | Affiliation | Role |
|---|---|---|
| DataFirst | University of Cape Town | Metadata Producer |
2020-03-29
Metadata version
Version 01: This metadata is identical to the same survey metadata (zaf-statssa-vcs-2017-2018-v1) available in the DataFirst website except Document ID and Study ID.