The World Values Survey (WVS) is a worldwide investigation of socio-cultural and political change and explores values relating to family, gender, work, politics, economics, religion and leisure time. The South African leg of this survey is administered by the Centre for International and Comparative Politics, University of Stellenbosch, in partnership with Markinor. The South African WVS has been conducted in 1982, 1990, 1995, 2001 and 2006, offering the opportunity to assess changing values over time.
Implementation of this research around the world allows for global crosscultural analysis.
The World Values Survey aims to attain a broad understanding of socio-political trends (i.e. perceptions, behaviour and expectations) among adults across the world.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
The sample was distributed as follows: 60% metropolitan (large cities with populations of 250 000+); 40% non-metropolitan (including cities, large towns, small towns, villages and rural areas)
The sample included adults 16 years+ in South Africa
Producers and sponsors
Centre for International and Comparative Politics
World Values Survey Association
The sample had to be representative of urban as well as rural populations. Roughly the distribution was as follows:
- South Africa: 60% metropolitan (large cities with populations of 250 000+); 40% non-metropolitan (including cities, large towns, small towns, villages and rural areas).
A standard form of sampling instructions was sent to each agency to ensure uniformity in the sampling procedure. Markinor stratified the samples for each country by region, sex and community size. To this end, statistics and figures that were supplied to us by the agencies were used. However, we requested the agencies to revise these where necessary or where alternatives would be more effective. The agencies then supplied the street names for the urban starting points, and made suggestions for sampling procedures in rural areas where neither maps nor street names were available. From sample-point level, the respondent selection was done randomly according to a selection grid used by Markinor (the first two pages of the master questionnaire).
Substitution was permitted after three unsuccessful calls. Six interviews were conducted at each sample point. The male/female split was 50/50. The urban sample included all community sizes greater than 500 and the rural sample all community sizes less than 500. This is the definition of urban and rural used in South Africa.
Remarks about sampling:
-Final numbers of clusters or sampling points: 500
-Sample unit from office sampling: Street Names
Weighting statistics were based on the AMPS (All Media and Products Survey) 200 figures. The ultimate source of these figures is the South African General Population Census, 1997. The data are kept by Statistics South Africa (www.statssa.com).
Dates of Data Collection
Data Collection Mode
Data Collection Notes
Interviewers were paid according to performance. The interviewer approached was through calls made at different times of day and calls made on different days of week. The minimum number of re-calls required were 3 and 20% of the interviews were back-checked.
The WVS questionnaire was translated from the English questionnaire by a specialist translator The translated questionnaire was pre-tested. The pre-tests were part of the general pilots. In total 20 pilots were conducted. The English questionnaire from the University of Michigan was used to make the WVS. Extra questions were added at the end of the questionnaire. Also, country specific questions were included at the end of the questionnaire, just before the demographics.The sample was designed to be representative of the entire adult population, i.e. 18
years and older, of your country. The lower age cut-off for the sample was 16 and there was not any upper age cut-off for the sample.
Some measures of coding reliability were employed. Each questionnaire is coded against the coding frame. A minimum of 10% of each coders work is checked to ensure consistency in interpretation. If any discrepancies in interpretation are World Values Survey (1999-2004) - South Africa 2001 v.2015.04.18 discovered, a 100% check is carried out on that particular coders work. Errors were corrected individually and automatically.
Estimates of Sampling Error
The error margins for this survey can be calculated by taking the following factors into account:
- all samples were random (as opposed to quota-controlled)
- the sample size per country (or segment being analysed)
- the substitution rate per country (or segment being analysed) - the rates were recorded on CARD 1; col. 805 of the questionnaire. From the substitution rate, the response rate can be calculated.
- Public use files, accessible to all
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.