The 2004 South African Social Attitudes Survey (the SASAS) is second survey (the first SASAS survey was conducted in 2003) conducted by the Human Sciences Research Council (HSRC) that will collect information on people's attitudes, beliefs and behaviour patterns in all nine provinces. The SASAS is intended to be a time series, run every year. It is designed to measure contemporary social attitudes, values and behaviour patterns and how they change over time. Such findings are of interest to government departments anxious to have more information about people's own assessments of social and political issues in their country. Subjects covered in the questionnaire include attitudes towards democracy and governance, poverty and social identity, amongst many other things. We expect that the data gathered will be used extensively by national and provincial governments, policy analysts, think tanks, politicians, journalists and academics, as well as being of interest to the general public. The data are also contributions to social history, to allow analysts in the future to discover what people thought and felt about the major social issues of today. The SASAS will therefore provide a unique long-term account of the social fabric of modern South Africa, of how its changing political and institutional structures interact over time with changing social attitudes and values. With data from the SASAS, people will be able to make detailed comparisons between provinces (or groups) on a wide range of social issues.
The primary objective of SASAS is to design, develop and implement a conceptually and methodologically robust study of changing social attitudes and values in South Africa to be able to carefully and consistently monitor and explain changes in attitudes amongst various socio-demographic groupings. The SASAS explores a wide range of value changes, including the distribution and shape of racial attitudes and aspirations, attitudes towards democratic and constitutional issues, and the redistribution of resources and power. Moreover, there is also an explicit interest in mapping changing attitudes towards some of the moral issues that confront and are fiercely debated in South Africa, such as gender issues, AIDS, crime and punishment, governance, and service delivery. The SASAS is intended to provide a unique long-term account of the social fabric of modern South Africa, and of how its changing political and institutional structures interact over time with changing social attitudes and values.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
The units of analysis in the study are households and individuals
Version 01: Edited, anonymised dataset for licensed distribution
The thematic content of the survey includes democracy, identity, public services, social values, crime, voting, demographics, families and family authority
The lowest level of geographic aggregation covered by the data is province
The population under investigation includes adults aged 16 and older in private households in South Africa
Producers and sponsors
Human Sciences Research Council (HSRC)
The South African Social Attitudes Survey has been designed to yield a representative sample of adults aged 16 and older. The sampling frame for the survey is the Human Sciences Research Council’s (HSRC) Master Sample, which was designed in 2002 and consists of 1 000 primary sampling units (PSUs). The 2001 population census enumerator areas (EAs) were used as PSUs. These PSUs were drawn, with probability proportional to size, from a pre-census 2001 list of EAs provided by Statistics South Africa.
The Master Sample excludes special institutions (such as hospitals, military camps, old age homes, school and university hostels), recreational areas, industrial areas and vacant EAs. It therefore focuses on dwelling units or visiting points as secondary sampling units, which have been defined as ‘separate (non-vacant) residential stands, addresses, structures, flats, homesteads, etc.’
As the basis of the 2004 SASAS round of interviewing, a sub-sample of 500 PSUs was drawn from the HSRC’s Master Sample. Three explicit stratification variables were used, namely province, geographic type and majority population group.
Within each stratum, the allocated number of PSUs was drawn using proportional to size probability sampling. In each of these drawn PSUs, two clusters of 7 dwelling units each were drawn. These 14 dwelling units in each drawn PSU were systematically grouped into 2 subsamples of seven, to give the two SASAS samples.
Number of units: Questionnaire 1: 2 497 cases realised from 3 500 addresses; questionnaire 2: 2 483 cases realised from 3 500 addresses; combined : 4980 cases
Data for the combined dataset is available with the individual weight (for analysis at individual level) as well as household weight (when analysis are done at household level).
The data were weighted to take account of the fact that not all the units covered in the survey had the same probability of selection. The weighting reflected the relative selection probabilities of the individual at the three main stages of selection: visiting point (address), household and individual.
Visiting points in the Northern Cape and KwaZulu-Natal were oversampled, because of the small population size in the former and a desire to ensure a large enough Indian/Asian sample in the latter. Because of this, weights had to be applied to compensate for the greater probability of being selected.
The resulting weight is called 'benchwgt' in the half samples and 'combiwgt' in the full sample. The sample is weighted to represent the South African population older than 16 years.
All analysis should be done on weighted data.
Dates of Data Collection
Data Collection Mode
To accommodate the wide variety of topics that was included in the 2004 survey, two questionnaires were administered simultaneously. Apart from the standard set of demographic and background variables, each version of the questionnaire contained a harmonised core module that will remain constant from round to round, with the aim of monitoring change and continuity in a variety of socio-economic and socio-political variables. In addition, a number of themes will be accommodated on a rotational basis. This rotating element of the survey consists of two or more topic-specific modules in each round of interviewing and is directed at measuring a range of policy and academic concerns and issues that require more detailed examination at a specific point in time than the multi-topic core module would permit.
In respect of the two SASAS questionnaires, the questions contained in the core module (demographics and core thematic issues) were asked of all 7 000 respondents, while the remaining rotating
modules were asked of a half sample of approximately 3 500 respondents each. The two different versions of the questionnaire were administered concurrently in each of the chosen sampling areas. Fieldworkers were required to complete a paper-based instrument while interviews were conducted face-to-face. Questions for the core module were asked of both samples (3 500 respondents each – 7 000) of which 5 583 were realised.
The International Social Survey Programme (ISSP) is run by a group of research organisations, each of which undertakes to field annually an agreed module of questions on a chosen topic area. SASAS 2003 represents the formalisation of South Africa's inclusion in the ISSP, the intention being to include the module in one of the SASAS questionnaires in each round of interviewing. Each module is chosen for
repetition at intervals to allow comparisons both between countries (membership currently stands at 40) and over time. In 2003, the chosen subject was national identity, and the module was carried in version 2 of the questionnaire (Qs.152-203).
The standard questionnaires dealt with democracy, identity, public services, social values, crime, voting, demographics, families and family authority
The rotating modules in the 2004 survey covered:
Questionnaire 1: Poverty, environment, democracy (part 2)
Questionnaire 2: ISSP module (citizenship), democracy (part 2)
Any publication, whether printed, electronic or broadcast, based wholly or in part on these materials, should acknowledge the original data creators or copyright holders, the funders of the Data Collections (if different) to acknowledge Copyright where appropriate. Any publication, whether printed, electronic or broadcast, based wholly or in part on these materials should carry a statement that the original data creators or copyright holders and the funders of the Data Collections (if different) bear no responsibility for their further analysis or interpretation.
Human Sciences Research Council. South African Social Attitudes Survey 2004 [dataset]. Version 1. Pretoria: Human Sciences Research Council [producer], 2010. Cape Town: DataFirst [distributor], 2012.
By accessing the data, you give assurance that
• the data and documentation will be used solely for educational, scholarly and nonprofit purposes,
• the data and documentation will not be duplicated or distributed without prior approval from the HSRC,
• the confidentiality of individuals/organisations in the data will be preserved at all times and that no attempt will be made to obtain or derive information relating specifically to identifiable individuals/organisations,
• the HSRC will be acknowledged in all published works based on the data and documentation
• the HSRC will be informed of any publications resulting from work based in whole or in part on the data and documentation, and
• the HSRC will not be held liable for the accuracy or comprehensiveness of the data. The data is provided on an "as is" basis and without warranty or liability of any kind.
Copyright, Human Sciences Research Council
DDI Document ID
University of Cape Town
Date of Metadata Production
DDI Document version
Version 02 (August 2013). Edited version based on Version 1.1 DDI (ddi-zaf-datafirst-sasas-2004-v1.1) that was done by DataFirst.