During 1999 the HSRC embarked on a self-funded multidisciplinary research project on the causes of migration in South Africa. The main purpose of the project was to analyse census and other secondary data and to undertake the necessary surveys with a view to determining the importance of economic and non-economic factors in the explanation of migration trends in South and southern Africa. The study highlighted the need for “purpose-made” data for the analysis of migration causes and patterns, which would be addressed to a large extent by sample surveys dealing specifically also with the social and psychological aspects of migration. To help address these issues a series of surveys was envisaged. The first of these would be aimed mainly at testing the utility of various items that would provide the required data for the migration research. A preliminary (initial) questionnaire survey among 911 households was was undertaken during the period 4 April to 25 June 2000 and the main survey was conducted in 2001-2002.
Kind of Data
Sample survey data
Unit of Analysis
Households and individuals
v1: Edited, anonymised data for public distribution
The survey collected data on age, gender, level of education, marital status, migration, use of health services, economic activity, and self-employment.
The lowest level of geographic aggregation covered by the data was suburb in the selected areas
The survey covered households in South Africa
Producers and sponsors
Human Sciences Research Council (HSRC)
Human Sciences Research Council
The 2001-2002 survey was based on a stratified, clustered random (complex) sampling design. The explicit strata used in the sampling design were the following: (a) census-based population group (African, coloured, Indian or white); (b) metropolitan/non-metropolitan location; (c) whether a so-called spatial development (SDI) area or not, and (d) census-based locality type (rural/urban). In non-metropolitan areas an implicit stratification of local government level (city, other local municipality or district municipality) was also used. Clustering took the form of randomly selecting (using systematic sampling) six households in each of the selected census enumerator areas (EAs) across the different strata.
Dates of Data Collection
Data Collection Mode
After a preliminary survey was undertaken in 2000 to evaluate the validity and potential reliability of 179 possible scale items for the questionnaire, with a view to reducing these items to an absolute minimum for the purpose of the main (2001-2002) survey. The results of the initial survey, as well as provisional findings from an analyses of census data and other secondary data sources were used to finalise the questionnaire to be used in the main survey. The design was largely based on the questionnaire used in the initial survey, sans those items deemed not significant for the construction of scale items.
Human Sciences Research Council. Migration in South Africa 2001-2002 [dataset]. Version 1. Pretoria: Human Sciences Research Council [producer], 2002. Cape Town: DataFirst[distributor], 2020. DOI: https://doi.org/10.25828/td2j-qr73