Data from the South Africa Integrated Household Survey

In order to use the data from the South Africa Integrated Household Survey Data, it is necessary to fill in the Data Use Agreement Form found at the bottom of this page. All data sets are available in either ASCII, SAS portable, or Stata formats. If you do not know which format you prefer, return to the page on Data Formats for a description of these formats.

PLEASE NOTE:  It has recently come to our attention that the data collected in clusters 217 and 218 should be viewed as highly unreliable and therefore removed from the data set.  The data currently available on the web site has been revised to remove the data from these clusters.  Researchers who have downloaded the data in the past should revise their data sets.  (26 October 2001)  For information on the data in those clusters, contact SALDRU.

The data files have been Pkzipped into four separate groups called A, B, C, and D. (Within each zipped group are several data files which have been zipped or compressed a second time. If you have difficulty unzipping the data files, see the instructions on decompressing the files in the section on data formats.)

• The Group A of data files contains all the data from sections 0 through 4 of the Household Questionnaire.
  A separate page contains details of the Group A files, including variable labels and summary statistics.
• The Group B of data files contains all the data from sections 5 through the end of the Household Questionnaire.
  A separate page contains details of the Group B files, including variable labels and summary statistics.
• The Group C of data files contains all of the data from the Community questionnaire, as well as data files containing general information such as: Demographic data, sampling frame information, sample weights, and basic data for descriptive statistics.
  A separate page contains details of the Group C files, including variable labels and summary statistics.
• The Group D of data files contains the constructed data files supplied with the raw data. (For example, these files contain measures of total household income and expenditure.)
  A separate page contains details of the Group D files, including variable labels and summary statistics.
  Note: We do not have the staff time to respond to questions about how or why certain assumptions were made in constructing the files. We provide them on an as is basis. In order to facilitate the use of these files, the essential program files used to create these data are available for downloading in the miscellaneous file in the documents section. We assume that most users of the data will not download these files, but will prefer to construct income and expenditure for themselves using the Group A and B data.

To obtain the data you will first need to fill out the Data Agreement Form. The form simply states that downloading the data obligates you to cite the source of the data and to send copies of papers to SALDRU and the LSMS Division of the World Bank. The form also requires you to state your name, address, and intended use of the data. Once you have submitted the form, you will be sent to a page containing links to the data files. If your browser doesn't support forms (or fails to forward you to the data page upon completing the form), contact LSMS and you will receive instructions on how to progress to the page with the data files.

Revised 10/26/01

Back to the South Africa Integrated Household Survey Page