South Africa's first Living Conditions Survey (LCS) was conducted by Statistics South Africa between September 2008 and August 2009. The main aim of this survey is to provide data that will contribute to better understanding of living conditions and poverty in South Africa for monitoring levels of poverty over time. Data was collected from 25 075 households across the country over a period of 12 months. The survey used a combination of the diary and recall methods. Households were required to complete their daily acquisitions in diaries provided by Stats SA for a period of a month and to answer a variety of questions from the household questionnaire administered by a Stats SA official on a variety of topics. These include household expenditure, subjective poverty, and income and anthropometry.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Units of analysis in the survey include households and individuals
v1.1: Edited, anonymised data for licensed distribution
This version, version 1.1 is the version provided to DataFirst by SSA in August 2012, with variable labels corrected.
The following corrections were made to variable labels in the Household file:
imp_rent - changed from "Imputed value of fuel" to "Imputed value of rent")
imp_Bother - changed from "Imputed value of cellular calls made" to "Imputed value of taxi travel"
imp_landcalls - changed from "Imputed value of cellular calls made" to "Imputed value of landline calls made"
imp_internet - changed from "Imputed value of fuel" to "Imputed value of internet usage"
The incorrect labels are still listed in the Record Layout and Metadata files.
The dataset received from SSA has in the Person data file a variable "Q621CASH" which has a value of 8 with no label (when the range of values is 1=Yes; 2= No; 3= Don't know; 9 = Unspecified).
DATA QUALITY NOTE:
In the Living Conditions Survey 2008/09 metadata document there are two different inflation adjustment dates: March 2006 (as described on page 77) and March 2009 (as described on page 81).
The information on page 77 is incorrect: The inflation adjustment date for this survey is March 2009.
DataFirst has corrected the metadata document and version 2 of this document, with the correct date, is available with the data.
There were seven modules in the household questionnaire. The first module dealt with establishing the composition and structure of the household, as well as capturing particulars of all household members. The second module collected information on health, disability, education and employment. The third module dealt with welfare, assets and information on dwellings and services. Modules four and five collected information on the different categories of consumption expenditure (including housing, clothing, furniture, appliances, transport, computer and telecommunication equipment, etc.), as well as information on subsistence and living circumstances. The sixth module dealt with savings, investments, debt, remittances and income. The seventh and last module collected anthropometric measurements (height, weight and waist) for all household members.
The survey had national coverage
The lowest level of geographic aggregation covered by the data is Province.
The survey covered all private dwelling units, workers' hostels, residential hotels, nurses' and doctors' quarters, but excludes patients in hospitals or clinics, guests in hotels and guesthouses, prisoners in prisons, scholars and students in school or student hostels and the aged in old age homes.
Producers and sponsors
Statistics South Africa
The sampling frame for the LCS was obtained from Statistics South Africa's Master Sample (MS) based on the 2001 Population Census Enumeration Areas. The scope of the Master Sample (MS) is national coverage of all households in South Africa. It was designed to cover all households living in private dwelling units and workers living in workers' quarters in the country. The MS consists of 3080 primary sampling units (PSUs) made up of enumeration areas. The PSU coverage comprises all settlement types, including urban formal, urban informal, rural formal and tribal areas. For the LCS, 3065 PSUs were sampled from the MS and roughly ten dwelling units (DUs) were sampled on average per PSU. In the case of multiple households, all households in the DU were included. The sample was evenly split into four rotations (quarters) with national representativity in each rotation. Each rotation (consisting of a sample for three months) was then evenly split into monthly samples. Ultimately, the sample was evenly spread over the 12 survey periods (one month each).
Dates of Data Collection
Data Collection Mode
The Living Conditions Survey 2008/2009 used a household questionnaire, a weekly diary, and a survey assessment questionnaire. The household questionnaire was a booklet of questions administered to respondents during the course of the survey month. The Weekly diary was a booklet that was left with the responding household to track all acquisitions made by the household during the survey month. The household (after being trained by the Interviewer) was responsible for recording all their daily acquisitions as well as information about where they purchased the item (source) and the purpose of the item. A household completed a different diary for each of the four weeks of the survey month. Interviewers then assigned codes for the classification of individual consumption according to purpose (COICOP) to reported items recorded in the weekly diary, using a code list provided to them. Finally the survey included a survey assessment questionnaire that was administered to households after the survey month was complete by either the district survey coordinator or provincial quality monitor. In addition to serving as a control questionnaire to verify information collected by the interviewers, the instrument was designed to evaluate data collection processes and the respondent's perceptions of Stats SA and the survey.
Statistics South Africa. Living Conditions Survey 2008-2009 [dataset]. Version 1.1. Pretoria: Statistics South Africa [producer], 2012. Cape Town: DataFirst [distributor], 2012. DOI: https://doi.org/10.25828/y3qr-2393