The LFS is a twice-yearly rotating panel household survey, specifically designed to measure the dynamics of employment and unemployment in South Africa. It measures a variety of issues related to the labour market,including unemployment rates (official and expanded), according to standard definitions of the International Labour Organisation (ILO).
All editions of the LFS have been updated (some more than once) since their release. These version changes are detailed in a document available from DataFirst (in the "external documents" section titled "LFS 2000-2008 Collated Version Notes on the South African LFS").
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Households (dwellings) and individuals
v1.1: Edited, anonymised dataset for licensed distribution
The South African September 2004 LFS dataset was originally released in March 2005 as 4 data files (worker, person, migrant and household). A second version was downloaded from the Statistics South Africa website in August 2011 by DataFirst. The only differences between these two versions is that 1.1 includes year and month variables and version 1.0 represented all variable names in lower case.
Household characteristics, household listing, demographics, education, economic activity, work for pay, business ownership, unemployment, employers, main work activity in the past week, wages, salary, employment, migration
Province (variable name: "Prov")
District Council (variable name: "DC")
The LFS sample covers the non-institutional population except for workers' hostels. However, persons living in private dwelling units within institutions are also enumerated. For example, within a school compound, one would enumerate the schoolmaster's house and teachers' accommodation because these are private dwellings. Students living in a dormitory on the school compound would, however, be excluded.
Producers and sponsors
Statistics South Africa
Statistics South Africa uses a rotating panel methodology for the labour force survey. The rotating panel methodology involves visiting the same dwelling units on a number of occasions (in this instance, five at most). After the panel is established, a proportion of the dwelling units is replaced each round (in this instance, 20%). New dwelling units are added to the sample to replace those that are taken out.
Enumeration Areas (EAs) that had a household count of less than twenty-five were omitted from the census 2001 frame that was used to draw the sample of Primary Sampling Units (PSUs) for the new Master Sample. Other omissions from the Master Sample frame included all institution EAs except workers, hostels, convents and monasteries. EAs from census 2001 were pooled in two stages, before and after sampling. Before sampling the criterion that was used to pool EAs was that they should contain a minimum of one hundred households. However, during listing it was discovered that there were discrepancies between the information on the database and what was on the ground.
Therefore, in the second stage of pooling, EAs that were found to have less than sixty dwelling units during listing were pooled. The Master Sample is a multi-stage stratified sample. The overall sample size of PSUs was 3000. The explicit strata were the 53 district councils/metros (DCs). The 3000 PSUs were allocated to these DCs using the power allocation method. The PSUs were then sampled using probability proportional to size principles. The measure of size used was the number of households in a PSU as calculated in the census. The sampled PSUs were listed with the dwelling unit as the listing unit. From these listings systematic samples of dwelling units per PSU were drawn. These samples of dwelling units form clusters. The size of the clusters differs depending on the specific survey requirements. The LFS uses one of the clusters that contain ten dwelling units.
The initial weights (household weights), based on the sample design, were equal to the inverse of the probability of selection. The initial weight for each member of the household was the same as the weight for the household itself. Further adjustment factors were then calculated within PSUs to account for non-response. To adjust for under-enumeration and to align survey estimates with independent population estimates, the weights were calibrated against Person benchmarks. A software package called CALMAR was used to perform this calibration. Using an iterative procedure, CALMAR adjusted the weights so that Person estimates conformed as closely as possible to external Person benchmarks. Gender, race and age group parameters were used for the Person cross-classification of the population.
Note that this was the LFS survey in which these benchmarks were performed according to the 2001 Census, with a new sample drawn in September 2004 and used for the proceeding Labour Force Surveys.