zaf-statssa-ohs-1998-v1.2
October Household Survey 1998
Name | Country code |
---|---|
South Africa | zaf |
Household Survey [hh]
Statistics South Africa. October Household Survey 1998 [dataset]. Version 1.2. Pretoria: Statistics South Africa (formerly Central Statistical Service) [producer], 2000. Cape Town: DataFirst [distributor], 2021. DOI: https://doi.org/10.25828/xc07-zq48
The October Household Survey is an annual survey based on a survey of a large number of households (ranging from 16 000 in 1996 through to 30 000 in 1997 and 1998, depending on the availability of funding). It covers a range of development indicators, including unemployment rates (official and expanded), according to standard definitions of the International Labour Organisation (ILO).
Sample survey data
Households and individuals
v1.2: Edited, anonymised dataset for public distribution.
Version 1 was received from Stats SA.
Version 1.1: Variables that were strings unnecessarily were converted to numeric variables.
Version 1.2: All files with weight variables (9 of the 11 files) had their weight variable divided by 10 000. This was following a note in Stats SA's metadata saying that the analyst should do this.
The scope of the OHS 1998 includes: employment, unemployment, informal sector, internal migration, services available by type of dwelling, access to health and social services, safety and well-bring of household, households by average household size and type of dwelling, level of education, quality of life, health statistics, vital statistics.
The survey had national coverage.
The lowest level of geographic aggregation for the data is province.
The survey covered households and household members in households in the nine provinces of South Africa
Name |
---|
Statistics South Africa |
A sample of 20 000 households was drawn in 2 000 enumerator areas (EAs) (that is 10 households per enumerator area). A two-stage sampling procedure was applied and the sample was stratified, clustered and selected to meet the requirements of probability sampling. The sample was based on the 1996 Population Census enumerator areas and the estimated number of people from the administrative records of the 1996 population Census. The sampled population excluded all prisoners in prisons, patients in hospitals, people residing in boarding houses and hotels (whether temporary or semi-permanent). The sample was explicitly stratified by province, Transitional Metropolitan Council (TMC)and District Council (DC). A square root method was used for the allocation of the sample EAs to the explicit strata.
Within each explicit stratum the EAs were stratified by simply arranging them in geographical order by magisterial district and, within the magisterial district, by EA. The allocated number of EAs was systematically selected with probability proportional to size in each stratum. The measure of size was the estimated number of people in Each EA. A systematic sample of 10 households was drawn.
The 1996 population Census was used as a basis for the weighting.
Household weights were calculated by using the reciprocal of the inclusion probabilities.
Since the sample selection was done in two stages (i.e. first stage - selection of an EA, second stage - selection of a household in the selected EA):
The inclusion probability of an EA (say p1):
Since this was done with probability proportional to size (size being the number of persons residing in the EA),
p1 = m . Ai /S Ai
mi - number of EAs in the sample in the i-th stratum (where stratum is the District Council in a province)
Ai - number of persons residing in the selected EA
S Ai - total number of persons in the population in the i-th stratum
The inclusion probability of the household (say p2):
Since ten (10) households (per EA) were selected systematically,
p2 = 10/number of households in the selected EA
Household weight = (1/p1.p2). Relative scaling was done on this weight to cater for the urban/non-urban split per province.
To calculate the person weight, the data was post-stratified by province, gender and age group (5 year age groups). The 1996 Census figures (adjusted for growth) were used as benchmarks. Relative scaling was also done on this weight to cater for the population group and urban/non-urban splits.
The data files in the October Household Survey 1997 (OHS 1997) correspond to the following sections in the questionnaire:
PERSON: Indivitual-level data from Section 1 and Section 4; BIRTHS: Data from Section 2; CHILDREN: Data from Section2; WORKER: Data from Section 3; MIGRANT: Data from Section 5; DEATHS: Data from Section 6; MIGRATION: Data from Section 7; DOMESTIC: Data from Section 8; HOUSE: Household-level data from Section 9
Errors in the marital codes in the original OHS 1998 questionnaire:
The questionnaire for the OHS 1998 originally provided by Statistics SA with the data files was incorrect. It was the OHS 1997 questionnaire with a OHS 1998 flap. The marital codes were different in the two surveys. In 1997, the codes for the variable Marital Status were:
1 Never married
2 Married - civil
3 Married - customary
4 Living together
5 Widowed
6 Divorced
In the 1998 survey, the codes for the variable Marital Status are:
1 Married - civil
2 Married - traditional (customary)
3 Living together
4 Widower/widow
5 Divorced/separated
6 Never married
DataFirst notified Statistics SA of this error on 13 July 2007 and they sent a corrected questionnaire. The correct questionnaire is version 2, available with the data since 2007.
Start | End |
---|---|
1998-10 | 1998-10 |
Errors in the marital codes in the original OHS 1998 questionnaire:
The questionnaire for the OHS 1998 originally provided by Statistics SA with the data files was incorrect. It was the OHS 1997 questionnaire with a OHS 1998 flap. The marital codes were different in the two surveys. In 1997, the codes for the variable Marital Status were:
1 Never married
2 Married - civil
3 Married - customary
4 Living together
5 Widowed
6 Divorced
In the 1998 survey, the codes for the variable Marital Status are:
1 Married - civil
2 Married - traditional (customary)
3 Living together
4 Widower/widow
5 Divorced/separated
6 Never married
DataFirst notified Statistics SA of this error on 13 July 2007 and they sent a corrected questionnaire. The correct questionnaire is version 2, available with the data since 2007.
Name | Affiliation | URL | |
---|---|---|---|
DataFirst | University of Cape Town | http://www.datafirst.uct.ac.za | info@data1st.org |
Public use files, available to all
Statistics South Africa. October Household Survey 1998 [dataset]. Version 1.2. Pretoria: Statistics South Africa (formerly Central Statistical Service) [producer], 2000. Cape Town: DataFirst [distributor], 2021. DOI: https://doi.org/10.25828/xc07-zq48
Copyright, Statistics South Africa
Name | Affiliation | URL | |
---|---|---|---|
DataFirst Helpdesk | University of Cape Town | support@data1st.org | http://support.data1st.org/ |
Name | Affiliation | Role |
---|---|---|
DataFirst | University of Cape Town | Metadata producer |
2021-03-24
Version 3