General Household Survey 2019
South Africa, 2019
Get MicrodataIdentification
ZAF_2019_GHS_v01_EN_M_v01_A_ESS
General Household Survey 2019
GHS 2019
| Name | Country code |
|---|---|
| South Africa | ZAF |
Other Household Survey [hh/oth]
The General Household Survey (GHS) is one of Statistics South Africa's longest-running surveys. It has been conducted for more than twenty years, with its first round conducted in July 2002, and it was originally designed to meet user need of a survey conducted regularly to measure the level of development and the performance of government programs and projects.
The General Household Survey is an annual household survey measuring the living conditions of households in South Africa. The survey collects data on education, health, and social development, housing, access to services and facilities, food security, and agriculture.
Sample survey data [ssd]
Households and individuals
Scope
The scope of the General Household Survey includes:
- Household characteristics: Dwelling type, home ownership, access to water and sanitation, access to services, transport, household assets, land ownership, agricultural production
- Individuals' characteristics: Demographic characteristics, relationship to household head, marital status, language, education, employment, income, health, fertility, mortality, disability, access to social services
Coverage
The General Household Survey has national coverage.
The lowest level of geographic aggregation for the data is Province (and metropolitan municipality, where this applies).
The survey covers all de jure household members (namely, the usual residents) of households in the nine provinces of South Africa, and residents in workers' hostels. The survey does not cover collective living quarters such as student hostels, old age homes, hospitals, prisons, and military barracks.
Producers and sponsors
| Name |
|---|
| Government of South Africa, Statistics South Africa (Stats SA) |
Sampling
From 2015, the General Household Survey uses a Master Sample frame, which was developed in 2013 as a general-purpose sampling frame to be used for all of Statistics South Africa's household-based surveys. This Master Sample has design requirements that are reasonably compatible with the General Household Survey. The 2013 Master Sample is based on information collected during the Census 2011 conducted by Statistics South Africa.
In preparation for the Census 2011, the country was divided into 103 576 enumeration areas (EAs). The census EAs, together with the auxiliary information for the EAs, were used as the frame units or building blocks for the formation of primary sampling units (PSUs) for the Master Sample, since they covered the entire country, and had other information that is crucial for stratification and creation of PSUs. There are 3324 PSUs in the Master Sample, with an expected sample of approximately 33 000 dwelling units (DUs). The number of PSUs in the current Master Sample (3324) reflects an 8.0 percent increase in the size of the Master Sample compared to the previous: the 2008 Master Sample had 3080 PSUs.
The larger Master Sample was selected to improve the precision (resulting in smaller coefficients of variation, known as CVs) of the General Household Survey estimates. The Master Sample is designed to provide survey estimates that are representative at provincial level, and within provinces, at metro/non-metro levels. Within the metros, the sample is further distributed by geographical type. The three geography types are Urban, Tribal and Farms. This implies, for example, that within a metropolitan area, the sample is representative of the different geography types that may exist within that metro.
The sample for the General Household Survey is based on a stratified two-stage design with probability proportional to size (PPS) sampling of PSUs in the first stage, and sampling of DUs with systematic sampling in the second stage. After allocating the sample to the provinces, the sample was further stratified by geography (primary stratification), and by population attributes using Census 2011 data (secondary stratification).
The sample weights were constructed in order to account for the following:
- the original selection probabilities (design weights)
- adjustments for PSUs that were sub-sampled or segmented
- population excluded from the sampling frame
- non-response
- weight trimming
- benchmarking to known population estimates from the Demographic Analysis Division within Statistics South Africa
Sampling weights for the data collected from the sampled households were constructed so that the responses could be properly expanded to represent the entire civilian population of South Africa. Design weights, which are the inverse sampling rate (ISR) for the province, were assigned to each of the households in a province.
Mid-year population estimates, produced by the Demographic Analysis Division, were used for benchmarking. Final survey weights were constructed using regression estimation to calibrate to national level population estimates cross-classified by 5-year age groups, gender and race, and provincial population estimates by broad age groups. The 5-year age groups are: 0-4, 5-9, 10-14, 55-59, 60-64, and 65 and over. The provincial level age groups are 0-14, 15-34, 35-64, and 65 years and over. The calibrated weights were constructed such that all persons in a household would have the same final weight.
The Statistics Canada software StatMx was used for constructing calibration weights. The population controls at national and provincial level were used for the cells defined by cross-classification of age by gender and by race. Records for which the age, population group or sex had item non-response could not be weighted and were therefore excluded from the dataset. No additional imputation was done to retain these records.
Household estimates that were developed using the UN headship ratio methodology were used to weight household files. The databases of Census 1996, Census 2001, Community Survey 2007, and Census 2011 were used to analyze trends and develop models to predict the number of households for each year. The weighting system was based on tables for the expected distribution of household heads for specific age categories, per population group and province.
Survey instrument
Statistics South Africa transitioned to electronic data collection in 2019, and the General Household Survey was redesigned to allow for data collection trough computer assisted personal interviews. Some of the variables were also renamed. See the document ghs-2019-variables-renamed for a correspondence between old and new names.
Data collection
| Start | End |
|---|---|
| 2019-01 | 2019-12 |
- Computer Assisted Personal Interview [capi]
Data appraisal
Statistics South Africa removed the EDU_SAME variable from the public release file of the General Household Survey 2019 because a coding error in the electronic questionnaire resulted in the data not being reliable. The coding error was in the enabling condition (skip instruction) for EDU21 which meant that most of the respondents who would otherwise have answered the question on whether they were doing the same grade as the year before were not asked the question. Once identified, the critical error was corrected in subsequent questionnaires.
Data Access
| Name | Affiliation | URL | |
|---|---|---|---|
| DataFirst | University of Cape Town | www.support.datafirst.org | [email protected] |
Public access data, available to all
Statistics South Africa. General Household Survey 2019 [dataset]. Version 1. Pretoria: Statistics SA [producer], 2019. Cape Town: DataFirst [distributor], 2019. DOI: https://doi.org/10.25828/vtvj-pv21
Contacts
| Name | Affiliation | URL | |
|---|---|---|---|
| DataFirst Helpdesk | University of Cape Town | [email protected] | www.support.data1st.org/ |
Metadata production
DDI_ZAF_2019_GHS_v01_EN_M_v01_A_ESS_FAO
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| DataFirst | University of Cape Town | Metadata producer | |
| Statistics Division | ESS | Food and Agriculture Organization of the United Nations | Metadata adapted for FAM |