Description:
The 2005 HSRC Master Sample was used for SABSSM 2008 and 2012, the SANHANES study in 2012 and SASAS 2007-2010 (adjacent EAs) to obtain an understanding of geographical spread of HIV/AIDS, perceptions and attitudes of people and other health related studies over time.
Abstract:
A sample can be defined as a subset containing the characteristics of a larger population. Samples are used in statistical testing when population sizes are too large for the test to include all possible members or observations. A sample should represent the whole population and not reflect bias toward a specific attribute.[1]
One of the most crucial aspects of sample design in household surveys is its frame. The sampling frame has significant implications on the cost and the quality of any survey, household or otherwise.[2]
The sampling frame .... in a household survey must cover the entire target population. When that frame is used for multiple surveys or multiple rounds of the same survey it is known as a master sample frame or .... master sample.[3]
A master sample is a sample drawn from a population for use on a number of future occasions, so as to avoid ad hoc sampling on each occasion. Sometimes the master sample is large and subsequent inquiries are based on a sub-sample from it.[4]
The HSRC compiles master samples in order to construct samples for various HSRC research studies. The 2005 HSRC Master Sample was used for SABSSM 2008 and 2012, SASAS 2007-2010 and the SANHANES study in 2012 to obtain an understanding of geographical spread of HIV/AIDS, perceptions and attitudes of people and other health related studies over time.
The 2005 HSRC Master Sample was created in the following way:
South Africa was delineated into EAs according to municipality and province. Municipal boundaries were obtained from the Municipal Demarcation Board. An Enumeration area (EA) is the smallest geographical unit (piece of land) into which the country is divided for census or survey enumeration.[5] The concepts and definitions of terms used for Census 2001 comply in most instances with United Nations standards for censuses.
A total of 1,000 census enumeration areas (EAs) from the 2001 population census were randomly selected using probability proportional to size and stratified by province, locality type and race in urban areas from a database of 80 787 EAs that were mapped using aerial photography to develop an HSRC master sample for selecting households.
The ideal frame would be complete with respect to the target population if all of its members (the universe) are covered by the frame.
Ideal characteristics of a master sample:
The master frame should be as complete, accurate and current as practicable.
A master sample frame for household surveys is typically developed from the most recent census, just as a regular sample frame is.
Because the master frame may be used during an entire intercensal (between census) period, however, it will usually require periodic and regular updating such as every 2-3 years.
This is in contrast to a regular frame which is more likely to be up-dated on an ad hoc basis and only when a particular survey is being planned[6]
[1] http://www.investopedia.com/terms/s/sample.asp
[2] http://unstats.un.org/unsd/demographic/meetings/egm/sampling_1203/docs/no_3.pdf
[3] http://unstats.un.org/unsd/demographic/meetings/egm/sampling_1203/docs/no_3.pdf
[4] A Dictionary of Statistical Terms, 5th edition, prepared for the International Statistical Institute by F.H.C. Marriott. Published for the International Statistical Institute by Longman Scientific and Technical. http://stats.oecd.org/glossary/detail.asp?ID=3708
[5] http://africageodownloads.info/128_mokgokolo.pdf
[6] http://unstats.un.org/unsd/demographic/meetings/egm/sampling_1203/docs/no_3.pdf
All enumeration areas (80 787 EAs) within the South African borders during the 2001 Census.
The whole country was delimited into EAs according to municipality and province. Municipal boundaries were obtained from the Municipal Demarcation Board.
A total of 1,000 census enumeration areas (EAs) from the 2001 population census were randomly selected using probability proportional to size and stratified by province, locality type and race in urban areas from a database of 80 787 EAs that were mapped in all surveys using aerial photography to develop all HSRC master sample for selecting households.
The first digit represents the province
The second and third digits represent the municipality
The last five digits represent the specific EA.
The sampled EAs formed primary sampling units (PSUs). Locality types were defined as urban formal, urban informal, rural formal (including commercial farms), and rural informal (tribal authority) areas. Oversampling of Coloureds and Indians or Asians was done to meet the required minimum sample size. Aerial photographs drawn from Google Earth were also employed to ensure that the most up-to-date information was available for the master sample.
In each sampled EA a total of 15 visiting points (VPs) or households were used as secondary sampling units (SSUs). Within each household selected for the survey, all household members (including consenting and non-consenting household members) constituted the ultimate sampling unit (USU). A VP was defined as a stand with an address that might have one or more than one residential household in which a group of people live and eat together 'from the same pot'. If multiple households existed in a visiting point,
A Kish grid (Kish, 1965) was used to randomly select a household or individual in the case where more than the required number of houses or individuals were eligible to participate.
