General Household Survey 2003 - South Africa

Abstract
---------------------------

Stats SA conducted the October Household Survey (OHS) annually from 1994 to 1999, based on a probability sample of a large number of households ranging from 16 000 to 30 000 households each year (depending on availability of funding). This survey was discontinued in 1999 due to the reprioritisation of surveys in the face of financial constraints. February 2000 saw the birth of the Labour Force Survey (LFS), which is a biannual survey conducted by Stats SA in March and September of each year. The LFS covers some areas previously covered by the OHS, but not all, since it is a specialised survey principally designed to measure the dynamics in the labour market. The September LFS each year does include a section designed to measure social indicators such as access to infrastructure, but again this section does not go into as much depth as the OHS used to. A need was therefore identified by our users for a regular survey designed specifically to measure the level of development and the performance of government programmes and projects. The General Household Survey (GHS) was developed for this purpose. While the survey replaces the October Household Survey (OHS), the indicators measured in the 13 nodal areas identified for the Integrated Rural Development Strategy (IRSD) formed the basis for the subject matter of the survey. The first round of the GHS was conducted in July 2002 and the second round in July 2003.

Geographic coverage
---------------------------

The scope of the General Household Survey 2003 was national coverage.

Analysis unit
---------------------------

The units of anaylsis for the General Household Survey 2003 are individuals and households.

Universe
---------------------------

The survey covered all de jure household members (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 students' hostels, old age homes, hospitals, prisons and military barracks.

Kind of data
---------------------------

Sample survey data [ssd]

Sampling procedure
---------------------------

For the GHS 2003 a multi-stage stratified sample was drawn using probability proportional to size principles.

The sample was drawn from the master sample, which Statistics South Africa uses to draw samples for its regular household surveys. The master sample is drawn from the database of enumeration areas (EAs) established during the demarcation phase of Census 1996. As part of the master sample, small EAs consisting of fewer than 100 households are combined with adjacent EAs to form primary sampling units (PSUs) of at least 100 households, to allow for repeated sampling of dwelling units within each PSU. The sampling procedure for the master sample involves explicit stratification by province and within each province, by urban and non-urban areas. Within each stratum, the sample was allocated disproportionately. A PPS sample of PSUs was drawn in each stratum, with the measure of size being the number of households in the PSU. Altogether approximately 3 000 PSUs were selected. In each selected PSU a systematic sample of ten dwelling units was drawn, thus, resulting in approximately 30 000 dwelling units. All households in the sampled dwelling units were enumerated. The master sample is divided into five independent clusters. In order to avoid respondent fatigue (the LFS is a rotating panel survey which is conducted twice yearly), the GHS sample uses a different cluster from the LFS clusters.

Mode of data collection
---------------------------

Face-to-face [f2f]

Research instrument
---------------------------

The GHS 2003 questionnaire collected data on:
Household characteristics: Dwelling type, home ownership, access to water and sanitation facilities, 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, disability, access to social services, mortality.
Women's characteristics: fertility

Response rate
---------------------------

Response codes Number of responses %
Completed 26 469 84.7
Non-contact 897 2.9
Refusal 645 2.1
Partly completed 18 0.1
Unusable information 1 0.0
Vacant 1 510 4.8
Listing error 246 0.8
Other 1 447 4.6
Total 31 233 100.0

摘要
---------------------------
南非统计局自1994年起每年进行一次十月份家庭调查（OHS），该调查基于16,000至30,000个家庭的概率样本（年度资金状况而定）。由于面对财务限制而重新调整调查优先级，该调查于1999年终止。2000年2月，劳动市场动态的测量专用调查——劳动力调查（LFS）诞生，Stats SA每年3月和9月进行此项双年度调查。LFS涵盖OHS所涉及的某些领域，但并非全部，因为它主要是一项针对劳动力市场的动态变化的专门调查。每年的9月LFS都包含一个专门用于测量社会指标（如基础设施获取情况）的部分，但这一部分并未像OHS那样深入。因此，我们的用户需要一项定期进行的、专门用于衡量政府项目与计划发展水平和绩效的调查。为此，我们开发了总体家庭调查（GHS）。虽然该调查取代了十月份家庭调查（OHS），但对于综合农村发展战略（IRSD）中确定的13个节点地区所测量的指标构成了调查主题的基础。GHS的第一轮调查于2002年7月进行，第二轮调查于2003年7月进行。
地理覆盖范围
---------------------------
2003年总体家庭调查的覆盖范围为国家层面。
分析单元
---------------------------
2003年总体家庭调查的分析单元为个人和家庭。
总体
---------------------------
该调查覆盖了南非九个省份的所有法定家庭成员（常驻居民）以及工人宿舍的居民。该调查不包括集体居住区，如学生宿舍、养老院、医院、监狱和军事营房。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
对于2003年GHS，采用了基于规模比例的概率多阶段分层抽样。
样本从主样本中抽取，主样本是南非统计局用于其常规家庭调查的样本。主样本来源于1996年人口普查划界阶段建立的调查区域数据库。作为主样本的一部分，由少于100户家庭组成的小型调查区域与相邻调查区域合并，形成至少有100户家庭的基本抽样单位（PSU），以便在每个PSU内对住宅单元进行重复抽样。主样本的抽样程序涉及按照省份和每个省份内的城乡区域进行明确分层。在每个分层内，样本分配是不成比例的。在每个分层中抽取了PPS样本的PSU，以PSU中的家庭数量为规模衡量标准。总共选取了大约3,000个PSU。在每个选定的PSU中，抽取了十个住宅单元的系统样本，从而产生了大约30,000个住宅单元。对所有抽样住宅单元中的家庭进行了统计。主样本分为五个独立的集群。为了避免受访者疲劳（LFS是一项每半年进行一次的轮换调查），GHS样本使用与LFS集群不同的集群。
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
2003年GHS问卷收集了以下数据：
家庭特征：住宅类型、房屋所有权、水卫生设施获取情况、服务获取情况、交通、家庭资产、土地所有权、农业生产
个人特征：人口统计特征、与家庭首脑的关系、婚姻状况、语言、教育、就业、收入、健康、残疾、社会服务获取情况、死亡率。
妇女特征：生育率
响应率
---------------------------
响应代码
响应数量
百分比

完成
26,469
84.7

未接触
897
2.9

拒绝
645
2.1

部分完成
18
0.1

信息不可用
1
0.0

空缺
1,510
4.8

列表错误
246
0.8

其他
1,447
4.6

总计
31,233
100.0