Demographic and Health Survey 2016 - South Africa

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

The primary objective of the South Africa Demographic and Health Survey (SADHS) 2016 is to provide up-to-date estimates of basic demographic and health indicators. Specifically, the SADHS 2016 collected information on fertility levels; marriage; sexual activity; fertility preferences; awareness and use of contraceptives; breastfeeding practices; nutrition; childhood and maternal mortality; maternal health, including antenatal and postnatal care; key aspects of child health, including immunisation coverage and prevalence and treatment of acute respiratory infection (ARI), fever, and diarrhoea; potential exposure to the risk of HIV infection; coverage of HIV counselling and testing (HCT); and physical and sexual violence against women. Another critical objective of the SADHS 2016 is to provide estimates of health and behaviour indicators for adults age 15 and older, including use of tobacco, alcohol, and codeine-containing medications. In addition, the SADHS 2016 provides estimates of the prevalence of anaemia among children age 6-59 months and adults age 15 and older, and the prevalence of hypertension, anaemia, high HbA1c levels (an indicator of diabetes), and HIV among adults age 15 and older.

The information collected through the SADHS 2016 is intended to assist policymakers and programme managers in evaluating and designing programmes and strategies for improving the health of the country’s population.

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

National

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

- Household
- Individual
- Children age 0-5
- Woman age 15-49
- Man age 15-59

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

The survey covered all de jure household members (usual residents), children age 0-5 years, women age 15-49 years and men age 15-59 years resident in the household.

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

Sample survey data [ssd]

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

The sampling frame used for the SADHS 2016 is the Statistics South Africa Master Sample Frame (MSF), which was created using Census 2011 enumeration areas (EAs). In the MSF, EAs of manageable size were treated as primary sampling units (PSUs), whereas small neighbouring EAs were pooled together to form new PSUs, and large EAs were split into conceptual PSUs. The frame contains information about the geographic type (urban, traditional, or farm) and the estimated number of residential dwelling units (DUs) in each PSU. The sampling convention used by Stats SA is DUs. One or more households may be located in any given DU; recent surveys have found 1.03 households per DU on average.

Administratively, South Africa is divided into nine provinces. The sample for the SADHS 2016 was designed to provide estimates of key indicators for the country as a whole, for urban and non-urban areas separately, and for each of the nine provinces in South Africa. To ensure that the survey precision is comparable across provinces, PSUs were allocated by a power allocation rather than a proportional allocation. Each province was stratified into urban, farm, and traditional areas, yielding 26 sampling strata.

The SADHS 2016 followed a stratified two-stage sample design with a probability proportional to size sampling of PSUs at the first stage and systematic sampling of DUs at the second stage. The Census 2011 DU count was used as the PSU measure of size. A total of 750 PSUs were selected from the 26 sampling strata, yielding 468 selected PSUs in urban areas, 224 PSUs in traditional areas, and 58 PSUs in farm areas.

For further details on sample design, see Appendix A of the final report.

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

Face-to-face [f2f]

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

Five questionnaires were used in the SADHS 2016: the Household Questionnaire, the individual Woman’s Questionnaire, the individual Man’s Questionnaire, the Caregiver’s Questionnaire, and the Biomarker Questionnaire. These questionnaires, based on The DHS Program’s standard Demographic and Health Survey questionnaires, were adapted to reflect the population and health issues relevant to South Africa. Input was solicited from various stakeholders representing government ministries and agencies, nongovernmental organisations, and international donors. After the preparation of the questionnaires in English, the questionnaires were translated into South Africa’s 10 other official languages. In addition, information about the fieldworkers for the survey was collected through a self-administered Fieldworker Questionnaire.

Cleaning operations
---------------------------

All electronic data files for the SADHS 2016 were transferred via the IFSS to the Stats SA head office in Pretoria, where they were stored on a password-protected computer. The data processing operation included secondary editing, which required resolution of computer-identified inconsistencies and coding of open-ended questions. The data were processed by a core group of four people; secondary editing was completed by 11 people. All persons involved in data processing took part in the main fieldwork training, and they were supervised by senior staff from Stats SA with support from ICF. Data editing was accomplished using CSPro software. Secondary editing was initiated in October 2016 and completed in February 2017. Checking inconsistencies in dates of immunisations was aided by the digital images of the immunisation page of the Road-to-Health booklet that had been collected on the tablet by fieldworkers at the time of the interview for that purpose.

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

A total of 15,292 households were selected for the sample, of which 13,288 were occupied. Of the occupied households, 11,083 were successfully interviewed, yielding a response rate of 83%.

In the interviewed households, 9,878 eligible women age 15-49 were identified for individual interviews; interviews were completed with 8,514 women, yielding a response rate of 86%. In the subsample of households selected for the male survey, 4,952 eligible men age 15-59 were identified and 3,618 were successfully interviewed, yielding a response rate of 73%. In this same subsample, 12,717 eligible adults age 15 and older were identified and 10,336 were successfully interviewed with the adult health module, yielding a response rate of 81%. Response rates were consistently lower in urban areas than in nonurban areas.

Sampling error estimates
---------------------------

The estimates from a sample survey are affected by two types of errors: nonsampling errors and sampling errors. Nonsampling errors are the results of mistakes made in implementing data collection and data processing, such as failure to locate and interview the correct household, misunderstanding of the questions on the part of either the interviewer or the respondent, and data entry errors. Although numerous efforts were made during the implementation of the SADHS 2016 to minimize this type of error, nonsampling errors are impossible to avoid and difficult to evaluate statistically.

Sampling errors, on the other hand, can be evaluated statistically. The sample of respondents selected in the SADHS 2016 is only one of many samples that could have been selected from the same population, using the same design and expected size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability among all possible samples. Although the degree of variability is not known exactly, it can be estimated from the survey results.

Sampling error is usually measured in terms of the standard error for a particular statistic (mean, percentage, etc.), which is the square root of the variance. The standard error can be used to calculate confidence intervals within which the true value for the population can reasonably be assumed to fall. For example, for any given statistic calculated from a sample survey, the value of that statistic will fall within a range of plus or minus two times the standard error of that statistic in 95% of all possible samples of identical size and design.

If the sample of respondents had been selected as a simple random sample, it would have been possible to use straightforward formulas for calculating sampling errors. However, the SADHS 2016 sample is the result of a multi-stage stratified design, and, consequently, it was necessary to use more complex formulas. Sampling errors are computed in SAS, using programs developed by ICF. These programs use the Taylor linearization method to estimate variances for survey estimates that are means, proportions, or ratios. The Jackknife repeated replication method is used for variance estimation of more complex statistics such as fertility and mortality rates.

A more detailed description of estimates of sampling errors are presented in Appendix B of the survey final report.

Data appraisal
---------------------------

Data Quality Tables
- Household age distribution
- Age distribution of eligible and interviewed women
- Age distribution of eligible and interviewed men
- Completeness of reporting
- Births by calendar years
- Reporting of age at death in days
- Reporting of age at death in months
- Height and weight data completeness and quality for children
- Completeness of information on siblings
- Sibship size and sex ratio of siblings

See details of the data quality tables in Appendix C of the survey final report.

摘要
---------------------------
南非人口与健康调查（SADHS）2016的主要目标是提供最新的人口与健康基本指标估计值。具体而言，SADHS 2016收集了关于生育水平、婚姻状况、性行为、生育偏好、避孕药具的认知和使用、母乳喂养实践、营养状况、儿童和孕产妇死亡率、孕产妇健康（包括产前和产后护理）、儿童健康的关键方面（包括免疫接种覆盖率、急性呼吸道感染（ARI）、发热和腹泻的流行率和治疗）、HIV感染风险的可能暴露、HIV咨询和检测（HCT）覆盖范围以及针对女性的身体和性暴力等信息。SADHS 2016的另一项重要目标是提供15岁及以上成年人的健康和行为指标估计值，包括烟草、酒精和含可待因药物的使用情况。此外，SADHS 2016还提供了6-59个月儿童和15岁及以上成年人贫血的患病率、高血压、贫血、高HbA1c水平（糖尿病指标）和HIV的患病率估计值。
地理覆盖范围
---------------------------
全国
分析单元
---------------------------
- 家庭
- 个人
- 0-5岁儿童
- 15-49岁女性
- 15-59岁男性

总体
---------------------------
调查涵盖了所有法定家庭成员（常住居民）、0-5岁儿童、15-49岁女性和15-59岁男性。
数据类型
---------------------------
样本调查数据 [ssd]
抽样程序
---------------------------
SADHS 2016使用的抽样框架是南非统计局主样本框架（MSF），该框架是基于2011年人口普查的普查区（EAs）创建的。在MSF中，可管理的EAs被视为一级抽样单位（PSU），相邻的小型EAs被合并形成新的PSU，大型EAs被分割成概念性PSU。该框架包含有关地理类型（城市、传统或农场）以及每个PSU中估计的住宅单元（DUs）数量的信息。Stats SA使用的抽样惯例是DUs。任何给定的住宅单元中可能有一个或多个家庭；最近的研究发现，平均每个住宅单元有1.03个家庭。
行政上，南非分为九个省。SADHS 2016的样本设计旨在为国家整体、城市和非城市地区以及南非的九个省份提供关键指标的估计值。为确保调查精度在各省之间可比，PSU的分配采用幂次分配而非比例分配。每个省份被分为城市、农场和传统地区，产生了26个抽样层。
SADHS 2016采用分层两阶段样本设计，第一阶段对PSU进行按规模成比例的抽样，第二阶段对DUs进行系统抽样。2011年人口普查的DUs计数被用作PSU的规模度量。从26个抽样层中选出了750个PSU，其中城市地区选出了468个PSU，传统地区选出了224个PSU，农场地区选出了58个PSU。
有关样本设计的更多详细信息，请参阅最终报告的附录A。
数据收集方式
---------------------------
面对面 [f2f]
研究工具
---------------------------
SADHS 2016使用了五份问卷：家庭问卷、个人女性问卷、个人男性问卷、照料者问卷和生物标志物问卷。这些问卷基于DHS项目标准人口与健康调查问卷，并根据南非的人口与健康问题进行了调整。征求了代表政府部门和机构、非政府组织和国际捐助者的各利益相关者的意见。在英语问卷准备就绪后，问卷被翻译成南非的10种其他官方语言。此外，还通过自填式调查员问卷收集了有关调查员的信息。
数据清理操作
---------------------------
所有SADHS 2016的电子数据文件都通过IFSS传输到比勒陀利亚的Stats SA总部，在那里它们被存储在密码保护的计算机上。数据处理操作包括二级编辑，需要解决计算机识别的不一致性以及对开放式问题的编码。数据处理由四名核心人员完成；二级编辑由11人完成。所有参与数据处理的人员都参加了主要实地工作培训，并由Stats SA的高级人员监督，ICF提供支持。数据编辑使用CSPro软件完成。二级编辑于2016年10月开始，并于2017年2月完成。通过免疫接种页面的数字图像（这是调查员在采访期间为该目的收集的）检查免疫接种日期的不一致性。
响应率
---------------------------
总共选择了15,292个家庭作为样本，其中13,288个被占用。在这些被占用的家庭中，11,083个家庭成功接受了访谈，响应率为83%。
在访谈的家庭中，确定了9,878名符合条件的15-49岁女性进行个人访谈；完成了8,514名女性的访谈，响应率为86%。在为男性调查选择的子样本中，确定了4,952名符合条件的15-59岁男性，其中3,618人接受了成功访谈，响应率为73%。在相同的子样本中，确定了12,717名符合条件的15岁及以上成年人，其中10,336人成功接受了成人健康模块的访谈，响应率为81%。城市地区的响应率通常低于非城市地区。
抽样误差估计
---------------------------
样本调查的估计值受到两种类型误差的影响：非抽样误差和抽样误差。非抽样误差是实施数据收集和数据处理过程中出现的错误的结果，例如未能找到和访谈正确的家庭、访谈员或受访者对问题的误解以及数据输入错误。尽管在实施SADHS 2016期间做出了众多努力以最大限度地减少此类错误，但非抽样误差是无法避免且难以进行统计分析的。
另一方面，抽样误差可以统计评估。SADHS 2016中选定的受访者样本只是从同一总体中选出的许多样本之一，使用相同的设计和预期规模。每个这样的样本都会产生与实际选定样本的结果略有不同的结果。抽样误差是衡量所有可能样本之间差异的指标。虽然变异的程度无法确切知道，但可以从调查结果中进行估计。
抽样误差通常以特定统计量（均值、百分比等）的标准误差来衡量，这是方差的平方根。标准误差可用于计算置信区间，在此区间内可以合理地假设人口的真实值。
例如，对于从样本调查中计算出的任何给定统计量，该统计量的值将在95%的所有可能样本中落在该统计量标准误差的两倍范围内。
如果受访者样本被选为简单随机样本，则可以使用简单的公式来计算抽样误差。然而，SADHS 2016的样本是多层分层设计的产物，因此有必要使用更复杂的公式。抽样误差使用ICF开发的程序在SAS中计算。这些程序使用泰勒线性化方法来估计均值、比例或比率等调查估计值的方差。对于更复杂的统计量，如生育率和死亡率，使用Jackknife重复复制法进行方差估计。
抽样误差估计的更详细描述请参阅调查最终报告的附录B。
数据评估
---------------------------
数据质量表
- 家庭年龄分布
- 符合条件和接受访谈的女性的年龄分布
- 符合条件和接受访谈的男性的年龄分布
- 报告的完整性
- 日历年度出生情况
- 死亡年龄报告（以天为单位）
- 死亡年龄报告（以月为单位）
- 儿童身高和体重数据完整性和质量
- 兄弟姐妹信息的完整性
- 兄弟姐妹的规模和性别比

有关数据质量表的详细信息，请参阅调查最终报告的附录C。