Demographic and Health Survey 2014 - Kenya

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

The 2014 Kenya Demographic and Health Survey (KDHS) was designed to provide information to monitor and evaluate population and health status in Kenya and to be a follow-up to the previous KDHS surveys. In addition, it provides new information on indicators previously not collected in KDHS surveys, such as fistula and men’s experience of domestic violence. The survey also aims to provide estimates for selected demographic and health indicators at the county level.

The specific objectives of the 2014 KDHS were to:
• Estimate fertility and childhood, maternal, and adult mortality
• Measure changes in fertility and contraceptive prevalence
• Examine basic indicators of maternal and child health
• Collect anthropometric measures for children and women
• Describe patterns of knowledge and behaviour related to transmission of HIV and other sexually transmitted infections
• Ascertain the extent and pattern of domestic violence and female circumcision

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

National coverage

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

- Household
- Individual
- Children age 0-5
- Woman age 15-49
- Man age 15-54

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

Sample survey data [ssd]

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

The sample for the 2014 KDHS was drawn from a master sampling frame, the Fifth National Sample Survey and Evaluation Programme (NASSEP V). This is a frame that the KNBS currently operates to conduct household-based surveys throughout Kenya. Development of the frame began in 2012, and it contains a total of 5,360 clusters split into four equal subsamples. These clusters were drawn with a stratified probability proportional to size sampling methodology from 96,251 enumeration areas (EAs) in the 2009 Kenya Population and Housing Census. The 2014 KDHS used two subsamples of the NASSEP V frame that were developed in 2013. Approximately half of the clusters in these two subsamples were updated between November 2013 and September 2014. Kenya is divided into 47 counties that serve as devolved units of administration, created in the new constitution of 2010. During the development of the NASSEP V, each of the 47 counties was stratified into urban and rural strata; since Nairobi county and Mombasa county have only urban areas, the resulting total was 92 sampling strata.

The 2014 KDHS was designed to produce representative estimates for most of the survey indicators at the national level, for urban and rural areas separately, at the regional (former provincial) level, and for selected indicators at the county level. In order to meet these objectives, the sample was designed to have 40,300 households from 1,612 clusters spread across the country, with 995 clusters in rural areas and 617 in urban areas. Samples were selected independently in each sampling stratum, using a two-stage sample design. In the first stage, the 1,612 EAs were selected with equal probability from the NASSEP V frame. The households from listing operations served as the sampling frame for the second stage of selection, in which 25 households were selected from each cluster.

The interviewers visited only the preselected households, and no replacement of the preselected households was allowed during data collection. The Household Questionnaire and the Woman's Questionnaire were administered in all households, while the Man's Questionnaire was administered in every second household. Because of the non-proportional allocation to the sampling strata and the fixed sample size per cluster, the survey was not self-weighting. The resulting data have, therefore, been weighted to be representative at the national, regional, and county levels.

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

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

Face-to-face [f2f]

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

The 2014 KDHS used a household questionnaire, a questionnaire for women age 15-49, and a questionnaire for men age 15-54. These instruments were based on the model questionnaires developed for The DHS Program, the questionnaires used in the previous KDHS surveys, and the current information needs of Kenya. During the development of the questionnaires, input was sought from a variety of organisations that are expected to use the resulting data. A two-day workshop involving key stakeholders was held to discuss the questionnaire design.

A total of five questionnaires were used in the 2014 KDHS: (1) a full Household Questionnaire, (2) a short Household Questionnaire, (3) a full Woman’s Questionnaire, (4) a short Woman’s Questionnaire, and (5) a Man’s Questionnaire. The 2014 KDHS sample was divided into halves. In one half, households were administered the full Household Questionnaire, the full Woman’s Questionnaire, and the Man’s Questionnaire. In the other half, households were administered the short Household Questionnaire and the short Woman’s Questionnaire. Selection of these subsamples was done at the household level - within a cluster, one in every two households was selected for the full questionnaires, and the remaining households were selected for the short questionnaires.

The Household Questionnaire was used to list all of the usual members of the household and visitors who stayed in the household the night before the survey. One of the main purposes of the Household Questionnaire was to identify women and men who were eligible for the individual interview. Some basic information was collected on the characteristics of each person listed, including age, sex, education, and relationship to the head of the household. The Household Questionnaire also collected information on characteristics of the household’s dwelling unit, such as the source of water, type of toilet facilities, materials used for the floor and roof of the house, ownership of various durable goods, and ownership and use of mosquito nets. In addition, this questionnaire was used to record height and weight measurements of women age 15-49 and children under age 5.

The Woman’s Questionnaires were used to collect information from women age 15-49.

The Man’s Questionnaire was administered to men age 15-54 living in every second household in the sample. The Man’s Questionnaire collected information similar to that contained in the Woman’s Questionnaire but was shorter because it did not contain questions on maternal and child health, nutrition, adult and maternal mortality, or experience of female circumcision or fistula.

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

Completed questionnaires were sent to the KNBS Data Processing Centre in Nairobi. Office editors who received the questionnaires verified cluster and household numbers to ensure that they were consistent with the sampled list. They also ensured that each cluster had 25 households and that all questionnaires for a particular household were packaged together.

Data entry began on May 28, 2014, with a four-day training session and continued until November 21, 2014. All data were double entered (100 percent verification) using CSPro software. The data processing team included 42 keyers, three office editors, two secondary editors, four supervisors, and one data manager. Secondary editing, which included further data cleaning and validation, ran simultaneously with data entry and was completed on January 28, 2015, in collaboration with ICF International. The KDHS Key Indicators Report was prepared and launched in April 2015.

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

A total of 39,679 households were selected for the sample, of which 36,812 were found occupied at the time of the fieldwork. Of these households, 36,430 were successfully interviewed, yielding an overall household response rate of 99 percent. The shortfall of households occupied was primarily due to structures that were found to be vacant or destroyed and households that were absent for an extended period of time.

As noted, the 2014 KDHS sample was divided into halves, with one half of households receiving the full Household Questionnaire, the full Woman’s Questionnaire, and the Man’s Questionnaire and the other half receiving the short Household Questionnaire and the short Woman’s Questionnaire. The household response rate for the full Household Questionnaire was 99 percent, as was the household response rate for the short Household Questionnaire.

In the households selected for and interviewed using the full questionnaires, a total of 15,317 women were identified as eligible for the full Woman’s Questionnaire, of whom 14,741 were interviewed, generating a response rate of 96 percent. A total of 14,217 men were identified as eligible in these households, of whom 12,819 were successfully interviewed, generating a response rate of 90 percent.

In the households selected for and interviewed with the short questionnaires, a total of 16,855 women were identified as eligible for the short Woman’s Questionnaire, of whom 16,338 were interviewed, yielding a response rate of 97 percent.

Response rates are lower in the urban sample than in the rural sample, more so for men. The principal reason for non-response among both eligible men and eligible women was failure to find them at home despite repeated visits to the household. The lower response rates for men reflect the more frequent and longer absences of men from the household

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

The estimates from a sample survey are affected by two types of errors: non-sampling errors and sampling errors. Non-sampling 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 2014 Kenya Demographic and Health Survey (2014 KDHS) to minimise this type of error, non-sampling 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 2014 KDHS 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 between 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 percent 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 2014 KDHS sample is the result of a multi-stage stratified design, and, consequently, it was necessary to use more complex formulae. Sampling errors are computed in either ISSA or SAS, using programs developed by ICF Macro. These programs use the Taylor linearisation method of variance estimation 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.

The Taylor linearisation method treats any percentage or average as a ratio estimate, r = y/x, where y represents the total sample value for variable y, and x represents the total number of cases in the group or subgroup under consideration.

Note: A more detailed description of estimate of sampling error is presented in APPENDIX B of the survey 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
- Nutritional status of children based on the NCHS/CDC/WHO International Reference Population
- Completeness of information on siblings
- Sibship size and sex ratio of siblings

Note: See detailed data quality tables in APPENDIX C of the report.

摘要
---------------------------
2014年肯尼亚人口与健康调查（KDHS）旨在为肯尼亚监测和评估人口与健康状况提供信息，并作为之前KDHS调查的后续研究。此外，它还提供了以前在KDHS调查中未收集的新信息，例如产道裂伤和男性遭受家庭暴力的经历。调查还旨在为县一级的选定人口与健康指标提供估计值。
2014年KDHS的具体目标如下：
• 估计生育率、儿童、母亲和成人死亡率
• 衡量生育率和避孕率的变化
• 检查母亲和儿童健康的基本指标
• 收集儿童和女性的体质测量数据
• 描述与HIV和其他性传播疾病传播相关的知识与行为模式
• 确定家庭暴力和女性割礼的广泛程度和模式

地理覆盖范围
---------------------------
全国覆盖

分析单位
---------------------------
- 家庭
- 个人
- 0-5岁儿童
- 15-49岁女性
- 15-54岁男性

数据类型
---------------------------
样本调查数据 [ssd]

抽样程序
---------------------------
2014年KDHS的样本是从主抽样框架——第五次全国样本调查与评估计划（NASSEP V）中抽取的。这是一个KNBS目前运营的框架，用于在肯尼亚进行基于家庭的调查。该框架的开发始于2012年，包含总计5,360个群组，分为四个相等的子样本。这些群组采用分层概率成比例抽样方法从2009年肯尼亚人口与住房普查中的96,251个统计区域（EAs）中抽取。2014年KDHS使用了2013年开发的NASSEP V框架的两个子样本。在这两个子样本中，大约一半的群组在2013年11月至2014年9月之间进行了更新。肯尼亚分为47个县，作为2010年新宪法中设立的行政自治单位。在NASSEP V的开发过程中，每个47个县都被划分为城市和农村层；由于内罗毕县和蒙巴萨县只有城市地区，因此总数为92个抽样层。
2014年KDHS旨在为国家层面的大多数调查指标产生代表性估计值，分别为城市和农村地区分别，以及地区（前省）层面，并针对选定指标在县一级。为了实现这些目标，样本被设计为有来自全国1,612个群组的40,300个家庭，其中农村地区有995个群组，城市地区有617个群组。每个抽样层独立选择样本，采用两阶段样本设计。在第一阶段，从NASSEP V框架中按等概率选择了1,612个EAs。清单操作中的家庭作为第二阶段选择的抽样框架，在该阶段，每个群组中选择了25个家庭。
调查员只访问预选的家庭，在数据收集过程中不允许替换预选的家庭。所有家庭都进行了家庭问卷和女性问卷的发放，而男性问卷则每两个家庭发放一次。由于抽样层的非比例分配和每个群组的固定样本量，该调查没有自我加权。因此，结果数据已被加权，以在全国、地区和县一级具有代表性。
有关样本选择的更多详细信息，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
---------------------------
2014年KDHS使用了家庭问卷、15-49岁女性问卷和15-54岁男性问卷。这些工具基于为DHS项目开发的模型问卷、之前KDHS调查中使用的问卷以及肯尼亚当前的信息需求。在问卷开发过程中，从预期将使用这些数据的各种组织中征求了意见。举办了一次为期两天的研讨会，涉及关键利益相关方，以讨论问卷设计。
2014年KDHS共使用了五份问卷：（1）完整的家庭问卷，（2）简短的家庭问卷，（3）完整的女性问卷，（4）简短的女性问卷，以及（5）男性问卷。2014年KDHS样本分为两半。在另一半中，家庭接受了简短的家庭问卷和简短的女性问卷。在群组层面进行这些子样本的选择——每个群组中每两个家庭中的一个被选为完整的问卷，其余家庭被选为简短问卷。
家庭问卷用于列出所有家庭中的常住成员和在前一晚调查中留宿的访客。家庭问卷的主要目的之一是识别符合个人访谈资格的女性和男性。收集了列出的每个人的基本特征信息，包括年龄、性别、教育和与户主的关系。家庭问卷还收集了家庭居住单位特征的信息，例如水源、厕所设施类型、房屋地板和屋顶使用的材料、各种耐用品的所有权以及蚊帐的所有权和使用情况。此外，此问卷还用于记录15-49岁女性和5岁以下儿童的身高和体重测量值。
女性问卷用于收集15-49岁女性的信息。
男性问卷用于调查样本中每第二个家庭中的15-54岁男性。男性问卷收集了与女性问卷中包含的信息类似的信息，但较短，因为它不包含有关母亲和儿童健康、营养、成人和母亲死亡率，或女性割礼或产道裂伤的经历的问题。

数据清理操作
---------------------------
完成后的问卷被发送到内罗毕的KNBS数据处理中心。收到问卷的办公室编辑核实了群组和家庭编号，以确保它们与抽样列表一致。他们还确保每个群组有25个家庭，并将所有特定家庭的问卷打包在一起。
数据输入于2014年5月28日开始，持续到2014年11月21日。所有数据均使用CSPro软件进行了双录入（100％验证）。数据处理团队包括42名键入员、3名办公室编辑、2名二级编辑、4名监督员和1名数据经理。二级编辑（包括进一步的数据清理和验证）与数据输入同时进行，并于2015年1月28日与ICF国际公司合作完成。KDHS关键指标报告于2015年4月准备和发布。

回应率
---------------------------
总共选择了39,679个家庭作为样本，其中在实地工作时发现了36,812个家庭被占用。在这些家庭中，有36,430个家庭成功接受了访谈，从而产生了99％的整体家庭回应率。家庭被占用的不足主要是由于发现空置或被破坏的结构以及长时间缺席的家庭。
如前所述，2014年KDHS样本分为两半，其中一半的家庭接受了完整的家庭问卷、完整的女性问卷和男性问卷，另一半的家庭接受了简短的家庭问卷和简短的女性问卷。完整的家庭问卷的家庭回应率为99％，简短的家庭问卷的家庭回应率也是如此。
在接受了完整问卷并接受访谈的家庭中，共确定了15,317名符合条件的女性接受完整的女性问卷，其中14,741人接受了访谈，产生了96％的回应率。在这些家庭中，共确定了14,217名符合条件的男性，其中12,819人成功接受了访谈，产生了90％的回应率。
在选择了简短问卷并接受访谈的家庭中，共确定了16,855名符合条件的女性接受简短的女性问卷，其中16,338人接受了访谈，产生了97％的回应率。
城市样本的回应率低于农村样本，男性尤其如此。合格男性和合格女性非回应的主要原因是在反复访问家庭后未能找到他们。男性的低回应率反映了他们更频繁和更长时间的家庭缺席。

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。非抽样误差是数据收集和数据处理中实施错误的结果，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2014年肯尼亚人口与健康调查（2014 KDHS）期间做出了许多努力以最大限度地减少此类错误，但非抽样误差是不可避免的，并且难以进行统计评估。
另一方面，抽样误差可以统计评估。2014 KDHS中选定的受访者样本只是从同一人口中，使用相同的设计和预期规模可以选出的许多样本之一。这些样本中的每一个都会产生与实际选定的样本结果略有不同的结果。抽样误差是衡量所有可能样本之间变异性的指标。尽管变异程度无法确切知道，但可以从调查结果中估计出来。
抽样误差通常以特定统计量（平均值、百分比等）的标准误差来衡量，它是方差的平方根。标准误差可用于计算置信区间，其中可以合理地假设真实值会落在该区间内。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（具有相同大小和设计）的标准误差的两倍范围内。
如果受访者样本是简单随机样本，则可以使用简单的公式来计算抽样误差。然而，2014 KDHS样本是多层次分层设计的产物，因此有必要使用更复杂的公式。抽样误差在ISSA或SAS中使用ICF Macro开发的程序进行计算。这些程序使用泰勒线性化方法对调查估计进行方差估计，这些估计值是平均值、比例或比率。对于更复杂的统计量，如生育率和死亡率，使用Jackknife重复复制方法进行方差估计。
泰勒线性化方法将任何百分比或平均值视为比率估计，r = y/x，其中y代表变量y的总样本值，x代表考虑的群体或子群体中的总案例数。
注意：有关抽样误差估计的更详细描述，请参阅调查报告的附录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
- 按日历年份的出生
- 死亡年龄按天的报告
- 死亡年龄按月的报告
- 基于NCHS/CDC/WHO国际参考人群的儿童营养状况
- 兄弟姐妹信息的完整性
- 兄弟姐妹的家族大小和性别比

注意：有关详细数据质量表的更多信息，请参阅报告的附录C。