Demographic and Health Survey 1992 - Egypt, Arab Rep.

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

The 1992 Egypt Demographic and Health Survey (EDHS) is the most recent in a series of surveys carded out in Egypt to provide information on fertility and child mortality levels, family planning awareness, approval and use and basic indicators of maternal and child health. The EDHS findings are important in monitoring trends in these variables and in understanding the factors which contribute to differentials in fertility and contraceptive use among various population subgroups. The EDHS also provides a wealth of health-related information for mothers and their children. These data are especially important for understanding the factors that influence the health and survival of infants and young children. In addition to providing insights into population and health issues in Egypt, the EDHS also hopefully will lead to an improved global understanding of population and health problems as it is one of more than 50 surveys implemented through the Demographic and Health Surveys program.

The primary objective of the EDHS is to provide data on fertility and mortality, family planning and maternal and child health. The survey obtained detailed information on these issues from a sample of ever-married women in the reproductive ages. In addition, a subsample of husbands was interviewed in an effort to obtain information on their fertility preferences and the role which they play in family planning decision making.

The EDHS information is intended to assist policymakers and administrators to evaluate existing programs and to design new strategies for improving family planning and health services in Egypt. A secondary objective is to enhance the capabilities of institutions in Egypt to collect process and analyze population and health data so as to facilitate the implementation of future surveys of this type.

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

National

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

- Household
- Children under five years
- Women age 15-49
- Men

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

Sample survey data

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

Sample Design

The 1992 Egypt Demographic and Health Survey covered over 11,000 households, which were scattered in 21 governorates. One of the key concerns in the design of the sample was the need 1o provide reliable estimates of fertility levels and contraceptive use for Egypt as a whole, and for urban and rural areas separately. Other domains for which reliable estimates were desired included the Urban Governorates, Upper Egypt, and Lower Egypt. In addition, estimates of key indicators for the women's sample were needed at governorate level. In order to allow for the governorate-level estimates, the number of households selected from each governorate is disproportionate to the size of the population in the governorate; thus, the EDHS sample is not self-weighting at the national level.

The 1992 EDHS sample was selected in three stages. The sampling units at the first stage were shiakhas/towns in urban areas and villages in rural areas. The frame for the selection of these primary sampling units (PSUs) was based on 1986 census data, which were provided by the Central Agency for Public Mobilization and Statistics (CAPMAS). During the first stage selection, 377 PSUs were sampled (169 in urban areas and 208 in rural areas).

The second stage of selection involved several steps. First, maps were obtained for each shiakha and village that had been selected at the first stage and divided into a number of roughly equal-sized parts. One of the pans was then selected from each PSU. In both urban and rural PSUs, a quick-count operation was carried out in the field to provide the information which was used to divide the selected pan into a number of segments of roughly equal size. Two segments from urban areas and one segment from rural areas were then chosen as the secondary sampling units.

After the secondary sampling units (SSUs) were selected, a household listing was obtained for each SSU. Using the household listing, a systematic random sample of households was chosen for the EDHS. A subsample of one-third of the households in every segment was selected for the husband survey. All ever-married women 15-49 who were present in the household on the night before the interview were eligible for the survey. The husbands' sample covered men who were currently married to eligible women.

Sample Implementation

Two different field operations were conducted during the sample implementation phase of the 1992 EDHS. A quick count for the PSUs selected in shiakhas/towns and villages was the first field operation. The objective of the quick count was to obtain an estimate of the number of households in the part to serve as the measure of size for the second stage selection.

Experience in the 1988 EDHS, in which a quick-count operation was carried out in only the urban PSUs, indicated that there was frequently significant variation between the target and the actual number of households in rural areas. This variation was largely due to the imprecision in assigning measures of size in some rural PSUs, which involved measuring the residential area on a map, many of which were out of date. Therefore, it was decided to carry out a quick count in both urban and rural areas. Prior to the quick-count operation, maps were obtained for each shiakha or town selected for the urban sample and for villages included in the rural sample that had more than 20,000 populations. These maps were divided into approximately equal-sized parts, and one part was randomly selected for the quick-count operation. For villages with less than 20,000 populations, the quick count was carried out for the entire village. It should be noted that the quick count for a rural area covered both the main village and all associated hamlets.

The one-week training course held prior to the quick-count field operation included both classroom instruction and practical training in shiakhas and villages not covered in the survey. The quick-count operation, which covered all 377 PSUs, was carried out between mid-May and mid-July 1993. A group of 52 field staff participated in the quick-count operation. The staff was divided into 15 teams, each composed of one supervisor, one cartographer and one counter.

As a quality control measure, 10 percent of the parts were selected, and a second count obtained. If the difference between the first and second counts was within 2 percent, the first count was accepted; otherwise, another visit was made to the field to resolve the discrepancy between the two counts. There were only a few cases in which a third visit was required.

The second field operation during the sample implementation phase involved a complete listing of all of the households living in the 546 segments chosen during the second stage of the sample selection. Prior to the household listing, 38 listing staff attended a one week training course, which involved both classroom lectures and field practice. After the training, 11 listing teams were formed. Each team consisted of a supervisor and two listers.

The listing operation started on September 10th and was completed by mid-October. Segments were relisted when the number of household in the listing differed markedly from that expected based on the quick count figures.

Note: See detailed description of sample design in APPENDIX B of the report which is presented in this documentation.

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

Face-to-face

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

The 1992 EDHS involved three types of questionnaires: a household questionnaire, an individual questionnaire for women, and an individual questionnaire for husbands. These questionnaires were based on the model survey instruments developed for the international Demographic and Health Surveys program. In particular, the household and women's questionnaires were built on the DHS model "A" questionnaire for high contraceptive prevalence countries. Additional questions on a number of topics not covered in the DHS model questionnaire were included in both the household and individual questionnaires.

The questionnaires were pretested in May 1992, following a two-week training course for supervisors and interviewers. Two supervisors, two field editors and ten interviewers participated in the pretest. Interviewer comments and tabulations of the pretest results were reviewed during the process of modifying the questionnaires.

The household questionnaire obtained a listing of all usual household members and visitors and identified those present in the household during the night before the interviewer's visit. For each of the individuals included in the listing, information was collected on the relationship to the household head, age, sex, marital status, educational level, occupation and work status. Finally, the household questionnaire also included questions on characteristics of the physical and social environment of the household (e.g., availability of electricity, source of drinking water, household possessions, etc.), which are assumed to be related to the health and socioeconomic status of the household.

The individual questionnaire for women was administered to all ever-married women age 15-49. It obtained information on the following topics:
- Background characteristics
- Reproduction
- Knowledge and use of family planning
- Other issues relating to contraception
- Fertility preferences
- Maternal care and breastfeeding
- Immunization and health
- Marriage
- Husband's background, residence and women's work

The women's questionnaire included a monthly calendar, which was used to record fertility, contraceptive use, marriage, spousal absence, migration, and employment histories for a nearly six-year period beginning in January 1987. In addition, the interviewing teams measured the height and weight of all children who were born since January 1987 and of their mothers.

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

Office Editing

The central office of the EDHS was responsible for collecting questionnaires from supervisors as soon as a cluster was completed. Questionnaires were coded and reviewed for consistency and completeness by office editors. To provide feedback for the field teams, the office editors were asked to write a summary report of problems for each PSU. The report was then reviewed by one of the two senior staff assigned to supervise the work of the office editors. When there were serious errors found in one or more questionnaires from a cluster, the team supervisor was contacted in order to ensure that the problem would not occur in other clusters in which the team was working.

Machine Entry and Editing

The data were entered and edited on microcomputers using the Integrated System for Survey Analysis (ISSA), a package program specially developed to process DHS data. ISSA allows range, skip and most consistency errors to be detected and corrected at the entry stage, substantially reducing the time required for the processing of data. The machine entry and editing phase began while interviewing teams were still in the field. The data entry personnel used eight IBM-compatible microcomputers to process the EDHS questionnaires. During the machine entry 20 percent of each segment was reentered for verification. One of the computers was assigned for this process. Working six days per week in two shifts, the data entry personnel completed the machine entry and editing of the data in three months.

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

A total of 11,304 households were selected; of these households, 10,760 were successfully interviewed. As noted, an eligible respondent for the women's survey was defined as ever-married women between the age of 15 and 49 years present in the household on the night before the interview. A total of 9,978 eligible respondents were identified, and of these women, 9,864 (99 percent) were interviewed.

A total of 3,027 men were identified as eligible for the husbands' survey, i.e., they were resident in a household selected for the husband subsample and married to a woman between the ages of 15 and 49. Of the eligible men, 2,466 were successfully interviewed (82 percent).

Note: See summarized response rates by place of residence in Table 1.2 of the report which is presented in this documentation.

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

The results from sample surveys are affected by two types of errors, non-sampling error and sampling error. Non-sampling error is due to mistakes made in carrying out field activities, such as failure to locate and interview the correct household, errors in the way the questions are asked, misunderstanding on the part of either the interviewer or the respondent, data entry errors, etc. Although efforts were made during the design and implementation of the EDHS to minimize this type of error, non-sampling errors are impossible to avoid and difficult to evaluate statistically.

Sampling errors, on the other hand, can be measured statistically. The sample of women selected in the EDHS is only one of many samples that could have been selected from the same population, using the same design and expected size. Each one would have yielded results that differed somewhat from the actual sample selected. The sampling error is a measure of the variability between all possible samples; although it is not known exactly, it can be estimated from the survey results.

Sampling error is usually measured in terms of standard error of 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, apart from non-sampling errors, 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 same statistic as measured in 95 percent of all possible samples with the same design (and expected size) will fall within a range of plus or minus two times the standard error of that statistic.

If the sample of women had been selected as a simple random sample, it would have been possible to use straightforward formulas for calculating sampling errors. However, the EDHS sample design depended on stratification, stages and clusters. Consequently, it was necessary to utilize more complex formulas. The computer package CLUSTERS, developed for the World Fertility Survey program by the International Statistical Institute, was used to assist in computing the sampling errors with the proper statistical methodology.

Note: See detailed estimate of sampling error calculation in APPENDIX C of the report which is presented in this documentation.

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

Data Quality Tables
- Household age distribution
- Age distribution of eligible and interviewed women
- Completeness of reporting
- Births by calendar year since birth
- Reporting of age at death in days
- Reporting of age at death in months

Note: See detailed tables in APPENDIX D of the report which is presented in this documentation.

摘要
---------------------------
1992年埃及人口与健康状况调查（EDHS）是该系列调查中最新的一个，旨在提供关于生育率和儿童死亡率、家庭计划意识和采纳情况，以及母婴健康的基本指标。EDHS的结果对于监测这些变量的趋势以及理解导致不同人口子群体间生育率和避孕使用差异的因素具有重要意义。EDHS还提供了大量与母亲及其孩子健康相关的信息。这些数据特别有助于理解影响婴儿和幼儿健康和生存的因素。除了为埃及的人口与健康问题提供洞见外，EDHS还有望促进全球对人口与健康问题的理解，因为它是由人口与健康状况调查计划实施的超过50项调查之一。

EDHS的主要目标是提供关于生育率和死亡率、家庭计划和母婴健康的数据。调查从样本中的已婚育龄妇女处获得了关于这些问题的详细信息。此外，还采访了一部分男性，以获取他们关于生育偏好以及在家庭计划决策中所起作用的详细信息。

EDHS的信息旨在帮助政策制定者和管理者评估现有计划，并为在埃及改进家庭计划和卫生服务设计新的战略。次要目标是通过提高埃及机构收集、处理和分析人口与健康数据的能力，以便促进此类未来调查的实施。

地理覆盖范围
---------------------------
全国

分析单元
---------------------------
- 家庭
- 5岁以下儿童
- 15-49岁妇女
- 男性

数据类型
---------------------------
样本调查数据

抽样程序
---------------------------
样本设计

1992年埃及人口与健康状况调查覆盖了超过11,000个家庭，这些家庭分布在21个省份。在设计样本时，一个关键关注点是提供关于埃及整体、城市和农村地区的生育率水平和避孕使用情况的可靠估计。其他需要可靠估计的领域包括城市省份、上埃及和下埃及。此外，还需要在省份层面获得妇女样本的关键指标估计。为了允许进行省份层面的估计，每个省份选定的家庭数量与其人口规模不成比例；因此，EDHS样本在国家层面不是自加权。

1992年EDHS样本的选定分为三个阶段。第一阶段的选择单元是城市地区的沙基亚/城镇和农村地区的村庄。选择这些一级抽样单位（PSU）的框架基于由公共动员和统计中央局（CAPMAS）提供的1986年人口普查数据。在第一阶段的选择中，共抽样377个PSU（城市地区169个，农村地区208个）。

第二阶段的选择涉及几个步骤。首先，为第一阶段选定的每个沙基亚和村庄获得了地图，并将其划分为若干个大致相等大小的部分。然后从每个PSU中选出一个部分。在城市和农村PSU中，进行了一次快速计数操作，以提供用于将选定的部分划分为若干个大致相等大小段的信息。然后从城市地区选择了两个段，从农村地区选择了两个段作为二级抽样单位。

在选择了二级抽样单位（SSU）后，为每个SSU获得了家庭清单。使用家庭清单，对EDHS选择了系统随机抽样的家庭。每个段中三分之一的家庭被选为丈夫调查的子样本。所有在采访前一天晚上在家庭中出现的15-49岁已婚妇女都有资格参加调查。丈夫样本涵盖了与合格妇女已婚的男性。

样本实施

在1992年EDHS样本实施阶段进行了两次不同的实地操作。对沙基亚/城镇和村庄中选定的PSU进行的快速计数是第一次实地操作。快速计数的目的是获得用于第二阶段选择的该部分家庭数量的估计值。

1988年EDHS的经验表明，在仅在城市PSU进行快速计数的调查中，目标和实际家庭数量之间经常存在显著差异。这种差异主要归因于某些农村PSU分配测量大小的不精确性，这涉及到在地图上测量住宅区域，其中许多已过时。因此，决定在城乡地区都进行快速计数。在快速计数操作之前，为城市样本中选定的每个沙基亚或城镇以及农村样本中包括的超过20,000人口的城市获得了地图。这些地图被划分为大致相等大小的部分，然后随机选择一个部分进行快速计数操作。对于人口少于20,000的村庄，对整个村庄进行了快速计数。应注意的是，农村地区的快速计数包括主村庄和所有相关村庄。

在快速计数实地操作之前举行的一周培训课程包括课堂指导和在调查未涉及的沙基亚和村庄的实地培训。快速计数操作在1993年5月中旬至7月中旬进行，覆盖了所有377个PSU。一支由52名实地工作人员组成的团队参与了快速计数操作。工作人员分为15个小组，每个小组由一名监督员、一名制图员和一名计数员组成。

作为一种质量控制措施，从10％的部分中选出了样本，并进行了第二次计数。如果第一次和第二次计数之间的差异在2％以内，则接受第一次计数；否则，将再次到实地解决两次计数之间的差异。只有少数情况需要第三次访问。

样本实施阶段的第二次实地操作涉及对第二阶段样本选择过程中选定的546个段中所有居住的家庭进行完整清单。在家庭清单之前，38名清单工作人员参加了一周培训课程，该课程包括课堂讲座和实地练习。培训后，组成了11个清单小组。每个小组由一名监督员和两名清单员组成。

清单操作于9月10日开始，于10月中旬完成。当清单中的家庭数量与根据快速计数数字预期的数量明显不同时，将重新进行清单。

注意：有关样本设计的详细描述，请参阅本文档中提供的报告附录B。

数据收集方式
---------------------------
面对面

研究工具
---------------------------
1992年EDHS涉及三种类型的问卷：家庭问卷、妇女个人问卷和丈夫个人问卷。这些问卷基于为国际人口与健康状况调查计划开发的模型调查工具。特别是，家庭和妇女问卷是基于DHS模型“A”问卷为高避孕率国家设计的。在家庭和个体问卷中都包括了DHS模型问卷未涉及的一些主题的附加问题。

问卷于1992年5月进行了预测试，当时为监督员和采访者举办了一周培训课程。两名监督员、两名实地编辑人员和十名采访者参加了预测试。在修改问卷的过程中，审查了采访者的评论和预测试结果的汇总。

家庭问卷获得了所有常规家庭成员和访客的清单，并确定了在采访者访问前一天晚上在家庭中的那些人。对于清单中的每个人，收集了与家庭户主的关系、年龄、性别、婚姻状况、教育水平、职业和工作状况等信息。最后，家庭问卷还包含了有关家庭物理和社会环境特征的问题（例如，电力供应、饮用水来源、家庭财产等），这些特征被认为与家庭的健康和社会经济状况有关。

妇女个人问卷对所有15-49岁的已婚妇女进行。它获取以下主题的信息：
- 背景特征
- 再生产
- 家庭计划知识和使用
- 与避孕相关的其他问题
- 生育偏好
- 母亲护理和母乳喂养
- 免疫接种和健康
- 婚姻
- 丈夫的背景、居住地和妇女的工作

妇女问卷包括一个月度日历，用于记录从1987年1月开始近六年的生育、避孕使用、婚姻、配偶缺席、迁移和就业历史。此外，采访团队还测量了自1987年1月出生的所有儿童及其母亲的高度和体重。

数据清理操作
---------------------------
办公室编辑

EDHS的中央办公室负责在集群完成后尽快收集监督员的手头问卷。办公室编辑对问卷进行编码并审查其一致性和完整性。为了为实地团队提供反馈，要求办公室编辑为每个PSU编写一份问题摘要报告。然后，由负责监督办公室编辑工作的两名高级工作人员之一审查报告。当发现一个或多个问卷存在严重错误时，将联系团队监督员，以确保不会在其他团队正在工作的集群中发生类似问题。

机器输入和编辑

使用集成调查分析系统（ISSA）在微计算机上输入和编辑数据，ISSA是一个专门为处理DHS数据开发的软件包程序。ISSA允许在输入阶段检测和纠正范围、跳过和大多数一致性错误，从而显著减少了数据处理所需的时间。机器输入和编辑阶段在采访团队仍在实地时开始。数据输入人员使用八台IBM兼容微计算机处理EDHS问卷。在机器输入过程中，每个段20％的数据被重新输入以进行验证。其中一台计算机被分配用于此过程。每周工作六天，分两班工作，数据输入人员在三个月内完成了数据的机器输入和编辑。

响应率
---------------------------
共选定了11,304个家庭；在这些家庭中，10,760个家庭成功接受了采访。如前所述，合格的女性调查受访者被定义为在采访前一天晚上在家庭中出现的15-49岁已婚妇女。共确定了9,978名合格受访者，其中9,864名（99％）的妇女接受了采访。

共确定了3,027名男性符合丈夫调查的资格，即他们居住在为丈夫子样本选定的家庭中，并与15-49岁的妇女结婚。在合格男性中，2,466名成功接受了采访（82％）。

注意：请参阅本文档中提供的报告中第1.2表总结的按居住地划分的响应率。

抽样误差估计
---------------------------
样本调查的结果受到两种类型误差的影响，非抽样误差和抽样误差。非抽样误差是由于在执行实地活动时犯的错误造成的，例如未能找到和采访正确的家庭、提问方式错误、采访者或受访者方面的误解、数据输入错误等。尽管在设计实施EDHS的过程中努力减少此类错误，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，抽样误差可以通过统计方法进行测量。EDHS中选定的妇女样本只是从同一人口中可能选出的许多样本之一，使用相同的设计和预期规模。每个样本都会产生与实际样本选定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；尽管它并不完全清楚，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，在此区间内，除了非抽样误差外，可以合理地假设人口的真正值。例如，对于从样本调查中计算出的任何给定统计量，该统计量在相同设计（和预期规模）的所有可能样本中测量的值在95％的情况下将落在加减该统计量标准误差两倍的范围之内。

如果妇女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，EDHS样本设计依赖于分层、阶段和集群。因此，有必要利用更复杂的公式。用于世界生育率调查计划的国际统计研究所开发的计算机程序CLUSTERS被用来协助使用正确的统计方法计算抽样误差。

注意：请参阅本文档中提供的报告中附录C的抽样误差计算的详细估计。

数据评估
---------------------------
数据质量表
- 家庭年龄分布
- 合格和接受采访的妇女年龄分布
- 报告的完整性
- 自出生以来的按日历年出生
- 死亡年龄按天报告
- 死亡年龄按月报告

注意：请参阅本文档中提供的报告中附录D的详细表格。