National Demographic Survey 1993 - Philippines

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

The 1993 National Demographic Survey (NDS) is a nationally representative sample survey of women age 15-49 designed to collect information on fertility; family planning; infant, child and maternal mortality; and maternal and child health. The survey was conducted between April and June 1993. The 1993 NDS was carried out by the National Statistics Office in collaboration with the Department of Health, the University of the Philippines Population Institute, and other agencies concerned with population, health and family planning issues. Funding for the 1993 NDS was provided by the U.S. Agency for International Development through the Demographic and Health Surveys Program.

Close to 13,000 households throughout the country were visited during the survey and more than 15,000 women age 15-49 were interviewed. The results show that fertility in the Philippines continues its gradual decline. At current levels, Filipino women will give birth on average to 4.1 children during their reproductive years, 0.2 children less than that recorded in 1988. However, the total fertility rate in the Philippines remains high in comparison to the level achieved in the neighboring Southeast Asian countries.

The primary objective of the 1993 NDS is to provide up-to-date inform ation on fertility and mortality levels; nuptiality; fertility preferences; awareness, approval, and use of family planning methods; breastfeeding practices; and maternal and child health. This information is intended to assist policymakers and administrators in evaluating and designing programs and strategies for improving health and family planning services in 'the country.

MAIN RESULTS

Fertility varies significantly by region and socioeconomic characteristics. Urban women have on average 1.3 children less than rural women, and uneducated women have one child more than women with college education. Women in Bicol have on average 3 more children than women living in Metropolitan Manila.

Virtually all women know of a family planning method; the pill, female sterilization, IUD and condom are known to over 90 percent of women. Four in 10 married women are currently using contraception. The most popular method is female sterilization ( 12 percent), followed by the piU (9 percent), and natural family planning and withdrawal, both used by 7 percent of married women.

Contraceptive use is highest in Northern Mindanao, Central Visayas and Southern Mindanao, in urban areas, and among women with higher than secondary education. The contraceptive prevalence rate in the Philippines is markedly lower than in the neighboring Southeast Asian countries; the percentage of married women who were using family planning in Thailand was 66 percent in 1987, and 50 percent in Indonesia in 199l.

The majority of contraceptive users obtain their methods from a public service provider (70 percent). Government health facilities mainly provide permanent methods, while barangay health stations or health centers are the main sources for the pill, IUD and condom.

Although Filipino women already marry at a relatively higher age, they continue to delay the age at which they first married. Half of Filipino women marry at age 21.6. Most women have their first sexual intercourse after marriage.

Half of married women say that they want no more children, and 12 percent have been sterilized. An additional 19 percent want to wait at least two years before having another child. Almost two thirds of women in the Philippines express a preference for having 3 or less children. Results from the survey indicate that if all unwanted births were avoided, the total fertility rate would be 2.9 children, which is almost 30 percent less than the observed rate,

More than one quarter of married women in the Philippines are not using any contraceptive method, but want to delay their next birth for two years or more (12 percent), or want to stop childbearing (14 percent). If the potential demand for family planning is satisfied, the contraceptive prevalence rate could increase to 69 percent. The demand for stopping childbearing is about twice the level for spacing (45 and 23 percent, respectively).

Information on various aspects of maternal and child health---antenatal care, vaccination, breastfeeding and food supplementation, and illness was collected in the 1993 NDS on births in the five years preceding the survey. The findings show that 8 in 10 children under five were bom to mothers who received antenatal care from either midwives or nurses (45 percent) or doctors (38 percent). Delivery by a medical personnel is received by more than half of children born in the five years preceding the survey. However, the majority of deliveries occurred at home.

Tetanus, a leading cause of infant deaths, can be prevented by immunization of the mother during pregnancy. In the Philippines, two thirds of bitlhs in the five years preceding the survey were to mothers who received a tetanus toxoid injection during pregnancy.

Based on reports of mothers and information obtained from health cards, 90 percent of children aged 12-23 months have received shots of the BCG as well as the first doses of DPT and polio, and 81 percent have received immunization from measles. Immunization coverage declines with doses; the drop out rate is 3 to 5 percent for children receiving the full dose series of DPT and polio. Overall, 7 in 10 children age 12-23 months have received immunization against the six principal childhood diseases---polio, diphtheria, ~rtussis, tetanus, measles and tuberculosis.

During the two weeks preceding the survey, 1 in 10 children under 5 had diarrhea. Four in ten of these children were not treated. Among those who were treated, 27 percent were given oral rehydration salts, 36 percent were given recommended home solution or increased fluids.

Breasffeeding is less common in the Philippines than in many other developing countries. Overall, a total of 13 percent of children born in the 5 years preceding the survey were not breastfed at all. On the other hand, bottle feeding, a widely discouraged practice, is relatively common in the Philippines. Children are weaned at an early age; one in four children age 2-3 months were exclusively breastfed, and the mean duration of breastfeeding is less than 3 months.

Infant and child mortality in the Philippines have declined significantly in the past two decades. For every 1,000 live births, 34 infants died before their first birthday. Childhood mortality varies significantly by mother's residence and education. The mortality of urban infants is about 40 percent lower than that of rural infants. The probability of dying among infants whose mother had no formal schooling is twice as high as infants whose mother have secondary or higher education. Children of mothers who are too young or too old when they give birth, have too many prior births, or give birth at short intervals have an elevated mortality risk. Mortality risk is highest for children born to mothers under age 19.

The 1993 NDS also collected information necessary for the calculation of adult and maternal mortality using the sisterhood method. For both males and females, at all ages, male mortality is higher than that of females. Matemal mortality ratio for the 1980-1986 is estimated at 213 per 100,000 births, and for the 1987-1993 period 209 per 100,000 births. However, due to the small number of sibling deaths reported in the survey, age-specific rates should be used with caution.

Information on health and family planning services available to the residents of the 1993 NDS barangay was collected from a group of respondents in each location. Distance and time to reach a family planning service provider has insignificant association with whether a woman uses contraception or the choice of contraception being used. On the other hand, being close to a hospital increases the likelihood that antenatal care and births are to respondents who receive ANC and are delivered by a medical personnel or delivered in a health facility.

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

National. The main objective of the 1993 NDS sample is to allow analysis to be carried out for urban and rural areas separately, for 14 of the 15 regions in the country. Due to the recent formation of the 15th region, Autonomous Region in Muslim Mindanao (ARMM), the sample did not allow for a separate estimate for this region.

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

- Household
- Women age 15-49

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

The population covered by the 1993 Phillipines NDS is defined as the universe of all females age 15-49 years, who are members of the sample household or visitors present at the time of interview and had slept in the sample households the night prior to the time of interview, regardless of marital status.

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

Sample survey data

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

The main objective of the 1993 National Demographic Survey (NDS) sample is to provide estimates with an acceptable precision for sociodemographics characteristics, like fertility, family planning, health and mortality variables and to allow analysis to be carried out for urban and rural areas separately, for 14 of the 15 regions in the country. Due to the recent formation of the 15th region, Autonomous Region in Muslim Mindanao (ARMM), the sample did not allow for a separate estimate for this region.

The sample is nationally representative with a total size of about 15,000 women aged 15 to 49. The Integrated Survey of Households (ISH) was used as a frame. The ISH was developed in 1980, and was comprised of samples of primary sampling units (PSUs) systematically selected and with a probability proportional to size in each of the 14 regions. The PSUs were reselected in 1991, using the 1990 Population Census data on population size, but retaining the maximum number of PSUs selected in 1980.

This sample is self-weighted in each of the 14 regions, but not at the national level. It was selected using a two-stage sample design; the first involved the selection of barangays,and the second, the selection of households in the sampled barangays. Barangays are the smallest political subdivisions. In general, the barangay corresponds to a census enumeration area. However, they vary widely in size, some covering more than 1,000 households. In the case when the barangay size was very large, it was segmented into several enumeration areas.

To maximize the efficiency of the sample design, the sample was allocated to the regions using a method called "power allocation procedure." This method optimizes the precision by taking into account sampling errors found in previous demographic surveys, in particular the 1978 Republic of the Philippines Fertility Survey. For this purpose, the following characteristics were considered: mean number of children ever born, proportion of women who want no more children, mean number of children desired, and proportion of married women who are using a family planning method.

A total of 2100 PSUs were selected for ISH, 750 of which were selected for the 1993 NDS. Individual households were selected with a probability of selection inversely proportional to the barangay's size to maintain a fixed overall sampling fraction within each region. An average of 20 completed interviews was targeted in each PSU.

In total, 750 PSUs and about 13,700 households were selected. The survey was well received by the respondents. Response rate for the household interview varies slightly by region (see Table A. 1). In some regions, all of the households in the sample were successfully interviewed. For the individual women's interview, Bicol women have the lowest overall response rate (93 percent).

A total of 15,029 women aged 15-49 years were successfully interviewed. The weighting factors to provide national estimates were calculated as the inverse of the overall sampling fractions, adjusted with the corresponding household and individual responses rates.

Sampling deviation
---------------------------

The household interviews identified 15,332 eligible women. Of these, 15,029 were successfully interviewed, giving a response rate of 98 percent. The principal reason for nonresponse among eligible women was the failure of interviewers to find them at home despite repeated visits to the household. Refusals were few in number (less than one percent).

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

Face-to-face

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

Three types of questionnaires were used for the 1993 NDS: the Household Questionnaire, the Individual Woman's Questionnaire and the Service Availability Questionnaire. The contents of the first two questionnaires were based on the DHS Model Questionnaire, which was designed for use in countries with high levels of contraceptive use. Additions and modifications to the model questionnaires were made after consultation with members of a Technical Working Group convened for the purpose of providing technical assistance to the NSO in the implementation of the survey.

The household and individual questionnaires were developed in English and then translated into and printed in six of the most widely spoken languages in the Philippines, namely: Tagalog, Cebuano, Ilocano, Hiligaynon, Bicol and Waray.

a) The Household Questionnaire was used to list all the usual members and visitors of selected households. Some basic information was collected on the characteristics of each person listed, including his/her age, sex, education, and relationship to the head of the household. The main purpose of the Household Questionnaire was to identify women who were eligible for individual interview. In addition, information was collected about the dwelling, such as the source of water, type of toilet facilities, materials used for the floor of the house, and ownership of various consumer goods.

b) The Individual Woman's Questionnaire was used to collect information from women aged 15-49. An important change from the past practice in large-scale demographic surveys in the Philippines is that the 1993 NDS covered all women 15-49 instead of limiting the interview to ever-married women. In keeping with past practice, the questionnaire contained a pregnancy history instead of the usual DHS birth history. Women were asked questions on the following topics:

- Background characteristics (education, religion, etc.)
- Reproductive behavior and intentions
- Knowledge and use of contraception
- Availability of family planning
- supplies and services
- Breastfeeding and child health
- Maternal mortality

c) The Health Service Availability Questionnaire was designed to collect information about health and family planning services available to the individual women respondents. This questionnaire was administered at the cluster level, that is, one questionnaire was filled for each of the 750 sample points. Combined with information collected in the main survey, data from the two surveys can identify subgroups of women who are underserved by the health and family planning providers.

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

Editing of the questionnaires was an integral part of the field data collection in the sense that questionnaires based on successful interviews were immediately edited by field editors. Further review and coding of some variables were done at the NSO central office. Machine processing was also done at the central office.

Processing of the NDS data was done with the use of the DHS computer program ISSA (Integrated System for Survey Analysis), from data entry to tabulation. Seven microcomputers were made available by NSO for data entry while Macro International provided four microcomputers for data management as well as for running edit and tabulation programs. Initial tabulations were generated by the end of August 1993, and a preliminary report was released in October 1993.

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

A total of 13,728 households was selected for the survey, of which 12,995 were successfully interviewed. The difference was due to one of the following reasons: some selected households had moved out or could not be located by the NDS team; there were no eligible respondents found for the selected household during the NDS team's visit; or the household simply refused to be interviewed.

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

Sampling errors, on the other hand, can be measured statistically. The sample of women selected in the 1993 NDS 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 nonsampling 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 1993 NDS sample was designed using stratification (region and urban/rural), clustering (barangay or a segment thereof) and stages of selection (barangay and household on the first and second stage, respectively). Consequently, it was necessary to utilize more complex formulas. The module on sampling errors in the ISSA package developed for the Demographic and Health Surveys program was used to assist in computing the sampling errors with the proper statistical methodology.

In addition to the standard errors, the program computes the design effect (DEFT) for each estimate, which is defined as the ratio between the standard error using the given sample design and the standard error that would result ifa simple random sample had been used. A DEFT value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a value greater than 1.0 indicates the increase in the sampling error due to the use of a more complex and less statistically efficient design. The ISSA program also computes the relative error and confidence limits for the estimates.

For each variable, the type of statistic (mean or proportion) and the base population are given in Table B.1 of the Final Report. Sampling errors are presented in Tables B.2.1-B.2.17 of the Final Report for variables considered to be of major interest. Results are presented for the whole country, divided into urban and rural areas, and for each of the 14 regions. For each variable, Tables B.2.1-B.2.17 present the value of the statistic (R), its standard error (SE), the number of unweighted (N) and weighted cases (WN), the design effect (DEFT), the relative standard error (SE/R), and the 95 percent confidence limits (R+2SE).

More complex estimates like the total fertility rate, infant mortality rate or medians are calculated using the Jackknife replication procedure incorporated in this ISSA module. Results are presented only for the whole country, divided into urban and rural areas, but not for each of the 14 regions, because these estimates need to have a large sample size to provide accurate precision.

The confidence limits have the following interpretation. For the proportion of married women currently using a contraceptive method (currently using any method), the overall average from the sample is 0.400 and its standard error is 0.006. Therefore, to obtain the 95 percent confidence limits, one adds and subtracts twice the standard error to the sample estimate, i.e., 0.400 + (2 x 0.006), which means that there is a high probability (95 percent) that the true proportion currently using is between 0.387 and 0.412.

The relative standard error for most estimates for the country as a whole is small, except for estimates of very small proportions. The magnitude of the error increases as estimates for subpopulations such as geographical areas are considered. For the variable currently using any method, for instance, the relative standard error (as a percentage of the estimated proportion) for the whole country and for urban and rural areas is 1.5 percent, 2.1 percent, and 2.3 percent, respectively.

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

Nonsampling 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 1993 NDS to minimize this type of error, nonsampling errors are impossible to avoid and difficult to evaluate statistically.

摘要
---------------------------
1993年全国人口普查（NDS）是一项针对15至49岁女性的全国代表性样本调查，旨在收集有关生育、计划生育、婴儿、儿童和孕产妇死亡率以及孕产妇和儿童健康的情报。该调查于1993年4月至6月进行。1993年NDS由国家统计局与卫生部门、菲律宾大学人口研究所及其他关注人口、健康和家庭规划问题的机构合作实施。1993年NDS的资金由美国国际开发署通过人口与卫生调查项目提供。

在调查期间，全国近13,000户家庭被访问，超过15,000名15至49岁的女性接受了访谈。结果显示，菲律宾的生育率持续缓慢下降。在当前水平下，菲律宾女性在其生育期内平均将生育4.1个孩子，比1988年记录的数字少0.2个。然而，与邻近的东南亚国家相比，菲律宾的总生育率仍然较高。

1993年NDS的主要目标是提供有关生育率和死亡率水平的最新信息；婚姻状况；生育意愿；对计划生育方法的认知、认可和使用；母乳喂养实践；以及孕产妇和儿童健康。这些信息旨在帮助政策制定者和管理者评估和设计改善国家健康和家庭规划服务的计划和政策。

主要结果

生育率在地区和社会经济特征方面存在显著差异。城市女性平均比农村女性少生育1.3个孩子，未受过教育的女性比受过大学教育的女性多生育一个孩子。比科地区女性平均比居住在马尼拉大都市的女性多生育3个孩子。

几乎所有女性都知道计划生育方法；避孕药、女性绝育、宫内节育器和避孕套为90%以上的女性所熟知。四分之一的已婚女性目前正在使用避孕措施。最受欢迎的方法是女性绝育（12%），其次是避孕药（9%），以及自然家庭规划和体外射精，这两种方法都被7%的已婚女性所采用。

在北棉兰老岛、中央维萨亚斯和南棉兰老岛、城市地区以及受过高于中学教育的女性中，避孕药的使用率最高。菲律宾的避孕药使用率明显低于邻近的东南亚国家；1987年泰国已婚女性使用计划生育的比例为66%，而印度尼西亚在1991年为50%。

大多数避孕药使用者从公共服务提供者那里获得他们的方法（70%）。政府卫生设施主要提供永久性方法，而村卫生站或卫生中心是避孕药、宫内节育器和避孕套的主要来源。

尽管菲律宾女性已经结婚的年龄相对较高，但她们继续推迟第一次结婚的年龄。一半的菲律宾女性在21.6岁时结婚。大多数女性婚后首次发生性行为。

一半的已婚女性表示她们不再想要孩子，12%的女性已经绝育。另外19%的女性希望至少再过两年才要孩子。几乎三分之二的菲律宾女性表示她们更喜欢生育3个或更少的孩子。调查结果指出，如果所有不希望出生的孩子都能避免，总生育率将降至2.9个孩子，这比观察到的比率低近30%。

菲律宾四分之一的已婚女性没有使用任何避孕方法，但希望推迟下一个孩子的出生两年或更长时间（12%），或者想要停止生育（14%）。如果满足潜在的计划生育需求，避孕药的使用率可能增加到69%。停止生育的需求大约是间隔生育的两倍（分别为45%和23%）。

1993年NDS收集了有关出生前五年内出生的婴儿的各种方面的信息，包括产前护理、疫苗接种、母乳喂养和食物补充以及疾病。结果显示，10个五岁以下儿童中有8个是由接受助产士或护士（45%）或医生（38%）产前护理的母亲所生的。

破伤风，婴儿死亡的主要原因之一，可以通过母亲怀孕期间接种破伤风类毒素疫苗来预防。在菲律宾，五年来有三分之二的新生儿是由在怀孕期间接受破伤风类毒素注射的母亲所生的。

根据母亲的报告和从健康卡中获得的信息，12至23个月大的90%的儿童都接种了BCG疫苗以及DPT和脊髓灰质炎的第一剂，81%的儿童接种了麻疹疫苗。随着剂量的增加，免疫覆盖率下降；接受DPT和脊髓灰质炎完整剂量系列的儿童的辍学率为3%至5%。总的来说，12至23个月大的10个儿童中有7个接种了针对六种主要儿童疾病的疫苗——脊髓灰质炎、白喉、百日咳、破伤风、麻疹和结核病。

在调查前的两周内，五岁以下儿童中有1/10患有腹泻。其中四分之一的儿童未接受治疗。在接受治疗的孩子中，27%的孩子接受了口服补液盐，36%的孩子接受了推荐的家用溶液或增加的液体。

与许多其他发展中国家相比，菲律宾的母乳喂养率较低。总体而言，在调查前五年内出生的13%的儿童从未接受母乳喂养。另一方面，在菲律宾普遍被 discourage 的瓶喂，相对而言比较常见。儿童很早就断奶；四分之一的2至3个月大的儿童完全母乳喂养，平均母乳喂养时间不到3个月。

在过去二十年里，菲律宾的婴儿和儿童死亡率显著下降。每1000名活产婴儿中，有34名婴儿在出生后第一年内死亡。儿童死亡率在母亲居住地和受教育程度方面存在显著差异。城市婴儿的死亡率比农村婴儿低约40%。没有受过正规教育的母亲的婴儿死亡概率是没有受过正规教育的婴儿的两倍。母亲年龄过小或过大、生育过多或生育间隔过短的孩子，死亡率风险较高。19岁以下母亲的婴儿死亡率最高。

1993年NDS还收集了使用姐妹法计算成年人和孕产妇死亡率所需的信息。对于男性和女性，在所有年龄段，男性的死亡率都高于女性。1980-1986年的孕产妇死亡率比估计为每10万出生213例，1987-1993年期间为每10万出生209例。然而，由于调查中报告的兄弟姐妹死亡人数很少，应谨慎使用年龄特异率。

从1993年NDS乡村居民可获得的健康和家庭规划服务的信息是从每个地点的一组受访者收集的。到达计划生育服务提供者的距离和时间与女性是否使用避孕药或所使用的避孕药的选择之间没有显著的关联。另一方面，靠近医院会增加接受产前护理和分娩的受访者接受ANC和由医务人员接生或在卫生设施分娩的可能性。

地理覆盖范围
---------------------------
全国。

1993年NDS样本的主要目标是允许对城市和农村地区分别进行分析，对国家15个地区中的14个地区进行分析。由于最近成立的第15个地区，穆斯林棉兰老自治区（ARMM），样本没有允许对该地区进行单独估计。

分析单元
---------------------------
- 家庭
- 15至49岁的女性

总体
---------------------------
1993年菲律宾NDS所涵盖的人口是所有15至49岁的女性，无论婚姻状况如何，只要他们是样本家庭的成员或在访谈时在场，并且在前一天晚上在样本家庭中过夜。

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

抽样程序
---------------------------
1993年全国人口普查（NDS）样本的主要目标是提供具有可接受精度的估计，用于社会学人口统计特征，如生育、计划生育、健康和死亡率变量，并允许对国家15个地区中的14个地区分别进行分析。由于最近成立的第15个地区，穆斯林棉兰老自治区（ARMM），样本没有允许对该地区进行单独估计。

样本是全国代表性的，总规模约为15,000名15至49岁的女性。综合家庭调查（ISH）被用作框架。ISH于1980年开发，由在14个地区中系统选择并具有与规模成比例的概率的初级抽样单位（PSU）样本组成。PSU在1991年重新选择，使用1990年人口普查的人口规模数据，但保留1980年选择的PSU的最大数量。

该样本在每个14个地区都是自我加权的，但在全国层面不是。它使用两阶段样本设计选择；第一阶段涉及选择村庄，第二阶段，在样本村庄中选择家庭。村庄是最小的政治单位。一般来说，村庄相当于一个人口普查登记区。然而，它们的大小差异很大，一些村庄覆盖超过1000户家庭。在村庄规模非常大的情况下，它被分割成几个登记区。

为了最大限度地提高样本设计的效率，使用了一种称为“功率分配程序”的方法将样本分配到地区。这种方法通过考虑在先前的人口调查中发现的抽样误差来优化精度，特别是在1978年菲律宾共和国生育调查中。为此，以下特征被考虑：平均生育子女数、希望不再生育的女性比例、希望生育子女数的平均值以及使用计划生育方法的已婚女性比例。

为ISH选择了2100个PSU，其中750个被选用于1993年NDS。使用与村庄规模成反比的概率选择个体家庭，以在每个地区内保持固定的整体抽样比例。每个PSU的目标是完成平均20次访谈。

总共选择了750个PSU和约13,700户家庭。调查得到了受访者的积极响应。家庭访谈的响应率在不同地区略有不同（见表A.1）。在一些地区，样本中的所有家庭都成功接受了访谈。对于个人女性的访谈，比科女性的总体响应率最低（93%）。

总共对15,029名15至49岁的女性进行了成功访谈。计算提供国家估计的加权因素的加权因子是整体抽样比率的倒数，并调整了相应的家庭和个人响应率。

抽样偏差
---------------------------
家庭访谈确定了15,332名有资格的女性。其中，15,029人接受了成功访谈，响应率为98%。有资格女性未响应的主要原因是在反复访问家庭后，访谈员未能找到她们。拒绝的人数很少（不到1%）。

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

研究工具
---------------------------
1993年NDS使用了三种类型的问卷：家庭问卷、个人女性问卷和服务可用性问卷。前两个问卷的内容基于DHS模型问卷，该问卷是为在高避孕率国家使用而设计的。在与为向NSO提供技术援助而召集的技术工作组成员协商后，对模型问卷进行了补充和修改。

家庭和个人问卷是用英语开发的，然后翻译成并印制了菲律宾最广泛使用的六种语言：他加禄语、宿务语、伊洛科语、希利盖农语、比科语和瓦莱语。

a) 家庭问卷用于列出所选家庭的全部常驻成员和访客。收集了有关列表中每个人特征的一些基本信息，包括他们的年龄、性别、教育和与家庭主人的关系。家庭问卷的主要目的是确定有资格接受个人访谈的女性。此外，还收集了有关住宅的信息，例如水源、厕所设施类型、房屋地板使用的材料以及各种消费品的所有权。

b) 个人女性问卷用于收集15至49岁女性的信息。与菲律宾以往的大规模人口调查实践相比，一个重要的变化是，1993年NDS涵盖了所有15至49岁的女性，而不是将访谈限制在已婚女性。与以往的做法一致，问卷包含怀孕史而不是通常的DHS出生史。女性被问及以下方面的问题：

- 背景特征（教育、宗教等）

- 生殖行为和意图

- 避孕知识的获取和使用

- 计划生育的可用性

- 供应和服务

- 母乳喂养和儿童健康

- 孕产妇死亡率

c) 健康服务可用性问卷旨在收集有关个人女性受访者可获得的健康和家庭规划服务的信息。该问卷在集群层面上进行管理，即每个样本点填写一份问卷。结合主要调查收集的信息，两个调查的数据可以确定由健康和家庭规划提供者服务不足的女性亚群体。

数据清理
---------------------------
问卷编辑是现场数据收集的一个组成部分，在意义上，基于成功访谈的问卷立即由现场编辑员进行编辑。在NSO中央办公室还进行了某些变量的进一步审查和编码。在中央办公室也进行了机器处理。

使用DHS计算机程序ISSA（综合调查分析系统）处理NDS数据，从数据输入到制表。NSO提供了7台微机用于数据输入，而Macro International提供了4台微机用于数据管理以及运行编辑和制表程序。初步制表在1993年8月底完成，初步报告于1993年10月发布。

响应率
---------------------------
总共选择了13,728户家庭进行调查，其中12,995户成功接受了访谈。差异是由于以下原因之一：一些被选中的家庭已搬出或NDS团队无法找到；在NDS团队访问期间，没有找到有资格的受访者；或者家庭简单地拒绝接受访谈。

抽样误差估计
---------------------------
抽样误差可以从统计上进行衡量。1993年NDS中选定的女性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。每个样本都会产生与实际样本选择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；虽然它不是确切知道的，但它可以从调查结果中估计。

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，这是方差的平方根。标准误差可以用来计算置信区间，在该区间内，除了非抽样误差外，可以合理地假设人口的真值将落在其中。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将落在加或减两倍标准误差的范围内。

如果女性的样本是简单随机样本，就可以使用简单的公式来计算抽样误差。然而，1993年NDS样本是使用分层（地区和城市/农村）、聚类（村庄或其部分）和选择阶段（村庄和第一阶段的家庭，第二阶段的家庭）设计的。因此，有必要使用更复杂的公式。用于人口与卫生调查项目的ISSA包中的抽样误差模块被用来帮助使用适当的统计方法计算抽样误差。

除了标准误差外，该程序还计算每个估计的设计效应（DEFT），它定义为使用给定样本设计计算的标准误差与如果使用简单随机样本将产生的标准误差之间的比率。DEFT值为1.0表示样本设计与简单随机样本一样有效，而大于1.0的值表示由于使用更复杂和统计效率较低的样本设计而增加的抽样误差。

ISSA程序还计算估计的相对误差和置信限。对于每个变量，表B.1中的最终报告中给出了统计量类型（平均值或比例）和基础人口。

抽样误差在最终报告的表B.2.1-B.2.17中给出，这些变量被认为是非常重要的。结果按整个国家、分为城市和农村地区以及每个14个地区进行呈现。对于每个变量，表B.2.1-B.2.17给出了统计量的值（R）、其标准误差（SE）、未加权（N）和加权案例（WN）、设计效应（DEFT）、相对标准误差（SE/R）和95%置信限（R+2SE）。

像总生育率、婴儿死亡率或中位数这样的更复杂的估计使用包含在此ISSA模块中的Jackknife复制程序计算。结果仅针对整个国家，分为城市和农村地区，但没有针对14个地区中的每个地区进行呈现，因为这些估计需要大样本量才能提供准确的精度。

置信限的解释如下。对于目前正在使用任何避孕方法的已婚女性的比例，样本的整体平均值为0.400，其标准误差为0.006。因此，为了获得95%置信限，将样本估计值加上和减去两倍标准误差，即0.400 + (2 x 0.006)，这意味着有很高的概率（95%）目前使用的是在0.387和0.412之间。

对于整个国家、城市和农村地区的估计，大多数估计的相对标准误差很小，除了非常小的比例的估计。随着考虑子群体，如地理区域的估计，误差的幅度增加。例如，对于目前使用任何方法这一变量，整个国家、城市和农村地区的相对标准误差（作为估计比例的百分比）分别为1.5%、2.1%和2.3%。

数据评估
---------------------------
非抽样误差是由于在执行现场活动时犯的错误造成的，例如未能找到和访谈正确的家庭、提问方式错误、访谈员或受访者的误解、数据输入错误等。虽然在设计和管理1993年NDS时努力尽量减少此类错误，但非抽样误差是无法避免的，也难以从统计上进行评估。