The 1993 Turkish Demographic and Health Survey (TDHS) is a nationally representative survey of ever-married women less than 50 years old. The survey was designed to provide information on fertility levels and trends, infant and child mortality, family planning, and maternal and child health. The TDHS was conducted by the Hacettepe University Institute of Population Studies under a subcontract through an agreement between the General Directorate of Mother and Child Health and Family Planning, Ministry of Health and Macro International Inc. of Calverton, Maryland. Fieldwork was conducted from August to October 1993. Interviews were carried out in 8,619 households and with 6,519 women.
The Turkish Demographic and Health Survey (TDHS) is a national sample survey of ever-married women of reproductive ages, designed to collect data on fertility, marriage patterns, family planning, early age mortality, socioeconomic characteristics, breastfeeding, immunisation of children, treatment of children during episodes of illness, and nutritional status of women and children. The TDHS, as part of the international DHS project, is also the latest survey in a series of national-level population and health surveys in Turkey, which have been conducted by the Institute of Population Studies, Haeettepe University (HIPS).
More specifically, the objectives of the TDHS are to:
Collect data at the national level that will allow the calculation of demographic rates, particularly fertility and childhood mortality rates; Analyse the direct and indirect factors that determine levels and trends in fertility and childhood mortality; Measure the level of contraceptive knowledge and practice by method, region, and urban- rural residence; Collect data on mother and child health, including immunisations, prevalence and treatment of diarrhoea, acute respiratory infections among children under five, antenatal care, assistance at delivery, and breastfeeding; Measure the nutritional status of children under five and of their mothers using anthropometric measurements.
The TDHS information is intended to assist policy makers and administrators in evaluating existing programs and in designing new strategies for improving family planning and health services in Turkey.
Fertility in Turkey is continuing to decline. If Turkish women maintain current fertility rates during their reproductive years, they can expect to have all average of 2.7 children by the end of their reproductive years. The highest fertility rate is observed for the age group 20-24. There are marked regional differences in fertility rates, ranging from 4.4 children per woman in the East to 2.0 children per woman in the West. Fertility also varies widely by urban-rural residence and by education level. A woman living in rural areas will have almost one child more than a woman living in an urban area. Women who have no education have almost one child more than women who have a primary-level education and 2.5 children more than women with secondary-level education.
The first requirement of success ill family planning is the knowledge of family planning methods. Knowledge of any method is almost universal among Turkish women and almost all those who know a method also know the source of the method. Eighty percent of currently married women have used a method sometime in their life. One third of currently married women report ever using the IUD. Overall, 63 percent of currently married women are currently using a method. The majority of these women are modern method users (35 percent), but a very substantial proportion use traditional methods (28 percent). the IUD is the most commonly used modern method (I 9 percent), allowed by the condom (7 percent) and the pill (5 percent). Regional differences are substantial. The level of current use is 42 percent in tile East, 72 percent in tile West and more than 60 percent in tile other three regions. "File common complaints about tile methods are side effects and health concerns; these are especially prevalent for the pill and IUD.
One of the major child health indicators is immunisation coverage. Among children age 12-23 months, the coverage rates for BCG and the first two doses of DPT and polio were about 90 percent, with most of the children receiving those vaccines before age one. The results indicate that 65 percent of the children had received all vaccinations at some time before the survey. On a regional basis, coverage is significantly lower in the Eastern region (41 percent), followed by the Northern and Central regions (61 percent and 65 percent, respectively). Acute respiratory infections (ARI) and diarrhea are the two most prevalent diseases of children under age five in Turkey. In the two weeks preceding the survey, the prevalence of ARI was 12 percent and the prevalence of diarrhea was 25 percent for children under age five. Among children with diarrhea 56 percent were given more fluids than usual.
Breastfeeding in Turkey is widespread. Almost all Turkish children (95 percent) are breastfed for some period of time. The median duration of breastfeeding is 12 months, but supplementary foods and liquids are introduced at an early age. One-third of children are being given supplementary food as early as one month of age and by the age of 2-3 months, half of the children are already being given supplementary foods or liquids.
By age five, almost one-filth of children arc stunted (short for their age), compared to an international reference population. Stunting is more prevalent in rural areas, in the East, among children of mothers with little or no education, among children who are of higher birth order, and among those born less than 24 months after a prior birth. Overall, wasting is not a problem. Two percent of children are wasted (thin for their height), and I I percent of children under five are underweight for their age. The survey results show that obesity is d problem among mothers. According to Body Mass Index (BMI) calculations, 51 percent of mothers are overweight, of which 19 percent are obese.
Kind of Data
Sample survey data
Unit of Analysis
- Women age 12-49
- Children under five
The Turkish Demographic and Health Survey (TDHS) is a national sample survey.
Unit of Analysis
- Women age 12-49
- Children under five
The population covered by the 1993 DHS is defined as the universe of all ever-married women age 12-49 who were present in the household on the night before the interview were eligible for the survey.
Producers and sponsors
Authoring entity/Primary investigators
Institute of Population Studies
General Directorate of Mother and Child Health and Family Planning
Ministry of Health
Macro International Inc.
U.S. Agency for International Development
The sample for the TDHS was designed to provide estimates of population and health indicators, including fertility and mortality rates for the nation as a whole, fOr urban and rural areas, and for the five major regions of the country. A weighted, multistage, stratified cluster sampling approach was used in the selection of the TDHS sample.
Sample selection was undertaken in three stages. The sampling units at the first stage were settlements that differed in population size. The frame for the selection of the primary sampling units (PSUs) was prepared using the results of the 1990 Population Census. The urban frame included provinces and district centres and settlements with populations of more than 10,000; the rural frame included subdistricts and villages with populations of less than 10,000. Adjustments were made to consider the growth in some areas right up to survey time. In addition to the rural-urban and regional stratifications, settlements were classified in seven groups according to population size.
The second stage of selection involved the list of quarters (administrative divisions of varying size) for each urban settlement, provided by the State Institute of Statistics (SIS). Every selected quarter was subdivided according tothe number of divisions(approximately 100 households)assigned to it. In rural areas, a selected village was taken as a single quarter, and wherever necessary, it was divided into subdivisions of approximately 100 households. In cases where the number of households in a selected village was less than 100 households, the nearest village was selected to complete the 100 households during the listing activity, which is described below.
After the selection of the secondary sampling units (SSUs), a household listing was obtained for each by the TDHS listing teams. The listing activity was carried out in May and June. From the household lists, a systematic random sample of households was chosen for the TDHS. All ever-married women age 12-49 who were present in the household on the night before the interview were eligible for the survey.
The results of sample implementation for the household and the individual interviews for the country as a whole, for urban and rural areas, and for the five regions of Turkey. The results indicate that of the 10,631 households selected, the TDHS fieldwork teams successfully completed interviews with 8,619 (81 percent). The main reasons fieldwork teams were unable to interview some households were that some of the listed dwelling units were found to be vacant at the time of the interview or the household was away for an extended period. Eight thousand nine hundred households were identified as being occupied, and 8,619 households were successfully interviewed. Consequently, the household response rate was calculated as 96.8 percent. The household response rate was higher in rural areas than in urban areas and highest in the Southern and Northern regions.
In the interviewed households, 6,862 eligible women were identified, of whom 95 percent were interviewed. Eligibility for the individual interview required that the woman be ever-married, be younger than 50 years of age, and be present in the household on the night before the interview. Among the small number of eligible women not interviewed in the survey, the principal reason for nonresponse was the failure to find the woman at home after repeated visits to the household. The eligible woman response rate was higher in rural areas than in urban areas and was higher in the Southern and Central regions than in the other three regions.
The overall response rate for the TDHS was calculated as 92 percent, ranging from 89 percent in the Eastern region to 95 percent in the Southern region.
Analysis has to be performed using weights. As mentioned earlier, the TDHS sampling plan is not a self-weighted one; in order to have sufficient numbers of observations for meaningful statistical analyses, more sample units were chosen from the Northern and Southern regions, which would have yielded inadequate numbers of observations if the target number of households had been allocated by PPS.
The number of households that were selected in each region according to power allocation as well as the expected numbers of households assuming a PPS distribution of the targeted 10,000 households can be seen in Table B.I of the Final Report.
The weight assigned to any stratum is simply the reciprocal of the sampling fraction employed in calculating the number of units in that particular stratum:
w (i) = / f (i) .
The term f(i), the sampling fraction at the i 'h stratum, is the product of the probabilities of selection at every stage in a stratum:
f (i) = P (i,l) * P (i,2) * ...... * P (i,s) where s is the stage.
The weights for the regions were assumed to be compensated for the nonresponse to the Household Questionnaire and to the Individual Questionnaire during fieldwork. The compensating factor for the nonresponse for the Household Questionnaire is the inverse value of:
R (i,2) = Completed households/Eligible households.
Eligible households include the households where interviews were completed, households where there were no competent respondents, households where interviews were postponed and eventually not completed, refusals, and those dwellings.that were not found by the fieldwork teams.
Similarly, the compensating factor for the nonresponse to the Individual Questionnaire is the inverse value of:
R (i,3) = Completed individual questionnaires/Eligible women.
Since selection was carried out proportionately in the urban/rural breakdown within the regions, and since there is almost no variation in nonresponse rates among the rural areas of the five regions, there was no need to calculate separate weights for rural and urban areas.
Weights should also include compensating factors for the missing clusters that were not visited at all for various reasons. Since sample selection was done in subregions, it would be better to have compensating factors in the subregional level.
The weights for the households were calculated by multiplying the above factors for each region and subregion. They were then standardized by multiplying these weights by the ratio of the number of interviewed households to the total weighted number of households. Standardization of the weights of individual women was undertaken by multiplying the individual weights by the ratio of the number of interviewed women to the total weighted number of women.
Dates of Data Collection (YYYY/MM/DD)
Mode of data collection
Type of Research Instrument
Two questionnaires were used in the main fieldwork for the TDHS: the Household Questionnaire and the Individual Questionnaire for ever-married women of reproductive age. The questionnaires were based on the model survey instruments developed in the DHS program and on the questionnaires that had been employed in previous Turkish population and health surveys. The questionnaires were adapted to obtain data needed for program planning in Turkey during consultations with population and health agencies. Both questionnaires were developed in English and translated into Turkish.
a) The Household Questionnaire was used to enumerate all usual members of and visitors to the selected households and to collect information relating to the socioeconomic position of the households. In the first part of the Household Questionnaire, basic information was collected on the age, sex, educational attainment, marital status and relationship to the head of household for each person listed as a household member or visitor. The objective of the first part of the Household Questionnaire was to obtain the information needed to identify women who were eligible for the individual interview as well as to provide basic demographic data for Turkish households. In the second part of the Household Questionnaire, questions were included on the dwelling unit, such as the number of rooms, the flooring material, the source of water, and the type of toilet facilities, and on the household's ownership of a variety of consumer goods.
b) The Individual Questionnaire for women covered the following major topics:
- Background characteristics
- Knowledge and use of family planning Other issues relating to contraception
- Maternal care and breastfeeding
- Immunisation and health
- Fertility preferences
- Husband's background
- Women's work and residence Values
- Attitudes and beliefs
- Materal and child anthropometry.
The woman's questionnaire included a monthly calendar, which was used to record fertility, contraception, postpartum amenorrhea and abstinence, breastfeeding, marriage, and migration histories for periods of more than five years, beginning in January 1988, up to the survey month. In addition, the fieldwork teams measured the heights and weights of children under age five and of their mothers, as well as mothers' as circumference.
Hacettepe Institute of Population Studies
Office Editing. The questionnaires were returned to the Institute of Population Studies by the fieldwork teams for data processing as soon as each provincial interview was completed. The office editing staff checked that the questionnaires for all the selected households and eligible respondents were returned from the field. The comparatively few questions that had not been precoded (e.g., occupation) were coded at this time.
Machine Entry and Editing. The data were entered and edited on microcomputers using the Integrated System for Survey Analysis (ISSA), a packaged program specifically developed to process DHS data. ISSA allows range, skip, and consistency errors to be detected and corrected at the data entry stage. The machine entry and editing activities were initiated within two days after the beginning of the fieldwork and were completed 10 days after the completion of the fieldwork. Advantage was taken of the fact that data processing activities ran concurrently with fieldwork. Field check tables from edited data were periodically produced for each interviewing team. These focused on such potential problems as high proportions of incomplete households and displacement of eligible respondents and were used to check the progress and quality of data from the field.
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.