The Malawi Demographic and Health Survey 1992 is the first survey of its kind to be conducted in Malawi.
The 1992 Malawi Demographic and Health Survey (MDHS) was a nationally representative sample survey designed to provide information on levels and trends in fertility, early childhood mortality and morbidity, family planning knowledge and use, and maternal and child health. The survey was implemented by the National Statistical Office during September to November 1992. In 5323 households, 4849 women age 15-49 years and 1151 men age 20-54 years were interviewed.
The Malawi Demographic and Health Survey (MDHS) was a national sample survey of women and men of reproductive age designed to provide, among other things, information on fertility, family planning, child survival, and health of mothers and children. Specifically, the main objectives of the survey were to:
- Collect up-to-date information on fertility, infant and child mortality, and family planning
- Collect information on health-related matters, including breastleeding, antenatal and maternity services, vaccinations, and childhood diseases and treatment
- Assess the nutritional status of mothers and children
- Collect information on knowledge and attitudes regarding AIDS
- Collect information suitable for the estimation of mortality related to pregnancy and childbearing
- Assess the availability of health and family planning services.
The findings indicate that fertility in Malawi has been declining over the last decade; at current levels a woman will give birth to an average of 6.7 children during her lifetime. Fertility in rural areas is 6.9 children per woman compared to 5.5 children in urban areas. Fertility is higher in the Central Region (7.4 children per woman) than in the Northem Region (6.7) or Southern Region (6.2). Over the last decade, the average age at which a woman first gives birth has risen slightly over the last decade from 18.3 to 18.9 years. Still, over one third of women currently under 20 years of age have either already given birlh to at least one child or are currently pregnant.
Although 58 percent of currently married women would like to have another child, only 19 percent want one within the next two years. Thirty-seven percent would prefer to walt two or more years. Nearly one quarter of married women want no more children than they already have. Thus, a majority of women (61 percent) want either to delay their next birth or end childbearing altogether. This represents the proportion of women who are potentially in need of family planning. Women reported an average ideal family size of 5.7 children (i.e., wanted fertility), one child less than the actual fertility level measured in the survey--further evidence of the need for family planning methods.
Knowledge of contraceptive methods is high among all age groups and socioeconomic strata of women and men. Most women and men also know of a source to obtain a contraceptive method, although this varies by the type of method. The contraceptive pill is the most commonly cited method known by women; men are most familiar with condoms. Despite widespread knowledge of family planning, current use of contraception remains quite low. Only 7 percent of currently married women were using a modem method and another 6 percent were using a traditional method of family planning at the time of the survey. This does, however, represent an increase in the contraceptive prevalence rate (modem methods) from about 1 percent estimated from data collected in the 1984 Family Formation Survey. The modem methods most commonly used by women are the pill (2.2 percent), female sterilisation (1.7 percent), condoms (1.7 percent), and injections (1.5 percent). Men reported higher rates of contraceptive use (13 percent use of modem methods) than women. However, when comparing method-specific use rates, nearly all of the difference in use between men and women is explained by much higher condom use among men.
Early childhood mortality remains high in Malawi; the under-five mortality rate currently stands at 234 deaths per 1000 live births. The infant mortality rate was estimated at 134 per 10130 live births. This means that nearly one in seven children dies before his first birthday, and nearly one in four children does not reach his fifth birthday. The probability of child death is linked to several factors, most strikingly, low levels of maternal education and short intervals between births. Children of uneducated women are twice as likely to die in the first five years of life as children of women with a secondary education. Similarly, the probablity of under-five mortality for children with a previous birth interval of less than 2 years is two times greater than for children with a birth interval of 4 or more years. Children living in rural areas have a higher rate ofunder-fwe mortality than urban children, and children in the Central Region have higher mortality than their counterparts in the Northem and Southem Regions. Data were collected that allow estimation ofmatemalmortality. It is estimated that for every 100,000 live births, 620 women die due to causes related to pregnancy and childbearing.
The height and weight of children under five years old and their mothers were collected in the survey. The results show that nearly one half of children under age five are stunted, i.e., too short for their age; about half of these are severely stunted. By age 3, two-thirds of children are stunted. As with childhood mortality, chronic undernutrition is more common in rural areas and among children of uneducated women.
The duration of breastfeeding is relatively long in Malawi (median length, 21 months), but supplemental liquids and foods are introduced at an early age. By age 2-3 months, 76 percent of children are already receiving supplements.
Mothers were asked to report on recent episodes of illness among their young children. The results indicate that children age 6-23 months are the most vulnerable to fever, acute respiratory infection (ARI), and diarrhea. Over half of the children in this age group were reported to have had a fever, about 40 percent had a bout with diarrhea, and 20 percent had symptoms indicating ARI in the two-week period before the survey. Less than half of recently sick children had been taken to a health facility for treatment. Sixty-three percent of children with diarrhea were given rehydration therapy, using either prepackaged rehydration salts or a home-based preparation. However, one quarter of children with diarrhea received less fluid than normal during the illness, and for 17 percent of children still being breastfed, breastfeeding of the sick child was reduced.
Use of basic, preventive maternal and child health services is generally high. For 90 percent of recent births, mothers had received antenatal care from a trained medical person, most commonly a nurse or trained midwife. For 86 percent of births, mothers had received at least one dose of tetanus toxoid during pregnancy. Over half of recent births were delivered in a health facility.
Child vaccination coverage is high; 82 percent of children age 12-23 months had received the full complement of recommended vaccines, 67 percent by exact age 12 months. BCG coverage and first dose coverage for DPT and polio vaccine were 97 percent. However, 9 percent of children age 12-23 months who received the first doses of DPT and polio vaccine failed to eventually receive the recommended third doses.
Information was collected on knowledge and attitudes regarding AIDS. General knowledge of AIDS is nearly universal in Malawi; 98 percent of men and 95 percent of women said they had heard of AIDS. Further, the vast majority of men and women know that the disease is transmitted through sexual intercourse. Men tended to know more different ways of disease transmission than women, and were more likely to mention condom use as a means to prevent spread of AIDS. Women, especially those living in rural areas, are more likely to hold misconceptions about modes of disease transmission. Thirty percent of rural women believe that AIDS can not be prevented.
Kind of Data
Sample survey data
Unit of Analysis
- Women age 15-49
- Men age 20-54
The Malawi Demographic and Health Survey 1992 covers the following topics:
- HIV Knowledge-Questions assess knowledge/sources of knowledge/ways to avoid HIV
- Maternal Mortality
- Men's Survey
- Service Availability
The population covered by the 1992 MDHS is defined as the universe of all women age 15-49 in malawi and all men age 20-54 living in the household.
Producers and sponsors
National Statistical Office (NSO)
Macro International Inc.
United States Agency for International Development
Based on the 1987 Malawi Population and Housing Census, the country is demarcated into 8,652 enumeration areas (EAs) of roughly equal population size. This sampling frame of census EAs was stratified by urban and rural areas within each of the three administrative regions, making six sampling strata in total. Within each sampling stratum, districts were geographically ordered, thereby providing additional implicit stratification.
The MDHS sample of households was selected in two stages. First, 225 EAs were selected from the 1987 census frame of EAs with probability proportional to population size. The distribution of selected sample points (EAs) is shown in the map of Malawi. The measure of EA size was based on the number of households enumerated during the 1987 census. NSO staff, after being trained in listing procedures and methods for updating maps, were sent to the selected EAs to list all households and produce maps which provided the orientation for later data collection teams in finding selected households. Households in refugee camps and institutional populations (army barracks, police camps, hospitals, etc.) were not listed. In the second stage, a systematic sample of households was selected from the above lists, with the sampling interval from each EA being proportional to its size based on the results of the household listing operation.
In these households, all women age 15-49 years were eligible lot interview. Further, a one-in-three systematic subsample of households was drawn, within which both eligible men age 20-54 years and women age 15-49 years were interviewed.
Because the objective of the survey was to produce region-level and urban/rural estimates of some indicators, an oversample of households in the Northern Region and in urban areas was necessary.
The results of household and individual interviews for the urban and rural sample and for Malawi as a whole shows a total of 5,811 households which was selected in the MDHS sample, of which 5,396 were currently occupied. Of the 5,396 occupied households, 5,323 were interviewed, yiclding a household response rate of 98.6 percent. Rural and urban response rates at the household level did not differ significantly.
Within the interviewed households, 5020 eligible women (15-49 years) were identified of which 4849 were interviewed, yielding an individual female response rate of 96.6 percent. In the one-in-three subsample of households, 1,288 eligible men were identified, of which 1,151 were successfully interviewed (89.4 percent response). The principal mason for nonresponse among both eligible men and women was the failure to find them at home despite repeated visits to the household. The lower response rate among men than women was due to more frequent and longer-term absence of men. The refusal rate in the MDHS was extremely low (0.1 percent).
Response rates were marginally better in rural areas than in urban areas in the male survey (rural, 90.3 percent; urban, 87.5 percent), but nearly the same in the female survey (rural, 96.5 percent; urban, 96.8 percent).
Thus the MDHS sample is not self-weighting at the national level, but it is self-weighting within each of the six region/urban-rural strata.
Dates of Data Collection
Data Collection Mode
Data Collection Notes
The Household Questionnaire was used to list all the usual members and visitors of selected households. A household is defined as one that consists of one or more persons, related or unrelated, who make common provisions for food, or who regularly take all their food from the same pot or same grainstore (Nkhokwe), or who pool their incomes for the purpose of purchasing food. Persons in a household may live in one or more dwelling units. 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, namely, those age 15-49 years.
For those women who were either absent or could not be interviewed during the first visit, a minimum of three visits were made before recording nonresponse. Women were interviewed with the individual female questionnaire. In the course of administering the household questionnaire, eligible men, namely, those age 20-54 were also identified. The individual male questionnaire was administered to all men age 20-54 living in every third household in the selected sample. The male questionnaire collected much of the same information found in the female questionnaire, but was considerably shorter because it did not contain questions on reproductive history, and maternal and child health.
During the household listing operation (i.e., before the main survey), one Health Services Availability Questionnaire was completed in each of the 225 MDHS sample points. Leaders in the community provided information that allowed an assessment of the availability of health and family planning services to persons living in the respective localities.
A three-week training course for the main survey was held in July and August of 1992. A total of 80 field staff was trained. The training course consisted of instruction in general interviewing techniques, field procedures, a detailed review of items on the questionnaires, instruction and practice in weighing and measuring children, mock interviews between participants in the classroom, and practice interviews in areas outside MDHS sample points. Only trainees who performed satisfactorily in the training programme were selected for fieldwork. Team leaders were NSO staff who had previously participated in the MDHS pretest.
The fieldwork for the MDHS was carried out by ten interviewing teams, each con- sisting of one team leader, one field editor, five female interviewers, one male interviewer and one driver. Additionally, senior NSO staff co- ordinated and supervised fieldwork activities. Data collection began 1 September and was completed on 10 November 1992.
National Statistical Office (NSO)
Four types of questionnaires were used: the Household Questionnaire, the Individual Female Questionnaire, the Individual Male Questionnaire, and the Health Services Availability Questionnaire. The contents of these questionnaires were based on the DHS Model B Questionnaire, which is designed for use in countries with a low level of contraceptive use. Modification of the questionnaires was undertaken by NSO in consultation with the Ministry of Health and Macro. The questionnaires were pretested in April 1992.
Completed questionnaires were retumed to NSO for data processing. The processing operation consisted of office editing, coding of open-ended questions, data entry and editing of errors found by the computer programs. Data entry, editing, and analysis were accomplished on personal computers, using a software program called ISSA (Integrated System for Survey Analysis). Data processing started on 14 September 1992 and was completed on 21 January 1993.
Estimates of Sampling Error
Sampling errors, on the other hand, can be measured statistically. The sample of women selected in the MDHS 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 one can be reasonably assured that, apart from nonsampling errors, the true value of the variable for the whole population falls. 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 MDHS sample design depended on stratification, stages and clusters. Consequently, it was necessary to utilise more complex formulas. The computer package CLUSTERS, developed by the International Statistical Institute for the World Fertility Survey, was used to assist in computing the sampling errors with the proper statistical methodology.
In addition to the standard errors, CLUSTERS 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 if a 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. CLUSTERS also computes the relative error and confidence limits for the estimates.
In addition to the standard errors, CLUSTERS program also 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 if a 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; 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.
Sampling errors are presented in Tables B.2-B.7 of the Final Report for variables considered to be of major interest. Results are presented for the whole country, for urban and rural areas, and for regions. For each variable, the type of statistic (mean or proportion) and the base population are given in Table B.1. For each variable, Tables B.2-B.7 present the value of the statistic (R), its standard error (SE), the number of unweighted (N) and weighted cases (WN), the design effect (DEBT), the relative standard error (SE/R), and the 95 percent confidence limits.
The confidence limits have the following interpretation. For the mean number of children ever bom (EVBORN), the overall average from the sample is 3.482 and its standard error is 0.049. Therefore, to obtain the 95 percent confidence limits, one adds and subtracts twice the standard error to the sample estimate, i.e., 3.482 + (2 x 0.049), which means that there is a high probability (95 percent) that the true average number of children ever born is between 3.383 and 3.580.
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 children ever born (EVBORN), for instance, the relative standard error (as a percentage of the estimated mean) for the entire country and its regional divisions are 1.4 percent, 2.5 percent, 1.9 percent and 2.4 percent, respectively.
Nonsampling error is the result of mistakes made in implementing data collection and data processing procedures, 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 numerous efforts were made during the design and implementation of the MDHS to minimise this type of error, nonsampling errors are impossible to avoid and difficult to evaluate statistically.
Data and Data Related Resources
Demography and Social Statistics Division, National Statistical Office
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.
DDI Document ID
World Bank, Development Economics Data Group
Generation of DDI documentation
Date of Metadata Production