Child Nutrition Status Module Bangladesh Household Expenditure Survey 1985-86
The Bangladesh Bureau of Statistics (BBS) included a nutritional status asbesament module to the 1985-86 Household Expenditure Survey (HES). The Nutritional Status Survey is the only national cross-sectional anthropometric based survey conducted in Bangladesh where data were periodically collected over a year's time. Since the anthropometric data were collected from the same households as the HES, national descriptions of nutritional status disaggregated by income and expenoiture information have been possible.
Since nutrition has a significant impact on growth patterns of young children, anthropometric measurement, in this case age, weight, height, and mid upper arm circumference give objective indicators of short and long term nutrition conditions.
The collection of data periodically over a year allowed for measure ments during the various seasons when food and health conditions might be expected to vary and affect nutritional status.
The fact that this module was attached to the Housebo.d Expenditure Survey permits relating nutritional status to a number of economic variables not always available in an anthropometric measurement survey.
In the HES, data were collected quarterly from sixteen households in each of 240 urban and rural areas. Households which had any children between the ages of 8 and 71 months were included in the nutritional status module.
The purpose of this report is to present the results of the Nutri tional Status Survey module of the 1985-86 HES.
The objectives of the HES Child Nutritional Status Module were as follows:
1 To describe the nutritional status of children 6-71 months of age by urban and rural areas and by selected health, water and sanitation variables. Looking at health, water, and sanitation variables gives additional information on the situation surrounding the nutritional status of young children. Health affects nutrition levels regardless of the availability of food, water quality and sanitation facilities..
2 To describe the nutritional status of children 6-71 months of age by selected HES variables. The HES collects information on the income and expenditure patterns of individual households. Respondents report daily purchase of food and other necessities using the daily diary system as well as larger, less frequent purchases. Total income, cash and in-kind, is also reported for the survey month. Having so much detailed economic data available for the same households in which child nutritional status is being measured affords a rare opportunity to study a broad range of relationships between nutrition and the resources available to the family.
3 To introduce nutritional status assessment an evaluation methodology to BBS.
Anthropometric measurements combined with various health and sanitary variables are an effective way to assess nutritional status of children. Collection and evaluation of this type of data are new to the BBS. Much research, and many surveys have resulted in very specific anthropometric measurement techniques, which must be followed consistently in an accurate way, to obtain reliable measurement/data. In addition, working with the measuring equipment and learning methods to physically handle young children takes training, practice, and confidence even for experienced statistical enumerators.
Kind of Data
Sample survey data [ssd]
Table 5 presents the numbers of PSU's, households, and children enumerated by time period. Nutrition survey teams identified if a 6-71 month old child was available for weighing and measuring, and if a parent/guardjan was available to interview.
Of the 960 households identified in each period, between 555 and 601 had at least one child 6-71 months of age. Between 504 and 550 households were interviewed for a mean coverage rate of 92.3% (range: 90.8% to 94.5%) and a corresponding mean non-response rate of 7.7% (range: 5.5% to 9.2%). Of the 176 households that were not interviewed, 90.0% of them had a survey child of 6-71 months who was not at home at the time of interview.
Producers and sponsors
Authoring entity/Primary investigators
BANGLADESH BUREAU OF STATISTICS
Statistics Division, Ministry of Planning
The "Integrated Multipurpose Sample" (IMPS) has been used by BBS for four surveys: the Household Expenditure Survey (HES), Labor Force Survey, Demographic Survey, and the Nutritional Status Survey, which was however carried out as a component of the HES, 1985-86.
The sample was designed to permit urban and rural, as well as national descriptions. There were 240 Primary Sampling Units (PSU), of which 108 were urban and 132 rural. There were four periods of data collection for both the HES and Nutrition Survey. Data were collected from 60 PSU's which were randomly selected each period. In each PSU, 16 households were selected by HES for interview and enumeration.
The following weights, which were adjusted for the non-response rate (see 4.2 below), were applied to produce a population proportionate sample for the Nutritional Status Survey:
Rural: 7721 (see Appendix -A)
A detailed description of the IMPS appears in Appendix B.
Dates of Data Collection (YYYY/MM/DD)
Mode of data collection
Ministry of Planning
Data editing was done at three different stages:
Field Editing: The survey teams checked completed questionnaires at the end of each household interview, at the end of the day and before departing from the PSU.
At BBS: At the end of data collection of each period, questionnaires were checked by enumerators and the two survey administrators. Enumerators were questioned on the difficulty of obtaining information on any particular variable.
Computer Editing: Data were edited and cleaned and any inconsistt-'nciez or missing values were identified. Missing anthropometric measurements were imputed for 20 children.
Results of four variables were transcribed from the HES questionnaires to the nutrition questionnaires at the end of each data collecticn The BBS IBM 5280 (microcomputer) was used for data entry, mainframe was used for the edit program, tabulation, and data Cobol, Fortran and Cents 4 were used to writa programme tabulations. The US National Center for Health Statistics (NCHS) standards for child growth were entered onto the BBS system for comparing survey data. Cross tabulations were created from the data where the unit of analysis was the survey child.