Bangladesh Population Census 1991
Bangladesh, 1991
BANGLADESH BUREAU OF STATISTICS
Identification
idno
BGD-BBS-BPC-1991-v01
Title
Bangladesh Population Census 1991
Country
| Name | Country code |
|---|---|
| Bangladesh | BGD |
Kind of Data
Census/enumeration data [cen]
Version
Version number
Bangladesh inherited the long tradition of conductiong the population census on decennial basis since 1872. The third population census of Bangladesh, which is thirteenth in the series, was conducted during March 12-15, 1991. The main objective of the census was to collect information about the basic characteristics related to housing, households and population for developing a comprehensive data base for development planning and human resource development programmes as well as economic management.
Version Date
1991-09
Producers and sponsors
Authoring entity/Primary investigators
| Agency Name | Affiliation |
|---|---|
| BANGLADESH BUREAU OF STATISTICS | Statistics and Informatics Division, Ministry of Planning |
Funding Agency/Sponsor
| Name | Abbreviation |
|---|---|
| Statistics and Informatics Division | SID |
Sampling
Sampling Procedure
Population Censuses have been conducted in the area now constituting Bangladesh, dates'back over one hundred years. In the early days data need was limited and very few data were collected and tabulated through census. In recent decades data demand is very high. Nation needs various types of data for its future planning. These data are collected through national census.It was decided by the National Statistical Council to conduct the third Population and Housing Census of Bangladesh in 1991 in three phases. The first phase is a complete count with a short questionnaire, the second phase is a post enumeration check (PEC) and the third phase is a sample count with a long questionnaire. Accordingly, the Population and Housing Census of 1991 was conducted in three phases. The first phase, that is, the complete count was done in March, 1991 and the third phase, the Sample Census was conducted in December,1991. This chapter will deals with the following aspects of sample census:
a) Sample Design.
b) Estimation Procedure.
c) Questionnaire
d) A Few Findings.
a) Sample Design.
b) Estimation Procedure.
c) Questionnaire
d) A Few Findings.
Deviations from the Sample Design
The sample design for sample census is a "Stratified Two-Stage Cluster Design". It was ,a self weighted realistic design. Population living in the households were the target population. In the sampling process, the target population was divided into 3 strata: a) RuralAreas (b) Statistical Metropolitan Areas (SMA) and (c) Other Municipal Areas.
At stage 1, the number of Primary Sampling Unit (PSU) i.e. number of enumeration areas were allocated to the strata as proportional to their population size. For operational convenience and for getting estimate at the lowest level feasible, the rural stratum was subdivided into 64 substrata, SMA stratum into 4 substrata and other municipality stratum into 58 substrata. The number of PSU5 allocated to each stratum were distributed among the substratum as proportional to the size of population as of 1991 census.
At stage 2 households were selected as SSU (Secondary Sampling unit or ultimate cluster). The sample design prepared on the basis of optimization of PSU and SSU considering births (frequent variable) anddisabled population (rare variable) of sample census, 1981.
The main features emerged are:
At stage 1, about 3.6% of the enumeration areas were selected as PSU. Where the number of PSU thus selected from each substratum fall short of 12, it was raised to 12 with consequential adjustment in the sizes of sample in large substrata.
At stage 2 about 25% of the households of selected EAs were selected as SSU. For operational convenience the size of ultimate cluster was fixed at 22 households irrespective of the size of PSU. Sampling units were selected systematically with a random start. On principle, the number of households of a selected PSU was divided by the
desired size of ultimate cluster (22) to come up with the interval. The initial number was selected randoinify by selecting a random number ( 1 <r < interval) and the successive SSU were selected by adding multiple of interval with the random number successively.
At stage 1, the number of Primary Sampling Unit (PSU) i.e. number of enumeration areas were allocated to the strata as proportional to their population size. For operational convenience and for getting estimate at the lowest level feasible, the rural stratum was subdivided into 64 substrata, SMA stratum into 4 substrata and other municipality stratum into 58 substrata. The number of PSU5 allocated to each stratum were distributed among the substratum as proportional to the size of population as of 1991 census.
At stage 2 households were selected as SSU (Secondary Sampling unit or ultimate cluster). The sample design prepared on the basis of optimization of PSU and SSU considering births (frequent variable) anddisabled population (rare variable) of sample census, 1981.
The main features emerged are:
At stage 1, about 3.6% of the enumeration areas were selected as PSU. Where the number of PSU thus selected from each substratum fall short of 12, it was raised to 12 with consequential adjustment in the sizes of sample in large substrata.
At stage 2 about 25% of the households of selected EAs were selected as SSU. For operational convenience the size of ultimate cluster was fixed at 22 households irrespective of the size of PSU. Sampling units were selected systematically with a random start. On principle, the number of households of a selected PSU was divided by the
desired size of ultimate cluster (22) to come up with the interval. The initial number was selected randoinify by selecting a random number ( 1 <r < interval) and the successive SSU were selected by adding multiple of interval with the random number successively.
Data Collection
Dates of Data Collection (YYYY/MM/DD)
| Start date | End date |
|---|---|
| 1991-03-12 | 1991-03-15 |
Mode of data collection
Face-to-face [f2f]
Supervision
To ensure coverage, effective management and close supervision the entire country was divided into 2,11 .538 enumeration areas (EA) by enumeration maps. For every 5 or 6 enumerators there was a supervisor. The geo-coding system developed in the previous censuses was checked, updated and computerized.Modif led Defacto method was adopted in counting the population. Zero hours of the night following 11th March, 1991 was considered as the census moment and the night from census moment to 5-00 a.m. was considered as the census night. To avoid omission and duplication people were counted where they were found during the census night.
Type of Research Instrument
For sample census, a long questionnaire was prepared for collecting the data. The questionnaire was not an OMR but most of the questions were precoded. The questionnaire had five modules, namely (a) Household module, (b) Individual module, (c) Disability module (d) Mortality module and (e) Fertility Module. In all, there were 68 questions. All the basic questions of main census were repeated, most of them were more detail. Most of the questions of sample census were not included in the short questionnaire of the main census. Three modules, namely, disability, mortality and fertility are completely new addition in the sample questionnaire over the census questionnaire. The statement given below shows summary of Sample Census
Questionnaire:
Name of Module Total No.of questions Total No.of questions No.of alternative answers
House and 20 5 70
household
Individual 22 9 55
Disability 03 - 12
Mortality 09 5 10
Fertility 14 8 16
Questionnaire:
Name of Module Total No.of questions Total No.of questions No.of alternative answers
House and 20 5 70
household
Individual 22 9 55
Disability 03 - 12
Mortality 09 5 10
Fertility 14 8 16
Data Collectors
| Name | Abbreviation | Affiliation |
|---|---|---|
| Statistics and Informatics Division | SID | Ministry of Planning |
Disclaimer and copyrights
Disclaimer
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.
Metadata production
Document ID
DDI-BGD-BBS-BPC-1991-v01
Producers
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| BANGLADESH BUREAU OF STATISTICS | BBS | Statistics and Informatics Division, Ministry of Planning | Documentation of the study |
Date of Production
2019-09-01