Poverty is a complex phenomenon responsible for economic backwardness of the country.Therefor allevation of poverty within the shortest possible time is the priority goal of the government for prescribing policy measures for poverty alleviation, the continuous flow of nauon regarding states and process dimension of poverty is inevitable. Considering this need, Bangladesh Bureau of Statistics has introduced a regular Poverty Monitoring Survey or System since 1994. The PMS was designed to establish and institutionalize a regular poverty monitoring system relating to multi-dimensional indicators of poverty and provide feedback to me policy makers to initiate appropriate poverty eradication strategies. The Centre on Integrated nulan Development for Asia and the Pacific (CIRDAP) has provided technical assistance to the survey easure programme. A large scale survey was conducted in May 95 with the assistance of CIRDAP. The survey sample of 1995 has produced poverty measures separately for urban and rural areas by desegregation at the level of 21 regions (**) of the country. The sample has also generated a wide range of poverty indicators for the 21 regions separately. The national level measures and indicators of poverty have been obtained by aggregating estimates of all regions together. The survey findings with as much disaggregations as possible have been presented in this report.
The poverty monitoring survey was initially started with a very small sample in the rural areas of the country. The first two rural surveys were conducted in October 94 and April 95 in 75 primary sampling units (PSU) of the country. These two surveys were conducted in 2250 households only.
From December 94 survey, the sample size was increased and the urban universe was brought under the purview of the survey programme. This larger sample comprises of 110 PSUs in the rural areas and 40 PSUs in the urban areas. The total number of households covered in this later sample were 4,500 of which 3,300 households were in the rural areas and 1200 households in the urban areas. The surveys conducted are statistically representative in the national level. The results of these surveys have been presented with urban-rural breakdown by national level desegregation only. In the recent time, the users of the poverty data demanded that
(**) The regions of Bandarban, Khagrachari and Rangamati are combined together to a single region as the sample size was small in these regions.
poverty indicators should be provided beyond the national levels, i.e, subnational levels (regions & districts) for identifying the spatial variation in the incidence of poverty over the administrative units of the country. These data are also needed for government and nongovernment organisations to adopt policies and programmes for poverty reduction. Considering the data need of the end-users the 99 poverty monitoring survey was conducted by adopting a very large sample.
Objectives of the Survey
The objectives of the survey are as follows:
Collection of information on household and housing structure of the population;
- Collection of demographic and socio-economic characteristics of the population;
- Collection of information on income and expenditure of the population;
- Collection of information on household access to land and other assets;
- Collection of information on household consumption and nutrition;
- Collection of information on household credit and expenditure saving;
- Collection of information on school attendance and
- Collection of information on household crises and crises copying measures.
Kind of Data
Sample survey data [ssd]
Producers and sponsors
Authoring entity/Primary investigators
BANGLADESH BUREAU OF STATISTICS
Statistics Division, Ministry of Planning
A two stage stratified sampling design was followed in drawing the sample for Poverty Monitoring Survey 95. The list of the enumeration areas (EA) of 1991 Population Census was used as the sample frame. Each EA of Population Census 91 is the cluster of around 100 households. These EAs were treated as primary sampling unit (PSU). First of all, the universe, i.e. the whole country was divided into two strata: urban & rural. Each of the two strata was further subdivided into 23 sub-strata, i.e. each of the 23 regions (greater districts) formed a separate sub stratum. Using the 1991 population census frame, a sample of 300 enumeration areas (PSUs) was allocated to the 23 regions of urban stratum at the first stage. Proportional allocation was made while allocating PSUs to the 23 regions. Table-1a and Table-1b show the distribution of PSUs (1st stage sample). Then the required number of EAs (PSUs) were drawn systematically using a random start. All the households in the selected EAs were listed completely in a listing form. Then a sub-sample of 20 households was drawn randomly from each selected EA using a random number table. Thus in the urban universe, there were 300*20, i.e. 6,000 (six thousand) households in the sample. Similar procedure was adopted for drawing sample from the rural universe except that the number of enumeration areas (EAS) selected for the rural universe was 500 and the una households 500*20.
Dates of Data Collection (YYYY/MM/DD)
Mode of data collection
Extensive training was planned for this extended survey. Training was imparted in two phases. In the first phase the Regional Statistical Officers (RSO) were imparted training at headquarters in Dhaka by the senior officers of BBS. Later, these RSOs (master trainers) imparted training to the Thana Statistical Officers and the local enumerators. The duration of the training was two days in each stage. The first day was devoted to class room lecture and the second day was used for field demonstration. The data collection work was supervised by the Thana Statistical Officers. Higher level supervision was carried out by the RSOs. In addition, the senior officers of BBS and CIRDAP visited some sample areas.
Type of Research Instrument
A manual questionnaire was canvassed in the survey. The format of the questionnaire included 13 separate sections. The survey individual questions pertained to all such household characteristics as could be broadly classified into, (i) housing, family size, demography and education (ii) land and other asset ownership (iii) income and expenditure (iv) credit and crisis management and (V) living condition
Ministry of Planning
Preliminary edit was done by TSOs at the thana level. Manual editing of data was performed in selected Regional Statistical Offices and the project headquarters by a standard editing manual. After editing, the data were entered through microcomputer. The entered data were also edited and verified to eliminate errors and omissions in the entry stage. Moreover, computer edit was done by a sophisticated edit programme for cleaning the data. Finally, the tables were run with the edited and cleaned data.