The purpose of the survey was to probe:
The results of the survey were used in informing the PRODIP project’s activities, strategies, and areas of opportunities. The results were also used to inform ongoing public information outreach and advocacy.
The survey was carried out with a nationally representative sample of Bangladesh population age 18 or older constituting the survey universe. The sample was drawn in two stages, with the households (basic sampling units) being selected in the second or final stage. The first-stage sampling unit was a cluster of 200-300 households. A cluster was formed with one or more Mauzas or part of a Mauza in rural areas. In urban areas it was made up of one or more Mohallahs or part of a Mohallah. Following the sampling strategies used in the Bangladesh Demographic and Health Survey (BDHS), 30 households were included into the sample from a cluster. More than 30 households were assumed to result in intolerable intra-cluster correlations. Thus, the sample was made up of 180 clusters with 146 from rural areas and 34 from urban areas, including 57 from the division of Dhaka, 31 from Chittagong, 25 from Rajshahi, 23 from Khulna, 22 from Rangpur, 12 from Barisal and 10 from Sylhet. The clusters were selected independently for each stratum, with PPES (Probability Proportional to Size) method of selection. The sampling frame for selection of the clusters for a stratum was created using the list of Mauzas/Villages by households, given in the 2001 Census Community Series Report for the districts making up the stratum.
In addition to the national sample, a special sample was separately drawn from the coastal, haor and hill regions of the country. The data from the special sample was analysed separately, without combining them with the data of the national sample. The list of upazilas constituting a region was supplied by The Asia Foundation. From the listed upazilas, the special sample was drawn in three stages. At the first stage, two upazilas will be randomly picked up in each region from among the upazilas not included in the national sample. At the second, seven clusters were randomly selected from the selected upazilas in a region, selecting 4 from their rural areas and 3 from their urban areas. At the third stage, the households, the basic sampling units, were selected drawing 30 households from a selected cluster. Thus, the special sample was made up of 630 households, including 210 households (120 rural and 90 urban) from each region.
Finally, the national sample and the special sample, together produced a total size of 6030 (=5400 + 630) households drawn from 201 (=180 + 21) clusters.
The tasks were carried out including:
Data was analyzed developing statistical estimates such as percentages, proportions, means, medians, and using such statistical tools as frequency distributions, cross tabulations, etc. and submission of clean data sets.
The survey was awarded by The Asia Foundation and funded by USAID.
Duration: May-December 2011.
Contact reference: Kazi Marful Islam, Email:firstname.lastname@example.org, Mobile # 01712-063-846, The Asia Foundation, House # 5, Road # 8, Baridhara, Dhaka-1212 and Phone: 882-6941-5.