Methodology

In September 2013, The Asia Foundation and The Daily Star implemented a rapid assessment of citizens' perception of recent events based on a national survey of Bangladesh citizens of voting age conducted in the second and third weeks of September 2013. The survey was conducted in 14 districts of the country (two in each of the seven divisions) and gathered impressions from a cross-section of ordinary citizens, civil society representatives, local businesspersons, and other (non-political) elites through informal individual interviews in (i) each divisional headquarter and a proximate rural upazila in the same district; and (ii) district headquarter of another randomly chosen district in each division and a proximate rural upazila in the same district. A formal questionnaire was administered.

The sample included 1,400 respondents and the survey design intended that 700 urban and 700 rural citizens would be interviewed. All respondents were 18 years old or above. The sample design included 100 respondents in each district (50 urban and 50 rural). Half of the interviewees in each location were to be ordinary citizens and the other half (non-political) elites. Ordinary citizens included marginal or small farmers; daily wage laborers in agriculture or non-agriculture; factory workers; construction workers; transport workers; shop employees; small businesspersons/traders; fishermen; roadside vendors; rickshaw/van pullers; etc. Full-time housewives from low income households were also considered as ordinary citizens. The (non-political) elite group included professionals such as teachers, lawyers, doctors and engineers; NGO officials; officials of private firms; businesspersons (producers and traders); religious leaders; and medium or large farmers. Housewives from high income households were included in the respondent base. Government officials, such as those in district or Upazila Nirbahi offices, were not interviewed; however, teachers in government schools, doctors in government clinics and similarly placed individuals were eligible for inclusion in the respondent base. The survey tried to ensure a wide variety of occupation types and a reasonable mix of men and women within each group. The survey also tried to include a proper representation of respondents from different age groups (18-29, 30-49, and 50+). Youth is defined in this report as those belonging to the age group 18-29.

In each district headquarter, 50 respondents (25 ordinary citizens and 25 elite) were to be chosen. In each rural upazila, likewise, 50 respondents (25 ordinary citizens and 25 elite) were to be chosen. The survey tried to ensure a mix of occupation types. The attempt to choose specific occupation types made it difficult to obtain the intended equal balance of males and females in the sample. Also, rural sampling fell short of the intended target of 700, while urban sample exceeded the target of 700. A suitable weighting procedure was then used to correct for the over and under sampling, and the effective weighted sample comprised of 1,054 rural (75%) and 346 urban (25%) respondents, of which 701 were males and 699 females.

You can read also...