WFP has been working in partnership with the Government of Bangladesh (GoB) to reduce poverty, enhance food security and nutrition, provide universal education and reduce vulnerability to recurrent shocks with the overall objective to support the GoB to achieve the Millennium Development Goals (MDGs) 1-5. The survey is started on October 2014 and will be finished on February 2015.
Project Description: The main objective of the Country Programme (2012-2014) is to improve the long-term food security and nutrition of ultra-poor households in the poorest and most food-insecure rural areas and urban slums. The CP was planned to assist 4,025,000 people in 15 selected districts over the five years (2012-2016). The programme includes the following four components:
Component 1. Improving Maternal and Child Nutrition (IMCN).This component applies a combined preventive and curative approach to IMCN in communities with a high prevalence of acute and chronic under-nutrition among children under 5.
Component 2.School Feeding (SF).The objective of SF is to increase access to primary and pre-primary education through attendance, enrolment, and completion rates.
Component 3.Enhancing Resilience to Disasters and the Effects of Climate Change (ER).This component focuses on protecting vulnerable communities and households in areas prone to recurrent shocks.
Component 4.Strengthening Government Safety Nets. This activity supports the GoB in reforming existing social safety nets programs that address hunger and household food insecurity through capacity development activities at national, divisional, district and upazila levels.
Objectives: The study will gather data to assess changes in food security, nutrition, livelihood, community resilience and primary school education since the baseline survey conducted in 2012. It will also focus on the household socio-economic profiles and overall context of the CP areas to sufficiently inform the midterm evaluation of the country programme. Concisely, the outcome survey is designed with the following specific objectives.
Tasks carried out by Mitra and Associates:
Methodology (sample design & tools): Sample size of household for IMCN, ER and ER+IMCN: 5,200 (Intervention 2600 & control 2600), and for schools for all WFP school feeding program, Control WFP, Govt. school feeding program, Control Govt.: 936 and sample size of unions:30.
Base on the selected villages/unions list provided by WFP a random sampling was done. Firstly, the villages were divided in clusters for North and Coastal areas on IMCN, ER, ER+IMCN for both intervention and control. For IMCN north & coastal area 20 clusters were selected for each intervention and control, total 80 clusters, for ER 120 clusters were selected and for ER+IMCN 60 clusters were selected and then 20 HHs were randomly selected from each cluster, total 5200 HHs (Intervention 2600 & control 2600) from 260 clusters.
For the school survey, following the WFP School lists Mitra and Associates selected the 936 schools from North, coastal, Dhaka and Bandarban by following systematic random sampling method.
For the Union validation survey, Mitra and Associates just followed the list of selected 30 unions.
For Resilience Enhancing (ER) and School Feeding, data in the outcome study will be collected by surveying the same households and the same schools, used in the baseline study. For the IMCN and convergence (ER + IMCN) areas, data will be collected by doing a cross-section survey of households to be sampled from the same sampled clusters (communities) used in the baseline study. The sample design that used in the baseline study is described below.
The total sample was designed to permit statically valid comparison across two geographic regions for households (North and Coast), and three regions for schools (North, Coast, and Dhaka). The sample size computation was based on the anthropometric indicator of stunting. Overall, the household survey of ER and IMCN components included 12 strata (three combinations of project components X programme/control). The minimum required sample size for each stratum was computed to be able to detect a difference of 25 percent in the stunting rate (from the national rate of 41%) with 90 percent confidence and 80 percent power, assuming a design effect of 2.0. The parameters for calculating sample size for school feeding programme were chosen to detect a change from 80 percent attendance (from WFP School Feeding Evaluation Report) to 90 percent (proposed target rate), with 95 percent confidence, and 80 percent power, and a design effect of 2.0.
Questionnaires and preparation of other survey manuals/forms: The questionnaires for household survey made in baseline is reviewed and formatted for the outcome survey, & made it user friendly & the questionnaire/guideline is translated, pre-tested and then finalized making necessary changes based on the pre-test results.
Prepared the following technical manuals and forms:
Training: A ten-day comprehensive training Program on data collection operation is organized for the field personnel comprising interviewers, supervisors, and quality control officers.
Data collection & quality control: Approximate 1 month been spent for the data collection. Quality control teams been employed for the quality control checking and they did re-interview some of the selected households, schools and check some of the reported non-response cases to see if they were all due to valid reasons and compared the re-interviewed data with the corresponding interviewed data.
Data processing& Data Analysis: Data processing is simultaneous with the data collection for the short time-frame of survey. It is started within two weeks after starting of the data collection and all data will be entered into computer with double entries, resolving discrepancies (100 percent verification) through referral to the original questionnaire. Data editing is included checking of range, structure and a selected set of checks for internal consistency.
Dummy tables been arranged as per the study objectives and after approval of WFP the actual tables will be constructed. Data is analyzed developing statistical estimates such as percentages, proportions, means, medians, and using such statistical tools as frequency distributions, and the cross tabulations discerning relations of knowledge and practices with socio-demographic characteristics. Data analyses been done by divisions and within a division it is aggregated in gender, rural-urban areas and in socio-economic/demographic characteristics.
Client Reference: The United Nations World Food Programme (WFP), IDB Bhaban, 14th Floor, E/8-A, Rokeya Sharani, Sher-e-Bangla Nagar, Dhaka-1207, Phone no: 880-2-9183022-33, E-mail: CountryDirector.Dhaka@wfp.org
Contact Person: Mr. Ezaz Nabi, Senior Programme Officer, World Food Programme, IDB Bhaban, 14th Floor, E/8-A Rokeya Sharani, Sher-e-Bangla Nagar, Dhaka-1207.