The baseline ‘Result and Impact (RIM)’ survey of the Market Infrastructure Development Project in Charland Regions (MIDPCR) was conducted in estuarine chars of 20 upazilas in the following five districts of Bangladesh-Barisal, Bhola, Patuakhali, Noakhali and Lakshmipur since February 2006 to June 2013. It is funded by International Fund for Agricultural Development (IFAD) and a grant from the Government of Netherlands (GoN) and the client is Local Government Engineering Department (LGED) Ministry of LGRD&C. It started on 29 January 2009 and ended on 30 April 2009.

 

Project Description: The purpose of the survey was to obtain the information to be used as benchmark for the mid-term and the final evaluation of the impact of the project. This survey collected information on selected socio-economic, demographic and anthropometric indicators, pertaining to households of MIDPCR groups.The project has three distinct target groups – small traders, primary producers and women labourers.

 

Four main components of project:

 

Rural Infrastructure Development

 

• Market Management Strengthening
• Market Infrastructure
• Transport Infrastructure

 

1. Production and Market Group Development

• Group Formation and Support
• Support for Marketable Crops
• Support for Income Generating Activities

 

1. Rural Enterprise Development and
2. Policy, Institutional and Management Support

 

Tasks carried out by Mitra and Associates:

 

Survey Methodology (sample design and tools): A random sample of 1200 households was selected from 60 clusters taking 20 households from a cluster. Each of the groups of beneficiary households was defined as a cluster. The sampling plan slightly differs from the RIMS guideline, which suggests that the sample be comprised of 900 households drawn from 30 clusters with 30 households selected from a cluster. This variation was made because the average size of a MIDPCR group includes fewer than 30 households and also because of having a better coverage of the project area.

 

In compliance with RIMS procedure, the sample was selected by applying a two-stage cluster sampling technique. At the first stage, clusters (groups) were selected. Households were selected at the second stage, selecting 20 households from each of the selected clusters. From a cluster, 20 households were randomly selected if it had more than 20 but for less than 20 households, additional households were interviewed from a neighboring group to complete the quota of 20 interviews.

 

In making the selection of clusters, listing activities was done in each project Upazila. 

 

The baseline survey was conducted using the questionnaire designed by IFAD to collect the data required to measure the anchor impact indicators. However, some new questions were added to the questionnaire to obtain measures of some additional variables considered to be useful in making a more robust assessment of the impact of the MIDPCR. 

 

Tool Development: Baseline survey was conducted using the questionnaire designed by IFAD but some new questions were added to the questionnaire to obtain measures of some additional variables considered to be useful. Bengali version of the questionnaire was prepared by Mitra and Associates to administer the survey.

 

Data Collection & quality control: Data was collected by interviewing & listing the group member of the household & households and by measuring the height and weight of the under-five children living in the household. It needed approximately 17 days. Research Officer and Control Officers visited the interviewers in the field and re-interviewed 10 percent of households on random basis to ensure the quality of the data. 

 

Training of interviewers and supervisors: The interviewing team members including both interviewers and supervisors were given a five-day training to prepare them for the data collection work. The training consisted of four days of class room training and one day of field training. 

 

Data entry and analysis: Data entry and analysis were carried out using MS Access program supplied by IFAD. In order to minimize data entry errors, data were entered in two different computers. All data base files were then compared to identify the entry errors. The observations that did not match were identified and manually corrected. 

 

Client Reference: Market Infrastructure Development Project in Char land Regions (MIDPCR), Local Government Engineering Department (LGED) Ministry of LGRDC, Bangladesh and funded by International Fund for Agricultural Development (IFAD), Italy