T.K. NATH
Institute of Forestry and Environmental Sciences, University of Chittagong, Chittagong, Bangladesh
M. INOUE
Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1- 1- Yayoi, Bunkyo-ku, 113-8657, Tokyo, Japan
Shifting cultivation, Settlement, Livelihood strategies, Rubber plantations, Social capital
Chittagong Hill Tracts, Bangladesh
Socio-economic and Policy
Livelihood
Selection of study villages Due to remote location, lack of accommodations, poor communication services, and civil unrest, the study was conducted in the USP second phase in Bandarban District. Development practitioners have long been aware that even if projects have the same level of overall assistance, results vary considerably from one location to another (Krishna 2004). Recent studies (Nath et al. 2005a, Nath and Inoue forthcoming) mention that the USP has both successes and failures in achieving project objectives. Some project villages seemed relatively successful in achieving the project’s objectives while others failed. Based on the opinions of the project manager and other staff members, discussions with some participants, and reviews of the project’s objectives, relatively successful (hereafter successful) and relatively unsuccessful (hereafter unsuccessful) project villages were defined as follows: • Successful: those villages where the project’s objectives are adequately achieved, and • Unsuccessful: those villages with inadequate progress in achieving the project’s objectives From field verification and discussions with the project authority and participants, it was concluded that four project villages were successful and six villages were unsuccessful. From among the four successful and six unsuccessful villages, one project village from each category was selected randomly for in-depth study. Brief comparative typologies and features of the sampled villages. Data and methods The first step of data collection consisted of key informant interviews and a walk through each sampled village. Eight elderly participants including village leaders, four from each village, and three project staff members were interviewed. Separate checklists were used for participants and project staff members to collect information. Participants told about their past livelihood strategies, current livelihood situations and jhum, local organization, participation in project activities, project benefits, and forest conditions. Project staff members told about site and participant selection, choice of production technologies, project outcomes, status and productivity of rubber plantations, benefit sharing, land tenure, and future plans. Village walks were used to explore forest conditions (growth, stock) of project villages, and general typology of the villages. In the second step, a household survey was conducted using a semi-structured questionnaire to collect quantitative data related to livelihood capitals, household income, and expenditure. Twenty-four households out of 48 and 19 households out of 38 were randomly selected from successful and unsuccessful villages, respectively, for the survey. The questionnaire was tested by a reconnaissance survey and refined as required. The questions were designed to collect data on five forms of capital of the surveyed participants based on the SL framework described by Scoones (1998) and DFID (2001). Parameters of five capitals included were: Human capital. Family size, education, members contributing cash income Physical capital. Housing conditions, household appliances, and livestock Natural capital. Landholding and forest conditions Financial capital. Credit and food security status Social capital. Groups and networks, trust and solidarity, collective actions. Quantitative data were summarised into averages and percentages, and determined standard deviation. A parametric test (one-way analysis of variance) was conducted to explore statistically significant differences among means of different variables between the two villages. Data of these variables between two villages were normally distributed. However, within each village extreme data sets were observed which were not normally distributed and hence one-sample Kolmogorov-Smirnov test was conducted to explore differences among participants of each village.
International Forestry Review Vol.11(3), 2009
Journal