Aims: In past, labour was extensively used in agriculture sector and there was huge surplus labour in agriculture sector. However, such trend has recently been changed where surplus labour in agricultural sector has reduced significantly and agriculture sector compete with non-agricultural sector in terms of hiring labour. Thus, the present study was undertaken to analyze the determinants of labour migration from agriculture sector to non-agricultural sector in Gopalganj district of Bangladesh.
 Place and Duration of Study: The study was conducted at 12 villages of three upazilas (Gopalganj Sadar, Tungipara and Kotalipara) in Gopalganj district. For the study, data were collected during the period from January to March in 2021.
 Methodology: To this end, primary data were collected from agricultural labours. Descriptive statistics and simple random sampling technique were used in this study. Binary logistic model was used to analyze the collected data. In addition, five point likert scale was used to rank the barrier towards internal labour migration.
 Results: Results found from the logit model indicate that factors like family size, education, past experience, access to available information, transportation facilities, and savings are positively related with the log of odd ratio in favor of labour migration from agriculture sector to non-agricultural sector while wage rate, age, off-farm income and farm holdings are inversely related with labour transfer from agriculture sector to non-agricultural sector. In addition, respondents in the study area have recognized lack of proper technical training as the major constraint in labour migration with a mean value of 4.48.
 Conclusion: The present study recommends that government should take initiatives to open skill development institutions in rural level so that agricultural labour can take training. Regarding necessary information on non-agricultural jobs, it can be recommended that government, local agents and NGOs, in case of migration, should take proper initiatives so that agricultural labours can easily get information about non-agricultural jobs.
Read full abstract