Socio-economic factors affecting the income of small-scale agroforestry farms in hill country areas in Yemen: A comparison of OLS and WLS determinants

Abstract

Yemen is a less-developed country in the Arabian Peninsula, with only 3% arable land. An agroforestry land-use system has been practiced traditionally by small-scale farmers, but is associated with low productivity and income. A study has been undertaken to determine the socio-economic attributes of farmers that influence the financial performance of agroforestry and non-agroforestry farms in the Bura’a Mountain region. A survey was conducted of 150 farmers involved in both agroforestry and non-agroforestry. Both OLS and WLS regression were applied, and coefficients compared in terms of consistency and goodness of fit. Incomes of farmers were found to be influenced by education, area of land, livestock holding, family size, and whether coffee is grown, but not farmer’s age. The WLS method produced efficient and consistent results, whereas OLS regression was affected by the heteroscedasticity. The findings of the study indicate that the farmers of the study area are in need of financial and technical support from government to increase their income. Infrastructural development and public intervention in developing farmers’ technical know-how could enhance production and ensure the optimum use of land as well as soil and water conservation.