Assessment of spatial and temporal dynamics of livelihoods: A methodological perspective

Abstract

Increased international attention to rural poverty alleviation and sustainable development underscores the need for better tools for analyzing the factors and conditions that shape livelihoods and for assessing the livelihood impacts of project- and policy-interventions. The first aspect encompasses important spatial dynamics, while the second addresses both temporal and spatial dynamics. To be effective, such approaches must accommodate the complex and multidimensional nature of livelihood systems by: i) using appropriate indicators of livelihoods outcomes and embracing multiple components of a livelihood system; ii) analyzing the influence of multiple and complex factors, including development interventions; iii) addressing differential impacts by taking appropriate aggregation at the village level. Powerful new geomatics technologies offer new ways to deal with spatial variability, and can be combined with innovative social-science approaches for more efficient socio-economic data collection and analysis. This paper discusses key principles for designing appropriate methods and reports lessons learned from our own experience in Jharkhand state, India and Kutai Barat district in East Kalimantan, Indonesia. In these two study areas, with relatively low levels of development and high forest cover, we assessed livelihood systems by: i) using available, broad range data of assets and socio-economic data in indices of development from secondary source; ii) using geomatics tools for sampling and analyses that encompass a range of theoretically important variables (e.g. road access; market access; proximity to large projects; tribal affiliation; topography; land suitability); iii) identifying key factors that characterize within-village stratification and designing household sampling accordingly; iv) aggregating unit of analysis to address differential impacts and relationships among livelihood components. Multilevel regression analysis is used to address hierarchical or differential structure in the data. The paper provides guidance for improved landscape-scale livelihoods analysis and targeting and identifies a way forward for further method improvement