Dealing with Increasing Dimensionality in Wellbeing and Poverty Measurement, Some Problems and Solutions

A recent literature has fomented considerable interest in multi-dimensional measurement of wellbeing and deprivation. However, increasing the number of dimensions over which various aspects of wellbeing are measured raises concerns regarding the robustness of such measures, the extent to which they relate to individual and societal preferences, and their diminishing ability to discriminate between alternative states of the world. Here these problems are investigated in a multi-dimensional context, and some solutions are proposed using a satisfaction survey as a test bed for exploring agent preferences over a range of wants and needs. An alternative way of viewing the robustness issue is proposed and ideas, based upon various notions of separable functions (borrowed from consumer choice theory and non-parametric statistics) are shown to facilitate measurement of wellbeing, and increase our ability to discriminate between alternative states of the world. The result is a modification to the classic AlkireFoster multi-dimensional impoverishment index which admits notions of substitutability and complementarity between the various aspects of impoverishment. The modified version of the index proved to have superior robustness properties compared to more restrictive versions of the index. JEL Code: