Dimensions of well-being and their statistical measurements

Nowadays, a relevant challenge regards the assessment of a global measure of well-being by using composite indicators of different features such as level of wealth, comfort, material goods, standard living, quality and availability of education, etc. In this paper, we focus on statistical methodologies designed to build composite indicators of well-being by detecting latent components and assessing the statistical relationships among indicators. We will consider some constrained versions of Principal Component Analysis (PCA) which allow to specify disjoint classes of variables with an associated component of maximal variance. Once that the latent components are detected, a Structural Equation Model (SEM) has been used to evaluate their relationships. These methodologies will be compared by using a data set from 34 member countries of the Organization for Economic Co-operation and Development (OECD).