Font Size: 

Composite Indicators, in the social sciences, are used more and more formeasuring very complex phenomena as the poverty, the progress and the well-being.The existing literature offers several alternative methods for obtaining a CompositeIndicators. The work focuses on building them through to Structural Equation Modeling,specifically with the use of Partial Least Squares-Path Modeling. In recent years manyadvances have been developed, in the context of these models to solve some problemsrelated to the role that the Composite Indicators play within that system; in particular onthe aspects linked to the high level of abstraction, when a Composite Indicator ismanifold, lacks its own manifest variables and is described by various underlying blocks.In this regard Partial Least Squares Component Regression Approach has been proposed,as alternative method for analyzing and studying higher-order construct.In this work we analyze in the framework of the same simulation design, its relativeperformance, analysing the bias and the variability of the estimates, comparing it withthe approaches known in the literature.Then an application case on the development of approach towards the set of CompositeIndicators of European sustainable development has been presented.