Developments in Higher-Order PLS-PM for the building of a system of Composite Indicators

Many phenomena are complex and therefore difficult to measure and to evaluate. Research, in the last years, has been focusing on the development and use of a system of Composite Indicators in order to obtain a global description of a complex phenomenon and to convey a suitable synthesis of information. The existing literature offers several alternative methods for obtaining a Composite Indicators. The work focuses on building them through to Structural Equation Modeling, specifically with the use of Partial Least Squares-Path Modeling. In recent years many advances have been developed, in the context of these models to solve some problems related to the role that the Composite Indicators play within that system; in particular, the research focuses on a particular aspect linked to the high level of abstraction, when a Composite Indicator is manifold, lacks its own manifest variables and is described by various underlying blocks. In this regard we have proposed two alternative methods for analyzing and studying higher-order construct Composite Indicator, on the calculation of the estimates for the determination of endogenous block, so as to be the best estimated and represented by the blocks below.