High order PLS path modeling to evaluate well-being merging traditional and big data: A longitudinal study

We propose using high order partial least squares path modeling (PLS-PM) to define a synthetic Italian well-being index merging traditional data, represented by the Quality of Life index proposed by “Il Sole 24 Ore”, and information provided by big data, represented by a Subjective Well-being Index (SWBI) performed extracting moods by Twitter. High order constructs allow to define a more abstract higher-level dimension and its more concrete lower-order sub-dimensions. These layered constructs have gained wide attention in applications of PLS-PM; many contributions in literature proposed their use to build composite indicators. The aim of the paper is to underline some critical issues in the use of these models and to suggest the implementation of a new adapted repeated indicator approach. Furthermore, following some recommendations proposed on the use of PLS-PM in longitudinal studies, we compare the situation in 2016 and 2017.