In this paper, we introduce some recent non-aggregative algorithms, for the construction of synthetic indicators on multi-indicators systems, and exemplify them on data pertaining to the sustainability of main Italian cities. The procedure employs tools from Partial Order Theory and computes the final synthetic scores, without the compensation effects of classical composite indicators, so better preserving the nuances of city sustainability. Since turning multi-indicator systems into single synthetic indicators is unavoidably forcing, the procedure provides also ways to assess the degrees of “rankability” of statistical units, so to help researchers evaluating the soundness of the synthesis process. The algorithms introduced in the paper can be easily implemented, using the package Parsec, freely available in the R ecosystem.
Partially Ordered Set Syntethic Indcators Multi-indicator Systems