Problemi e metodologie statistiche di misurazione di fenomeni complessi tramite indicatori e indici sintetici

Abstract

The use of empirical indicators and composite indices for the measurement of multidimensional phenomena (i.e. quality of life, health, social welfare, etc.) implies a series of conceptual and methodological problems here analysed in view to construct composite indices statistically significant i.e. valid and reliable. In applied research in different domains one needs only one measure, an index of a multidimensional phenomenon to compare it with others and put it in complex statistical models. After a statistical definition of index (composite indicator) the A. describes with examples and discusses in detail four different methods of constructing composite indices. At the end advantages and disadvantages of the four methods are briefly examined.