Uncovering regional disparities - the use of factor and cluster analysis

The aim of this paper is to present a new methodology for the classification of spatial-economic entities in order to support a regional development policy. Such a classification may be obtained through the use of multivariate statistical methods - factor and cluster analysis, and is based on eleven economic and social indicators. The local government units of three Croatian counties are used as the working sample. Those are as follows: the County of Istria, the Lika-Senj County and the Primorje-Gorski Kotar County. Factor analysis leads to identification of a small number of socio-economic dimensions that summarise adequately the information contained in the original set of variables. Cluster analysis is used to look for groups of towns and municipalities with similar levels of socio-economic development. Multivariate techniques were successful in identifying the main axes of socio-economic characterisation and the regions of the observed counties with differing degrees of development. The new methodology for spatial-economic entities' classification enables a much more useful characterisation of the territory for policy-making purposes.