Comparative Analysis of the Ordering of Polish Provinces in Terms of Social Cohesion

The article describes an assessment of the social cohesion of Polish provinces. The assessment was based on classical metric and interval-valued data using a hybrid approach combining multidimensional scaling with linear ordering. In the first step, after applying multidimensional scaling, the objects of interest were represented in a two-dimensional space. In the second step, the objects were linearly ordered based on the Euclidean distance from the pattern object. Interval-valued variables characterize the objects of interests more accurately than do metric data. Classic data are of an atomic nature, i.e. an observation of each variable is expressed as a single real number. By contrast, an observation of each interval-valued variable is expressed as an interval. Interval-valued data were derived by aggregating classic metric data on social cohesion at the level of districts to the province level. The article describes a comparative analysis of the results of an assessment of the social cohesion of Polish provinces based on classical metric data and interval-valued data.