Cluster Analysis as an Analytical Tool of Population Policy Cover Image

Кластерный анализ как аналитический инструментарий политики народонаселения
Cluster Analysis as an Analytical Tool of Population Policy

Author(s): Oksana Mikhaylovna Shubat, Irina Viktorovna Shmarova
Subject(s): Economy
Published by: Институт экономики Уральского отделения Российской академии наук
Keywords: population trends; Russian regions; family institution; population policy; family policy; differentiated approach; cluster analysis; Ward’s method; Euclidean distance; multidimensional data classifica

Summary/Abstract: The predicted negative trends in Russian demography (falling birth rates, population decline) actualize the need to strengthen measures of family and population policy. Our research purpose is to identify groups of Russian regions with similar characteristics in the family sphere using cluster analysis. The findings should make an important contribution to the field of family policy. We used hierarchical cluster analysis based on the Ward method and the Euclidean distance for segmentation of Russian regions. Clustering is based on four variables, which allowed assessing the family institution in the region. The authors used the data of Federal State Statistics Service from 2010 to 2015. Clustering and profiling of each segment has allowed forming a model of Russian regions depending on the features of the family institution in these regions. The authors revealed four clusters grouping regions with similar problems in the family sphere. This segmentation makes it possible to develop the most relevant family policy measures in each group of regions. Thus, the analysis has shown a high degree of differentiation of the family institution in the regions. This suggests that a unified approach to population problems’ solving is far from being effective. To achieve greater results in the implementation of family policy, a differentiated approach is needed. Methods of multidimensional data classification can be successfully applied as a relevant analytical toolkit. Further research could develop the adaptation of multidimensional classification methods to the analysis of the population problems in Russian regions. In particular, the algorithms of nonparametric cluster analysis may be of relevance in future studies.

  • Issue Year: 13/2017
  • Issue No: 4
  • Page Range: 1175-1183
  • Page Count: 8
  • Language: Russian