The article deals with the problem of determining socio- economic differences within the constituent entities of the Siberian Federal District. The authors have conducted comprehensive research of the degree of socioeconomic development difference using cluster analysis methods based on the operational data of the Altai Federal Statistics Service. The data characterize economic and social situation in the Altai Territory, Krasnoyarsk Territory, Zabaykalsk Territory, Novosibirsk Oblast, Omsk Oblast, Tomsk Oblast, Irkutsk Oblast, Kemerovo Oblast, the Republics of Altai, Tuva, Khakassia and Buryatia by the end of 2014. The following indicators of socio-economic development are used: per capita income, per capita consumption, the average monthly gross payroll, agricultural output, the amount of construction work performed, the retail trade turnover, total employment rate and unemployment rate in the economically active population. As a measure of similarity, the authors use the Euclidean distance in the normalized variables space. The normalized variables have been calculated as a ratio of the difference between the initial values and the mean value and the variance for each variables set. The use of k-means method led to building the cluster structure of the Siberian Federal District with 2- to- 6- cluster division. The authors define the location, composition, statistical characteristics of different levels of clusters, the degree of similarity and dissimilarity between the constituent entities of the Siberian Federal District.