Compounded assessment of heavily polluted areas can be a challenge, especially in mostly heavy industrial region such Upper Silesia (Poland). The main and the biggest river flowing through this region is Kłodnica River.This study presents an application of multivariate data analysis in the field of river sediments pollution. The dataset consists of As, Cd, Cr, Cu, Fe, Ni, Mn, Pb, and Zn contents in sediment samples collected from the Kłodnica River (Poland) in sampling campaigns (2013 and 2014). As chemometric statistic tools cluster analysis (CA), multivariate analysis of variance and discriminant analysis (MVDA) and factor analysis (FA) were used to investigate the matrix of 32 sampling points.The cluster analysis presents that pollution can be distinguished into three groups which were strongly dependent of the contamination's concentrations (As, Cd, Cu, Pb, and Zn) and the localization of the samples. Multivariate analysis of variance and discriminant analysis confirms the results from the CA. This method was used also to apply real reduction in the tested matrix dimension by use of forward strategy. In the 4 th step of that strategy the variables As, Cd, Cu, Pb, and Zn are sufficient to describe the variability in the river sediments and to separate the groups of sampling points.Two factors obtained from the factor analysis explained approximately 61 % of the total variance of the river system and allow distinguish of the dominant anthropogenic pollution sources in the river system. Factor 1 describes 36.74 % of the common variance and is highly loaded by As, and Cd. Factor 2 induces pollution with Cu, Pb and Zn and explains 24.67 % of the common variance. The interpretation of factor analysis was presented by the representation of factor scores as a function of the Kłodnica River-kilometers.