The possibilities of application of three multivariate methods: multiple regression analysis, principal component analysis, and factor analysis for examination of relationships between the height growth of forest stands and soil properties are compared. The case study was based on soil samples taken from 148 experimental plots established in Norway spruce stands. Nineteen physical and chemical properties were recorded in two horizons, and their relationships to the site index based on the top height were studied. A combination of different multivariate techniques in proposed as an optimum solution. Factor analysis with an oblique axis rotation allows to identify the soil factors as a composite expression of measured and observed soil characteristics. Subsequently, after calculating the factor scores, multiple regression can be used for the construction of a growth prediction formula.