Modelling wheat yields variability in Polish voivodships

The paper presents an analysis of wheat yields variability in the voivodeships of Poland. The main aim of the study was to find out what are the statistical relationships between the wheat yield variability and the following factors: arable area, size of wheat production area, share of arable land used for wheat production, land quality and average yield. For that purpose a multiple linear regression was applied. It was found out that the detected spatial autocorrelation of wheat yields variability measured by standard deviations can be explained in 75% by the fitted model. Two of the considered variables showed a significant negative effect on this variability: the logarithm of arable area and the land quality, while the other two: the average wheat yield and the wheat production area displayed a significant positive effect on the variability. The effect of share of arable land used for wheat production itself was not significant.