The aim of this study was to determine the size of the sample collected to describe the quality of feed. The research hypothesis was formulated as follows: collecting one sample from the bag of feed is sufficient to properly determine the feed quality. The work included conducting two research experiments for feed mixtures (8 and 12 ingredients), sampling during the bagging of 20 bags from three levels of bags, determining the content of samples and comparing the content of samples taken from three levels. Statistical analysis was based on the use of multivariate analysis of variance to determine whether there were significant differences in percentage for three levels of sampling. Based on the statistical analysis it can be concluded that there are no statistically significant differences between the percentages of individual components for three levels of samples taken from bags (about 3% of volume), so it seems that one sample (about 1% of volume) is sufficient to assess the mixture quality on this basis.