Abstract
In recent years, a lot of analysis was carried out to investigate data on chemical compositions of (non-)alcoholic beverages (whiskey, wine, beer, etc.) and food. However, no study has ever considered the compositional pitfalls inherent in such studies, which may lead to arbitrary results. Two approaches were compared, compositional data analysis (CoDa) and classical statistical analyses. The outcome of compositional data analysis methods provided very well-interpretable results. The method we are proposing is novel in the field of food chemistry, with demonstrated ability to analyse the chemical composition of drinks and food.