Different forms found in the literature to express the error between calculated phase equilibrium variables using equation of state models and experimental data are analyzed in this note. High pressure PTxy data of binary mixtures containing supercritical carbon dioxide are used as study cases. The common practise of using the solvent concentration, instead of the solute concentration, to analyze the accuracy of a thermodynamic model is critically discussed. Several statistical parameters usually employed in the literature are considered to be misleading and it is shown that they do not give a clear indication about how accurate are the model results.