<p>The aim of this paper is to describe a metrological approach for measurement uncertainty evaluation, applied to the quantification of melamine in a food matrix, being melamine a well-known food adulterant with harmful effects on health. The work provides insights on the calibration process of a Raman spectroscope, carried out in the melamine amount fraction range (0.16–5.0) mg&#xa0;L⁻¹ and based on the application of Weighted Total Least-Squares regression. The focus is on the evaluation of the measurement uncertainty associated to the recovery efficiency of the method. As a case study, an experimental method previously developed and validated at INRiM and based on the Surface Enhanced Raman Spectroscopy was considered as a starting point. The Law of Propagation of Uncertainty was applied to evaluate the measurement uncertainty of the results. The proposed approach for the uncertainty evaluation could be applied to the detection of the levels of melamine in real samples, as milk or milk-derived ingredients. </p>