Summary: Calibrating is measuring an object as a function of another measure of this very object. Though cheap and easy to obtain, however, this last measure is not accurate. The true measure can be obtained with great accuracy but it takes time and effort. We synthesize controlled calibration and random calibration and also we make a comparative study of classical and reverse calibration methods. A synthesis of multivariate calibration introduced by P. J. Brown [J. R. Stat. Soc., Ser. B 44, 287–321 (1982; Zbl 0511.62083)] is carried out. The most substantial paragraphs are 5 and 7, devoted respectively to the creation of a universal methodology to estimate uncertainty in univariant calibration and to the numerical comparison of several calibration techniques.