The calibration of space solar cells can be performed under extraterrestrial conditions (in space) or Earth-bound (either in the laboratory or outdoors). The corresponding calibration procedures and correction methods to obtain the calibration value are described for measurements on a balloon or in a jet as well as in a solar simulator or under direct or global outdoor illumination. Special emphasis is given to bias and random errors that may in¯uence the measurement uncertainty. For terrestrial calibration methods utilizing a spectral mismatch correction, an ambiguity in the calibration value arises due to dierences between two well-known sources for the extraterrestrial spectral irradiance, as given by Thekaekara or the World Radiation Centre and to dierent estimates of the solar constant. Some calibration methods use light sources with spectra that are very dierent from the extraterrestrial spectrum. The impact of this on the size and error of the mismatch correction is quantized for all relevant cell technologies. As the mismatch correction depends both on calibration spectrum and cell spectral response, there is not a single terrestrial calibration method with best uncertainty for `all' cell technologies. Rather, good control of measurement errors in spectral response and simulator spectrum measurement determine the applicability of the various calibration methods.