This paper deals with the development of a fully electrical method to estimate the sensitivity of capacitive MEMS accelerometers in batch fabrication. The objectives are to test and calibrate them without the need of costly and time-consuming mechanical test equipments. Alternate electrical test measurements are introduced and two methods are evaluated. The first one is based on the analytical study of the relations between the electrical test parameters and the sensitivity. The second one is based on a regression analysis of these relations on a learning batch of sensors. Both are evaluated on a case study accelerometer with Monte Carlo simulations. Applying the methods for calibration shows that the dispersion of the sensitivity can be improved. Results show that the second method is more promising.