Standardization methods in activation analysis with charged particles are studied critically. Several approximate standardization
methods that do not require knowledge of the excitation function are compared with the “numerical integration method” using
excitation function data from the literature. It is shown that these methods yield accurate results if the threshold energy
of the considered reaction is high and if sample and standard have a comparable Z value. A method that gives a rapid estimate
of the maximum possible error is also presented. It is shown that for the “numerical integration method” the accuracy of the
excitation function data has only a small influence on the overall accuracy. The influence of the accuracy of stopping power
data and of possible deviations from Bragg's rule for light element standards is also considered.