We propose and analyze in detail a method to measure the in-air spatial spread parameter of clinical electron beams. Measurements are performed at the center of the beam and below the adjustable collimators sited in asymmetrical configuration in order to avoid distortions due to the presence of the applicator. The main advantage of our procedure lies in the fact that the dose profiles are fitted by means of a function which includes, in addition to the Gaussian step usually considered, a background which takes care of the dose produced by different mechanisms for which the Gaussian model does not account. As a result, the spatial spread is obtained directly from the fitting procedure and the accuracy permits a good determination of the angular spread. The way the analysis is done is alternative to that followed by the usual methods based on the evaluation of the penumbra width. Besides, the spatial spread found shows quadratic-cubic dependence with the distance to the source predicted by the Fermi-Eyges theory. However, the corresponding values obtained for the scattering power are differing from those quoted by ICRU nr. 35 by a factor of approximately 2 or larger, which requires of a more detailed investigation.