[en] Monte Carlo simulations have been used to assess the accuracy of the retarding field method used to unfold experimental electron energy distributions. It is shown that electron reflection at the electrode and extrapolation to zero energy of experimental data can explain, at least in part, the observed agreement between experimental and calculated energy distributions. Reasons for the good but not perfect agreement between theory and experiments in Ar at elevated E/N are discussed, together with the intrinsic limitations of the experimental retarding-field technique