Cylindrical ionization chambers produce perturbations (gradient and fluence) in the medium, and hence the point of measurement is not accurately defined in electron beam dosimetry. The gradient perturbation is often corrected by a shift method depending on the type of ion chamber. The shift is in the range of 0.33-0.85 times the inner radius (r) of the ion chamber, upstream from the centre of the chamber, depending upon the dosimetry protocol. This variation in shift causes the surface dose to be uncertain due to the high dose gradient. An investigation was conducted to estimate the effective point of measurement of cylindrical ion chambers in electron beams. Ionization measurements were taken with the ion chamber in air and in a phantom at source to chamber distances of <100 cm and >100 cm respectively. The data in air and in the phantom were fitted with the inverse square and electron depth dose functions, respectively. The intersection of the two functions provides an accurate estimate of the ion chamber shift and the surface dose. Our results show that the shift correction for an ion chamber is energy dependent. The measured shifts vary from 0.9r to 0.5r between 6 MeV and 20 MeV beams respectively. The surface dose measured with the ion chambers and mathematically determined values are in agreement to within 3%. The method presented in this report is unambiguous, fast and reliable for the estimation of surface dose and the shift needed in electron beam dosimetry.