How is the ERI affected by the removal of the data related to one electrode?

For a profile with N electrodes, we plan to invert N+1 datasets: the dataset with all available electrodes and N datasets from which we will remove the data related to the ith electrode. We will then compare the result of each "image" relative to the first one. Using this technique, we will be able to evaluate the importance of the data related to an electrode in the inversion process. If the influence is very low, it could mean that the subsurface is quite homogeneous; a larger electrode spacing or lower measurements could be done. But if the influence is large, it could mean that the subsurface is showing small scaled features, or that the electrode site is causing trouble. We plan on applying this methodology to the synthetic data sets proposed on the NSG website and possibly on some real data recently acquired in Belgium.