We have conducted a comparative study to determine the most efficient and reliable way to calculate the depth of investigation (DOI) index to assess the quality of electrical resistivity tomography models. We compared the results of using resistivity and logarithm of resistivity after testing them on four synthetic models by direct modeling and a field case, in which the resistivity model was validated by auger drillings. We tested the two most commonly used acquisition arrays, dipole-dipole, and Wenner-Schlumberger. The index calculated with the logarithm of resistivity clearly appears to be more satisfactory than the resistivity-based index. The method based on resistivity systematically overestimates risk (high DOI) in areas of high resistivity, and it underestimates risk in conductive zones. As a result, we strongly recommend the use of the logarithm of inverted resistivity to calculate the DOI index.