Summary Recent advances in understanding the induced polarization (IP) method have led to improvements in interpreting hydraulic properties from electrical measurements. Distinguishing the effect of surface conduction from conduction through the electrolyte filling the interconnected pore spaces has been an ongoing challenge in interpreting field-scale electrical resistivity datasets. Previously proposed mechanistic models have suggested that this limitation can be overcome by utilizing the coefficient that describes the ratio between IP measurements and surface conductivity. In this study, we examine this proportionality coefficient ($\mathcal{l}$) through the relationship between IP parameters (imaginary conductivity and normalized chargeability) and surface conductivity for a sample group of 98 sedimentary rocks, composed of sandstones, carbonates, and mudstones. A strong linear relationship is observed between the IP parameters and surface conductivity. However, values of $\mathcal{l}$ vary significantly across each sample group such that low-salinity estimates of formation factor (F) using the single universal estimate of $\mathcal{l}$ are poor. Estimates of F for a single rock type (sandstone, carbonate, or mudstone) are improved using $\mathcal{l}$ values unique to that rock type, although F estimates for mudstones show high sensitivity to changes in $\mathcal{l}$. Using $\mathcal{l}$ coefficients calibrated on a sample group with similar lithological properties to the investigated group moderately improves the estimation of F. Consistent with theory, no relationship is observed between the proportionality coefficient and the measured petrophysical parameters of the porous medium. Our results suggest that, although IP measurements provide a valuable field-scale proxy for surface conductivity, improving petrophysical predictions (i.e. in this case, estimating F) remains a challenge.