Electrode polarization analysis is frequently used to determine free-ion diffusivity and number density in ionic conductors. In the present study, this approach is critically examined in a wide variety of electrolytes, including aqueous and nonaqueous solutions, polymer electrolytes, and ionic liquids. It is shown that the electrode polarization analysis based on the Macdonald-Trukhan model [J. Chem. Phys. 124, 144903 (2006); J. Non-Cryst. Solids 357, 3064 (2011)] progressively fails to give reasonable values of free-ion diffusivity and number density with increasing salt concentration. This should be expected because the original model of electrode polarization is designed for dilute electrolytes. An empirical correction method which yields ion diffusivities in reasonable agreement with pulsed-field gradient nuclear magnetic resonance measurements is proposed. However, the analysis of free-ion diffusivity and number density from electrode polarization should still be exercised with great caution because there is no solid theoretical justification for the proposed corrections.