Four-paint impedance spectroscopy of solid materials systems is severely hampered by unavoidable voltage-divider effects associated with the reference electrodes. As demonstrated by test circuit studies and experiments with Pt/YSZ/Pt cells (with embedded sliver reference electrodes), high impedance reference electrodes can produce distorted and erroneous impedance data. The relationships between these data and the sample properties (conductivity, dielectric constant) and instrument Limitations (input impedance/capacitance) are derived. Successful four-point impedance measurements on conductive systems require large effective dielectric constants, which may be unattainable in bulk solids, but are often associated with internal interfaces.