First of all, impedance is a circuit that consists of either resistance, inductance or capacitance. Impedance can be purely resistive, purely reactive, or purely capacitive or a combination of any of them. Impedance that is inductive or capacitance will change depending on the frequency. The impedance, or ohmic value increases with frequency in a reactor. In a capacitor, the ohmic value decreases with increase in frequency. It is a possibility that the variation in measured value you are referring to is due to a unspecified test frequency. Headphone impedance is reported at 1 kHz to allow comparison between apples with apples instead of apples and oranges. To really know the true nature or the impedance of a circuit, measurements need to be made using an oscilloscope to detect any phase shift between voltage and current for a wide range of frequencies. In a purely resistive circuit the voltage will be in phase with the current. In a capacitive circuit the current leads the voltage. In a reactive circuit, current lags the voltage. I know the ER4B cable uses capacitors, so I wanted to throw that out as a possibility regarding the P->S cable. I will at one point verify the nature of the impedance in the P->S cable. I have a multichannel oscilloscope. But I think that to avoid problems, folks should use the P->S cable.
