Yes, if you have a milliohmeter, or the device has quite high resistance to begin with.
As mentioned, the voltage drop under load is probably a more repeatable figure; it also accounts for the somewhat nonlinear Rds vs. Id curve, so you know what voltage drop to expect in place.
An aside: if the voltage drop, at the test current, is less than a diode drop (< 0.5V?), it doesn't matter which direction you measure it in.
Tim