Assuming this case and the mixer works fine (and has NO internal leakage faults). Most mixers are two wire devices. That means the mixer has power. If that's the case, i see two possibilities. either the hot and ground are reversed and the mixer uses a polarized plug, or the ground on the receptacles (which is NOT used on a two wire appliance) is NOT grounded back at the service entrance. And the op is trying to use this ground which is not really hooked up to anything for his tests. Now, either of these should be an electrician issue. Maybe the house was a older house and was upgraded from two pin to three pin sockets without actually putting in and connecting the ground wire. It would not be the first time this has happened. I think you can legally do this in some retrofit cases by using a 3 pin GFI outlet with no separate ground wire back to the service panel.
Accuracy has nothing to do with it. Its the input IMPEDANCE of the volt meter. A digital fluke has a typical input z or 10M ohm or so. It can do this because it has a very high impedance OP amp buffer on the input side.
That kind of high z input will allow a phantom voltage to read. so it has NOTHING to do with accuracy, and every thing to do with the load the meter puts on what is being measured. Older analog meter have a mid level impedance. from 1000 ohms to 50K or so. Depends on the model. Most have no input buffer amps (Older VTVM excepted), so there input impedance is a combinations of the meter movement and the voltage divider resistors used for the voltage ranges. This mid level impedance is enough of a load to make these phantom voltages disappear. Phantom voltages have voltage potential, but little current drive potential. A phantom voltage in black box form would be a voltage source with a VERY high resistor in series with the voltage source. So it cannot drive much current into a load.
This phantom voltage is like getting shocked off the carpet by static electricity. The voltage might be there but its not enough to cause any harm.