(don't you hate it when answers are unclear, or contradictory)?
The uncertainty comes from the fact that the information you gain from the meter reading will need to be interpreted with the effect of a polarizing filter in mind.
To understand this, I'd suggest an experiment. With the filter in place on the camera, take a number of meter readings, each with the polarizer set at a different position on its ring. You should see variation in the readings, because the polarizer tends to filter out the polarized light reflecting from the scene. You have to consider, however, whether it is the polarized light portions of your scene that you want to key into when you are choosing your exposure. In many cases, you are most interested in the other parts of the scene.
So which of the various meter readings do you choose?
If you meter separately, without the polarizer in place, and then apply the standard filter factor, you will get consistent exposure of the parts of the scene unaffected by the setting of the ring on the filter. That may lead to better results.
If it is a TTL meter, yes.
But if you are unsure, run a roll of film through the camera with the polariser in place, at midline and maximum effect. Then you will be in no doubt as to what works and does not. Be aware that a polariser set to maximum effect on overcast days e.g. in forests, can make the scene very drab, dark and generally underexposed, which is where you have to be on guard and leverage the exposure with exposure compensation. It all returns to the suggestion of running a roll of film through the camera and actively experimenting. That is the only way you will be sure.