[Glennbo]

It's prolly all incorrect, but I use a visual and auditory method to set the level of my final mixes. When rendering to wave or MP3, if I don't see some air space in visual image of the wave being rendered, I cancel and make some adjustments until I do see some significant variation.

Likewise, if the wave image has tons of whitespace, I'll cancel and go the other way until I have a wave image that looks both like it is hitting the fringe, and also leaving a fair amount of breathing room.

Finally after getting a full render that looks OK visually, I'll listen to it, and compare it with both commercial media, and my own previous mixes to see if it seems in line volume wise. I like slamming mixes, but not to the degree that they have no space left in them and feel crowded.
-