I am using wavelab. I usually use their resampling and a L3 when I mix my own things. The L3 does dithering to. Should I resample and then use the L3 or do them both together. Actually Im going to start recording at 44 instead of 48 soon so I wont have to worry about it to often but occasionally I will master something for someone else so I still want to know the correct procedure

If I needed to do both, I'd probably just do it at the same time - Resample, limit (again if necessary, as you'll almost undoubtedly have overs after resampling if it's a relatively hot signal), dither.