I tried to import a 22.5kHz, 16bit mono wave file into the audio timeline, synchronize it with the video clip and then export it again as a 22.5kHz, 16bit mono wave file (no filters, no changes in volume levels, nothing). That works in general without any problems but i realized that the amplitude in the audio file is actually 3dB lower than the input.
Repeated the same steps with 250Hz sine wave and white noise with the same results. The sine wave was 3dB attenuated and the white noise's RMS was 3dB lower as well (not the peak value, though).
Any ideas why this is happening and how to fix it, apart from applying a +3dB gain to everything I do?