I would not sweat it too much. It is more likely due to the fact the output meters in the Quad capture may be of a different type and they may also have a different response compared to Sonar's meters
Sonar's meters are showing peak levels hence they are always going to be reading higher than a meter that might be showing showing something else eg rms or a combination of rms /peak.
Put Sonar's output meters into peak and rms mode and have a look at them and see what the differences are.
Also you are only talking the Quad Capture output metering in software, there are no actual output meters on the unit itself.
The only real way you can calibrate meters properly is to use a continuous sine wave test signal and play that back on a Sonar track and observe the output level (rms) in Sonar and on the Quad Capture. I think the you may find they will agree much more so. Music is not necessarily an accurate way of comparing metering.