Your testing confirms that your recording has correct polarity.
The other source is the one which would seem polarity inverted, which fits DS's experience as well as mine with the positive going side typically registering higher values.
Another way to tell is to zoom way, way in on the waveform of an otherwise isolated transient (stick click, clap or the like) and look for which direction the waveform goes first. It should go up (positive voltage) before down, but can be hard to see if buried amongst other sounds.
PFAs can be wired for either inverted or non-inverted output. Jon informed me that those intended for powering low-voltage electrets are generally inverting unless he knows what mics they are intended for, as many of those mics have inverted output so the PFA inversion corrects that. The initial pair I received from him early on for DPA 4060 (which have a positive polarity output) inverted polarity, and I sent them back to him when ordering a few other pairs so that they would all have non-inverted output.