Way too complicated and complex to shortly explain.
Have done it many times before, at RC planes and jets tournaments when all radio transmitters had to be checked for band deviation (bleeding) and power Tx output.
Contestant will put his radio on the table, turn it ON and move the sticks. I was at about 150 yards away and look for Tx RSSI, power, band, center peak and deviation. I had no idea of who's radio was, what type and model, I was just looking for his radio transmit data. You have no idea of how much difference is between radios of the same brand, type and make.
I am sorry I do not understand these complex things but.....
If you are simply looking at carrier amplitude together with deviation on a FHSS system you are not testing the system.
First of all, apart from the fact that the frequency is shifting within the band so to is the power output of the transmitter dependant upon the received signal strength. The telemtry sent back from the RX will call for more power from the TX if certain criteria are met. This is not based on amplitude of carrier alone but also on the parity data check on the packets sent within the data stream. Furthermore, the data is sycronised and sent in frames and this information needs to be received, validated, decoded and interpreted by the RX.
What you are doing is a simple rf strength and a crude quality test on the transmitters regulated and non controlled output. Switching two identical transmitters on and testing them this way will indeed give differing results since.
1) You are not in a controlled environment and therefore have no control over the noise floor, multipath situation, or power output from the TX final stage
2) You have no way of generating or confirming the checksum data required with the packets sent and/or received and therefore have absolutely no way of determining the quality of the signal (a lower powered signal can easily have far better quality)
3. You have no control over the people in the environment, their cellphones, other transmitting devices within close proximity
4. Simply watching the waveform while someone waggles the sticks a bit won't actually tell you anything since a stick input is directly translated into packet data and then transmitted within the data stream. Without knowing the integrity of that data it's a fairly pointless exercise.
I will bow to your superior knowledge on the subject but fail to see how simply setting up a spectrum analyser in the field and looking at a FHSS based signal is doing a meaningful test such as we are talking about here.