There ARE standards, but this gets as subtle (as in the usual manner there is more then ONE).
There are for example multiple weighting curves that will give different numbers, especially when you consider that CCIR 468 uses a defined quasi peak detector.
My usual go to instrument is either an Audio Precision test set or a Prism Sound (Now 'Spectral measurement I believe) DScope III, the AP is a better instrument but the DScope is way more user friendly and easier to script for repetitive measurements.
AP have a wide range of generally very good app notes on audio measurement, worth your time even if you cannot afford their latest and greatest.
The biggest favour you can do people is to SPECIFY what the measurement conditions are:
Frequency response : 20Hz - 20Khz is useless, it tells me nothing except that the vendor is likely playing games.
Frequency response : 20Hz - 20Khz +-0.5dB, phase shift less then 10 degrees, tells me what to actually expect.
Input/output impedance is usually not a big deal, you can do a load pull if you really want to know, but for reasonably contemporary line level you are typically expecting input to be a few k (differential mode) and output to be ~100R (again differential mode). Outputs ARE sometimes expected to drive 600R but that is largely a legacy thing.
THD and THD+N you can hit with a good computer sound card, possibly combined with an active notch filter to get you more fundamental rejection (Build it in a cake tin and run it off batteries to try to keep the mains hum out).