CD, here's what I'd do if it were me:

1) Get a measurement rig (either have someone come over who has one, or you can put one together - a "turnkey" option is the Omnimic, but you can put together your own rig for less than half the price of that).

2) As VOTT suggests, take everything in the signal path, write it all down, and build a test plan that allows for understanding what each component is doing to the measurements. For example,
* start with raw drivers (free air impedance sweeps),
* then add in the cabinets (cabinet loaded impedance sweeps),
* then add in the amp with no EQ or crossover (raw driver FR measurements)
* then add in the crossover
and so on. Time consuming, yes. But fun, and only with that methodical approach can you get down to the nitty gritty of how all the pieces are changing the sound.

3) If you can't get hold of a FR measurement rig, you can still go through the steps in #2 with listening tests and get some value from it, and at a minimum you'd probably want to keep a "listening journal" of sort to be able to keep track of everything as more of the system components get layered in.

just my 2?