If you wanted to do a theoretical analysis of how much feedback is introduced by a given feedback resistor size in a global feedback loop scenario, you would build a mathematical model using the basic idea you suggested. The core concept in this kind of approach is to model how much voltage is applied at voltage divider junction of the feedback resistor and the lower resistor in the feedback circuit. This is kinda complicated and only as accurate as the assumptions made in developing the model.
On the other hand, if you have the amp on the bench and you have an oscilloscope and a function generator, I find it easier just to set the amp up with a proper dummy load resistor on the speaker outputs, run a 1KHz sine wave through the amp and set the output to about 1/4 watt with feedback connected. Then measure the voltage across the load resistor with and without the feedback line connected. Take the ratio of those two voltage measurements, then convert to dB. This is a reasonably accurate measure of the actual amount of feedback applied.
