How to level match when comparing components?

I agree with Elizabeth and would take it one step further. Leave A in for a week or a month, then replace with B. Limit music during that time to that which you are intimately familiar with and best represents the range of your musical preferences with some emphasis of material that is demanding of the system. Otherwise A/B is useful on in knee-jerk evaluations of spot differences. How those differences actually play out in the long run enjoyment of your system are not fully understood in a quick A/B comparison, IMO, though some really blatant ones might be easy to eliminate some components from consideration entirely.

As far as matching levels, the original question, you can also get level meter apps for an iPhone and probably other similar devices that would be useful for such a simple task. I think the iPhone app that I have is by Studio Six Digital. One other note, it would be useful to have a recording of white noise (or pink noise) to actually use in association with the meter in order to match the level. It will be easier to use that then a recording of music that is probably changing rapidly. You can probably download something by doing a google search.

Then you have the real problem. Which do I *like* better, which *is* better

I hope you are joking. If those two are different in your mind, then you do have a real problem. If you start trying to please someone else, or second guess some "absolute" or objective then you will never be happy with what you choose. The only thing that should be of concern is what sounds best to you, assuming you are going to be paying for it and living with it and enjoying your music through it. I guess if you are buying as a gift for say, Micheal Fremmer, then you should probably be concerned with what he thinks....otherwise it's up to you.

The OP asked a simple question. The answer pink noise and a db meter.
He/she is not asking whether one should be analytic or emotional regarding component comparisons or what the requisite period of time for a valid comparison.

And (s)he was answered, and it turned into a more interesting discussion. The direction of the discussion is entirely pertinent to the direction of his question. What's the problem?

Is there a good pink noise generating test disc out there to help with this?

Stereophile's test discs have pink noise.

You can also download a file from the Internet for free like those available here and other places. If you only have a CD player, download it and rip a CD from the file.