OK, I fooled around with it a little bit and found that scanning was helpful in the sense of getting rid of the tonal distraction. The meter calibrates in increments of 1/12th stop, so I only need limited numerical precision for the tone matching. For some reason I find it very difficult to match tones by eyeballing them, so it was comforting to use a numerical value. Also, it didn't take very long. I could, of course, rerun the test strips at a finer step size and get even closer, but the scanner allowed me to estimate intermediate values, so I will run with this and see how it works.
Thanks for the opinions. Also, thanks for the link, JaZ. I downloaded ImageJ and will check it out Aside from doing low res scans to put on Flickr, I don't do any picture scanning, so I'm pretty ignorant in that regard. I don't even know if my scanner will do 16-bit linear.