I think Arne has raised an interesting issue. Seems there could be an accuracy standard that all spine testers, draw weight scales, etc should be required to to meet if they want a certification by AMO, or some other recognized standard.
I don't own a spine tester, but it would be nice (if I did) if I could adjust it to match a calibratiion at the mid-range of the spines I'm interested in. Testing a dozen shafts of the same rated spine by several different makers should put the value close I would think.
Of course, unless you are manufacturing arrows, the readings can be used for their relative value if the tester is consistent. If I have an arrow that is supposed to be .500 spine, but reads .520 and it shoots perfectly as set up for my bow, then I will know I want arrows that spine .520 on my tester. Still, if you pay your money for what is supposed to be an accurate device, then you want your money's worth.