Any 'test kit' that doesn't come with a listed resolution and error probably isn't trustworthy.
"They" say their testing method is +/- 10% ... but its measuring a vague "Compounds by Volume" ... it's not selective for thca or any specific cannibinoid.
If doing the method for "OVER" error, also have to factor in error of the scale, too, so that will compound with the 10%. e.g. 1.1 * 1.05 = 15.5% error if the scale is contributing 5% measuring error on volume.
I'd wager there are all sorts of other molecules in solution that could confuse this test. It's fine for entertainment purposes, but i woudln't have much faith in the test results. Seems like it's using some sort of chromotagraphy.. it's not isolating a particular molecule, measuring, then dividing by original mass or volume etc...
in the right conditions it might work okay, but it's depending on correlation, not causation.