So, as far as I understand, g-factor is some theoretical value, that affects an ability to perform all mental tasks, IQ tests are the way to measure g-factor, and the Flynn effect is the phenomenon of increasing IQ scores through last decades without actual increasing of underlying g-factor. Correct me if I'm wrong somewhere.

Assuming all of it is true, my questions are following:
> 1) Do I understand it correctly, that we can't measure the pure g-factor itself, and can only measure it approximately through IQ tests? And we know how good IQ test at measuring g-factor by looking at how strong results of subtests correlates with each other?
> 2) How exactly do we know, that changes in IQ between decades are NOT caused by changes in g-factor?
> 3) Can we measure, what percentage of IQ change is due to non-g_factor reasons?
> 4) If we can, why can't we adjust IQ value to real g-factor value given correction for non-g_factor reasons? And if we can't, why, and how can we even trust IQ tests then, if we don't know, how much of it is affected by real intelligence?
> 5) How do we know that racial IQ gap is correlates with g-factor?
> 6) What caused Flynn effect then if not changes in g-factor?