> having two separate studies for A-B and B-C does not prove there's any A-C relationship
You're making a claim about the world. The evidence points in the opposite direction. If you have a problem with cognitive function relating to critical thinking, put forward some evidence about it. Because the link between those seems much more intuitive than your unsubstantiated hypothesis.
I don't have problem with either. But you're claiming A-C(education/critical thinking relationship) based on the above two without any justification other than your intuition.
B is not even the same thing in the two studies you mentioned. One uses "Lawson CTSR" measure secifically, the other uses nothing in particular, it's just a summary of research in which you didn't point to anything in particular. When it uses something it measures cognitive decline (which is not the same thing as the cognitive development measure in the first study, it's difference between two measures of cognitive development at different times).
Even the example table in the study pretty nicely disproves what you're claiming:
Someone with 22 years of education has cognitive performance of 10 while someone with 10 years has performance of 15 at the same age in that table. Lower educated just decline faster over time, but are still "smarter" in the end.
You also cite no numbers for correlations. So if A-B has correlation 0.2 B-C has correlation 0.3, A-C will have correlation 0.06 in general, which is nothing that proves A-C has any meaningful relationship.
Again, I'm not saying there's no A-C relationship. If there is it just needs to be studied directly. Pointing to two random studies proves absolutely nothing.
It may be true that there's some A-C relationsip, but having two separate studies for A-B and B-C does not prove there's any A-C relationship.
> So yes, voters with educational credentials are more likely to be better critical thinkers as well
Not based on your argument, no.