For the first questions, I think the reason most people answer like they do is that when you ask a question in the form "if we change A, what happens?", the implication is that no other relevant variable has changed. And sure enough, if we just build more houses, housing will get cheaper. The article then says "actually, the reason we built more houses is because population is increasing, so housing gets more expensive". This is the kind of details that should be in the question.

It is like asking "if you press the gas pedal harder, will your car go faster or slower". Of course people will say "faster", but then you say "got you! if the reason you pressed the gas is because you are climbing a slope, you are going slower".

Yeah, a common problem in cognitive bias research...