explain what is wrong with the way regression is used in the

explain what is wrong with the way regression is used in the following question.
using data from several cities on x=% of residents with college educations and y = median price of home,you get a strong positive correlation. you conclude that having a college education causes you to be more likely to buy an expensive home.

Solution

Data is not paired. We cannot use regression in the following question.