How does "Thing decreased in half" sound like the effect you'd expect from population independent variables? If, in the last century, the average number of toes had decreased in half I'd have a hell of a lot of questions.
"the average number of toes had decreased in half" is very hard to parse. I have no idea what you intended by this proposition, but here is my interpretation, and yes, I believe you do expect to see one half of the population average a lower number of toes.
Say I measure the average number of toes in the population twice. Once in 1990, and once in 2020, and there is no difference in the average number of toes between these two populations. Now instead imagine, for my second measurement, that I sort the people by how many toes they have, pick midpoint of that list, and take the average number of toes in each group. The average number of toes has now indeed decreased in this half. Mind you, I'm just responding to what you said. I have no idea what you meant.
