A lot of the talk is about how the mean is changing, but from the data it looks like the biggest change is often the spread. This is important because, eg, the 35-44 cohort aren't making an average coffee, everyone in that group is making their own coffee. In all cases across all groups, the 2:1 ratio is clearly dominant. What changes is the size of the minority that makes it much shorter or longer. In other words: the mean isn't the only thing that matters, the mode show that there's really very little change between all the variables looked at.