Today on a morning show, I was appalled at the graph used to make a point. The general point being made, by the usually reliable Steve Ratner, was that while more people than in the past are educated, their median income actually dropped...a counterintuitive conclusion for sure (if it were true!)
But look more closely...the graph on the left shows a higher percentage of the 18-34 age cohort is educated than in the past. This graph is fine in that the scaling of percentages across time is accurate.
The problem occurs when juxtaposing the left-hand graph with the right-hand graph. The right-hand graph shows median income across time, however it is not scaled from 0 which dramatically overemphasizes the income difference in recent years.
A 2nd, and potentially more important, problem with the right-hand graph is that it is for all people in the 18-34 age cohort, not just those with college education. In a way, you are truly comparing apples and oranges.
It is entirely plausible that the 21% of the 18-34 age population that is educated has a higher income than in the past but this increase is more than offset by the 79% of the 18-34 age group, without higher education, that may have dramatically lower incomes.
If that is true, than the real point of the two graphs, when properly scaled, might very well be that an education is even more important today than it ever has been.
Shame on Steve Ratner of Morning Joe...big MathFail!