I think (i.e. infer from context) that they mean that in 2000, the people who were of an age to be getting PhDs[0] made up that percent of their age "group"[1].

I think (i.e. infer from context) that they mean that in 2000, the people who were of an age to be getting PhDs[0] made up that percent of their age "group"[1]. On the axes, I'm assuming due to poor choice of sig figs that "0, 1, 1, 2, 2" represents "0, 0.5, 1, 1.5, 2, ...", because that seems plausible. But yes, that figure (all the figures in the article, really) could have used a bit of editorial guidance. [0] vague! [1] other people "of an age to be getting PhDs"? also vague!