Alas, set theory and the Venn diagram approach has been off the UK syllabus for some time as well as in the US.
I take your point about the correctly drawn/direct proportion version of the graphic being more quickly assimilated. I just think that with such a 'dynamic range' of data (0.1% compared with 0.2% in one category compared with 17% vs 14% in another) any kind of graphic will require very high resolution to convey the overall picture. I suspect that may have been the reason for the non-proportionality in the original Nielsen graphic.
I take your point about the correctly drawn/direct proportion version of the graphic being more quickly assimilated. I just think that with such a 'dynamic range' of data (0.1% compared with 0.2% in one category compared with 17% vs 14% in another) any kind of graphic will require very high resolution to convey the overall picture. I suspect that may have been the reason for the non-proportionality in the original Nielsen graphic.