Matthew's Journal

Interesting, yes, and no doubt fairly objective, but I'm left with a feeling there must be more revealing ways to present the same underlying data.

For starters, it feels potentially misleading to map the modal causes of death: with nine categories, the cause they show could represent anything from 11% to 100% of deaths! More octogenartans than teenagers could be dying on the roads and the maps would leave us none the wiser.

Secondly, the outline doesn't look entirely like the UK to me. Is it some kind of magic stretch so that each hex has the same population, or has something altogether stranger happened to the map? In any case, it makes it very hard for me, at least, to overlay the map with my general geographical knowledge of the UK: I couldn't reliably point to Cambridge on the map, let alone figure out the extent of the London conurbation.

But mainly, it feels like someone ought to be doing some number crunching to build graphics that illustrate trends. The BBC's article says cancer dominates in rural areas in late middle age: it would be interesting to see an aggregate line graph for all areas judged rural, plotting prevalence of each cause of death against age. Similarly, such graphs for a few latitudinal bands might better illustrate the North-South disparities.