Jeni's Musings

I think these comments are WAY too harsh. Most methods of representing data are misleading in one way or another. Anyone who “can’t read” something because a graph is misleading is going to: (a) miss out on a lot of interesting stuff; and (b) be misled by all sorts of people when they DON’t notice that a graph is misleading. “Depressed zero” is a recognised, if always to-be-suspected, way of showing better the differences between a set of data points, rather than their absolute values. Look how the differences between Boroughs are more apparent on the start-at-70 graph above. You’ll find depressed zeros in reputable sources as well as disreputable ones. Reading a graph, one should ALWAYS look at the axes and see what they reveal about units, position and so on : the really bad graphs don’t give you enough info to know what’s happening. Jeni’s first graph did give enough info. Even if, like me, one was initially “misled”, it only takes some care to recover and realise one’s mistake. In doing this, one has engaged with the data and is likely to understand it even better than from a more superficial immediate acceptance. Dismissing it (and accompanying material) because it could have been presented differently does not strike me as sensible.