Friday 18 November 2011

Forbes map of American Migration

This map by Forbes (and IBM) on patterns of American Migration caught my eye this week (Go to the web site here). At first glance it's an elegant design and has the right sort of graphics and interactivity you'd want in a clean, authoritative web map. In cartographic terms though, it flatters to deceive. Here are a few thoughts on why the message of the map is compromised.

The lines between areas add to the initial impact of the map but since they don’t actually represent quantity of migratory flow (through variable width/transparency etc) they provide very little and tend to obscure the choropleth underneath - particularly when there are a lot of lines. It’s good that they can be hidden but the idea of migration would be best served by making proportional line widths. So if they don't actually serve a purpose they're just chart-junk.

I have a suspicion they are mapping totals rather than adjusting by area or population (we don;t learn what the choropleth units are anywhere). The pop-ups seems to support this notion so the map itself doesn’t illustrate sensible distributions. It’s cognitively confusing. A better choropleth would be to show the difference between in- and out-migration, per capita which would show the relationship between the two variables as well as being proportional to the overall population count.

The inclusion of income per capita is a little confusing. What’s the message in relation to migration?...that people move to richer areas perhaps? It seems to have just been included as an additional variable. It would help if this was more clearly explained so the map reader isn’t left trying to decipher numbers and their relationship spatially, statistically or conceptually.

So all in all it looks nice but actually offers very little and challenges the map reader to make any sensible interpretation.

1 comment:

  1. @vicchi and I were gushing about mapporn at GeoCommunity when Kimberly Kowal treated us to a quick journey through the British Library Map Collection's treasures but perhaps we should refactor the term for mapgizmology without purpose