29 Jan 2018 21:17 GMT
A few weeks ago I posted this scatterplot on Twitter. It shows Office for National Statistics estimates of net internal migration between local authorities in England and Wales in the year ending June 2016. The local authorities are grouped into quartiles based on how urban or rural they are.
As you can see, there is an interesting pattern: the local authorities with the highest net outflows are among the most urban. But not all urban areas have large net outflows, and there are not many rural areas with correspondingly large net inflows.
I wanted to find a way of representing these migration flows that would let me explore how people were moving between local authorities: which local authorities had the largest flows in each direction, and what was the balance of flows between these and other local authorities?
With that in mind, I built an interactive hexmap that shows the ONS internal migration data.
How it works
The map shows the 348 district and unitary local authorities in England and Wales, and shades each of them according to the magnitude of their internal migration flows. By default the map shows the net flow in each area (these are the flows shown in the scatterplot above), but the tabs let you switch to see the gross inflows and outflows too.
If you click on a local authority (or tap on it twice on a touchscreen) you can see the flows between that local authority and other areas. Mousing over an area (or tapping on it once on a touchscreen) pops up a label showing the name of the local authority and the flows in question.
The map stores the currently selected local authority and flow in the URL, so you can link directly to a given combination of local authority and flow. Here are the net migration flows between the London Borough of Barnet and other local authorities, for example. You can see that there are net flows into Barnet from local authorities in central London, and net flows out from Barnet to less urban areas. And here is the map for Sevenoaks, which shows a similar pattern.
Strengths and weaknesses
I think this approach works well in some respects and less well in others. It effectively condenses a very large amount of data into a relatively simple interface that allows you to easily explore geographical patterns of internal migration, which is what I set out to do.
However, one weakness of this design is that the variation in the size of the migration flows between local authorities is so great that it's not really possible to represent the data using the same shading scale for every area.
Consequently, while each type of flow is represented using the same set of colours, the scale associated with those colours changes depending on which local authority is selected.
This ensures that you see an appropriate level of variation in the data for each local authority, but it also means that the maps for each area are not directly comparable with one another: you always have to glance at the colour key to check the magnitude of the flows shown in each case.
Effectively, this interface provides 348 different local authority maps for each type of migration flow. The maps all use the same grammar to communicate, but they show the data at different scales.
A new hexmap of local authorities
One final feature of the visualisation worth mentioning is the hexmap itself. This is an entirely new hexmap of local authorities in England and Wales that was built from scratch. I intend to write more about the process of making it, and share some tools I have developed to make creating hexmaps easier, so look out for those in forthcoming posts.