Saturday 3 September 2016

SIMD 2016: Concentration and Contrasts

The 2016 version of the Scottish Index of Multiple Deprivation was released this week. It's very similar to indices used elsewhere in the UK and other parts of the world. The interactive mapping is great, so I'm not going to do anything on that. Instead, I looked at areas where there are contrasts or concentrations relating to the most and least deprived deciles. With a relative deprivation measure, some places always have to be at the bottom or top, but it's the location, level of spatial clustering, proximity and total number that I was interested in exploring. Here are some maps, starting with Glasgow and then going on a tour round the country. Red areas are in the most deprived 10% of Scotland and blue areas are in the 10% least deprived.

This is probably the bleakest map, in terms of concentrated deprivation

The famous Drumchapel/Bearsden cross-boundary contrast visible here

A bit more mixed towards the south east of Glagsgow's urban area

Alness and Invergordon - some of the more 'hidden' areas on most SIMD maps

Ayr is very notable for its north/south deprivation split

Edinburgh is often contrasted with Glasgow - and you can see why

Again, a somewhat mixed pattern in Falkirk and nearby

Hamilton and Motherwell are also quite mixed in a 'most-least' sense

Inverness - persistent pockets at both ends of the spectrum

A little bit of Fife - which is a bit split in this view

I did another 20 or so maps, in addition to the ones above. They can be seen on a separate Google Drive page if you are interested.

These are all in high-resolution - feel free to use them

I'm not planning to do any more SIMD mapping as there are enough other people looking at it. I just wanted to explore a little bit in relation to how the most and least deprived locations were spread across Scotland. Clearly, very little has changed (and I wouldn't have expected it to) but there have been some recent reports relating to the suburbanisation of poverty in Scotland, including this Conversation piece by Nick Bailey and Jonathan Minton.

Finally, a little SIMD 2016 map insight for you. According to my analysis, there are 31 data zones in the most deprived decile which have a neighbouring data zone in the least deprived decile. Conversely, there are 26 data zones in the least deprived decile with a neighbour in the most deprived decile. In a lot of these, the buildings in neighbouring areas are some way from each other, but in some examples (see below) neighbouring streets are in opposite deciles.

Some neighbouring 'most-least' areas in SIMD 2016

Notes: when you rank places some will always be at the top and some at the bottom, obviously. All other things being equal, however, we might not expect to see such large concentrations of deprivation in single locations. But we know that things are not equal and history and geography have an important role to play in the formation and persistence of these patterns. We can easily blame the politicians of today or those of the past, but I don't think it's that simple. These patterns have persisted for decades and in some cases even longer. For a really good guide to the SIMD2016, read this introduction from the Scottish Government. 

Mapping: I've coloured entire data zones but added in building outlines. Why? Well, the SIMD relates not only to household attributes (such as income) but also to area characteristics (e.g. crime) so the wider environment is important. I wanted to strike a kind of compromise here in relation to what's displayed. Also, the place names are Ordnance Survey open data so the hierarchy in terms of capitalisation and font size is taken directly from the original data. I used QGIS for the mapping. The building outlines are quite generalised so you can't pick out individual houses - this was a deliberate choice as I want to avoid giving the impression that this data says anything about individual people - it's about areas. For reference, there are a total of 6,976 data zones in Scotland.