Interpreting political maps

I recently tweeted a couple of maps showing the 100 most and least deprived constituencies in England. I used the 2010 English Indices of Deprivation to calculate this, aggregating the data from smaller areas to parliamentary constituencies. The method is not perfect but on the whole the areas identified are either among the poorest or richest in England. There are 533 constituencies in England so the figure of 100 is roughly the 20% most and least deprived (18.76% to be more precise). I shaded the maps using red for Labour, blue for Conservative, Yellow for Liberal Democrats and so on. The most obvious thing about the maps is, of course, the fact that the most deprived map shows nearly all Labour constituencies and the least deprived shows almost all Conservative constituencies. Click the caption below the images to see interactive versions.

100 most deprived constituencies

100 least deprived constituencies

These kinds of maps often make a big impact and are shared widely but people tend to draw conclusions from the patterns they see that are not necessarily correct - and often conclusion which mirror pre-existing biases and perspectives. For example, some people see these maps and claim that voting Labour makes you poor or that only rich people vote Conservative. Some would even claim that this proves that Labour has failed the constituencies they serve. Opponents would argue that coalition cuts have merely deepened spatial inequalities and hit Labour-voting areas hardest. This is all a bit dramatic, but you don't have to search online long to find such views.

Other people might say that if you want to be richer you should vote Conservative. Other people would tell you not to be so simplistic and point to the way in which voting patterns are formed at the local level. Still others might point to the longstanding economic differences between north and south in England and say that this has something to do with it. Perhaps others will say that the Conservatives are the party of the rich and that Labour are the party for the poor. There are varying degrees of truth in all these views but the point I want to make here is that none of this can be proven just by looking at a political map.

For me, such maps are a starting point for a conversation about what these patterns might mean, whether they are a problem and what might be done about it, if anything. I'm not making these maps because I'm pro-Labour or pro-Conservative or because I think that they prove anything in particular but because I want to draw attention to the patterns and what they might mean. Finally, some observations from the maps...

• There are no Labour constituencies amongst the 100 least deprived in England.
• There are 2 Conservative constituencies amongst the 100 most deprived in England.
• Sheffield Hallam (Liberal Democrat, Nick Clegg) is amongst the 100 least deprived constituencies in England. 
• There are 5 Liberal Democrat constituencies amongst the 100 most deprived in England.
• There are 7 Liberal Democrat constituencies amongst the 100 least deprived in England.

It will be very interesting to see how these patterns change (if at all) after the General Election this year.

Scottish Index of Multiple Deprivation 2009 - Google Map

Recently I've been experimenting with google fusion tables and the mapping and database opportunities they present. One dataset I used for this was the new English Indices of Deprivation for 2010 - something the Guardian picked up. This was only really a quick experiment (literally done over an hour or so) but I was impressed with the capabilities.

Since, I'm from Scotland I also did a Scottish version - click the image below to go to the test site. Again, this is only experimental but I find it useful for exploring the data in an easy to use manner.

Technical note: this also led me to discover the amazing shpescape aka Shape to Fusion for converting shapefiles to fusion format - code which was written by Josh Livni at google. Enough nerdspeak for now.

A Nearest Neighbour Index of Multiple Deprivation

I've blogged before about my attempts to add a more spatial dimension to the existing Indices of Deprivation for England. In this post, I have decided to share the data for my calculations on a Nearest Neighbour Index of Multiple Deprivation, following the method described in this paper.

The image below shows the IMD2007 (left) and NNIMD2007 (right) for Birmngham, with red most deprived and blue least deprived.

I have put together a spreadsheet containing the original IMD data plus my NNIMD data for all 32,482 LSOAs in England. If you click on the links below you can access the data - you'll need to click the 'download' link to get the data.

NNIMD dataset - Excel 2003 version (.xls)

NNIMD dataset - Excel 2007 version (.xlsx)

Area Effects - do they exist?

Much has been written in the academic literature about 'area effects', or how concentrations of poverty (or disadvantage, or deprivation, or whatever you want to call it) can act to create further disadvantage in terms of all sorts of things; from health to employment, to crime and so on.

Another issue here is the extent to which the geography of deprivation has largely remained the same in cities across the United Kingdom. So, if we look back in time and follow neighbourhoods through to the present day many of the areas that we target with policy today were seen as 'problems' decades ago. Despite lots of money being spent on them, things have not got much better.

The Department of Communities and Local Government's recently published Regeneration Framework has more to say on these issues but it also acknowledges the above two problems. But what can we do about it? Well, the Framework has three priority outcomes:

• improving economic performance in deprived areas;
• improving rates of work and enterprise in deprived areas; and
• creating sustainable places where people want to live and can work, and businesses want to invest.

Some good ideas but have we any reason to think it will be any more effective than previous approaches? Maybe. There is a growing realisation that, after 40+ years of area-based initiatives in England, something just isn't right in the approach we take.

The point of this post is to say that I concur with most of what's in the Framework but I don't there there is enough emphasis on the extent to which deprivation is a spatial phenomenon, as well as an attributional one. What I mean by this is that the spatial manifestation of deprivation also needs to be tackled head on. But first, it needs to be understood - and this is what my current research focus is on. I've blogged on this before, but I just wanted to re-emphasise it here since the Framework was on my mind. There's also a reference to a paper that I co-wrote (bottom of page 55) with some of the spatial context material in it.