Sunday, 7 February 2016

Spotlight on Higher Education participation and deprivation

There has been quite a lot in the news recently about participation rates in higher education in the UK, with universities defending their access records after criticism from the Prime Minister. Part of this is about gender and ethnicity but it's also about socio-economic class, so I thought I'd take a little look at higher education participation rates in relation to deprivation, as a kind of proxy for socio-economic class. I took the the participation of local areas data from HEFCE (the dataset known as POLAR3) and data on deprivation using the official indices from across the UK (IMD, NIMDM, SIMD, WIMD). I looked at how participation in higher education varies in the most deprived fifth of the country - bearing in mind that the deprivation data is unique to each country of the UK and that there are differences between the data in Scotland and elsewhere. So, it's best to compare within rather than between nations in the maps below. Having said all this, it's time for some maps, first for England.

Only 20% most deprived are in the spotlight here

We can see from the first map, of London, that across the 20% most deprived areas participation in HE is quite variable and actually in the highest quintile in places. But this is something of an anomaly and London is a bit different in this respect, as noted by scholars such as Sol Gamsu. The situation is a little different in Bristol, below, where areas amongst the 20% most deprived nearly all have the lowest HE participation rates. A couple of points to note here, though. This says nothing about cause and effect (i.e. this isn't a finger-pointing exercise, even if the HE sector in general can do more). Also, the geographical scales are different - POLAR3 is available at ward level and deprivation data at lower scales (LSOA in England and Wales, SOA in Northern Ireland and Data Zone in Scotland). So, let's call this 'indicative' and move on.

HE participation and 20% most deprived - Bristol

Moving from the South West to the North East of England, there is slightly more variation in HE participation among the most deprived 20% of areas, but not much. There are many good reasons why we shouldn't be surprised about it (some historic, some socio-economic) but the patterns below do indicate that there is a need (or opportunity) to widen access.

HE participation and 20% most deprived - North East

Now it's time to run through a few more maps from the rest of England - West Yorkshire, South Yorkshire, Merseyside, Greater Manchester, East Midlands and the West Midlands.

HE participation and 20% most deprived - West Yorkshire

HE participation and 20% most deprived - South Yorkshire

HE participation and 20% most deprived - Merseyside

HE participation and 20% most deprived - Greater Manchester

HE participation and 20% most deprived - East Midlands

HE participation and 20% most deprived - West Midlands

As you can see from the maps above (remember, only the 20% most deprived of areas are highlighted - the rest have the 'lights off' effect), there is a good bit of variation in HE participation rates in some urban areas. I've shown the LSOA boundaries in white and it's a bit frustrating/tantalising with the POLAR3 data being at the larger ward scale, but I think these patterns merit further investigation. Maybe this has already been done, in which case do let me know. To complete the UK picture, here are some more maps, starting with Glasgow and Edinburgh. Just remember that in some cities (such as Glasgow) there are more areas in the most deprived 20% than others - and Glasgow and Edinburgh provide an obvious contrast in this regard. 

HE participation and 20% most deprived - Glasgow
HE participation and 20% most deprived - Edinburgh

To Northern Ireland now with a map of Belfast, followed by Cardiff. In both cases there is a little variation but most of the areas shown - all within the 20% most deprived -  overlap with wards with the lowest HE participation rates. 

HE participation and 20% most deprived - Belfast

HE participation and 20% most deprived - Cardiff

What does this all mean? Well, at first glance, it appears that in some cities - and in particular in London - there are higher levels of HE participation in some of the most deprived areas. In others, the opposite appears to be the case. Overall, it appears HE participation is low in the 20% most deprived areas. But, there are many caveats here. Even in areas where the participation is higher, we don't know where that participation is taking place - or why. These are things I'd like to know more about. In the first instance, I was most interested in the patterns. Also, some areas have a lot more deprived areas so this affects how we might interpret the maps - a point reinforced if I include maps for Oxford and Cambridge, which have very few areas in the 20% most deprived in England. 

HE participation and 20% most deprived - Oxford
HE participation and 20% most deprived - Cambridge

Notes: as I said above, the datasets used here are at different geographical scales, so this makes it all a bit frustrating. The within-ward variation in HE participation we might be interested in cannot be discerned from these maps, but these patterns appear consistent with results reported elsewhere. The best way to interpret these maps would be to say that the areas shown are within wards with a specific HE participation quintile, rather than to read anything more into it from a statistical point of view. I think this is a useful initial exercise but ideally we'd have more fine-grained POLAR3 data to do this with - maybe it already exists, but I couldn't find it. If you want to look at the POLAR3 data yourself - including in an interactive map - go to the POLAR web pages.