This blog was kindly contributed by Tej Nathwani, Principal Research (Economist) at HESA.
Enhancing social mobility by ensuring there is equal opportunity for all and delivering more equitable growth across regions are policy objectives of governments in all UK nations. Higher education is expected to have an important role in helping to achieve these ambitions. For example, by encouraging and supporting those residing in more disadvantaged localities into education, providers can help raise local skill levels. However, in order for providers to be able to do this, they need access to data that can assist them in identifying those areas where they should prioritise their outreach and widening participation activities.
Today, HESA has published a report detailing a new area-level measure of socioeconomic disadvantage that we have developed using educational qualifications and occupation data contained within Census 2011. Compared to existing area-level variables available within the sector, it brings the additional benefits of being:
- based on a smaller geographical domain; and
- able to pick up deprivation in parts of the country, which present measures are unable to capture effectively
As this is an area-level measure, it is intended that this could be utilised in carrying out activities such as outreach work in areas of disadvantage. We are not advocating that it be used to make individual-level decisions such as contextual admissions for which more suitable measures are beginning to emerge within the sector (as we note in the report).
Why is HESA doing this?
Firstly, as a producer of statistics, we conduct our activities according to the Code of Practice for Statistics. As a result, one of the quality dimensions that we must work towards where possible is for our data and statistics to be comparable across geographic domains (for example, by region).
However, current UK-wide measures that are drawn upon in widening participation activity vary in the extent to which they are useful in different parts of the country. For example, the limitations of POLAR (participation of local areas) in places such as Scotland and London are well-documented. As a result, home nations have increasingly utilised country-specific variables (for example, Indices of Multiple Deprivation or IMD) in supporting their widening participation agenda.
Secondly, we must also try to ensure the data we disseminate continues to have relevance and value to users. Present policy objectives in the UK centre around ensuring there is equal opportunity for all and that nobody is left behind. There is also an ambition for the benefits of growth to be distributed more evenly across the country. Higher education is expected to play a key role in meeting these aspirations, with outreach activity that helps those from disadvantaged communities consider and access study being an important first step.
Yet, for providers to be able to achieve such aims, they need to be able to identify those localities where their influence will be most beneficial. This is not always feasible with existing area-level measures. For example, one of the issues with IMD is that it does not adequately pick up deprivation outside of major conurbations (for example, in rural areas).
Consequently, we have drawn upon the Census to try and create an area-level measure that can be applied UK-wide and that can capture socioeconomic disadvantage across a greater breadth of the country.
How have HESA created this new measure?
The organisations responsible for collecting and disseminating Census information in the UK release key statistics at output area level (referred to as ‘small areas’ in Northern Ireland). This is the smallest level of geography at which data is made available to the public (typically comprising of no more than a few hundred households). One of the other criticisms often made of area-level measures is that even those localities that are classified as being more deprived will contain households experiencing lower levels of socioeconomic disadvantage. To try to alleviate this concern as best we can, we have relied upon a smaller geographic territory than either POLAR or IMD (irrespective of the nation being considered) do.
Having downloaded these statistics, we then ranked all 274,611 UK output areas based on the educational qualifications and occupations of its residents. For the purposes of the onward analysis, we conducted in the context of higher education, those in the bottom quintile of the ranking were defined as being the most socioeconomically disadvantaged group.
What analysis has been done using this new measure?
Having formed our measure, we linked our various external sources to HESA records. In particular, we focused on UK domiciled full-time first degree entrants aged 18 to 20 in the 2011/12 academic year (corresponding to the year the Census took place). The main purpose behind our exploration was to see how the geographical distribution of students who sit within quintile 1 of these three area-level variables (our measure, POLAR and IMD) varies for our population of interest and hence to examine whether our measure can be shown to provide additional advantages over existing area-based measures (aside from being UK-wide).
The general pattern we observe is that our measure does indeed appear to encompass a greater span of each nation when compared with POLAR and IMD (as demonstrated by the detailed maps on our website). As we are also able to incorporate information into our dataset on how urban or rural the area a student lived in prior to starting higher education, we illustrate that in both Scotland and Northern Ireland in particular, our measure is capturing a greater proportion of students from more remote small towns/rural vicinities.
This consequently provides further evidence of the potential benefits of our measure and how it may help to overcome the known limitations of existing area-level variables. Please note that we do recognise that other area-level measures still have their merits in being used for widening participation activity (which we allude to in our paper).
What are the next steps?
We firstly welcome feedback on the potential usefulness of this measure. Should a positive response be received, we will then focus our attention on how this information can be ingested into our collection, as well as subsequently provided to users in a form that supports them with their activities and decision-making processes. In our report, we have also outlined other ways in which we will look to extend the work carried out here (e.g. by updating our exploration using Census 2021 data).
Please do feel free to submit your comments and questions to us by emailing email@example.com.