srhe

The Society for Research into Higher Education

Tag Archives: HESA


by Leave a comment

Moving with the times: The growing need for better graduate mobility data

by Tej Nathwani

Introduction

As SRHE noted in their summary of the theme of the 2022 conference, one of the current areas of discussion is the relationship between student mobility and outcomes. For example, the Institute for Fiscal Studies (IFS) have used the Longitudinal Education Outcomes dataset to explore trends in graduate mobility and earnings in England. While mobility is correlated with individual destinations, there are also wider macroeconomic consequences resulting from the extent to which graduates move around the country.

In a separate paper by the Resolution Foundation and the Centre for Economic Performance, researchers at the two organisations highlighted how one of the key factors that explains variations in productivity across areas are human capital levels – measured by the share of graduates in the locality. Hence, while providers can help with widening participation and upskilling the labour force in our most deprived regions, the full benefits of this for the vicinity may only be realised if those individuals who study in higher education choose not to move out of the area or region. One of the consequences of this is that providers are increasingly working with employers to try and ensure graduates can utilise their skills in the local economy (for example at Sheffield Hallam).

Given the state of the UK economy and the role mobility may have on individuals and growth, this is a topic that will remain salient in forthcoming years. However, even before we think about the association between mobility and outcomes, the first question to consider is how data might help us to better understand the extent to which graduates move for study and/or work. Historically, exploration of graduate movements has been at a regional level, which has become less relevant and valuable at a time when interest also lies in inequalities within regions, as well as between them. This blog will thus focus on a new marker HESA has generated to help our users gain more detailed insights into mobility.

The current problem

Patterns of regional migration and the categorisation of graduates into different groups based on this was first explored by Prospects back in the mid-2000s. One of the limitations of using such an aggregated level of geography, however, is that Wales, Scotland and Northern Ireland are all classified as individual regions. This means we are unable to examine what mobility is like within these nations. To see the drawbacks for investigating mobility in England using region, consider the neighbouring areas of Bradford and Leeds – both of which are within Yorkshire and The Humber. As the ONS regional economic activity data illustrates, there has been a divergence in the economic performance of these two places over the last twenty years. Hence, a graduate originally from Bradford who studies at the local university, but then moves for work to Leeds would be allocated to the same group in a regional analysis as one who initially lives, studies and is then employed in Bradford. With the graduate share being a key factor in understanding the differences in economic performance between areas, the possibility of distinguishing between graduates who remain in areas of low economic activity and those who move out of such localities for work is growing in importance.

A potential solution

HESA collect the postcode at which the individual resides prior to starting higher education and also request similar data from the graduate in the Graduate Outcomes survey regarding their location of employment (if they don’t know the postcode for their employment location, we ask the graduate to provide the town/city/area in which they work). There is therefore the potential to map these postcodes to local authority data (and their equivalents in Scotland and Northern Ireland). Using local authority of residence/work and region of study, we have created a mobility marker consisting of the following seven categories:

  1. Stays in same region for study and finds work in the same local authority as original location of residence
  2. Returns to the same local authority for work as original location of residence, having left region/country for study
  3. Stays in same region for study, but finds work in different local authority (in the same region) to original location of residence
  4. Returns to a different local authority (of the same region) for work when compared with original location of residence, having moved region/country for study
  5. Moved region/country for work, but did not move region for study
  6. Moved region/country for study, but did not then move region/country again for work
  7. Moved region/country for study and then moved region/country again for work (with the region/country being different to their original region/country of residence)

Going back to our original example of the two graduates from Bradford (one who moves for work and one who doesn’t), this new classification ensures they are no longer placed in the same group. Rather, one is allocated to category A, while the other is assigned to C. Such distinctions will help improve our awareness of overall patterns of mobility across time.

Concluding thoughts

Our initial exploration into mobility and job quality suggests that migrating for employment is correlated with graduates finding a role that fits better with their career plans. With similar findings on the benefits of moving for work from a salary perspective also being reported by the IFS, this could potentially leave those aiming to reduce disparities in economic performance between areas with a conundrum. Policies aiming to upskill the labour force in more deprived areas and help reduce spatial inequalities require these individuals to remain in such neighbourhoods. Yet current evidence suggests that moving for work is associated with more positive outcomes for these people. Given the relevance to policy aims, as we continue to collect increasing amount of data on graduates through our annual Graduate Outcomes survey, we shall be exploring the potential to map how mobility differs by area (eg by investigating whether we have adequate sample size at more granular levels of geography). If this does prove feasible, this will help end users with ascertaining the extent to which localities with lower output are gaining/losing graduates.

High levels of inequality and poor growth are two key concerns for the UK economy. We hope that the development of new measures on deprivation and graduate mobility can help the higher education sector with tackling these issues and assist providers in capturing the wider impact they are making in society.

Feedback on our mobility marker is most welcome. Please send these to pressoffice@hesa.ac.uk.

To learn more about Graduate Outcomes, visit www.graduateoutcomes.ac.uk or view the latest national level official statistics.

To be kept updated on our publication plans and latest research releases, please join our mailing list.

Tej Nathwani is a Principal Researcher (Economist) at HESA, which is now part of Jisc.


by Leave a comment

Deprivation data: Introducing a new UK-wide area-based measure

by Tej Nathwani

Introduction

The 2020s will be a pivotal period in determining the UK’s economic future. That’s the primary message of a recent report published by the Resolution Foundation and Centre for Economic Performance at LSE. While major events such as the pandemic and Brexit have undoubtedly played a part in this, there are also longer-term factors that have contributed to the country reaching this position. Examples noted by the researchers include stagnant productivity levels, large disparities in economic performance between areas and inequalities in our education system.

Naturally, one of the questions being increasingly asked of the UK higher education sector is how it is helping to resolve some of the latter issues. Yet being able to tackle these matters successfully, as well as understand the outcomes from various interventions, requires the provision of suitable data. As the body responsible for the collection and dissemination of information about UK higher education, HESA has a role to play in supplying appropriate variables and statistics to our users that support them in their decision-making. Hence, the past few years have seen us develop new fields designed to be relevant and valuable in meeting the current needs of our customers. Across two separate blogs we will be outlining what these are and the potential value they can deliver. In this first piece we begin with a focus on our work relating to socioeconomic disadvantage.

The uses of data on deprivation in higher education

One of the ways in which providers seek to improve equality of opportunity in education is through outreach activity. These are initiatives that aim to raise aspiration and attainment among those from disadvantaged backgrounds, as well as helping to inform them of the potential benefits that studying for a degree can offer. Area-based data on deprivation will typically be used in two ways. Firstly, as part of the eligibility criteria that an individual must meet to participate (for example, at Surrey). Secondly, it can help providers determine the areas of the country which they believe would be most useful to target given their strategic ambitions (for example, at King’s College London).

The current problem

The most commonly used area-based measure of disadvantage across the sector in each of the four nations is the index formed from the Indices of Deprivation. However, while this is especially effective in capturing deprivation in major urban areas, it is known to be less useful in identifying this in rural locations. For example, Na h-Eileanan Siar in Scotland has no localities that emerge in the bottom quintile of the Scottish Index of Multiple Deprivation (SIMD), despite income levels being below the national average. (Indeed, local government looking at poverty in the area highlight that ‘There are difficulties in using the SIMD in rural areas. Areas such as the Outer Hebrides are sparsely populated, socially heterogeneous and less sensitive to area-based measures such as SIMD. This can lead to a situation where households in rural areas are omitted from policy and targeting by national interventions designed to address poverty and inequalities’.) Furthermore, the size of the areas used to derive the index can also make it difficult to fully understand the levels of deprivation within localities. For example, there may be pockets within a zone that are experiencing higher levels of disadvantage compared with other vicinities, but the use of a more aggregated geographic domain can lead to this being masked. The consequence of this for the higher education sector is that there may be some prospective students who live in deprived neighbourhoods, but due to the limitations of existing data, find themselves unable to participate in outreach activity (eg as a result of not meeting the eligibility criteria or providers not targeting their place of residence).

Comparability is also an important aspect of high-quality statistics. Each nation of the UK, however, adopts a different approach in generating its index from the Indices of Deprivation. This means it is not a UK-wide variable and does not enable statistics to be evaluated across nations. Both the Office for Statistics Regulation and the latest State of the Nation report by the Social Mobility Commission (see p20) have highlighted this as an existing data gap that inhibits our understanding of wider societal trends in social mobility.

A potential solution

The question we therefore asked ourselves was ‘Can we create a UK-wide area-based measure of deprivation that can also address some of the drawbacks of existing indicators?’. To do so, we relied upon the 2011 Census, given the questions asked across the nations are harmonised as far as possible, meaning a UK-wide metric can be created. Data are also released at ‘output area’ level (output areas are often referred to as ‘small areas’ in Northern Ireland), which is a smaller level of geography than is used for the Indices of Deprivation. Output areas will typically contain less than 500 individuals.

With earnings data not available in the Census, our measure of deprivation was derived using the qualifications and occupations of residents in output areas, given these two factors are key determinants of low income. Having generated this, and to understand the potential value it could bring, we compared the bottom quintile of our measure to the equivalent group in the index produced from the Indices of Deprivation (ie the most deprived neighbourhoods). In each of the four nations, we found that our measure picked up a greater proportion of rural areas, albeit to varying degrees. Furthermore, when looking at those output areas that emerged in the lowest fifth of our measure, but a higher quintile of the index developed using the Indices of Deprivation, we observe that the most prevalent localities are based in local authorities/council areas/local government districts where there appear to be lower levels of economic activity (eg County Durham in England, North Lanarkshire in Scotland, Rhondda in Wales, as well as Armagh City, Banbridge and Craigavon in Northern Ireland).

Concluding thoughts

In summary, our measure does seem to overcome some of the existing shortcomings of area-based indicators of deprivation. Over the next few years, we shall therefore be looking to supply the measure to users in an accessible format, alongside updating it using information from the most recent Census. As well as supporting equality of opportunity, if the measure can help to raise participation and skill levels in some of our most deprived neighbourhoods, there is also the possibility that this will assist with reducing spatial disparities in output. For example, the study by the Resolution Foundation and the Centre for Economic Performance notes that the ability of the Shared Prosperity Fund to successfully increase growth may well depend on the levels of human capital in the area. Through upskilling local residents living in disadvantaged localities, providers may therefore be able to facilitate the creation of the conditions needed for growth-enhancing initiatives to succeed. Of course, this rests on the assumption that these areas do not subsequently see residents move to other parts of the country. Understanding the geographical mobility of graduates will thus be the topic of our next blog.

Read more about our measure, its correlation with income and how it compares to the Indices of Deprivation https://www.hesa.ac.uk/insight/08-11-2022/new-area-based-measure-deprivation-summary.

Feedback on our measure of deprivation is most welcome. Please send this to pressoffice@hesa.ac.uk.

To be kept updated on our publication plans and latest research releases, please join our mailing list.

Tej Nathwani is a Principal Researcher (Economist) at HESA, which is now part of Jisc.

Ian Mc Nay


by Leave a comment

Lies, damned lies and spin; never mind the statistics

By Ian McNay

Press reports, 31 January, on UCAS statistics on the 15 January deadline showed remarkable unanimity around telling, shall we say…not the whole truth:

–         Girls lead the way as degree applications hit record levels – Times

–         Record numbers of 18-year-olds apply to university – Telegraph

–         University applications hit record high – Guardian

The Telegraph had a second story claiming the number of applicants aged 20 and over had increased by 5%.

All this gave comfort to [English] ministers who claim that high fees have had no long term effect on applications. So, let us look at the longer term and compare the cycle for 2014 entry Continue reading →