Educational advantage and employability of UK university graduates

Louise Pigden (School of Engineering and Technology, University of Derby, Derby, UK)
Andrew Garford Moore (Department of Corporate Planning and Performance, University of Derby, Derby, UK)

Higher Education, Skills and Work-Based Learning

ISSN: 2042-3896

Article publication date: 20 February 2019

Issue publication date: 20 September 2019

Abstract

Purpose

In the UK, the majority of university students specialise and study just one subject at bachelor degree level, commonly known in the UK as a single honours degree. However, nearly all British universities will permit students if they wish to study two or even three subjects, so-called joint or combined honours degrees, internationally known as a double major. The purpose of this paper is to explore whether educational advantage, measured by the “Participation of Local Areas” (POLAR) classification, correlated with rates of graduate destinations for joint and single honours graduates. This study focused particularly on Russell Group and Post-92 Universities.

Design/methodology/approach

The authors analysed the complete data set provided from the Higher Education Statistics Agency Destination of Leavers from the Higher Education survey, and combined this with data from the POLAR4 quintiles, which aggregate geographical regions across the UK based on the proportion of its young people that participate in higher education. The data were analysed to establish whether there was a difference in the highly skilled graduate employability of the joint honours students, focusing particularly on Russell Group and Post-92 Universities, in order to build on previous published work.

Findings

Single honours and joint honours graduates from higher participation POLAR4 quintiles were more likely to be in a highly skilled destination. However at both the Russell Group and the Post-92 universities, respectively, there was no trend towards a smaller highly skilled destinations gap between the honours types for the higher quintiles. For the highest POLAR4 quintile, the proportion of joint honours graduates was substantially higher at the Russell Group than at Post-92 universities. Furthermore, in any quintile, there were proportionately more joint honours graduates from the Russell Group, compared with single honours graduates, and increasingly so the higher the quintile.

Research limitations/implications

This study focused on joint honours degrees in the UK where the two or three principal subjects fall into different Joint Academic Coding System (JACS) subject areas, i.e. the two or three subjects are necessarily diverse rather than academically cognate. This excluded the class of joint honours degrees where the principal subjects lie within the same JACS subject area, i.e. they may be closer academically, although still taught by different academic teams. However, the overall proportion of joint honours graduates identified using the classification was in line with the UCAS (2017) data on national rates of combined studies acceptances.

Practical implications

All Russell Group graduates, irrespective of their POLAR4 quintile, were far more likely to be in a highly skilled destination than single or joint honours graduates of Post-92 universities. Even the lowest quintile graduates of the Russell Group had greater rates of highly skilled destination than the highest quintile from Post-92 universities, for both single and joint honours graduates. This demonstrated the positive impact that graduating from the Russell Group confers on both single and joint honours graduates.

Social implications

This study could not explain the much smaller gap in the highly skilled destinations between single honours and joint honours graduates found in the Russell Group, compared with the Post-92. Why do a higher proportion of joint honours graduates hail form the upper POLAR4 quintiles, the Russell Group joint honours graduates were more disproportionately from the upper POLAR4 quintiles and the joint honours upper POLAR4 quintiles represented such a larger proportion of the Russell Group overall undergraduate population? Other student characteristics such as tariff on entry, subjects studied, gender, age and ethnicity might all contribute to this finding.

Originality/value

This study demonstrated that, averaged across all universities in the UK, there was a trend for both single honours and joint honours graduates from higher participation POLAR4 quintiles to be more likely to be in a highly skilled destination, i.e. the more educationally advantaged, were more likely to be in a highly skilled destination, as a proportion of the total from each honours type. This accorded with HESA (2018b) data, but expanded those findings to include direct consideration of joint honours graduates.

Keywords

Citation

Pigden, L. and Moore, A.G. (2019), "Educational advantage and employability of UK university graduates", Higher Education, Skills and Work-Based Learning, Vol. 9 No. 4, pp. 603-619. https://doi.org/10.1108/HESWBL-10-2018-0101

Publisher

:

Emerald Publishing Limited

Copyright © 2019, Emerald Publishing Limited


1. Introduction

In the UK, the vast majority of university students specialise throughout their undergraduate degree and study just one academic subject area at bachelors’ degree level (UCAS, 2017). This is commonly known in the UK as a single honours degree. This is in contrast to many other university educational systems globally, for example in North America, where students must achieve a breadth of knowledge across several academic disciplines, combined with a depth of knowledge in their major subject. Notwithstanding the emphasis on early specialisation in the UK, nearly all British universities (UCAS, 2016) will actually permit students, if they wish, to study two or even three academic subjects in parallel. These are referred to as joint or combined honours degrees (hereafter referred to simply as “joint honours degrees”), and would be known as a double major internationally.

Pigden and Moore (2018) has a more detailed account of the characteristics of joint honours degrees in the UK, and other studies in the literature expand on the learning experience of joint honours students (Hodgson, 2011; Pigden, 2016; Pigden and Jegede, 2016, 2018) and also the combinations of subjects most likely to lead to highly skilled destinations (graduate-level employment or further study) following graduation (Pigden and Moore, 2017).

The focus on the proportion of graduates in highly skilled destinations is important because the proportion of young people (under the age of 30 years) participating in a UK university education continues to rise steadily, and reached 49 per cent by 2015/2016 (DfE, 2017a). Whether the UK university system continues to represent efficiency, effectiveness and value for money is the subject of discussion amongst students, parents, educators, the UK Government and leaders of UK universities (Browne, 2010; Universities UK, 2015; Dowling, 2015).

More recently, this continued growth in student numbers is partly due to the removal of the university student number cap in 2015/2016 in England, UK, meaning that universities are now free to recruit as many students as they wish without financial penalty. Despite the expansion in participation, the positive median earnings differential between graduates and non-graduates have remained remarkably constant over the period of 2006–2017 (DfE, 2017b), with graduate median earnings consistently around £10k more than non-graduate median earnings.

This demonstrates the overall effectiveness of the UK higher education system, from an earnings perspective, but does not necessarily apply evenly across all subjects studied. For example, “Medicine, mathematics and economics graduates all typically earn at least 30% more than the average graduate, while creative arts graduates earn around 25% less on average” (Belfield et al., 2018). Unfortunately, direct earnings data are not available for joint honours graduates.

In England, UK, university students have been charged much higher tuition fees since 2012/2013, when annual student fees were trebled to £9,000, in response to the Browne (2010). This transferred almost the entire cost of tuition onto the students themselves, to be paid for via loans. Furthermore in 2015, government-funded, means-tested maintenance grants and covering the students’ cost of living were also removed and replaced with loans. The outcome of this has been to leave the poorest graduates with an average debt of £57,000 (Belfield et al., 2017).

The UK Government and social commentators continue to be concerned with fair access to a university education for disadvantaged groups, particularly given that a university education confers a positive earnings differential on graduates compared with non-graduates. There is an intention to counteract some of the earnings disadvantage affecting lower socioeconomic groups, compared with those from better-off backgrounds (Belfield et al., 2018). There are, however, large differences in participation rates across the UK; for example, participation, categorised by whether the student had been in receipt of Free School Meals (FSM), varied between 14 per cent (FSM) and 48 per cent (non-FSM) of the population within a single local authority area, Reading, England, in this case (DfE, 2017c).

There is therefore continued political debate on how universities can achieve a more equitable balance of student admission and fair access for all students irrespective of their social characteristics. For example, the UK Government’s Office for Fair Access, safeguards and promotes fair access to higher education by approving and monitoring so-called “access agreements”, the approval of which permits individual universities to charge higher tuition fees.

One methodology used to evaluate the fairness of access to UK universities is via the “Participation of Local Areas” (POLAR) classification, which aggregates geographical regions across the UK based on the proportion of its young people that participate in higher education. POLAR is used to inform the targeting, and to support the analysis, of widening participation activities designed to increase social mobility amongst low-participation groups. POLAR quintile 1 represents the lowest participation areas (most educationally disadvantaged) and POLAR quintile 5 represents the highest participation areas (most educationally advantaged).

The most recent version of the classification is POLAR4. It is based on the combined participation rates of 18-year olds entering university between 2009/2010 and 2013/2014, and 19-year olds entering university between 2010/2011 and 2014/2015. This version superseded POLAR3, which is, however, still used in many current analyses, and which is the proportion of young people entering university by the age of 19 years between 2005/2006 and 2010/2011. On average, participation rates have increased nationally and POLAR4 is the first POLAR classification to have no geographical areas in the UK with a participation rate of 0 per cent (HESA, 2017).

University admissions profiles in the UK are often not well spread across the POLAR quintiles, potentially indicating a lack of equity in access to some universities. In the 2016/2017, statistics recently published by the HESA (2018a), Oxford and Cambridge Universities have 2.8 and 3 per cent of their new student admissions, respectively, from the lowest low-participation neighbourhood (POLAR3 quintile 1).

Among the Russell Group (research-driven universities which are highly selective of their students) more generally, the proportion from POLAR3 quintile 1 is low. For example, University College London, Imperial College, and Durham follow Oxford and Cambridge, with 3.2, 3.4 and 4.2 per cent of their new student admissions, respectively, from the lowest participation neighbourhood (POLAR3 quintile 1). The University of Liverpool scores the highest among the Russell Group for 2016/2017, with 9.7 per cent of their new student admissions from quintile 1. Nationally, 11.4 per cent of new entrants into higher education come from the lowest participation neighbourhoods, as defined by the POLAR3 classification (HESA, 2018a), so the Russell Group are admitting proportionately far fewer students from educationally disadvantaged areas.

Admission to Britain’s top universities is a theme of interest in the UK Parliament, and according to a written question to the Department for Education (Evennett, 2018), for 2017 entry, Oxford and Cambridge Universities received 48.5 and 48.3 per cent of their student applications from POLAR3 quintile 5 (most educationally advantaged), with University College London, Imperial College and Durham not far behind with 45.0, 44.8 and 48.0 per cent of their student applications, respectively, from POLAR3 quintile 5. On average, the Russell Group received 41.4 per cent of all UK 18-year old student applications from POLAR3 quintile 5.

Within the UK university sector, so-called “Post-92” universities are typically former vocationally oriented polytechnics that converted to universities shortly after the 1992 expansion of the UK university sector. In a recent report from the Higher Education Policy Institute (Martin, 2018), the Gini index is used to demonstrate how evenly students are distributed across the POLAR3 quintiles at universities in the UK. It is striking to note the clustering at the bottom of the table for the Russell Group, and the top of the table, the universities with the most equitable admissions profiles, is dominated by Post-92 universities.

As a measure of university graduates’ ability to find suitable work, the UK Destination of Leavers from Higher Education (DLHE) survey data, provided by the Higher Education Statistics Agency (HESA), are used to identify the proportion of graduates in highly skilled destinations (graduate jobs or further study). According to HESA (2018b), analysis of the 2015/2016 DLHE data set shows that POLAR3 quintile 1 graduates have the lowest percentage of graduates in highly skilled employment or further study at 71 per cent of the total, while quintile 5 graduates have the highest proportion of graduates in highly skilled destinations or further study at 75 per cent of the total. This analysis implies that educational disadvantage, at the national level, persists over the course of a university education and affects the ability of graduates to secure graduate-level employment or go into further study.

The current study sought to build on the previous work of Pigden and Moore (2018) which finds nationally an approximate −3 per cent point negative gap between the proportion of joint honours graduates in highly skilled destinations compared with single honours graduates. Pigden and Moore (2018) find, however, that both single and joint honours graduates from the Russell Group are more employable compared with the national average and “both single and joint honours graduates of the Russell Group exceeded the national average (of graduates in highly skilled destinations) by +9.04% points for single honours and +10.59% points for joint honours” for their respective honours type. Contrasting this, at Post-92 universities, single honours graduates are −8.3 per cent points lower and joint honours graduates −11.9 per cent points lower than the national average for highly skilled destinations for their respective honours type.

Furthermore, Pigden and Moore (2018) find that the gap between the proportion of joint honours graduates in highly skilled destinations compared with single honours graduates is much smaller at the Russell Group, compared with Post-92 universities. At the Russell Group, the gap between single and joint honours graduates is just −1.52 per cent points, but at Post-92 universities the gap is −7.13 per cent points.

The purpose of the current study was to evaluate the highly skilled destination rates of joint honours graduates compared with single honours graduates, and to correlate this with a measure of educational disadvantage, POLAR4 quintiles. By adding in POLAR4 quintile data, the main research question explored in the current study was whether this participation factor correlated with highly skilled destinations for joint honours graduates, who had studied at the Russell Group or Post-92 universities. The intention behind the study was to analyse whether the effects found in Pigden and Moore (2018) whereby the Russell Group joint honours graduates are far more likely to be in highly skilled destinations than their Post-92 university counterparts, and with a much smaller gap, were due, in part, to an association with POLAR4 participation rates.

2. Methodology

The current study specifically built upon the methodology and analysis of highly skilled destinations (either graduate employment or further study) for joint honours graduates used in Pigden and Moore (2017, 2018). In order to identify the proportion of graduates in a highly skilled destination, the DLHE survey data, provided by HESA, were analysed via a unique, customised data set incorporating additional, publicly non-published data on the academic subjects studied by the graduate. By analysing the subjects studied, joint honours graduates could be identified analytically. In the generic analyses of DLHE published publicly by HESA, the joint honours graduates are apportioned across the subjects studied, and so cannot be evaluated and scrutinised directly. Therefore, the current study provided a mechanism for identifying joint honours graduates and directly exploring their rates of highly skilled destinations, an approach which is not possible in the publicly available DLHE data.

The current study specifically considered the outcomes of full-time undergraduates in the UK and utilised a consistent, analytic approach for analysing the DLHE data set, as deployed in the previous work of Pigden and Moore (2017, 2018). However, in the current study, additional social mobility data, namely the POLAR4 quintiles, were included, in order to facilitate an analysis of highly skilled destinations correlated with participation rates.

While Pigden and Moore (2018) comprise the DLHE data set from academic years 2011/2012–2014/2015, the current study added in a further two years of data and spanned 2011/2012–2016/2017. The additional POLAR4 social mobility data were added for the three most recent years of the survey: 2014/2015–2016/2017. The method for identifying joint honours graduates was via the Joint Academic Coding System (JACS), used by HESA to classify academic subjects.

The customised DLHE data set used in the current study included up to three JACS principal subjects studied by the graduate, not normally published in the HESA annual analysis of the DLHE survey. Where a degree comprised academic subjects studied from a single JACS subject area, then this was deemed a single honours degree, and where the subjects studied were drawn from different JACS subject areas, then these were defined in our study as joint honours degrees.

Pigden and Moore (2018) have further details and the limitations of this approach to define and identify joint honours degrees via the DLHE survey data. As in Pigden and Moore (2018), we considered whether graduates had studied at one of the Post-92 universities, or at a Russell Group university, in order to reflect on the differences in graduate outcomes between these two groups of universities.

To analyse the effect of completing a single honours degree compared with directly related joint honours degrees, in most of the analyses “single honours only” subjects were removed, i.e. academic subjects were removed that did not feature in any of the joint honours degrees in the DLHE data set, for example, JACS B5 Opthalmics and JACS A4 Clinical Dentistry (see Table I). The rationale was that the current study sought to establish whether there was an observable impact in studying two or three subjects as a joint honours degree that were also available to study as single honours, i.e. the impact was inherent in this mode of study, rather than in the actual subjects studied.

So that our study complemented the recent teaching quality assessment of UK universities under the teaching excellence framework (Higher Education Funding Council for England, 2017), we used the same criteria for highly skilled employment or further study as defined by the Higher Education Funding Council for England (2015), namely that the definition of highly skilled employment was any occupation within categories 1–3 of the Standard Occupational Classification (Office for National Statistics, 2010). All further studies were also considered to be highly skilled and were therefore included wherever highly skilled destinations were referred to.

3. Results

3.1 Proportion of graduates by honours type in the UK between 2011/2012 and 2016/2017

Tables II and III update the analysis of Pigden and Moore (2018), with two additional academic years of DLHE data. Using the same definition of a joint honours degree, the analysis of DLHE showed that from 2011/2012 to 2015/2016, there was a year-on-year decline in the proportion of graduates with a joint honours degree, compared with the proportion of all single honours graduates (including subjects not available to study as joint honours).

As seen in Table II, in 2011/2012, 10.76 per cent of the total number of graduates had a joint honours degree; this then fell approximately 0.5 per cent points each year, and was down to 8.83 per cent of the total in 2015/2016. However, in the most recent year of survey data available, 2016/2017, there was an increase in the proportion of joint honours graduates, up to 9.16 per cent of the total. This modest 0.33 per cent point increase in the proportion of joint honours graduates was too small to as yet to recommend a review of institutional policy around joint honours, but would be reviewed in subsequent years to check for sustained growth.

It was noted that the absolute number, rather than proportion, of all graduates with a joint honours degree had actually increased year on year, albeit at a slower rate than the growth in graduates with a single honours degree. In the UK, as in the rest of the developed and developing world, there had been a “massification” of university participation (Marginson, 2016), with a steady rise in the proportion of the population entering higher education year on year, so it was unsurprising to also see this growth in joint honours numbers.

3.2 Proportion of graduates by honours type in highly skilled destinations in the UK six months after graduating

Excluding subjects not available to study as part of a joint honours degree (see Table I), we found that there had been a year-on-year increase in the proportion of graduates in highly skilled destinations, at the six months point following graduation (the point at which the DLHE survey is administered). For single honours graduates, in 2011/2012, just 64.30 per cent had secured a highly skilled destination six months after graduating, but by 2016/2017 this had risen to 76.26 per cent of the total. Similarly, joint honours graduates also had a year-on-year improvement in the proportion in highly skilled destinations, rising from 60.98 per cent of the total in 2011/2012 to 73.55 per cent in 2016/2017 (see Table III).

Table III shows that nationally, graduates with a joint honours degree had a year-on-year negative gap in the proportion in highly skilled destinations six months after graduating, compared with those who had a single honours degree. This fluctuated slightly from year to year, but was approximately minus 2.99 per cent point points, taken from a straight average of the last three years, the period for which we had POLAR4 data.

However, as previously mentioned, this national averaging masked substantial variation between the Russell Group and Post-92 universities found in Pigden and Moore (2018). To reiterate those findings, at the Russell Group, the gap between the proportion of single and joint honours graduates in highly skilled destinations was much lower at −1.52 per cent points.

However, joint honours graduates from Post-92 universities were much less likely to be in highly skilled destinations compared with the proportion of single honours graduates from Post-92 universities. There was a gap of −7.13 per cent points between the respective proportions of graduates from each honours type in highly skilled destinations (Pigden and Moore, 2018). The following analyses sought to explore these gaps further.

3.3 Proportion of graduates by honours type in highly skilled destinations by POLAR4 quintile

In Table IV, we analysed the proportion of graduates in highly skilled destinations, six months after graduating, by POLAR4 quintiles. Quintile 1 represented the lowest participation regions (most educationally disadvantaged), and quintile 5 represented the highest participation regions (most educationally advantaged). The DLHE data were summed over the three years 2014/2015–2016/2017, in order to smooth any year-on-year variation in graduate destinations from the respective honours type.

Table IV demonstrates that, averaged across all universities in the UK, there was a trend for both single honours and joint honours graduates from higher participation POLAR4 quintiles to be more likely to be in a highly skilled destination, i.e. the more educationally advantaged, were more likely to be in a highly skilled destination, as a proportion of the total from each honours type. This accorded with HESA (2018b) data, but expanded those findings to include direct consideration of joint honours graduates.

We already knew from Table III to expect, for the period 2014/2015–2016/2017, an average −2.99 per cent point gap between the proportion of joint honours graduates in a highly skilled destination compared with single honours graduates; Pigden and Moore (2018) showed that there was a much smaller gap at the Russell Group compared with Post-92 universities. However, previous work did not examine whether this gap was consistent across POLAR4 quintiles, and whether the gap varied depending on the level of educational advantage at the point of entry to university.

The difference in the proportion of graduates in highly skilled destinations between joint honours graduates and single honours graduates decreased substantially across the quintiles, with a −5.31 per cent point gap in quintile 1, falling to a −1.97 per cent point gap for quintile 5. In other words, joint honours graduates hailing from the most educationally advantaged regions in the UK had a much smaller, albeit still negative, highly skilled destinations gap compared with single honours graduates.

Therefore, Table IV appears to suggest that the lower quintiles were proportionately contributing less into the gap between the honours types, otherwise the national gap of −2.99 per cent points would have been higher, given that a straight average of the gap across the POLAR4 quintiles (assuming an equal distribution of graduates) was −3.96 per cent points. This suggestion was tested further on in the analysis in Table VII.

3.4 Proportion in highly skilled destinations of the university population of graduates, by honours type, by POLAR4 quintile

We looked at the Russell Group (Table V) and Post-92 universities (Table VI), to analyse whether the observed variation in highly skilled destinations across the POLAR4 quintiles (Table IV) also occurred within these two different groups of university.

Tables V and VI demonstrate that all Russell Group graduates, irrespective of their POLAR4 quintile, were far more likely to be in a highly skilled destination than single or joint honours graduates of Post-92 universities. Even the lowest quintile graduates of the Russell Group had greater rates of highly skilled destination than the highest quintile from Post-92 universities, for both single and joint honours graduates.

Interestingly, Tables V and VI also show that at both the Russell Group and the Post-92 universities, there was no trend towards a smaller gap between the honours types for the higher quintiles, as we had observed at the national level over all universities in Table IV. The gap between single honours and joint honours graduates from both the Russell Group and Post-92 universities was fairly uniform across all quintiles, and certainly not decreasing as in Table IV. However, the gap between the rates of highly skilled destination between single and joint honours graduates at the Russell Group was much smaller, in every quintile, than the gap at Post-92 universities.

These features seemed to suggest that the pattern found in Table IV was not only due to a higher proportion of graduates coming from the upper quintiles, but also could be in part because a greater proportion of upper POLAR4 quintile and a smaller proportion of the lower POLAR4 quintile joint honours graduates, compared with single honours, were from the Russell Group. This would magnify the effect of the small Russell Group highly skilled destinations gap between the honours types, found in the highest quintile. We examined this in Table VII.

3.5 Proportion of the overall national single and joint honours population, for each POLAR4 quintile at the Russell Group or Post-92 university

In Table VII, we analysed for each quintile the proportion of single and joint honours graduates, across the national population, from the Russell Group and Post-92 universities, in order to investigate the results found in Tables IV–VI. Table VII shows the proportion from each honours type and quintile only from Russell Group and Post-92; the graduates from other types of university made up the remainder. For example, for POLAR 1, 13.30 per cent of single honours came from Russell Group, 67.28 per cent from Post-1992s and the remainder of the single honours graduates from POLAR 1 were from other types of university.

Table VII shows that for the highest POLAR4 quintile, the proportion of joint honours graduates was substantially higher at the Russell Group than at Post-92 universities. With 49.42 per cent of the POLAR4 quintile 5 (most educationally advantaged) joint honours graduates coming from the Russell Group and only 26.99 per cent of the quintile 5 joint honours graduates coming from Post-92 universities. In contrast, 33.46 per cent of the quintile 5 single honours graduates came from a Russell Group and 43.89 per cent from Post-92 universities.

Furthermore, in any quintile, there were proportionately more joint honours graduates from the Russell Group, compared with single honours graduates, and increasingly so the higher the quintile. In the Russell Group, the rate of increase in the proportion of joint honours graduates for the higher quintiles was faster than the rate of increase in the proportion of single honours graduates in the higher quintiles. These two observations would explain the reducing highly skilled destinations gap found the higher the quintile in Table IV.

To give a sense of the scale of the different populations, we included in Table VIII a mirror of Table VII, but showing the absolute graduate numbers in each POLAR4 quintiles, for each honours type, at the Russell Group and Post-92 universities.

3.6 Proportion of the respective population within the Russell Group or post-92 universities, of single or joint honours graduates, split by POLAR4 quintile

In Table IX, we analysed the proportion of graduates split down by POLAR4 quintile who had studied a single or a joint honours degree, within either a Russell Group, a Post-92 university and over all universities. This analysis illustrated several points.

First, over the entire population in all universities, and relative to single honours, there were proportionately fewer joint honours graduates in the lower quintiles, and proportionately more in the upper quintiles. However, because Table VII shows that proportionately, more of the joint honours graduates were from the Russell Group, this is why we saw the trend in Table IV, which showed a decreasing gap between single and joint honours graduate rates of highly skilled destinations in the upper quintiles.

Second, that the range of participation across the quintiles was much smaller within the Post-92 universities than within the Russell Group. It was clear that quintile 5 (most educationally advantaged) dominated within the Russell Group, while quintile 1 (most educationally disadvantaged) were highly under-represented in within the Russell Group; this was the case for both single and joint honours.

While this pattern also existed within the Post-92 universities, the spread was much more even and this group of universities had a far more equitable admissions profile compared with the Russell Group. This echoed the data from HESA (2018a) and also demonstrated similar distributions across the POLAR4 quintiles for joint honours as well as single honours graduates.

Third, within both groups of universities, the proportion of their single honours graduates was broadly comparable to the proportion of their joint honours graduates, for any particular quintile. However in the Russell Group, ranging between the lowest participation quintile (1) and the highest (5), there was a shift proportionately to a higher proportion of their joint honours graduates, as a fraction of the Russell Group population of joint honours graduates. In other words, the higher up the participation quintile, the more likely were graduates to hold a joint honours degree as a proportion of the Russell Group’s entire joint honours population, compared with the likelihood of holding a single honours degree as a proportion of the Russell Group’s entire single honours population.

Across the Russell Group, there was a greater likelihood for their joint honours graduates to be from the highest participation, most educationally advantaged, quintile, than the proportion of their single honours graduates from the highest quintile. This was not seen in the Post-92 universities, where comparable proportions of their single and joint honours came from each quintile. A reflection of this point was followed up in the Discussion section, and also as a basis for future work.

3.7 Proportion of the university population of graduates, by honours type, split by POLAR4 quintile

In order to explore further any relationship between graduates from different groups of university and the POLAR4 quintiles, we analysed whether joint honours graduates were a large part of the university groups’ overall populations. This analysis included “non-joint” subjects, to enable an analysis of the entire set of graduates, irrespective of subject studied.

Table X clearly demonstrates that the higher the quintile (more educationally advantaged), the higher the proportion of joint honours graduates compared with single honours graduates, averaged over all universities in the UK. The educationally advantaged were more likely to have elected to study a joint honours degree as a proportion of that quintile, compared with lower quintiles.

This national finding hid, however, an even more prominent and very interesting difference between the Russell Group and Post-92 universities, shown in Tables XI and XII. At the Russell Group (Table XI), the relative proportion of joint honours graduates in any quintile was higher than the national average, and in quintile 5 the proportion of the total Russell Group graduates with a joint honours degree was substantial at 13.49 per cent.

At Post-92 universities (Table XII), there was little variation in the proportion of joint honours graduates, irrespective of their POLAR4 quintile, and overall the proportion relative to single honours at these universities was much lower in every quintile than the national average.

Tables IX–XI demonstrate that the joint honours graduates had a larger “footprint” within the Russell Group, compared with at the Post-92 universities, and the higher quintile joint honours graduates even more so. A reflection on this interesting point was included in the Discussion section and also formed in part the proposal for future work.

4. Discussion

Over a long period of expansion in the participation in a university education in the UK, the numbers graduating with a joint honours degree have grown year on year, albeit with a slightly declining overall share of the graduate population (Table II). The graduates of 2016/2017 bucked this trend, with a modest 0.33 per cent point rise in the proportion of all graduates with a joint honours degree. It will be interesting to observe in future years whether this represents a turning point and whether the market share for joint honours continues to grow. In planning their future portfolios, university leaders may wish to consider the durability of this Honours type, with an eye on the potential for future growth and increase in market share.

Pigden and Moore (2018) found a national, year-on-year, negative gap between the proportion of joint honours graduates in highly skilled destinations compared with single honours graduates. In the current study, we added a further two years of DLHE survey data to the analysis of Pigden and Moore (2018) and found that the negative gap persisted (Table III), averaging −2.99 per cent points between the highly skilled destinations rates of the two honours types over the most recent three years of data available (2014/2015–2016/2017).

In seeking to understand this gap better and build on previous published work, in the current study we explored any potential relationship between the POLAR4 classification of participation and the rates of highly skilled destinations, especially in combination with the graduate having studied at the Russell Group or a Post-92 university. The main research question being explored was whether there was a correlation between POLAR4 quintiles and highly skilled destinations, and whether the negative highly skilled gap for joint honours graduates was attributable to particular quintiles, perhaps magnified by the effect of studying at the Russell Group or Post-92 university.

We already knew from Pigden and Moore (2018) that joint honours graduates of the Russell Group were more likely to be in highly skilled destinations than Post-92 university joint honours graduates, and also that the gap between single honours and joint honours graduates was much smaller at the Russell Group than in Post-92 universities. Might categorisation of the graduate by their POLAR4 quintile be correlated with these results for joint honours graduates? We knew already that the Russell Group were overall admitting disproportionately more students from quintile 5 (HESA, 2018a), but not whether upper quintile joint honours graduates were more or less likely to be from the Russell Group. We also knew that nationally the lower quintiles generally had lower rates of highly skilled destination (2018b), but what was the relationship between POLAR4 quintiles and the highly skilled destinations of joint honours graduates?

Table IV demonstrates that at the national level, there was a trend for both single honours and joint honours graduates from higher participation POLAR4 quintiles to be more likely to be in a highly skilled destination, i.e. the more educationally advantaged were more likely to be in a highly skilled destination. This accorded with HESA (2018b) data, but expanded upon that to include direct consideration of joint honours graduates.

Furthermore, Table IV shows that joint honours graduates hailing from the most educationally advantaged regions in the UK had a much smaller, albeit still negative, highly skilled destinations gap compared with single honours graduates. We showed that nationally there was a trend towards a smaller highly skilled destinations gap between single honours and joint honours graduates, the higher the level of educational advantage. The smallest gap for quintile 5 and the largest gap for quintile 1.

Tables V and VI demonstrate that all Russell Group graduates, irrespective of their POLAR4 quintile, were far more likely to be in a highly skilled destination than single or joint honours graduates of Post-92 universities. Even the lowest quintile graduates of the Russell Group had greater rates of highly skilled destination than the highest quintile from Post-92 universities, for both single and joint honours graduates.

However, at both the Russell Group (Table V) and Post-92 universities (Table VI), there was no observable trend as seen in Table IV towards a smaller highly skilled destination gap for the upper quintiles. The gap between single and joint honours rates of highly skilled destination remained fairly consistent irrespective of the quintile, and was much larger at Post-92 universities than the Russell Group. So the national trend seen in Table IV seemed instead to have related to a higher proportion of joint honours graduates in the upper quintiles, combined with a higher proportion also having studied at the Russell Group.

Table VII shows that the proportion of quintile 5 joint honours graduates from the Russell Group was disproportionately high, and in any quintile there were proportionately more joint honours graduates from the Russell Group, compared with single honours graduates, and increasingly so the higher the quintile. In the Russell Group, the rate of increase in the proportion of joint honours graduates for the higher quintiles was faster than the rate of increase in the proportion of single honours graduates in the higher quintiles. These two observations would explain the reducing highly skilled destinations gap found the higher the quintile in Table IV.

To reiterate, Tables V and VI show that at both the Russell Group and Post-92 universities, there was no trend towards a smaller highly skilled destinations gap across the range of POLAR4 quintiles. So although both single and joint honours graduates of both universities were more likely to be in a highly skilled destination, the higher their POLAR4 quintile, neither the Russell Group nor Post-92 universities were more or less likely to impact on the relative performance of single or joint honours graduates, for any particular quintile.

We also verified and expanded in Table IX upon the HESA (2018a) data and showed that the Post-92 universities had a far more equitable admissions profile for both joint and single honours graduates, compared with the Russell Group. This analysis also highlighted that the joint honours graduates from the Russell Group were far more likely to have come from educationally advantaged (quintile 5) regions.

It was interesting to note in Table X that the increasing proportion of joint honours graduates compared with single honours graduates as educational advantage increased, with POLAR4 quintile 1 having the lowest proportion of joint honours graduates to quintile 5 having the highest proportion, relative to the single honours graduates. This national finding hid, however, an even more prominent difference between the Russell Group and Post-92 universities, shown in Tables XI and XII.

At the Russell Group (Table XI), the relative proportion of joint honours graduates in any quintile was higher than the national average, and in quintile 5 the proportion of the Russell Group graduating with joint honours was substantial at 13.49 per cent of the total population of graduates. At Post-92 universities (Table XII), there was little variation in the proportion of joint honours graduates, irrespective of their POLAR4 quintile, and overall the proportion relative to single honours at these universities was much lower in every quintile, and lower than the national average. Might this particular feature contribute towards the smaller highly skilled destinations gap between single honours and joint honours graduates at the Russell Group, which we were unable to explain in the current study?

Why was the gap between single and joint honours highly skilled destinations so much larger at Post-92 universities than at the Russell Group, as found in Pigden and Moore (2018), and also shown to exist irrespective of the POLAR4 quintile in the current study? Examining the distribution of graduates across the POLAR4 quintiles in the current study had not explained this difference, and so further work would be required exploring other university, student and graduate characteristics and demographic profiling.

For example, it is known (Telhaj et al., 2015; Feng and Graetz, 2017; Walker and Zhu, 2018) that there is a positive relationship among achieving “good honours” (a First Class or Upper Second Class degree), university selectivity and securing graduate employment. Furthermore, certain demographics and characteristics, such as tariff on entry, age, gender and ethnicity, are correlated with class of degree achieved and academic outcomes (Richardson, 2018; Mountford-Zimdars et al., 2015; Naylor and Smith, 2004). Therefore, further exploration of this literature and data might help to understand the highly skilled destinations gap between single and joint honours graduates, and why the gap is so much smaller at the Russell Group.

5. Conclusion and future work

The current study confirmed the previous work of Pigden and Moore (2017, 2018) in showing a negative highly skilled destinations gap between joint and single honours graduates, at the national level, updated to include the most recent two years of data available from the DLHE survey.

The current study also showed that at the national level, graduates who had come from the higher POLAR4 quintiles (more educationally advantaged at the point of admission to university) were more likely to be in a highly skilled destination post-graduation. This was true for both single and joint honours graduates, and demonstrated the lasting effect of educational advantage on individuals, even following a university education, at the national level.

However, the impact of the type of university at which the graduate had studied was demonstrated in Tables V and VI, and this highlighted that all Russell Group graduates, irrespective of their POLAR4 quintile, were far more likely to be in a highly skilled destination than single or joint honours graduates of Post-92 universities. Even the lowest quintile graduates of the Russell Group had greater rates of highly skilled destination than the highest quintile from Post-92 universities, for both single and joint honours graduates. Our study could not explain this difference, and other student characteristics such as tariff on entry, subjects studied, gender, age and ethnicity might all contribute to this finding.

The gap between single and joint honours graduates decreased the higher the quintile, at the national level. However, in both the Russell Group and Post-92 universities, the gap remained fairly constant, irrespective of the quintile, albeit with a much smaller gap at the Russell Group than the Post-92 universities.

This was thought to be because the proportion of quintile 5 joint honours graduates from the Russell Group was disproportionately high (Table VI). The joint honours upper POLAR4 quintiles also represented a larger footprint within the Russell Group overall undergraduate population (Table XI), although the impact of this last point is unclear without further exploration.

At Post-92 universities, the admissions profile was more even across the quintiles than at the Russell Group, although still with a preponderance of upper POLAR4 quintiles (Table IX). However, there was an even spread of joint honours graduates across the POLAR4 quintiles at Post-92 universities (Table XII), in contrast to the Russell Group where there was an increase in the proportion of joint honours graduates the higher the quintile.

These educationally advantaged joint honours graduates at the Russell Group have had more of a footprint at the UK’s top performing, highly selective universities would be an interesting point to explore in future work, and may assist in explaining the relative success of the Russell Group joint honours graduates. It may be that the graduates’ confidence levels are higher, given the relatively high prevalence of higher quintiles for this type of honours degree at the Russell Group.

Future work will also seek to understand why a higher proportion of joint honours graduates hail from the upper quintiles, why the Russell Group joint honours graduates were more disproportionately from the upper POLAR4 quintiles and why the joint honours upper POLAR4 quintiles represented such a larger proportion of the Russell Group overall undergraduate population. This may be because of differences in the careers advice and guidance provided in independent or higher educational advantaged schools or, say, differences in university marketing and admissions strategies.

Our study of POLAR4 quintiles could not explain the much smaller gap in the highly skilled destinations between single honours and joint honours graduates found in the Russell Group, compared with the Post-92 universities. Other student characteristics and demographic profiling, such as tariff on entry, subjects studied, age, gender and ethnicity, could all be contributing factors, particularly in relation to the classification of honours degree achieved, and will also form the basis of future work.

Non-joint honours subjects

JACS code Principal subject
A1 Pre-clinical Medicine
A2 Pre-clinical Dentistry
A9 Others in Medicine and Dentistry
B5 Ophthalmics
G02 Broadly based programmes in computer science (2011/2012 only)
D1 Pre-clinical Veterinary Medicine
D2 Clinical Veterinary Medicine & Dentistry
D9 Others in Vet Sci., Ag & related subjects
H9 Others in Engineering
I5 Health Informatics
J1 Minerals Technology
K0 Architecture, Build & Plan: any area
K9 Others in Architecture, Build & Plan
W0 Creative Arts & Design: any area
A3 Clinical Medicine
A4 Clinical dentistry

Proportion of graduates, includes “non-joint” subjects

Proportion of graduates
DLHE year Single honours (%) Joint honours (%)
2011/2012 89.24 10.76
2012/2013 89.61 10.39
2013/2014 90.22 9.78
2014/2015 90.71 9.29
2015/2016 91.17 8.83
2016/2017 90.84 9.16

Proportion of graduates in highly skilled destinations, excludes “non-joint” subjects

Highly skilled destinations (TEF methodology)
DLHE year Single honours (%) Joint honours (%)
2011/2012 64.30 60.98
2012/2013 66.00 63.60
2013/2014 68.29 65.82
2014/2015 71.31 67.78
2015/2016 73.20 70.48
2016/2017 76.26 73.55

Proportion of graduates in highly skilled destinations by POLAR4 quintile, excludes “non-joint” subjects

Highly skilled destinations (TEF methodology)
POLAR4 Single honours (%) Joint honours (%) Difference (%)
1 71.29 65.98 5.31
2 72.22 67.30 4.92
3 72.82 68.88 3.95
4 72.94 69.31 3.64
5 75.60 73.63 1.97

Proportion in highly skilled destinations of the Russell group population of graduates, by honours type, split by POLAR4 quintile, excludes “non-joint” subjects

Highly skilled (TEF methodology)
POLAR4 Single honours (%) Joint honours (%) Difference (%)
1 78.59 77.55 1.04
2 78.85 76.23 2.61
3 79.33 77.96 1.37
4 79.16 76.61 2.56
5 80.04 78.95 1.09

Proportion in highly skilled destinations of the Post-92 university population of graduates, by honours type, split by POLAR4 quintile, excludes “non-joint” subjects

Highly skilled (TEF methodology)
POLAR4 Single honours (%) Joint honours (%) Difference (%)
1 69.71 62.73 6.99
2 70.14 62.83 7.31
3 70.42 62.52 7.91
4 69.82 63.60 6.22
5 71.94 65.36 6.58

Proportion in each POLAR4 quintile of the overall national single and joint honours population for that quintile, showing the Russell Group and Post-92 universities only

Proportion of national population
Russell Group Post-92 universities
POLAR4 Single honours (%) Joint honours (%) Single honours (%) Joint honours (%)
1 13.30 18.61 67.28 58.08
2 15.84 23.69 63.60 50.32
3 18.68 28.43 59.32 44.83
4 22.41 33.46 55.09 39.99
5 33.46 49.42 43.89 26.99

Number in each POLAR4 quintile of the overall national single and joint honours population for that quintile, showing the Russell Group and Post-92 groups of universities only

Russell Group Post-92 universities
POLAR4 Single honours Joint honours Single honours Joint honours
1 8,635 1,035 43,675 3,235
2 14,540 1,945 58,380 4,135
3 21,360 3,095 67,830 4,880
4 31,220 4,750 76,735 5,680
5 63,980 11,425 83,930 6,240

Proportion of graduates from the Russell Group and Post-92 universities by POLAR4 quintile, excludes “non-joint” subjects

Russell Group Post-92 universities Total population
POLAR4 Single honours (%) Joint honours (%) Single honours (%) Joint honours (%) Single honours (%) Joint honours (%)
1 5.94 4.65 13.19 13.38 10.53 8.98
2 10.18 8.75 17.64 17.11 15.01 13.26
3 15.14 13.90 20.51 20.18 18.84 17.55
4 22.32 21.35 23.20 23.50 23.13 22.91
5 46.42 51.35 25.46 25.82 32.49 37.31

Proportion by honours type in a university population, by POLAR4 quintile, includes “non-joint” subjects

Proportion
POLAR4 Single honours (%) Joint honours (%)
1 92.22 7.78
2 91.96 8.04
3 91.56 8.44
4 91.07 8.93
5 89.79 10.21

Proportion by honours type in the Russell Group populations, by POLAR4 quintile, includes “non-joint” subjects

Proportion
POLAR4 Single honours (%) Joint honours (%)
1 90.06 9.94
2 89.20 10.80
3 88.54 11.46
4 88.12 11.88
5 86.51 13.49

Proportion by honours type in Post-92 universities populations, by POLAR4 quintile, includes “non-joint” subjects

Proportion
POLAR4 Single honours (%) Joint honours (%)
1 93.12 6.88
2 93.40 6.60
3 93.31 6.69
4 93.13 6.87
5 93.12 6.88

References

Belfield, C., Britton, J., Deardon, L. and van der Erve, L. (2017), “Higher education funding in England: past, present and options for the future”, IFS Briefing Note No. BN211, ISBN: 978-1-911102-48-9, The Institute of Fiscal Studies, London.

Belfield, C., Britton, J., Buscha, F., Deardon, L., Dickson, M., van der Erve, L., Sibieta, L., Vignoles, A., Walker, I. and Zhu, Y. (2018), The Relative Labour Market Returns to Different Degrees, ISBN: 978-1-78105-912-8, The Institute of Fiscal Studies, London.

Browne, J. (2010), “Securing a sustainable future for higher education: an independent review of higher education funding and student finance”, available at: http://webarchive.nationalarchives.gov.uk/+/hereview.independent.gov.uk/hereview/report/ (accessed 1 August 2018).

DfE (2017a), Participation Rates in Higher Education: Academic Years 2006/2007-2015/2016, Department for Education, London, available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/648165/HEIPR_PUBLICATION_2015-16.pdf (accessed 1 August 2018).

DfE (2017b), Graduate Labour Market Statistics 2017, Department for Education, London, available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/701720/GLMS_2017.pdf (accessed 1 August 2018).

DfE (2017c), Widening Participation in Higher Education: 2017, Department for Education, London, available at: www.gov.uk/ government/uploads/system/uploads/attachment_data/ file/635104/SFR39-2017-MainTables.xlsx (accessed 1 August 2018).

Dowling, A. (2015), “The Dowling review of business-university research collaborations”, Open Government Licence, London, available at: www.gov.uk/bis (accessed 1 August 2018).

Evennett, D. (2018), “Higher education: admissions: written question – 130854”, available at: www.parliament.uk/business/publications/written-questions-answers-statements/written-question/Commons/2018–03-05/130854/ (accessed 31 July 2018).

Feng, A. and Graetz, G. (2017), “A question of degree: the effects of degree class on labor market outcomes”, Economics of Education Review, Vol. 61 No. C, pp. 140-161.

HESA (2017), “POLAR4 classification: a local geography classification for young participation in higher education”, available at: www.hefce.ac.uk/pubs/year/2017/201729/ (accessed 31 July 2018).

HESA (2018a), “Widening participation: UK performance indicators 2016/17”, available at: www.hesa.ac.uk/news/01–02-2018/widening-participation-tables (accessed 31 July 2018).

HESA (2018b), “Differences in student outcomes – the effect of student characteristics”, Data analysis March 2018/05, available at: www.hefce.ac.uk/pubs/year/2018/201805/ (accessed 31 July 2018).

Higher Education Funding Council for England (2015), “2014-15 funding and monitoring data overview”, available at: www.hefce.ac.uk/data/famd/latest/2014-15,overview/#section1 (accessed 30 July 2018).

Higher Education Funding Council for England (2017), “Teaching excellent framework (TEF)”, available at: www.hefce.ac.uk/lt/tef/ (accessed 30 July 2018).

Hodgson, J. (2011), “The experience of joint honours students of English in UK higher education”, Report Series No. 26, ISBN 978-1-905846-55-9, The Higher Education Academy, English Subject Centre, York, June.

Marginson, S. (2016), “High participation systems of higher education”, Journal of Higher Education, Vol. 87 No. 2, pp. 243-271.

Martin, I. (2018), “Benchmarking widening participation: how should we measure and report progress?”, HEPI Policy Note No. 6, Oxford, available at: www.hepi.ac.uk/wp-content/uploads/2018/04/HEPI-Policy-Note-6-Benchmarking-widening-participation-FINAL.pdf (accessed 1 August 2018).

Mountford-Zimdars, A., Sabri, D., Moore, J., Sanders, J., Jones, S. and Higham, L. (2015), Causes of Differences in Student Outcomes, Higher Education Funding Council for England, Bristol, available at: www.hefce.ac.uk/media/HEFCE,2014/Content/Pubs/Independentresearch/2015/Causes,of,differences,in,student,outcomes/HEFCE2015_diffout.pdf (accessed 25 November 2018).

Naylor, R.A. and Smith, J.P. (2004), “Determinants of educational success in higher education”, in Johnes, G. and Johnes, J. (Eds), International Handbook of the Economics of Education, Chapter 11, Edward Elgar.

Office for National Statistics (2010), “Standard occupational classification”, available at: www.ons.gov.uk/methodology/classificationsandstandards/standardoccupationalclassificationsoc (accessed 30 July 2018).

Pigden, L. (2016), “Understanding the lived experiences of joint honours graduates: how can educators best enable student success?”, International Journal of Arts and Sciences, Vol. 9 No. 3, pp. 467-483.

Pigden, L. and Jegede, F. (2016), “Combined degrees & employability: a comparative analysis of single and joint honours graduates of UK universities”, West East Journal of Social Sciences, Vol. 5 No. 2, pp. 11-19.

Pigden, L. and Jegede, F. (2018), “Understanding the educational needs of joint honours degree students in a post Brexit United Kingdom higher education sector”, PEOPLE: International Journal of Social Sciences, Vol. 4 No. 1, pp. 383-404.

Pigden, L. and Moore, G. (2017), “Does subject choice in a joint degree affect highly skilled graduate employment?”, PUPIL: International Journal of Teaching, Education and Learning, Vol. 1 No. 1, pp. 93-114.

Pigden, L. and Moore, A.G. (2018), “Employability outcomes for university joint honours graduates”, Higher Education, Skills and Work-Based Learning, Vol. 8 No. 2, pp. 195-210, doi: 10.1108/HESWBL-11-2017-0088.

Richardson, J.T.E. (2018), “Understanding the under-attainment of ethnic minority students in UK higher education: the known knowns and the known unknowns”, in Arday, J. and Mirza, H. (Eds), Dismantling Race in Higher Education, Palgrave Macmillan, Cham, pp. 87-102.

Telhaj, S., Naylor, R. and Smith, J. (2015), “Does degree class matter? Graduate earnings and student achievement in UK universities”, Oxford Economic Papers, Vol. 68 No. 2, pp. 525-545.

UCAS (2016), “Applicants and acceptances for universities and colleges – 2016”, Provider by subject group, available at: www.ucas.com/data-and-analysis/ucas-undergraduate-releases/ucas-undergraduate-end-cycle-data-resources/applicants-and-acceptances-universities-and-colleges-2016 (accessed 30 July 2018).

UCAS (2017), “End of cycle report 2017 – patterns by subject”, available at: www.ucas.com/data-and-analysis/ucas-undergraduate-releases/ucas-undergraduate-analysis-reports/2017-end-cycle-report (accessed 30 July 2018).

Universities UK (2015), “Efficiency, effectiveness and value for money”, ISBN: 978-1-84036-332-6, available at: www.universitiesuk.ac.uk/policy-and-analysis/reports/Documents/2015/efficiency-effectiveness-value-for-money.pdf (accessed 10 February 2019).

Walker, I. and Zhu, Y. (2018), “University selectivity and relative returns to higher education: evidence from the UK”, Labour Economics, Vol. 53, August, pp. 230-249.

Further reading

UCAS (2018), “2017 entry UCAS undergraduate reports by sex, area background and ethnic group”, available at: www.ucas.com/corporate/data-and-analysis/ucas-undergraduate-releases/ucas-undergraduate-reports-sex-area-background-and-ethnic-group/2017-entry-ucas-undergraduate-reports-sex-area-background-and-ethnic-group (accessed 31 July 2018).

Corresponding author

Louise Pigden can be contacted at: l.pigden@derby.ac.uk