We know a lot about what graduates earn.
Earnings data—especially at subject level—has become key to debates about the value of higher education.
But we know far less about how graduates themselves experience their early careers. Until now, subject-level data on graduate job quality—how meaningful their work is, how well it aligns with their goals, and whether it uses their university-acquired skills—has been missing from the policy debate.
My new study (co-authored with Fiona Christie and Tracy Scurry and published in Studies in Higher Education) aims to fill this gap. Drawing on responses from the 2018-19 graduation cohort in the national Graduate Outcomes survey, we provide the first nationally representative, subject-level analysis of these subjective graduate outcomes.
What we find has important implications for how we define successful outcomes from higher education—and how we support students in making informed choices about what subject to study.
What graduates tell us
The Graduate Outcomes survey includes a set of questions—introduced by HESA in 2017—designed to capture core dimensions of graduate job quality. Respondents are asked (around 15 months after graduation) whether they:
- find their work meaningful
- feel it aligns with their future plans
- believe they are using the skills acquired at university
These indicators were developed in part to address the over-reliance on income as a measure of graduate success. They reflect a growing international awareness that economic outcomes alone offer a limited picture of the value of education—in line with the OECD’s Beyond GDP agenda, the ILO’s emphasis on decent work, and the UK’s Taylor Review focus on job quality.
Subject-level insights
Our analysis shows that most UK graduates report positive early-career experiences, regardless of subject. Across the sample, 86 per cent said their work felt meaningful, 78 per cent felt on track with their careers, and 66 per cent reported using their degree-level skills.
These patterns generally hold across disciplines, though clear differences emerge. The chart below shows the raw, unadjusted proportion of graduates who report positive outcomes. Graduates from vocational fields—such as medicine, subjects allied to medicine, veterinary science, and education—tend to report particularly strong outcomes. For instance, medicine and dentistry graduates were 12 percentage points more likely than average to say their work was meaningful, and over 30 points more likely to report using the skills they acquired at university.
However, the results also challenge the narrative that generalist or academic degrees are inherently low value. As you can see, most subject areas—including history, languages, and the creative arts, often targeted in these debates—show strong subjective outcomes across the three dimensions. Only one field, history and philosophy, fell slightly below the 50 per cent threshold on the skills utilisation measure. But even here, graduates still reported relatively high levels of meaningful work and career alignment.
Once we adjusted for background characteristics—such as social class, gender, prior attainment, and institutional differences—many of the remaining gaps between vocational and generalist subjects narrowed and were no longer statistically significant.
This chart shows the raw proportion of 2018-19 graduates who agree or strongly agree that their current work is meaningful, on track and using skills, by field of study (N = 67,722)
Employment in a highly skilled occupation—used by the Office for Students (OfS) as a key regulatory benchmark—was not a reliable predictor of positive outcomes. This finding aligns with previous HESA research and raises important questions about the appropriateness of using occupational classification as a proxy for graduate success at the subject level.
Rethinking what we measure and value
These insights arrive at a time when the OfS is placing greater emphasis on regulating equality of opportunity and ensuring the provision of “full, frank, and fair information” to students. If students are to make informed choices, they need access to subject-level data that reflects more than salary, occupational status, or postgraduate progression. Our findings suggest that subjective outcomes—how graduates feel about their work—should be part of that conversation.
For policymakers, our findings highlight the risks of relying on blunt outcome metrics—particularly earnings and occupational classifications—as indicators of course value. Our data show that graduates from a wide range of subjects—including those often labelled as “low value”—frequently go on to report meaningful work shortly after graduation that aligns with their future plans and makes use of the skills they developed at university.
And while job quality matters, universities should not be held solely accountable for outcomes shaped by employers and labour market structures. Metrics and league tables that tie institutional performance too closely to job quality risk misrepresenting what higher education can influence. A more productive step would be to expand the Graduate Outcomes survey to include a wider range of job quality indicators—such as autonomy, flexibility, and progression—offering a fuller picture of early career graduate success.
A richer understanding
Our work offers the first nationally representative, subject-level insight into how UK graduates evaluate job quality in the early stages of their careers. In doing so, it adds a missing piece to the value debate—one grounded not just in earnings or employment status, but in graduates’ own sense of meaning, purpose, and skill use.
If we are serious about understanding what graduates take from their university experience, it’s time to move beyond salary alone—and to listen more carefully to what graduates themselves are telling us.
DK notes: Though the analysis that Brophy et al have done (employing listwise deletion, examining UK domiciled first degree graduates only) enhances our understanding of undergraduate progression and goes beyond what is publicly available, I couldn’t resist plotting the HESA public data in a similar way, as it may be of interest to readers:
