Roughly four in five people who start a full-time UK PhD are expected to finish it: England’s former funding council, HEFCE, projected in 2013 that 72.9% of the 11,625 UK and EU students who began full-time doctorates in 2010–11 would qualify within seven years, and the figure most often quoted for eventual completion is 80.5%. Separately, a Freedom of Information exercise covering 26,076 candidates who sat a viva at 14 UK universities between 2006 and 2017 found that 96% passed. HESA, the official custodian of UK higher education statistics, does not publish an equivalent headline non-continuation figure for postgraduate researchers the way it does for undergraduates — here is exactly what does and does not exist.
The Two Sources Behind the Headline Figures
The numbers you will see quoted come from two different kinds of data, measuring two different things, and they are easy to blur together. The first is HEFCE’s projection of qualification rates for doctoral entrants — a forward-looking estimate for a cohort of starters. The second is a Freedom of Information analysis by DiscoverPhDs, an independent PhD-advice publisher, which asked 14 universities for the outcomes of candidates who reached examination: Cardiff, Imperial College London, LSE, Queen Mary University of London, UCL, Aberdeen, Birmingham, Bristol, Edinburgh, Nottingham, Oxford, York, Manchester and Sheffield, with vivas held between 2006 and 2017. DiscoverPhDs then combined the two to estimate the overall journey from enrolment to outcome:
| Figure | Value | Where it comes from |
|---|---|---|
| Full-time UK/EU 2010–11 starters projected to qualify within seven years | 72.9% | HEFCE report, July 2013 (11,625 entrants) |
| Starters completing eventually (within 25 years) | 80.5% | HEFCE projection as cited by DiscoverPhDs |
| Candidates who sat the viva and passed | 96% (25,063 of 26,076) | DiscoverPhDs FOI analysis, 14 universities |
| Viva failures awarded an MPhil instead | ~97% of those who failed | DiscoverPhDs FOI analysis |
| Estimated share of all starters who leave before the viva | 16.2% | DiscoverPhDs’ own combination of the HEFCE and FOI figures |
| Estimated share of all starters who fail the viva | 3.3% | DiscoverPhDs’ own combination of the HEFCE and FOI figures |
Three things are worth stating plainly about these figures, in keeping with treating them as evidence rather than as a settled national statistic. First, the 96% viva pass rate describes candidates who reached examination at 14 research-intensive universities, several of them Russell Group — not a random or representative sample of the roughly 160-plus UK institutions that admit doctoral researchers, and it says nothing on its own about how many people left before the viva. Second, the HEFCE figures are projections for full-time UK and EU entrants, made before that cohort had finished; part-time candidates and later cohorts are not covered by them. Third, the 16.2% and 3.3% splits are an estimate built by combining a projection for one population with viva outcomes from another, so they are best read as an approximate picture of the journey, not a measured breakdown.
The FOI half carries its own caveat: Freedom of Information responses depend on how each university’s own student records system codes an outcome, and DiscoverPhDs notes that not every university reported the detailed outcome (for example, whether a failed candidate received an MPhil), so the finer breakdowns rest on the data that was provided. Neither source is built from one single, standardised national definition the way HESA’s own undergraduate non-continuation measure is.
Why Doesn’t HESA Publish a Simple PGR Non-Continuation Figure?
HESA does publish non-continuation performance indicators, but its most prominent, regularly cited non-continuation measure is built around first-year undergraduate entrants continuing into a second year of study — a metric that does not map cleanly onto a multi-year, often part-time, frequently interrupted doctoral candidature. A PhD does not have a single “second year” continuation checkpoint in the way a three-year undergraduate degree does, and outcomes that would count as “non-continuation” for an undergraduate — an intermission for ill health, a change from full-time to part-time registration, a transfer between institutions following a supervisor’s move — are routine, often recoverable, parts of a normal PGR candidature rather than dropout in the everyday sense. That mismatch is a likely reason no single, prominently reported HESA headline for PGR non-completion exists in the way it does for undergraduate non-continuation.
What HESA does publish reliably, and what this site has already covered directly from HESA’s own figures, is the population count itself: how many PhD students are there in the UK and the registration-length data underlying how long does a UK PhD actually take. Both of those are population and duration measures, not completion-rate measures, and the distinction matters if you are citing this article’s figures alongside either of those pieces: none of the three answers the same question as the others. A genuine national, year-by-year, definition-standardised PGR completion rate would need a regularly published measure on the same basis as the undergraduate one — until then, the HEFCE projections and the 14-university viva data are the most useful substitutes available.
Full-time and part-time registration add a further wrinkle. A part-time candidature routinely runs longer, and the HEFCE projections quoted above cover full-time starters only, so the completion picture for part-time PGR study specifically could differ from the figures above, in a direction the published data does not let you determine.
Why 72.9% and 80.5% Are Not Evidence of a Trend
The gap between 72.9% and 80.5% is a difference of measurement window, not evidence that completion improved: the first counts only those projected to qualify within seven years, the second counts eventual completion over a much longer horizon. A seven-year cutoff will always show a lower rate than a measure with a longer or no cutoff, because it counts a candidate who eventually finishes in year eight as a non-completer. For context, the same 2013 HEFCE report put the equivalent seven-year figure at 70.1% for 2009–10 starters and 70.5% for 2008–09 starters, so year-to-year movement in the seven-year measure was small.
What Actually Predicts Whether an Individual Candidature Completes?
Neither source above breaks its figures down by cause, but the doctoral-education literature and sector guidance consistently point to a small number of structural factors that supervisors, graduate schools and funders treat as the practical levers: funding security across the whole expected candidature rather than just the first year or two, a supervisory team that meets regularly and gives structured, actionable feedback rather than being difficult to access, a project scope that was realistic at the outset rather than expanding as new interesting questions appear, and an institution’s own progress-review process actually being used to flag a struggling candidature early rather than as a formality. None of these is a guarantee in either direction, and this article does not attach a specific percentage to any one of them, because no dataset we are aware of breaks the non-completion figure down by cause with enough rigour to support one.
What This Means If You Are Choosing Between a PhD and Another Route
Read plainly, roughly four in five people who start a UK PhD are expected to finish it, and of those who reach the viva, the great majority pass, with most of those who do not still leaving with an MPhil rather than nothing. Leaving before examination accounts for most of the gap, and the available data does not break that group down by reason — funding running out, a change in personal circumstances, a change of career direction, or a genuine mismatch between the project and the candidate are all folded into the same estimate. If your own candidature runs into difficulty, how to rebuild your thesis for a resubmission after a referral covers the specific process for the smaller group who reach examination and are asked to revise and resubmit rather than being awarded outright, which is a different, more common outcome than outright failure, and the actual examiner-report patterns behind viva outcomes generally are covered in PhD viva outcomes in the UK: what the data from 127 institutions actually shows.
None of these figures should be read as a prediction about any individual candidature. A projected rate for a national cohort, or a pass rate across 26,076 vivas, describes the group, not the specific combination of funding security, supervision quality and project scope that determines any one person’s actual outcome — the same caveat that applies to any population-level statistic used to reason about an individual case.
Frequently Asked Questions
What percentage of PhD students in the UK complete their doctorate?
Roughly four in five. HEFCE projected that 72.9% of full-time UK and EU students starting in 2010–11 would qualify within seven years, and the commonly quoted eventual-completion figure is 80.5%. HESA does not publish an equivalent headline completion or non-continuation rate for postgraduate researchers.
Why doesn’t HESA publish an official PhD completion rate?
HESA’s most prominent non-continuation measure is built around undergraduate first-to-second-year progression, a concept that does not map cleanly onto a multi-year doctoral candidature with routine intermissions, mode changes and transfers. That mismatch is a likely reason no single equivalent PGR headline exists.
What share of PhD students leave before the viva?
DiscoverPhDs estimates about 16.2% of starters, by combining the HEFCE completion projection with its own viva-outcome data. It is an estimate rather than a measured figure, and it is not broken down by the reason for leaving.
What share of PhD students fail the viva outright?
About 4% of those who sit the viva, per the 14-university FOI data — roughly 3.3% of all starters on DiscoverPhDs’ combined estimate. Of those who fail, roughly 97% are awarded an MPhil rather than leaving with no qualification at all.
What is the viva pass rate for those who actually sit it?
96%: 25,063 of the 26,076 candidates who sat a viva at the 14 universities in the FOI analysis passed.
Which universities were included in the FOI analysis?
Cardiff, Imperial College London, LSE, Queen Mary University of London, UCL, Aberdeen, Birmingham, Bristol, Edinburgh, Nottingham, Oxford, York, Manchester and Sheffield — a sample of 14 research-intensive universities, not a representative sample of every UK institution awarding PhDs.
Is the 80.5% figure nationally representative?
It is a projection for full-time UK and EU starters, not a measured outcome for every doctoral student, and it does not cover part-time candidates. Treat it as the best available estimate rather than a standardised national statistic.
What is the 72.9% figure sometimes cited alongside this?
HEFCE’s 2013 projection that 72.9% of the 11,625 UK and EU students who began full-time doctorates in 2010–11 would qualify within seven years. A fixed seven-year cutoff will always show a lower rate than an eventual-completion measure, so the two figures are not directly comparable.
Does HESA publish any PGR data at all?
Yes — population counts by year and mode of study, which this site covers directly in how many PhD students are there in the UK, and registration-length data. What it does not publish as a single headline is a completion or non-continuation rate specifically for postgraduate researchers.
What happens if I leave a PhD before finishing?
Outcomes vary by institution and stage reached; some universities offer an exit award (such as an MPhil) depending on how much work has been examined, while others do not. Check your own institution’s research degree regulations for the specific rule that applies to your registration.
Should I use this figure to decide whether to start a PhD?
Use it as context, not a prediction. A projected rate for a national cohort describes the group, not the specific combination of funding, supervision and project scope that will determine your own outcome.