Viva results and candidate experience need separate scrutiny

Updated Sep 07, 2026

Examination records and candidate accounts answer different questions about a viva. Alistair McCulloch, Cassandra Loeser and Rachel Bageas combine them in an evaluation at the University of South Australia, published online in Assessment & Evaluation in Higher Education on 3 October 2025.

Evidence scope: this briefing uses the publisher’s abstract and accessible indexed methods, results and discussion passages. The complete article and table image were not accessible during review.

What the evaluation found

UniSA mandated an oral defence in 2016. A December 2024 survey invited 357 candidates to respond; 70 did so, a 19.6% response rate, and 45 provided at least one comment. The authors note that invitations went to student email addresses that graduates might no longer monitor.

The outcome analysis covered 357 candidates examined during 2019–23. Compared with preliminary recommendations, 232 had the same result, 120 a higher result and 5 a lower result after oral defence. Five out of 357 is approximately1.4%, although the paper’s narrative labels it less than 1%. In total, 125 results differed. This records change across stages of one examination model; it does not isolate the viva from every other influence on the final decision.

Most survey respondents described the experience positively. The accessible results also describe administrative delays, examiner conduct concerns and five respondents raising concerns about inadequate or insufficiently impartial chairing. The discussion identifies feedback and ongoing contact with examiners as developmental possibilities. These accounts should not be treated as proof of effects on research careers or as evidence that training alone produced the positive experiences.

Questions for doctoral colleges

Our suggestion is to ask about preparation, chairing, administration, examiner conduct and the usefulness of feedback separately. An overall satisfaction question may not explain where a candidate experienced difficulty. Offer an appropriate confidential route for individual concerns alongside any aggregate feedback collection; a rare concern may still need attention.

Review examination records and candidate accounts together while keeping their purposes clear. An altered result does not establish fairness, and a positive experience does not determine the academic merits of a decision. Keep information about who answered and how long after the examination they responded.

Teams could use de-identified examples to discuss chair and examiner responsibilities. Any change to preparation or training needs its own evaluation. A PGR comment framework can support local discussion of themes, while the governance checklist addresses handling and review.

FAQ

Q: Did the viva improve results for one-third of candidates?

A: The authors report 120 higher outcomes after oral defence out of 357 records. This comparison shows differences between preliminary and final stages; it does not establish a causal effect separate from the rest of the examination process.

Q: Were lower outcomes below 1%?

A: No. The paper reports five lower outcomes, which is approximately 1.4% of 357. We have corrected that percentage in this briefing.

Q: Do positive responses rule out problems?

A: No. The survey includes positive accounts and specific negative experiences, with a 19.6% response rate at one institution. It cannot establish that the model works equally well for every candidate.

Correction, 7 September 2026: corrected the percentage for five lower outcomes, clarified the limited source access and distinguished recorded outcome changes from causal effects.

Reference

Alistair McCulloch, Cassandra Loeser and Rachel Bageas. “Does the viva matter? PhD student experiences of the oral examination and its contribution to examination outcome and researcher development.” DOI:10.1080/02602938.2025.2566652.