Writing attracts wide but shallow comment in examiner reports: around 73 per cent of examiners raise it at least once, and it accounts for roughly an eighth of the average report, against 43 per cent for analysis and reporting. That is the shape of the thing you are being judged on — and the studies that establish it are Australian, because no equivalent open UK content analysis exists.
What the report is actually made of
The most detailed published content analysis of doctoral examiner reports covers 101 candidates and 303 examiners. Its distribution of attention is the single most useful thing a candidate can know before submitting.
| What examiners commented on | Share of reports mentioning it | Share of the average report’s text | Source and year |
|---|---|---|---|
| Analysis and reporting | 89% | 43.1% | Holbrook, Bourke, Lovat and Dally, 2004 |
| Contribution and significance | 70% | 9% | Holbrook et al., 2004 |
| Editorial matters (typographical, mechanical, clerical or technical errors) | — | 8% | Holbrook et al., 2004 |
| Substantial issues of communicative competence | 73% | 5% | Holbrook et al., 2004 |
| Scope of the study | 45% | 3% | Holbrook et al., 2004 |
In the authors’ own words: “73 per cent of examiners mentioned substantial issues of communicative competence at least once, while 50 per cent identified specific corrections”, and “On average in this study they devoted 5 per cent of text units or lines to substantial issues of communicative competence (351), but 8 per cent to editorial matters (typographical, basic mechanical, clerical or technical errors).”
Read that table carefully, because it cuts both ways. Writing is not what your report will mostly be about — analysis and reporting takes ten times the space. But writing is what almost every examiner will mention, and more of the writing commentary is spent on typos than on expression.
An earlier study quoted in the same paper puts the ubiquity more strongly still: Johnston “found that ‘almost without exception examiners commented on the writing and presentation’.” Hansford and Maxwell recorded the specific frequencies — “typographical and spelling errors to be noted by examiners in 42 per cent of theses, both writing style and format and presentation each to be noted in 22 per cent, grammatical and spelling errors in 17 per cent”.
A necessary caveat about where this data comes from
The three content-analysis studies above — Holbrook and colleagues, Mullins and Kiley, and Dally and colleagues — are all Australian. We searched for an openly available UK equivalent and did not find one. Several relevant UK and international papers are behind paywalls we could not open, and we have not quoted anything we could not read.
So treat the percentages as evidence about doctoral examination in a closely comparable system, not as UK figures. What is UK evidence, and appears below, is the published criteria that UK examiners are formally asked to apply — from Newcastle, Warwick and Reading. Those are directly authoritative for a UK candidate; the Australian percentages tell you how examiners behave in practice, which no UK institution publishes.
Why a typo costs more than a typo
The most consequential finding in this literature is not about proportions. It is about inference.
Mullins and Kiley report that “One of the most common descriptors of a poor thesis, across all disciplines, was ‘sloppiness’”, and then explain why examiners care so much: “The concern with sloppiness was that examiners considered it was an indicator that the research itself might not be rigorous and the results and conclusions could not be trusted.” One examiner put the mechanism plainly: “If there are mistakes in calculations then that starts one thinking. If you find something suspicious in a thesis then you start to read it differently.”
Another described the effect on their own reading, and this is worth reading twice because it is an examiner describing a bias they know they have:
“I give my students strong advice on how not to ‘flip’ an examiner from ‘reasonable’ to ‘unreasonable’ by having irritating things in the thesis such as typos and other careless textual mistakes that indicate lack of attention to detail. Once flipped (and I am aware of this happening), I am irritated and I have to work very hard at overcoming this irritation and not letting it influence my view of the thesis, although this is not easy.”
That is the real return on a proofreading pass. Not the marks attached to the errors, but the reading posture the errors produce.
Overclaiming: the criticism that recurs across every study
Where examiners do engage substantively with your prose, one fault comes up repeatedly. Mullins and Kiley, summarising Hansford and Maxwell, report that examiners focused on “format and presentation, the literature review, the theoretical/philosophical framework of the thesis and the problem of unsubstantiated and/or over-generalised conclusions.”
UK examiner guidance encodes the same concern as a question examiners are told to ask. Reading’s good practice guide instructs examiners to consider whether “the discussion illuminate[s] the results, and do the stated conclusions follow from the results”. That is an overclaiming test, written into a UK examination document.
The practical implication is a sentence-level discipline: every claim in your discussion should be traceable to a specific result, and any claim that is not should either acquire the evidence or lose the force. How to write limitations so they strengthen rather than weaken the case is covered in our guide to the discussion chapter, which sets out the three-part structure a limitation needs.
Reading also reproduces a list of the characteristics of a poor thesis, drawn from Rowena Murray, whose final item is the most quietly devastating in the whole literature: “not being able to explain at the end of the thesis what has actually been argued in it.”
What UK examiners are formally asked to apply
This is the UK half of the evidence, and it is published openly by the institutions concerned.
Newcastle, in its handbook for examiners of research degrees, sets four criteria. Alongside authentic, scholarly and professional sits: “Well-structured, written, and presented — The thesis should demonstrate skill in writing and presenting research similar to scholarly work in their field. It should be clearly structured and orderly in arrangement, and well-written and presented.” Note the benchmark — scholarly work in your field, not general good English.
Warwick states the requirement more tersely, and attaches a power to it: “The literary form must be satisfactory and the thesis should not be of unnecessary length. Examiners may require a candidate to confirm the length of the thesis.”
Reading names four attributes: “contribution to knowledge / originality and creativity / integration and coherence / presentation and clarity”. Two of those four are properties of the writing rather than of the research.
Set against the Australian percentages, this is the tension a candidate has to hold. Formally, presentation is one of four headline attributes. Behaviourally, it takes about an eighth of the report. Both are true, and the resolution is that clear writing is mostly invisible when it works — it earns no commentary and simply lets the analysis be assessed.
How the report is produced, and what its length tells you
Three procedural facts from UK guidance that most candidates never learn.
- The reports are independent. Warwick: “The examiners should prepare independent reports, without consultation, before any oral examination or conferral with the other examiner has taken place.” Your internal and external examiner form their views separately, which is why they sometimes differ in the viva.
- They are short. Reading: examiners “produce an independent report (typically around 1-2 pages)”, having been advised to “set aside at least two days to read the thesis and draft the pre-viva report.”
- Length is itself a signal. Newcastle tells its examiners that if the criteria “have clearly been fully met, the report will normally be very brief. On the other hand, if Examiners have serious doubts or concerns about whether criteria have been met, a fuller account would be expected.” A long pre-viva report is not a compliment.
Who these people are and how they are appointed is covered in who examines a PhD thesis in the UK; what outcomes they may recommend, and on what evidence, in PhD viva outcomes in the UK.
The criteria you are given may not be the criteria applied
One finding deserves separating out, because it changes how you should read any assessment guidance including your own institution’s. Mullins and Kiley found that examiners “had established their own criteria, and noted but did not use guidelines provided.” Dally and colleagues restate it: “even when examiners are provided with specific assessment criteria, they do not necessarily follow these when making judgements about thesis quality.”
Dally and colleagues also record the substantive standard examiners tend to apply instead: “the thesis (or parts thereof) had to be suitable for publication and the literary presentation was expected to be clear, discernible, coherent, accurately and cogently written, concise and authoritative.”
And the level of that standard is lower than doctoral anxiety assumes. The title of the Mullins and Kiley study is a direct quotation from an examiner: “A PhD is a stepping stone into a research career. All you need to do is to demonstrate your capacity for independent, critical thinking. That’s all you need to do. A PhD is three years of solid work, not a Nobel Prize.”
What to do with this before you submit
- Proofread as a rigour signal, not as tidiness. The sloppiness finding means mechanical errors are read as evidence about your analysis. This is the cheapest defence available.
- Audit every conclusion against a specific result. Reading’s test — do the stated conclusions follow from the results — is the one you can run on yourself.
- Make it readable in two days. That is the reading time Reading advises, and navigability is what makes it achievable.
- Be able to state what the thesis argues, in one sentence. The inability to do so is the last item on the poor-thesis list.
If corrections do follow, the procedure for working through an examiners’ report is set out in our guide to completing your corrections and writing the response letter, and the administrative sequence before all this begins is in submitting your thesis for examination.
Getting a thesis into that condition
Everything above points the same way: examiners spend their attention on analysis, and mechanical failures degrade their confidence in it. The clerical layer of a thesis is therefore worth automating precisely because it is not where the judgement is.
Draft your thesis in Tesify: keep the whole document in one workspace so the argument can be checked end to end, and let the bibliography and citation formatting look after themselves rather than becoming the errors an examiner reads as sloppiness. The scholarship stays 100% yours — the analysis is the 43 per cent, and it has to be. There is a free tier.
Frequently asked questions
What do examiners write most about in a thesis report?
Analysis and reporting. In a study of examiner reports for 101 doctoral candidates, comments on analysis and reporting appeared in 89 per cent of reports and took 43.1 per cent of the average report’s text. No other category comes close. Note that this is Australian data — no equivalent open UK analysis exists.
How much of an examiner report is about writing quality?
Roughly an eighth. The same study found examiners “devoted 5 per cent of text units or lines to substantial issues of communicative competence… but 8 per cent to editorial matters”. Writing attracts wide but shallow comment — many examiners mention it, briefly.
Do examiners comment on writing at all if the thesis is good?
Almost always. That study found “73 per cent of examiners mentioned substantial issues of communicative competence at least once, while 50 per cent identified specific corrections”. An earlier study found that “almost without exception examiners commented on the writing and presentation”.
Why do typos matter so much more than they should?
Because examiners read them as evidence about the research. Mullins and Kiley found that sloppiness worried examiners because they “considered it was an indicator that the research itself might not be rigorous and the results and conclusions could not be trusted.” One examiner: “If you find something suspicious in a thesis then you start to read it differently.”
What is the overclaiming test an examiner applies?
Reading’s guidance for examiners phrases it as a direct question: “does the discussion illuminate the results, and do the stated conclusions follow from the results”. If a conclusion cannot be traced to a specific result, it will not survive that question.
What standard does a UK thesis actually have to meet on writing?
Newcastle publishes it as one of four criteria: a thesis should be “Well-structured, written, and presented”, demonstrating “skill in writing and presenting research similar to scholarly work in their field” and being “clearly structured and orderly in arrangement, and well-written and presented.”
Is there a rule about thesis length in the examination criteria?
Warwick includes one, and it is bluntly phrased: “The literary form must be satisfactory and the thesis should not be of unnecessary length. Examiners may require a candidate to confirm the length of the thesis.”
How long is an examiner report?
Short, if all is well. Reading says examiners “produce an independent report (typically around 1-2 pages)”. Newcastle makes the length itself informative: if the criteria are clearly met “the report will normally be very brief”, whereas serious doubts call for “a fuller account”.
Do the two examiners confer before writing?
No, and this is a formal requirement rather than a courtesy. Warwick states that “The examiners should prepare independent reports, without consultation, before any oral examination or conferral with the other examiner has taken place.”
Do examiners follow the criteria they are given?
Not reliably. Mullins and Kiley found that examiners “had established their own criteria, and noted but did not use guidelines provided.” A later review restates it: examiners “do not necessarily follow these when making judgements about thesis quality.”
How long does an examiner spend on the thesis?
Reading advises its examiners to “set aside at least two days to read the thesis and draft the pre-viva report.” Two days for eighty thousand words is a fast read, which is why navigability and clarity carry weight out of proportion to their share of the report.
What is the standard being applied overall?
Competence rather than brilliance. The Mullins and Kiley title quotes an examiner directly: “All you need to do is to demonstrate your capacity for independent, critical thinking. That’s all you need to do. A PhD is three years of solid work, not a Nobel Prize.”
