---
name: doctoral-research-positioning-and-originality
description: >
  Establishes what is genuinely original about a doctoral project, tests that
  claim against the literature designed to disprove it, and articulates it as a
  contribution statement that survives examination. Use for "what is original
  about my PhD", "has this been done before", "how do I state my contribution",
  "my contribution has shrunk", "the examiner will ask what is new here",
  "I cannot find a gap", "is this original enough for a doctorate",
  "positioning my research", "how do I know my study is a contribution".
category: 15 Academic University Research
ref: "15.13"
tier: 3
inherits: [K2, K3, K4, K5]
---

# Doctoral Research Positioning and Originality

## 1. One-line description

A method for identifying, testing and stating the original contribution to knowledge a doctoral project makes, in a form specific enough to be defended against an expert examiner who will actively try to show it has been made already.

## 2. What this skill is used for

**The research problem it solves.** Doctoral candidates fail at originality in two opposite ways and both are avoidable. The first is paralysis: a belief that originality means an idea nobody has ever had, which is a standard almost no completed doctorate meets and which stops good projects before they start. The second is over-claim: a contribution asserted in language so broad ("this study contributes to our understanding of leadership") that it names no knowledge and can be refuted by pointing at any of forty prior studies. Between them sits a third failure that appears late and does the most damage: a project that has produced a great deal of work and cannot say what is known now that was not known before. This skill separates the several legitimate forms originality takes, builds a search designed to disprove novelty rather than confirm it, and forces the contribution into a single sentence that names the knowledge added, the population or setting it holds in, and the evidence that establishes it.

**Where it sits in the research lifecycle.** At the start of a doctorate, where it converts an area of interest into a defensible position, and again at least twice: after the literature is properly mapped, and in the final year when the contribution as delivered must be re-articulated against the contribution as proposed.

**Typical use cases.**
- Converting a topic into a positioned contribution before a proposal or upgrade viva.
- Testing a claimed gap that may be an absence of interest rather than an absence of knowledge.
- Re-stating a contribution that has shrunk, changed shape, or moved during fieldwork.
- Preparing the answer to the examiner's question about what is actually new.
- Deciding whether a replication, a method transfer or a new-context study is enough for a doctorate in this discipline.
- Distinguishing the contribution from the quantity of work done, when a candidate is exhausted and conflating the two.

**Who uses it.** Doctoral candidates in any discipline and any paradigm, at proposal, mid-candidature and pre-submission; supervisors testing a candidate's positioning before it meets an examiner; upgrade and confirmation panels.

## 3. When to use it

- You can describe your topic but cannot say in one sentence what will be known because of your work.
- A supervisor, panel or reviewer has asked "what is the contribution?" and the answer felt thin when you gave it.
- You have found a gap in the literature and need to establish whether it is a real gap or a well-founded silence.
- Your project has changed since the proposal and the original contribution statement no longer describes it.
- You are writing an upgrade, confirmation or annual review document that must state the contribution.
- You are approaching submission and need the contribution claim to be exactly as large as the evidence supports, and no larger.
- Someone has published something adjacent to your work and you need to establish whether you have been scooped, partly scooped, or not scooped at all.
- You are preparing for an examination in which the first substantive question will be about novelty.

## 4. When NOT to use it

- **The work is at masters level.** A masters dissertation is normally assessed on competent independent research, not on an original contribution to knowledge. Applying a doctoral originality test to it manufactures anxiety and can push a student into an unfeasible design. Use **15.11 Discussion and Contribution Development**, which states contribution at the level a masters actually requires.
- **You need a literature review, not a positioning argument.** If the task is to find, appraise and synthesise what exists, that is **15.06 Systematic Literature Review** or **15.14 Advanced Systematic Review and Meta-Analysis**. This skill consumes a literature map; it does not produce one. Positioning built on an unsystematic reading of the field is the single most common way a candidate discovers at viva that the contribution was made in 2014.
- **The theory itself is the object of work.** Where the task is to define constructs, specify mechanisms and set boundary conditions, that is **15.15 Theory Building and Conceptual Contribution**. This skill establishes that a theoretical contribution would be original; it does not build the theory.
- **The claim is empirical and the data does not yet exist.** A contribution cannot be finalised before the evidence that carries it. Draft it as a claim to be tested, mark it provisional, and revisit. Stating a contribution in the past tense before fieldwork is complete produces a thesis written to defend a prediction rather than to report a result (K4 §3.4).
- **You want a search for confirming evidence.** If what is wanted is a list of citations supporting the novelty claim, this skill will not provide it, because a search designed to confirm novelty is worthless as evidence of novelty. The method here is adversarial by construction.
- **The discipline's originality conventions are unknown and unobtainable.** Fields differ profoundly: what counts as an original contribution in analytic philosophy, in clinical trials, in ethnography and in computational modelling has very little in common. If the candidate cannot supply examples of accepted theses or the institution's own criteria, do not assume a standard. Ask, and if no answer is available, produce a positioning that is explicit about which convention it assumes. `RESEARCHER DECISION REQUIRED` (K5 §2.7).
- **The real problem is thesis structure or examination strategy, not novelty.** Holding the argument together across chapters is **15.18 Thesis Architecture and Chapter Coherence**; hardening the thesis against attack is **15.19 Examiner-Proofing and Defence Strategy**.
- **Academic integrity.** Per §12.1: this skill does not draft a contribution statement, an introduction or a positioning passage for submission as the candidate's own unaided writing. It interrogates a claim the candidate makes and returns the weaknesses in it. Where an institution prohibits AI assistance for a task, this skill must not be used for it.

## 5. Required inputs

**Required.**
- **A statement of what the candidate believes the contribution is, in their own words.** However rough. The method operates on a candidate's claim; it does not generate one. If nothing can be stated, the prior task is topic work, not positioning.
- **The discipline and the degree's stated criteria for originality**, or an honest statement that these are unknown. Institutional regulations usually contain a sentence defining the doctoral standard, and it is the sentence the examiner is formally applying.
- **Access to the relevant literature the candidate has actually read**, with the ability to open sources. A novelty claim tested against remembered literature is untested (K4 §2.4).

**Optional, and what each one adds.**
- **Examples of recently accepted theses in the same department.** The most reliable available calibration of what this discipline in this institution accepts as sufficient. Nothing else substitutes.
- **The proposal or upgrade document.** Lets the current contribution be compared against the promised one, which is where honest re-articulation starts.
- **The data collected so far, or a description of it.** Converts the contribution claim from aspiration into something testable against evidence.
- **Names of likely examiners and their published positions.** Turns generic defence into specific defence, and reveals whether the contribution contradicts an examiner's own published work, which changes how it must be argued.
- **The intended thesis format (monograph or publication-based).** Publication-based theses distribute the contribution across papers and need an overarching contribution that is more than the sum of the papers.
- **A supervisor's written feedback on the contribution.** Identifies where the supervisor already doubts the claim, which is where an examiner will press.

## 6. Questions to ask before starting

1. **What will be known after this thesis that was not known before it?** The whole skill reduces to this question. If the answer describes activity ("a survey of 400 nurses was conducted") rather than knowledge, the contribution has not yet been identified. Default if unanswered: treat the project as pre-positioning and work backwards from the research questions.
2. **Which form of originality is being claimed?** New evidence, new method, new theory, new context, new synthesis, a replication that matters, or a method transferred across fields. Each carries a different evidentiary burden and a different attack. Default: identify the form from the design rather than letting the candidate choose the most flattering one.
3. **What is the discipline's convention for sufficiency?** In some fields a single well-designed empirical study in a new context is a doctorate; in others it is a chapter. Default: state the assumed convention explicitly in the output and flag it as unverified.
4. **Who else is working on this right now?** Not only what is published. Conference programmes, preprints, registered protocols and doctoral registers reveal work in progress that will be published before submission. Default: assume at least one other group is close, and design the contribution so that it survives their publication.
5. **What would have to be true for this contribution to be worth nothing?** Forces the candidate to name the disconfirming condition rather than defend a position. Default: construct the strongest case against the claim and present it.
6. **Has the project changed since the contribution was last stated?** Scope reductions, access failures and null results all move the contribution. Default: assume it has moved and re-derive it from what actually exists.
7. **Is the contribution claim sized to the evidence?** A single qualitative study in one organisation does not establish a general mechanism. Default: cap the claim at what the design licenses (K3 §3.6).

## 7. Step-by-step methodology

**Step 1. Separate the topic, the questions, the work and the contribution.**
Write four short statements and put them side by side: the topic (the area), the research questions (what is asked), the work (what will be or was done), and the contribution (what is known afterwards). Candidates routinely have three of these and believe they have four. The commonest substitution is the work standing in for the contribution: "I interviewed 32 clinicians across four hospitals" is effort, not knowledge. The second commonest is the topic standing in for the contribution: "this thesis contributes to the literature on clinician burnout" names a shelf, not an addition to it. *Correct result: four distinct statements, with the contribution one containing a claim that could be true or false.*

**Step 2. Classify the form of originality, and accept that most doctorates take a modest one.**
Place the claim in one of these, or in a named combination:

| Form | What is new | Principal attack |
|---|---|---|
| New evidence | An empirical result not previously established | It has been established, or the design cannot establish it |
| New method or instrument | A way of measuring, sampling, analysing or observing | It is a variant of an existing method, not a new one |
| New theory | Constructs, relationships, mechanism or boundary | It is existing theory relabelled (see 15.15) |
| New context for established theory | Whether and how a theory holds in a setting it has not been tested in | The transfer is trivial: nobody expected it to differ |
| New synthesis | An integration that reorganises what is known | It is a literature review, not a contribution |
| Replication that matters | A result that needed independent verification and did not have it | The original was not important or contested enough to be worth repeating |
| Method transfer across fields | An established technique applied where it has not been used | The application is mechanical and produces nothing the field did not already believe |

Naming the form is not bureaucratic. It determines what evidence must be assembled in Step 4, and it removes the paralysis caused by comparing a normal doctorate against the standard of a paradigm shift. A new context study is a legitimate doctorate in most fields provided the transfer is non-obvious and the reason to expect difference is theorised in advance. *Correct result: a named form, a named principal attack, and a one-line statement of why this form is sufficient in this discipline.*

**Step 3. Convert the gap claim into a testable proposition, then look for the reason the gap exists.**
Rewrite "there is a gap in the literature on X" as "no study has established Y in population Z using a design capable of showing it". That version can be falsified by a single citation, which is the point. Then run the question every good examiner runs: why has nobody done this? There are four honest answers, and only one of them supports a doctorate. It has been done and the candidate has not found it. It cannot be done, for reasons of access, measurement or ethics that the candidate has not yet met. It is not worth doing, because the answer is either known by implication or of no consequence. Or it genuinely has not been done and is worth doing, usually because a new condition, technology, dataset, population or theoretical development has made it answerable or interesting only recently. State which of the four applies and why. A gap with no account of why it persisted is an unexamined gap. *Correct result: a falsifiable novelty proposition and an explicit answer to why the gap exists.*

**Step 4. Run the disconfirmation search, designed to destroy the claim.**
This is the step candidates skip and examiners do not. Build a search whose success condition is finding the study that makes the thesis unnecessary. Practically: search the vocabulary of every community that could have done this work, including the ones the candidate does not read, because the same construct is named differently across fields and cross-disciplinary duplication is the most frequent source of the unpleasant surprise at viva. Search non-English literature where the phenomenon is not language-bound. Search grey and unpublished sources: dissertations and theses in national and institutional repositories, preprints, conference proceedings, registered protocols and trial registries, working paper series. Search backwards through the reference lists of the closest studies and forwards through their citations, which is how the near-miss study is usually found. Search by method and by population as well as by topic, since a study that answers your question may be filed under a topic you would never search. Log every search, string and date, so that the absence claim is defensible rather than asserted (K2 §4.3). *Correct result: a search log, a shortlist of the closest existing studies, and for each the specific reason it does not make the contribution redundant.*

**Step 5. Build the nearest-neighbour table.**
Take the five to ten closest studies and tabulate them against your own on the dimensions that matter: research question, population, setting, theoretical frame, design, measures, analysis, and what each concluded. Then write, for each, one sentence on what your study does that it did not. If any cell reads "same" across every dimension, the contribution is a replication and must be argued as one, which is legitimate but a different argument. If the only difference is the country or the sector, the contribution is a context claim and must carry a theorised reason to expect a different answer, otherwise it is an unmotivated repetition. This table is the single most useful artefact this skill produces: it is the evidence base for the novelty claim, the answer to the examiner's "how does this differ from", and a section of the literature chapter. *Correct result: a completed comparison table where every row has a stated point of difference and none of the differences is purely geographic without a reason.*

**Step 6. Write the contribution in one sentence, then attack the sentence.**
The sentence must contain: what is now known, the population or setting in which it holds, the evidence that establishes it, and the theoretical or practical consequence. Vague verbs are where contributions hide, so ban "explores", "investigates", "sheds light on", "contributes to our understanding of", and "highlights the importance of". Each of them describes an activity or an attitude, not a piece of knowledge. Then attack the sentence four ways. Is every noun in it defined? Could a reasonable expert say "we knew that"? Does the evidence named actually establish the claim, or merely be consistent with it? Would the sentence still be true if the study's main result had come out the other way, which would mean the contribution is the doing rather than the finding? Rewrite until it survives. *Correct result: one sentence, under forty words, that a hostile expert could disagree with but not dismiss as empty.*

**Step 7. Position within a conversation, not against a void.**
Identify the specific scholarly conversation the thesis joins: the two to four positions currently held, who holds them, what is at issue between them, and where your work enters. Then write the entry sentence: this thesis speaks to the disagreement between A and B about C, and its contribution is D. Positioning against a void ("little research has examined") is weak because it invites the examiner to fill the void with a study you did not cite, and because it locates the work nowhere. Positioning inside a live disagreement is strong because the contribution's value is visible without argument. Where the field has no live disagreement on the point, say so, and position against the assumption everyone is making instead. *Correct result: a named conversation, its participants, the point at issue, and your entry.*

**Step 8. Size the claim to the evidence, and write the version you can defend.**
Take the contribution sentence and ask what design would be needed to support it. Then compare that with the design you have. Almost always the claim is a size or two too large: a mechanism claimed from an associational design, a general statement from a single-site study, a causal verb on cross-sectional data (K4 §3.2). Rewrite at the size the evidence carries, and note separately what the larger claim would need, because that becomes the future-research section and it is more valuable than the generic version. A smaller claim defended completely beats a larger one defended partially, and examiners reward the first. *Correct result: a contribution statement whose scope matches the design, with the over-claimed version recorded alongside as what remains to be established.*

**Step 9. Handle the shrinking contribution honestly.**
Contributions shrink. Access is refused, a sample is smaller than promised, an instrument fails, a result is null, a competitor publishes half of it. The wrong response, and the common one, is to defend the original size with weakening evidence, which produces exactly the thesis an examiner takes apart. The right response is to re-derive the contribution from what actually exists, using Steps 1 to 8 again, and then to say plainly in the thesis what changed and why. Two things are worth knowing here. A null or negative result is a contribution when the question mattered and the design was capable of detecting the effect; state it as a finding about the world, not an apology about the study. And a shrunk contribution honestly stated is examinable, while an intact contribution dishonestly maintained is not. `RESEARCHER DECISION REQUIRED` where the re-derived contribution may fall below the doctoral threshold, since that judgement belongs to the supervisor and the institution, not to this analysis (K5 §2.7). *Correct result: a current contribution statement, a record of how it differs from the proposed one, and a short honest account of why.*

**Step 10. Rehearse the examiner's three questions.**
Write the answers out. What is new here? Who says it is not, and why are they wrong? What would you say to someone who claims this was settled by the closest prior study? Each answer runs under ninety seconds spoken, names specific literature, and concedes what is genuinely conceded. An answer that concedes nothing reads as a candidate who has not understood the field. Then hand the contribution statement to someone who has not read the thesis and ask them to state what is new: if they cannot, the sentence has failed regardless of how it reads to its author. *Correct result: three written answers, each naming specific work, each with a real concession in it.*

## 8. Analytical framework

The positioning argument is a chain, and each link is separately attackable:

    Field conversation → Specific unresolved point → Why it is unresolved →
    What this thesis establishes → The evidence that establishes it →
    The scope within which it holds → What follows for the conversation

Applying it: work the chain in both directions. Forwards it produces the introduction and the contribution statement. Backwards, from the evidence you actually have to what it can establish to which point in the conversation it settles, it produces the honest version, which is usually smaller and always more defensible. Where the two directions disagree, the backwards pass wins.

A second structure sits underneath, the originality test itself:

    Is the claim specific enough to be false?
      → Has a designed disconfirmation search failed to falsify it?
        → Do the nearest neighbours differ on a stated dimension?
          → Does the design support the claim at the size stated?
            → Would an expert in the field call this an addition to knowledge?

A "no" at any level sends the claim back for rewriting, not the project back for redesign. Most originality problems are statement problems, not research problems, and the diagnostic value of this ladder is that it says which one you have.

## 9. Output format

**1. Contribution statement.** One sentence, plus a paragraph of expansion, plus the named form of originality.

**2. The four statements.** Topic, research questions, work, contribution, side by side.

**3. Novelty proposition and gap account.** The falsifiable version of the gap claim, and which of the four reasons explains why the gap exists.

**4. Disconfirmation search log.** Sources searched, strings, dates, coverage limitations, and what was found. Including the searches that returned nothing.

**5. Nearest-neighbour table.**

| Study | Question | Population and setting | Theory | Design | Measures | Conclusion | Point of difference from this thesis |
|---|---|---|---|---|---|---|---|

**6. Positioning statement.** The conversation, its positions, and where this thesis enters.

**7. Claim sizing.** The claim as stated, the design that would support it, the design in hand, and the resized claim. With the residual claim recorded as future work.

**8. Contribution movement log.** Where the contribution now differs from the proposal, and why.

**9. Anticipated challenges.** The three examiner questions with drafted answers, and the concessions to be made.

**When the evidence is thin, the format must not force fabrication (K4 §1).** If the disconfirmation search finds a study that makes the contribution redundant, that is the output, and it is a good outcome at month twelve and a survivable one at month thirty-six. The nearest-neighbour table is never padded with studies that were not read, and no row is completed from a title and an abstract alone. If the contribution cannot be stated in one sentence, the correct output says so and names what is missing, rather than producing an elegant sentence that means nothing.

## 10. Quality checks

Run before any positioning claim is presented or submitted. These sit on top of K4 §8.

1. Does the contribution sentence name a piece of knowledge rather than an activity, an intention or a topic?
2. Could the novelty claim be falsified by a single citation, and has a search actively tried to find that citation?
3. Was the search built to disconfirm, and does the log include the vocabularies of adjacent fields?
4. Were theses, preprints, conference material and registered protocols searched, not only published journal articles?
5. Has every nearest-neighbour study been read in full, rather than screened from its abstract?
6. Does every point of difference in the comparison table state a substantive difference rather than a geographic or sectoral one alone?
7. Is there a stated reason why the gap persisted, and is it the one that supports doing the work?
8. Does the design named actually support the claim at the size stated, with no causal language on a non-causal design?
9. Is the contribution positioned inside a named conversation with named participants?
10. If the study's main result had gone the other way, would the contribution still stand? If yes, is the contribution the doing rather than the finding, and is that intended?
11. Has the contribution been re-derived from what exists rather than carried forward from the proposal?
12. Would a competent expert outside the immediate sub-field understand what is new from the statement alone?

## 11. Common failure modes

| Failure | How to recognise it | How to prevent it |
|---|---|---|
| **Effort mistaken for contribution** | The claim describes fieldwork volume, years spent or sample size | Separate the four statements (Step 1) |
| **Topic as contribution** | "Contributes to the literature on X" | Ban the phrase; require a knowledge claim (Step 6) |
| **The unexamined gap** | A gap asserted with no account of why it exists | Answer the four reasons question (Step 3) |
| **Confirmation search** | Every source found supports the novelty claim | Build the search to falsify (Step 4) |
| **Single-vocabulary search** | The whole reference list comes from one field's journals | Search adjacent communities' terminology (Step 4) |
| **Geographic originality** | The only difference from prior work is the country | Require a theorised reason to expect difference (Step 5) |
| **Over-claimed scope** | A general mechanism claimed from one site or one wave | Size the claim to the design (Step 8, K4 §3.3) |
| **Defending the dead contribution** | The proposal's claim maintained after the project changed | Re-derive from what exists (Step 9) |
| **The invisible concession** | Answers to challenge that concede nothing | Rehearse with a real concession in each (Step 10) |
| **Scoop panic** | A superficially similar publication treated as fatal | Complete the nearest-neighbour row before concluding |
| **AI-generated adjacent citations** | Plausible-looking references to studies that fit the argument suspiciously well | Nothing enters the table unopened. See §12 |
| **Contribution inflation in the abstract** | The abstract claims more than Chapter 7 delivers | Check the abstract against the resized claim, not the proposal |

## 12. AI guardrails

Skill-specific. The universal prohibitions in K4 apply in full and are not repeated.

1. **Academic integrity, and it binds this whole skill.** These skills assist a researcher's thinking, structure and rigour. They do not produce work to be submitted as the student's own unaided output. The user must comply with their institution's AI use policy and its declaration requirements, which vary by institution and by assessment. Where an institution prohibits AI assistance for a task, this skill must not be used for it. The skill never writes a passage for submission as though the student wrote it; it interrogates, structures, critiques and teaches. Operationally here: the contribution sentence is drafted by the candidate and stress-tested by the skill. Where a rewrite is offered to demonstrate a principle, it is labelled as a demonstration of the principle and the candidate writes their own. A candidate who cannot produce the contribution sentence themselves cannot defend it in a viva, and the viva is the point at which this becomes visible.

2. **Never assert that something is original.** Novelty is an absence claim, and an absence claim can only be supported by a documented search. The most this skill may say is that the searches recorded in the log did not find prior work establishing the claim, naming what was searched and what was not. Model knowledge of a literature is not a search, and it is systematically weakest exactly where it matters: recent work, non-English work, theses, and small-field literatures.

3. **Never produce a "closest prior work" list from model knowledge.** Every entry in the nearest-neighbour table is a source that has been opened and read. A fabricated near-neighbour is worse than a missing one, because the candidate will build the difference argument against a study that does not exist and be dismantled for it.

4. **Never manufacture a gap.** If the searches do not support a gap, say the gap was not established. Do not construct a narrower and narrower framing until an absence appears, since a gap produced by adding qualifiers ("in this sector, in this age band, during this period") is usually an artefact of the framing and an examiner will say so.

5. **Never talk a candidate out of a shrunk contribution or into an inflated one.** Both are forms of telling the candidate what they want to hear. Report the size the evidence supports and let the supervisor and the institution judge sufficiency (K5 §2.7).

6. **Never state what a discipline requires for a doctorate as though it were universal.** Standards vary by field, country and institution. Where a convention is assumed, label it as an assumption and name what would confirm it.

7. **Never resolve a scoop question without reading the competing work.** "Someone has published something similar" is not a finding until the two studies have been compared on the dimensions in Step 5. Both premature panic and premature dismissal are failures here.

8. **Cap confidence in any novelty claim at moderate unless the search was systematic and reported** (K3 §3.4). A purposive search supports "no prior work was identified", not "this is original".

## 13. Best-practice principles

- **Originality is a property of the claim, not of the effort.** Two candidates can do identical work and one has a contribution and the other has a report, because one has identified what is now known and the other has described what was done.
- **The disconfirmation search is the cheapest insurance in a doctorate.** A week spent trying to destroy your own novelty claim in year one is worth more than any other week in the degree, and it is the week that prevents the worst outcome in the field, which is discovering the duplicate at the viva.
- **Most doctorates are contributions of the modest kinds, and that is the standard, not a compromise.** New evidence in a defensible design, or an established theory tested where it had not been tested with a good reason to expect difference, is what the majority of accepted theses actually contain.
- **A contribution positioned inside a disagreement needs no rhetoric.** If two camps disagree and your work bears on the disagreement, the value is self-evident. If your work sits in an empty field, you have to argue that the field should not be empty, which is a harder argument.
- **The word "gap" is doing too much work in most theses.** Gaps are cheap to assert and expensive to defend. Ask why it is empty before claiming the right to fill it.
- **Say what you are not claiming.** Explicitly bounding the contribution is read by examiners as command of the material, not as weakness, and it forecloses the attack that you have over-reached.
- **A null result is a contribution when the question mattered and the design could have found the effect.** The failure is not the result; it is presenting it apologetically, which invites the examiner to treat it as a failed study rather than a finding.
- **Watch for the contribution that is really your supervisor's.** Where a project sits inside a supervisor's programme, the candidate's own contribution must be separable and statable, because the viva asks what *you* contributed.
- **Re-state the contribution at least three times during the doctorate.** At proposal, after fieldwork, and at write-up. Contributions move, and the cost of noticing late is a rewritten introduction under deadline.
- **Volume is not a defence.** A three-hundred-page thesis with no statable contribution fails, and a lean one with a clear contribution passes. Examiners read for the claim, not for the weight.
- **The abstract is where over-claim is caught.** It is written last, read first, and compared by every examiner against the conclusions chapter. Make it match the resized claim.

## 14. Worked example

Generic fictional scenario, academic, education.

**INPUT**

A doctoral candidate in education, two years in, has interviewed 26 secondary school teachers about how they adapt a national numeracy curriculum in practice, and has classroom observation from six of them. Her proposal claimed the contribution as "a new understanding of teacher agency in curriculum implementation". A panel member has asked what is new. She cannot answer in under three minutes, and the answer she gives describes her fieldwork.

**PROCESS**

*Step 1.* The four statements are written out. Topic: teacher agency in curriculum implementation. Questions: how do teachers adapt a prescribed numeracy curriculum, and what shapes the adaptations? Work: 26 interviews, six observation sequences, thematic analysis. Contribution: currently a restatement of the topic. The gap between statement three and statement four is the whole problem.

*Step 2.* The candidate initially wants to claim new theory. Examined against her design, the claim available is a combination: new evidence, plus a new context for an established theory of professional agency that has been developed and tested almost entirely in a different national system with different accountability pressures. Naming the form immediately clarifies the burden: she must show the transfer is non-obvious and had a theorised reason to differ.

*Step 3.* "There is a gap in the literature on teacher agency in this system" becomes "no study has established how accountability pressure conditions the form teacher adaptation takes in a high-stakes, centrally prescribed numeracy curriculum". Why is it empty? Not because it cannot be done. The honest answer is that the accountability regime in question was introduced four years ago, and the theory predates it. That is a strong reason for a gap and it becomes the opening of her positioning.

*Step 4.* The disconfirmation search is deliberately hostile. It adds the vocabulary of two adjacent communities, curriculum studies and public administration, where the same behaviour is discussed as "street-level discretion". This turns up a body of work she had not read. It also searches national thesis repositories and a conference programme, which surfaces an unpublished thesis in a neighbouring system covering similar ground.

*The judgement call.* The unpublished thesis is close. The instinct is to conclude she has been scooped, and her second instinct, once she reads it, is to dismiss it because it is unpublished. Both are wrong. The nearest-neighbour row is completed properly: same theoretical frame, similar interview method, different accountability regime, no observational data, and a conclusion that adaptation is primarily individual. Her observational data shows adaptation being negotiated between teachers in departmental meetings, which the interview-only design could not have seen. That is the point of difference, and it is a real one. The scoop is not a scoop; it is her closest interlocutor, and citing it strengthens her position considerably.

*Steps 5 to 8.* The comparison table is built for eight studies. The contribution sentence goes through five drafts. The version claiming "teacher agency is collectively rather than individually constituted" is cut, because 26 teachers in one region cannot establish a general claim about how agency is constituted (K4 §3.3). The surviving version: "In a centrally prescribed, high-accountability numeracy curriculum, teachers' adaptations are negotiated collectively at departmental level rather than exercised individually, a pattern visible in observation but absent from interview accounts in this study, which suggests existing agency theory has under-specified the collective unit of adaptation because it was developed under lower-accountability conditions."

*Step 9.* Her contribution has shrunk from "new understanding of teacher agency" to a specified, bounded claim. It is smaller and it is defensible, and the interview-observation divergence is a methodological point worth more than the original vague claim.

**OUTPUT**

A contribution statement of thirty-eight words with its scope stated, a nearest-neighbour table of eight studies, a search log covering three disciplinary vocabularies and two repositories, a positioning statement that enters a live disagreement about whether agency is individual or collective, and three rehearsed answers. The panel's question is now answerable in fifty seconds.

`RESEARCHER SIGN-OFF REQUIRED` from the supervisor on whether the resized claim meets the doctoral threshold in this department (K5 §2.7).

## 15. Advanced usage

**Positioning against a known examiner.** Where examiners are known or likely, read their recent work and locate your contribution relative to their published positions. Three cases arise. Your work supports their position, in which case the risk is that they accept it too readily and the internal examiner does the challenging. Your work is neutral, which is the easiest case. Or your work contradicts them, which is entirely acceptable and requires that you have engaged their argument at its strongest, cited it accurately, and can concede its merits before disagreeing. The failure is not contradicting an examiner; it is contradicting them without having read them properly. See **15.19 Examiner-Proofing and Defence Strategy**.

**The publication-based thesis.** Where the thesis is a collection of papers, each paper carries a contribution and the thesis must carry one that is more than their sum. Run this method twice: once per paper, once for the overarching claim. The overarching claim is where these theses are most often weak, and it is what the linking chapters exist to establish. See **15.18 Thesis Architecture and Chapter Coherence**.

**Contribution portfolios.** Some theses contribute on more than one front, typically an empirical finding plus a methodological or instrument development. Keep them separate, size each independently, and rank them, because an examiner will ask which is the principal contribution and a candidate who answers "both equally" has usually not tested either.

**When the field is genuinely crowded.** In dense literatures the contribution is rarely a gap and is usually a resolution: reconciling two incompatible findings by identifying the moderator or the definitional difference that produced them. This is a strong doctoral contribution and it requires the disagreement to be diagnosed properly rather than averaged (see **10.01 Literature Review and Desk Research**, Step 9 for the diagnostic).

**When the field is nearly empty.** An empty field is more often a warning than an opportunity. Before claiming it, establish that the question is answerable and consequential, and expect to spend a chapter justifying that the silence is an oversight rather than a judgement.

## 16. Skill chain

**Recommended previous skills:**
- **15.06 Systematic Literature Review** or **15.14 Advanced Systematic Review and Meta-Analysis.** Hand over the mapped literature without which no novelty claim can be tested.
- **15.01 Academic Research Topic Selection.** Hands over a topic narrow enough to be positioned.
- **10.04 Research Gap Identification.** Hands over candidate gaps for adversarial testing here.

**Recommended next skills:**
- **15.15 Theory Building and Conceptual Contribution.** Takes a claimed theoretical contribution and builds it properly.
- **15.16 Doctoral Methodology Justification and Rigour.** Takes the sized contribution and designs the study capable of supporting it.
- **15.17 Longitudinal and Multi-Study Design.** Where the contribution needs more than one study.
- **15.18 Thesis Architecture and Chapter Coherence.** Takes the contribution statement and makes it the golden thread.
- **15.19 Examiner-Proofing and Defence Strategy.** Takes the contribution and the anticipated challenges into examination preparation.

**Runs well alongside:**
- **15.03 Referencing and Citation Management**, which verifies every source the novelty argument rests on.
- **13.04 Bias Detection**, which audits whether the search assembled itself around the candidate's prior.
- **15.24 Journal Article Development**, where the same contribution claim becomes a paper's positioning.

---
A Yazi Supplied Skill and resource.