The Ultimate Guide to Writing, Evaluating and Defending a PhD Thesis
A PhD thesis is more than a long academic document. It is a structured argument showing that you can identify an important problem, investigate it rigorously, interpret evidence critically and make an original contribution to knowledge. This guide explains the complete journey—from shaping the research problem to preparing for the viva—using practical examples and examiner-focused checks.
Executive Summary
A strong PhD thesis has seven connected qualities: a significant research problem, a defensible gap, focused research questions, an appropriate methodology, transparent analysis, an evidence-based contribution and whole-document coherence. Weak theses usually fail not because every chapter is poor, but because these elements do not align.
This Version 1 guide gives you a practical framework for planning, writing, evaluating and defending your thesis. It will be expanded over time with deeper discipline-specific examples and links to the complete Averon Research methodology, data analysis and viva resources.
How to Use This Guide
This pillar supports four different reading paths:
- Early-stage researchers: begin with the research problem, gap, questions and contribution.
- Mid-stage researchers: focus on literature, theory, methodology and alignment.
- Final-year researchers: prioritise evaluation, coherence, submission checks and viva preparation.
- Journal authors: use the research-logic, methodology and discussion sections to strengthen manuscripts.
The doctoral research journey
Key Takeaways
- A PhD thesis must make and defend an original contribution—not merely describe a topic.
- The research problem, gap, questions, methodology, findings and conclusion must form one logical chain.
- Each chapter has a distinct function, but examiners judge the thesis as a complete argument.
- Methodological justification matters as much as methodological description.
- The discussion chapter is where findings become a contribution to knowledge.
- Evaluation should begin before submission and include research logic, evidence, coherence and viva readiness.
Guide Contents
Use the five-part navigation below to move through this handbook without leaving the page.
Understanding and Planning a PhD Thesis
Doctoral standards, research logic, topic selection, problems, gaps, questions and contribution.
1. What Is a PhD Thesis?
A PhD thesis is a substantial, original and independently produced work of research submitted for the award of a doctoral degree. Its precise form differs by country, university and discipline. Some theses are monographs organised into conventional chapters. Others are thesis-by-publication documents built around several journal papers. Practice-based doctorates may include creative or professional outputs accompanied by a critical commentary.
Despite these differences, the central doctoral requirement remains similar: the candidate must demonstrate a contribution to knowledge supported by rigorous and transparent research.
Examiners are not simply asking whether the thesis contains useful information. They are asking whether the work demonstrates doctoral-level judgement: Can the candidate justify the problem, choices, interpretation and contribution?
How a PhD thesis differs from a master's dissertation
A master's dissertation normally demonstrates that a student can understand existing knowledge and apply an appropriate method to a defined problem. A PhD thesis goes further. It must show that the researcher can extend, challenge, refine or reframe existing knowledge.
| Dimension | Master's dissertation | PhD thesis |
|---|---|---|
| Purpose | Demonstrate advanced understanding and application | Make and defend an original contribution |
| Scope | Usually limited and tightly defined | Broader intellectual depth, even when the empirical study is focused |
| Independence | Guided research competence | Sustained independent scholarly judgement |
| Contribution | May apply existing knowledge in a new context | Must add meaningful knowledge or understanding |
| Defence | Sometimes assessed without oral defence | Commonly defended through a viva or oral examination |
2. What Do PhD Examiners Expect?
Examination criteria vary, but examiners commonly assess originality, rigour, critical engagement, methodological credibility, coherence and contribution. They also evaluate whether the candidate understands the limitations of the research and can defend the decisions made.
A useful way to understand examiner expectations is to imagine six questions running through the entire thesis:
- What important problem does this research address?
- What is not adequately known?
- Why is this study capable of addressing that gap?
- What did the research find?
- How does that finding change or deepen existing understanding?
- Can the candidate defend the reasoning and evidence?
Typical examination outcomes
Universities use different labels, but common outcomes include pass, pass with minor corrections, pass with major corrections, revise and resubmit, award of a lower degree, or fail. The difference between minor and major corrections usually depends on whether changes affect presentation or the intellectual substance of the thesis.
Read your university's formal doctoral examination regulations before finalising the thesis. Do not rely only on informal advice because word limits, formatting, declarations and examination outcomes differ significantly.
What Makes Research Doctoral?
Doctoral quality does not come from technical vocabulary, excessive length or the number of theories cited. Research becomes doctoral when it makes a meaningful contribution supported by rigorous, independent and critically reasoned scholarship.
Most candidates do not need to invent an entirely new theory. Originality often comes from explaining an existing phenomenon better, integrating theories usefully, producing important evidence or showing why accepted explanations fail in particular conditions.
| Test | Examiner question | Evidence required |
|---|---|---|
| Significance | Why was this problem worth investigating? | Problem, gap and rationale |
| Originality | What does the thesis add? | Contribution statement and discussion |
| Rigour | Why should the findings be trusted? | Design, sampling and analysis |
| Criticality | Were alternatives evaluated? | Literature review and discussion |
| Independence | Does the candidate demonstrate scholarly judgement? | Justified choices and limitations |
| Defensibility | Can the central decisions be defended? | Coherent argument and viva readiness |
3. Characteristics of an Outstanding PhD Thesis
An outstanding thesis is not necessarily the longest or most technically complicated. It is a thesis in which the central argument is visible, the research decisions are justified and the contribution is credible.
Originality
Originality can take several forms. A study may produce new empirical evidence, develop a new conceptual explanation, test a theory in a meaningful way, integrate previously disconnected literatures, introduce a methodological innovation or reveal an overlooked mechanism.
Rigour
Rigour means that the study was designed and conducted carefully enough for readers to trust the findings. In quantitative research, this may involve measurement quality, assumptions, statistical validity and robustness. In qualitative research, it may involve reflexivity, transparency, credibility and depth of interpretation.
Critical thinking
Criticality is not constant disagreement. It is the ability to examine assumptions, compare evidence, identify limitations, consider alternatives and reach proportionate conclusions.
Coherence
Coherence means every major part serves the same research purpose. A thesis can contain excellent individual chapters and still be weak if the chapters do not connect.
Defensibility
A defensible thesis does not claim perfection. It shows that choices were reasoned, limitations were understood and conclusions remain credible within appropriate boundaries.
The Research Logic Chain
The strongest theses make the logic of the study visible from beginning to end. Every element should lead naturally to the next.
A break anywhere in this chain creates risk. A question may ask “why,” while the methodology produces only descriptive statistics. A framework may appear in the literature chapter but play no role in analysis. Findings may show association while the conclusion claims causality.
Major corrections often arise because the research questions, methodology, findings and conclusions do not address the same intellectual problem.
4. Running Example: From Topic to Defensible Thesis
Throughout this guide, consider a fictional doctoral researcher, Sarah, who is studying how digital transformation is reshaping human resource management in public organisations.
Sarah's initial topic is:
Digital transformation and human resource management in the public sector.
This is a topic, not yet a research problem. After reviewing literature and public-sector reports, she identifies a more precise issue:
Public organisations increasingly adopt digital systems, but existing research does not adequately explain how these changes alter HR decision-making, employee discretion and institutional accountability simultaneously.
She then develops the following main research question:
How does digital transformation reshape human resource management practices and decision-making in public organisations?
Her contribution is not simply that she studies a new location. She integrates institutional theory, technology affordance theory and public value theory to explain how digital systems influence HR structures, practices and outcomes.
Students often describe the topic, context and participants but never convert them into a precise intellectual problem. Examiners then struggle to see what the thesis is actually trying to resolve.
5. Choosing and Refining a Research Topic
A suitable doctoral topic sits at the intersection of significance, feasibility, originality and researcher capability. A topic can be important but impossible to investigate within available time, data and access. It can also be feasible but too narrow to support a meaningful doctoral contribution.
Questions to test a potential topic
- Is the issue academically or practically significant?
- Is there a genuine unresolved question rather than merely limited local research?
- Can the required data be accessed ethically and realistically?
- Can the study be completed within the doctoral timeframe?
- Does the researcher have or can develop the necessary methodological capability?
- Is the topic focused enough to investigate deeply?
Example: too broad, too narrow and appropriately focused
Too broad: Artificial intelligence in healthcare.
Too narrow: Nurses' opinions about one software interface in one ward during one month.
Appropriately focused: How clinical decision-support systems influence nurses' professional judgement and escalation decisions in acute-care settings.
6. Writing the Research Problem
Before: broad topic
“Artificial intelligence is becoming important in recruitment, and more research is needed.”
After: defensible problem
“AI-enabled recruitment systems increasingly influence screening, yet research does not adequately explain how algorithmic recommendations affect transparency, accountability and human discretion in public-sector hiring.”
A strong research problem normally contains four elements:
- The current situation or established knowledge.
- The specific limitation, contradiction or unresolved issue.
- The consequence of that limitation.
- The need for the proposed research.
Weak example
Employee engagement is important, and more research is needed in developing countries.
Improved example
Although employee engagement is widely associated with organisational performance, existing models assume relatively stable employment relationships. In rapidly digitising public organisations, automated monitoring, hybrid work and algorithmic decision support may change how employees experience autonomy and trust. The absence of an integrated explanation limits both theory and managerial decision-making.
The improved version does not merely claim that the context is understudied. It shows why existing explanation may be inadequate.
7. Identifying a Defensible Research Gap
A research gap is not simply the absence of studies. It is an important limitation in existing knowledge that your study can meaningfully address.
Common types of research gaps
- Empirical gap: evidence is incomplete, contradictory or unavailable.
- Theoretical gap: existing theories do not adequately explain the phenomenon.
- Methodological gap: previous methods cannot capture an important dimension.
- Contextual gap: existing knowledge may not transfer to a materially different setting.
- Integration gap: related bodies of knowledge remain disconnected.
- Temporal gap: major technological or institutional change has made earlier evidence insufficient.
A geographical gap becomes doctoral only when the new context matters theoretically or empirically. “No study has been conducted in Country X” is usually weaker than explaining why Country X challenges assumptions embedded in existing research.
8. Developing Research Questions and Objectives
Weak research question
How does AI affect human resources?
Improved research question
How does the use of AI-enabled recruitment systems influence decision transparency and professional discretion in UK public-sector organisations?
Weak objective
To study digital transformation in HR.
Improved objective
To examine how AI-enabled recruitment tools redistribute decision authority between HR professionals, line managers and automated systems.
Research questions determine what evidence is required. Their wording should match the intended contribution and methodology.
Question types
- Descriptive: What patterns or characteristics exist?
- Exploratory: How do participants experience or understand a phenomenon?
- Explanatory: Why or through what mechanisms does something occur?
- Comparative: How and why do cases differ?
- Evaluative: How effective is an intervention or policy?
- Causal: What effect does one factor have on another under defined conditions?
Sarah's primary question is explanatory and exploratory. A cross-sectional survey measuring attitudes could describe associations, but it might not explain institutional mechanisms. She therefore selects a comparative qualitative case-study design supported by document analysis.
Alignment test
For every research question, complete this sentence:
This question will be answered using ______ data, collected through ______, analysed using ______, because ______.
If you cannot complete the sentence convincingly, the question and method may not align.
9. Defining the Original Contribution
“This is the first study in my country” is not automatically a strong contribution. The context must reveal something theoretically, empirically or practically important.
Your contribution should state what the thesis adds, to whom it matters and how it changes understanding. It should be more precise than “the study is novel.”
Four contribution categories
- Theoretical contribution: develops, extends, combines, qualifies or challenges theory.
- Empirical contribution: provides important new evidence about a phenomenon.
- Methodological contribution: introduces or validates a useful way of studying the issue.
- Practical or policy contribution: supports better decisions, interventions or institutional design.
Weak contribution statement
This is the first study of digital HR in these organisations.
Stronger contribution statement
The study develops an integrated explanation of how digital systems reshape HR practices through three interacting mechanisms: redistribution of decision authority, increased data visibility and redefinition of public-value accountability.
Writing the Core Thesis Chapters
Thesis structure, introduction, literature review and theoretical or conceptual frameworks.
10. How to Structure a PhD Thesis
There is no universal structure, but a conventional monograph thesis often includes the following chapters:
- Introduction
- Literature review
- Theoretical or conceptual framework
- Methodology
- Findings or results
- Discussion
- Conclusion
Some disciplines combine literature and theory, combine results and discussion, or include several empirical chapters. The correct structure is the one that communicates the argument clearly and complies with institutional conventions.
11. Writing the Introduction Chapter
The introduction gives the reader an intellectual map. It should explain the context, problem, gap, purpose, questions, significance, scope and structure of the thesis.
A practical introduction sequence
- Introduce the broad context without excessive history.
- Narrow to the specific research problem.
- Summarise the relevant knowledge limitation.
- State the aim, questions and objectives.
- Explain the study's significance and anticipated contribution.
- Define important boundaries and terms.
- Provide a concise chapter roadmap.
Some introductions contain extensive background but delay the research problem for many pages. Examiners should understand the central problem early.
12. Writing the Literature Review
What critical synthesis looks like
Descriptive version
Smith found that monitoring improved productivity. Khan found that employees disliked monitoring. Lee studied public organisations.
Critical synthesis
Although digital monitoring is associated with productivity gains, evidence remains inconsistent regarding autonomy and trust. Most studies use private-sector cross-sectional data, limiting understanding of how public accountability shapes outcomes.
- Listing: studies are mentioned separately.
- Grouping: studies are organised into themes.
- Comparison: agreements and contradictions are identified.
- Evaluation: assumptions, methods and limitations are examined.
- Synthesis: literature is used to build a new argument and justify the study.
“The candidate demonstrates familiarity with the literature, but the chapter remains descriptive and does not sufficiently establish the intellectual basis for the research questions.”
The literature review establishes what is known, how it is known, where disagreements exist and why your research is necessary. It should synthesise rather than catalogue sources.
Descriptive writing
Smith found X. Jones found Y. Ahmed studied Z.
Critical synthesis
Existing studies consistently associate digital monitoring with productivity, but their reliance on short-term private-sector surveys limits understanding of how monitoring affects discretion and accountability in public organisations.
The second example compares evidence, identifies a methodological limitation and connects it to the research problem.
Literature review checklist
- Are the most relevant debates identified?
- Are sources organised by ideas rather than authors?
- Are assumptions and methods critically evaluated?
- Are contradictions explained rather than ignored?
- Does the review lead logically to the gap?
- Is the theoretical framework grounded in the reviewed literature?
13. Theoretical and Conceptual Frameworks
A theoretical framework uses established theory to guide explanation. A conceptual framework may combine concepts, propositions and relationships developed for the specific study. In practice, terminology differs across disciplines, so clarity is more important than labels.
The framework should influence:
- how constructs are defined;
- which relationships are examined;
- what data is collected;
- how data is analysed;
- how findings are interpreted;
- where theoretical contribution is claimed.
Sarah's integrated framework is valuable only if the three theories are genuinely connected. She must explain what each theory contributes, where they overlap, where they differ and how integration improves explanatory power.
Methodology, Analysis and Findings
Method selection, sampling, research quality, results and analytical reporting.
14. Writing the Methodology Chapter
Choosing the broad methodological approach
This is only a starting point. Theory, access, ethics, time, disciplinary conventions and contribution also matter.
Research philosophy without unnecessary jargon
Philosophical positioning matters when it changes what counts as knowledge, how evidence is interpreted and which claims are legitimate. It should not become a detached catalogue of ontology and epistemology.
Weak philosophy section
“This study adopts interpretivism because interpretivism is commonly used in qualitative research.”
Stronger justification
“The study adopts an interpretive position because it investigates how HR professionals construct and negotiate the meaning of algorithmic recommendations within specific institutional settings.”
Sampling: adequacy rather than arbitrary numbers
Sampling should be justified by the research purpose. Quantitative studies may require power analysis or precision targets. Qualitative studies may justify adequacy through information richness, diversity, conceptual depth or saturation.
- Define the target population or case universe.
- Explain inclusion and exclusion criteria.
- Describe recruitment and access.
- Identify likely selection bias.
- Explain why the achieved sample is adequate.
Validity, reliability and trustworthiness
| Research tradition | Typical concerns | Possible responses |
|---|---|---|
| Quantitative | Validity, reliability, bias, assumptions | Validated measures, pilots, assumption checks, robustness analysis |
| Qualitative | Credibility, dependability, reflexivity | Audit trail, triangulation, negative cases, reflexive memoing |
| Mixed methods | Quality of each strand and integration | Clear rationale, integration points, joint displays, meta-inferences |
“The candidate describes the selected methods in detail, but the rationale linking those methods to the research questions remains underdeveloped.”
Ethics and data protection
Explain consent, confidentiality, risk, data handling, power relationships and discipline-specific concerns. For AI, digital traces or sensitive data, explain privacy safeguards and limits on secondary use.
The methodology chapter explains and justifies how the research produced credible evidence. It should not become a textbook chapter describing every possible philosophy or method.
Core methodology components
- Research philosophy or paradigm, where relevant
- Research design
- Population, setting and case selection
- Sampling strategy
- Data collection procedures
- Measures, instruments or interview protocols
- Analytical procedures
- Quality criteria
- Ethics and data protection
- Methodological limitations
“I used thematic analysis because it is flexible” is not enough. Flexibility may explain convenience, but the thesis must explain why thematic analysis is suitable for the research question and epistemological position.
Quantitative example
An engineering researcher tests whether a new composite coating improves corrosion resistance. The methodology should justify experimental controls, sample preparation, measurement reliability, statistical tests and assumptions.
Qualitative example
A healthcare researcher studies how nurses make escalation decisions. The methodology should justify participant selection, interview design, reflexivity, coding, theme development and credibility procedures.
Mixed-methods example
An education researcher surveys 500 students and interviews 30 participants. The methodology must explain not only both methods, but also how quantitative and qualitative evidence will be integrated.
Unsure Whether Your Methodology Is Defensible?
The Averon Research Evaluator can identify possible weaknesses in alignment, sampling, analysis and methodological justification before submission.
Evaluate Your Thesis →15. Writing Results and Findings
How to organise a findings chapter
Choose an organising principle that makes answers visible: research questions, hypotheses, themes, cases, phases or analytical models.
Students sometimes organise findings according to the order in which analysis was performed rather than the order that best answers the research questions.
Weak quantitative reporting
“There was a significant relationship between autonomy and engagement.”
Stronger reporting
“Perceived autonomy was positively associated with engagement after controlling for tenure and role level. The effect was moderate, and the confidence interval excluded zero.”
Weak qualitative theme
Theme 1: Technology.
Stronger analytical theme
“Algorithmic recommendations became a new source of authority, but managers resisted them when professional accountability remained personal.”
The results or findings chapter presents what the analysis produced. Its structure should help the reader understand the answers to the research questions.
For quantitative research
- Report data screening and assumptions where relevant.
- Present descriptive results before complex tests.
- Include effect sizes and confidence intervals where appropriate.
- Use tables and figures strategically.
- Do not claim causality unless the design supports it.
For qualitative research
- Explain the organisation of themes or categories.
- Use quotations as evidence, not decoration.
- Show variation, contradiction and complexity.
- Maintain a clear distinction between participant statements and researcher interpretation.
- Connect findings to the research questions.
Discussion, Evaluation and Submission
Discussion, conclusion, quality scorecards, examination, evaluation and submission readiness.
16. Writing the Discussion Chapter
The seven-part discussion sequence
- State the finding.
- Explain what it means.
- Compare it with prior research.
- Explain agreement, contradiction or novelty.
- Consider alternatives.
- Show implications.
- State the boundary of the claim.
Results repeated
“Participants reported that digital systems increased monitoring.”
Doctoral-level discussion
“The increase in monitoring did not uniformly reduce autonomy. Its effect depended on whether performance data was advisory or tied to accountability processes, suggesting that institutional use—not technology alone—reshapes discretion.”
The discussion should show precisely how the thesis changes, extends or qualifies existing understanding.
The discussion explains what the findings mean. This is where the thesis moves from results to contribution.
A strong discussion should:
- answer the research questions;
- interpret the main findings;
- compare them with previous research;
- explain agreements and contradictions;
- consider alternative explanations;
- show theoretical implications;
- identify practical or policy implications;
- state the contribution without exaggeration.
Example
Sarah finds that digital HR systems increase central visibility but do not automatically centralise decisions. In some organisations, local managers retain discretion; in others, algorithmic recommendations become effectively mandatory. Her discussion explains this variation through institutional rules and public accountability pressures. The contribution is therefore not “technology changes HR,” but an explanation of when and why digital systems redistribute authority differently.
Repeating findings in different words is not discussion. Interpretation requires explaining meaning, relationships, mechanisms and implications.
17. Writing the Conclusion
A practical conclusion structure
- Restate the problem and purpose.
- Answer the questions.
- State the contribution.
- Explain implications.
- Discuss limitations.
- Identify future research.
- End with significance.
A conclusion is not stronger because it makes broader claims. It is stronger when it clearly states what the study established and where that knowledge remains bounded.
The conclusion should bring the entire thesis together. It normally revisits the research problem, summarises answers, states the contribution, discusses implications, acknowledges limitations and recommends future research.
Avoid two extremes
- Too little: a short summary that does not explain contribution.
- Too much: a new discussion chapter introducing evidence or arguments not developed earlier.
The Averon Research Quality Scorecard
Use this as a diagnostic rather than a grade.
| Dimension | 1 — Weak | 3 — Developing | 5 — Strong |
|---|---|---|---|
| Research problem | Broad topic | Visible but partly justified | Precise, evidenced and significant |
| Research gap | Asserted absence | Some synthesis | Demonstrated and consequential |
| Questions | Vague | Mostly focused | Focused and contribution-led |
| Methodology | Described only | Generally appropriate | Aligned and defensible |
| Analysis | Disconnected | Competent | Rigorous and transparent |
| Discussion | Repeats findings | Some interpretation | Explains meaning and implications |
| Contribution | Vague novelty | Present but underdeveloped | Precise and integrated |
| Coherence | Chapters conflict | Mostly aligned | One clear argument |
How Universities Actually Examine a PhD Thesis
Processes differ, but doctoral examination commonly involves an internal examiner, an external examiner and an oral defence or viva. Some institutions also appoint an independent chair.
Internal examiner
The internal examiner understands institutional regulations and the academic context but should not have supervised the research.
External examiner
The external examiner provides independent subject expertise and helps ensure comparable doctoral standards.
The viva
The viva tests authorship, understanding, judgement and defensibility. Candidates must explain choices, respond to criticism and demonstrate awareness of limitations.
Possible outcomes
- Pass: the thesis meets the standard.
- Minor corrections: limited changes that do not alter the central research.
- Major corrections: substantial revisions to analysis, argument or methodological explanation.
- Revise and resubmit: significant work followed by further examination.
- Lower award or fail: used where doctoral standards are not met, subject to regulations.
Major corrections do not always mean the study is invalid. Often the contribution is potentially acceptable but insufficiently supported, explained or integrated.
18. How to Evaluate Your Thesis Before Submission
The eight-pass evaluation method
- Purpose pass: problem, gap and contribution.
- Alignment pass: questions through method and conclusion.
- Evidence pass: claims versus evidence.
- Methodology pass: appropriateness and transparency.
- Criticality pass: alternatives and counterarguments.
- Contribution pass: comparison with closest prior work.
- Presentation pass: structure, references and consistency.
- Viva pass: likely challenges.
What the Averon Research Evaluator Checks
Evaluation should move beyond proofreading. Use a whole-thesis audit.
Research logic audit
Map the problem, gap, questions, framework, method, findings and contribution on one page. Draw arrows between them. Any weak connection requires attention.
Evidence audit
Review every major claim and identify the evidence supporting it. Reduce, qualify or remove claims that exceed the evidence.
Methodology audit
Check whether each methodological choice is justified, transparent and aligned with the questions.
Contribution audit
Confirm that the claimed contribution appears consistently in the introduction, discussion, conclusion and abstract.
Coherence audit
Read the abstract, introduction and conclusion consecutively. They should describe the same study and contribution.
For a detailed checklist, read How to Evaluate Your PhD Thesis Before Submission.
19. Why PhD Theses Receive Major Corrections or Fail
Common reasons include unclear contribution, weak alignment, insufficient methodological justification, inadequate analysis, descriptive discussion and unsupported conclusions. Presentation issues matter, but conceptual and methodological weaknesses create greater examination risk.
See the detailed guide: 15 Common Reasons PhD Theses Fail Examination—and How to Avoid Them.
20. Complete Pre-Submission Checklist
- The research problem is explicit and evidenced.
- The gap is demonstrated through critical literature synthesis.
- Research questions are focused and answerable.
- The contribution is precise and proportionate.
- The methodology is aligned and justified.
- Sampling and data procedures are transparent.
- Analysis is appropriate and reproducible or auditable.
- Findings answer the questions.
- The discussion interprets rather than repeats.
- Limitations are acknowledged honestly.
- The conclusion resolves the original problem.
- Terminology is consistent throughout.
- Tables, figures, references and appendices are complete.
- Institutional requirements have been checked.
- The final exported file has been reviewed.
Viva, AI and Final Resources
Viva preparation, responsible AI, FAQs, myths, printable checks and final advice.
21. Preparing to Defend the Thesis in the Viva
Four categories of viva questions
| Category | Examples |
|---|---|
| Foundation | Why this problem? |
| Methodology | Why this design? |
| Interpretation | What is the strongest alternative? |
| Contribution | What changes because of this thesis? |
Defensive response
“My supervisor approved this method.”
Defensible response
“I selected a comparative case-study design because the question concerns how institutional context shapes the process. A survey could identify patterns but not the mechanisms central to the contribution.”
The viva tests both the thesis and your understanding of it. Preparation should focus on reasoning, not memorised scripts.
Questions you should be able to answer
- What is your original contribution?
- Why is the research problem important?
- Why did you choose this methodology?
- What is the strongest alternative explanation?
- Which finding is most important?
- What are the most serious limitations?
- What would you change if you repeated the study?
- How does your work change the field?
Practical viva preparation
- Re-read the thesis critically.
- Create a one-page contribution summary.
- Review examiner publications where permitted and appropriate.
- Prepare responses to methodological challenges.
- Conduct at least one realistic mock viva.
- Mark important pages in the thesis.
- Practise explaining complex ideas simply.
22. Using AI Responsibly During Thesis Writing and Evaluation
AI use decision checklist
- Does your institution permit this use?
- Could confidential data be exposed?
- Can every claim and reference be verified?
- Are you supporting rather than replacing judgement?
- Could the output misrepresent authorship?
- Is disclosure required?
AI tools can help generate questions, improve clarity, organise notes and identify possible inconsistencies. They can also produce fabricated references, oversimplified methodology and confident but incorrect interpretations.
Responsible uses
- Brainstorming alternative explanations
- Testing whether an argument is understandable
- Creating a revision checklist
- Identifying inconsistent terminology
- Improving sentence clarity while preserving meaning
High-risk uses
- Generating references without verification
- Producing analysis of data the model has not properly inspected
- Writing methodological justification without understanding it
- Submitting generated arguments as original scholarly reasoning
- Uploading confidential research without checking privacy terms
AI should support judgement, not replace it. You remain responsible for accuracy, originality, confidentiality and compliance with university policy.
23. Frequently Asked Questions
How long should a PhD thesis be?
There is no universal length. Institutional limits vary by discipline and thesis format. A thesis should be long enough to establish and defend the contribution without unnecessary repetition.
How many research questions should a PhD thesis have?
Many theses have one central question supported by two to four sub-questions, but quality and alignment matter more than number.
Can a PhD thesis use only qualitative research?
Yes. Qualitative research can support a rigorous doctoral contribution when the design, sampling, analysis and claims are appropriate.
Can a PhD thesis use only quantitative research?
Yes. Quantitative research is suitable when the questions require measurement, comparison, prediction or causal estimation and the design supports the claims.
Is mixed methods always stronger?
No. Mixed methods is stronger only when integration of qualitative and quantitative evidence is necessary and executed rigorously. Adding a second method without a clear purpose can weaken the study.
What is the most important thesis chapter?
No single chapter works independently. The introduction frames the problem, methodology supports credibility, findings provide evidence and discussion establishes contribution.
Can good writing compensate for weak methodology?
No. Clear writing can expose or explain reasoning, but it cannot repair an inappropriate design or unsupported analysis.
When should I start evaluating my thesis?
Evaluate chapters during drafting, but conduct a complete whole-thesis review once the full argument is stable and before the final submission deadline.
How do I know whether my contribution is original?
Compare your findings and explanation with the closest existing studies. State precisely what your work adds and why that addition matters.
What if my findings are not statistically significant?
Non-significant findings can still be valuable. Interpret them in relation to design, statistical power, theory and prior evidence without overstating conclusions.
What if qualitative participants disagree?
Disagreement may be analytically important. Explore variation rather than forcing all data into a single narrative.
Can a thesis pass with limitations?
Yes. Every study has limitations. Examiners expect awareness, proportionate claims and a credible contribution within stated boundaries.
Should the conclusion include new references?
Usually the conclusion synthesises arguments already developed. Limited referencing may be appropriate, but major new literature should not appear for the first time.
How many times should I proofread?
Use separate passes for argument, structure, evidence, references, language and final formatting. Combining all checks in one reading is ineffective.
Can an evaluator guarantee that my thesis will pass?
No legitimate evaluator can guarantee an examination result. Evaluation can identify risks and areas for improvement, while final outcomes depend on the thesis, institutional criteria and examiner judgement.
Common Myths About PhD Theses
Reality: Coherence and contribution matter more than page count.
Reality: Critical synthesis matters more than citation volume.
Reality: A method is strong only when it fits the question.
Reality: Mature researchers identify limitations and bound their claims.
Reality: Language cannot repair poor logic or design.
Reality: They expect a credible, original and defensible contribution.
Additional Frequently Asked Questions
Can I change research questions after data collection?
Yes, when justified and transparently explained—not merely rewritten to fit results.
How recent should the literature be?
Use foundational work where necessary and current work where the field has evolved.
How many theories should a thesis use?
Only as many as genuinely needed. Multiple theories require a clear integration logic.
Can unexpected findings strengthen a thesis?
Yes, when analysed carefully rather than treated as errors.
What if data does not support a hypothesis?
Report it honestly and interpret it in relation to theory, power and prior evidence.
What is the difference between contribution and implication?
Contribution is the new knowledge added; implication is what that knowledge means.
Printable Final Checklist
24. Final Advice
Do not measure thesis readiness by how tired you are of editing it. Measure readiness by whether the research problem, evidence, reasoning and contribution can be followed and defended.
A strong thesis does not pretend to be flawless. It demonstrates mature academic judgement: important questions, appropriate methods, honest limitations and conclusions that are both meaningful and proportionate.
Evaluate Your Thesis Before Submission
The Averon Research Evaluator helps identify possible weaknesses in research logic, methodology, analysis, argumentation, contribution and presentation. Use the feedback to prioritise revisions and prepare more confidently for examination.
Start Your Thesis Evaluation →Important: This guide provides general educational information. Doctoral requirements differ by university, country and discipline. Always follow institutional regulations and supervisory guidance. No review or evaluation can guarantee a particular examination outcome.
“The thesis contains substantial work, but the candidate has not made sufficiently clear what is original or how the findings advance existing understanding.”