Dissertation Writing, Defence and Research Impact

Introduction and context

A dissertation is more than an extended report of activities. It is a coherent scholarly argument that identifies a significant problem, positions that problem within existing knowledge, explains a defensible research design, presents credible evidence, and draws conclusions that follow from the evidence. In human resource management (HRM), a dissertation should also demonstrate practical sensitivity: recommendations must be relevant to organisations while remaining proportionate to the data and attentive to employee rights.

A strong dissertation therefore connects five elements:

1. The problem—what requires investigation and why it matters.
2. The literature—what is already known, contested or missing.
3. The method—how evidence was generated and why the choices are appropriate.
4. The findings—what the evidence shows.
5. The contribution and implications—what changes in theory, research, practice or policy.

The dissertation defence is the public examination of this argument. Research impact is the process through which findings influence knowledge, decisions, practice, policy or future research. These three activities—writing, defending and creating impact—are distinct but interdependent.

Learning objectives

By the end of this chapter, students should be able to:

• organise a dissertation into a logically connected structure;
• develop a clear argument supported by appropriate evidence;
• use tables, figures, citations and appendices responsibly;
• distinguish description, analysis, interpretation and recommendation;
• avoid plagiarism and manage sources ethically;
• prepare and deliver an oral or poster defence;
• communicate findings to academic and managerial audiences; and
• formulate realistic future research and organisational-impact plans.

1. The dissertation as an integrated argument

1.1 The alignment principle

Every major section should answer a different question while contributing to the same research purpose.

A useful quality test is alignment. The research question should lead to the objectives; the objectives should determine the data required; the data should be analysed using methods suited to the question; and the conclusions should not exceed the evidence.

For example, a study asking, “How do line managers experience the implementation of hybrid-work policies?” is primarily exploratory and qualitative. It should not be presented as measuring the causal effect of hybrid work on organisational productivity unless the design and data support that claim.

1.2 The argument map

Before drafting, prepare an argument map. It may contain:

• the central claim;
• three to five supporting claims;
• evidence for each claim;
• relevant limitations; and
• the practical or theoretical consequence.

A paragraph normally performs one main function. A reliable structure is claim–evidence–reasoning–link:

• Claim: state the point.
• Evidence: cite literature or present data.
• Reasoning: explain why the evidence supports the point.
• Link: connect the point to the research question or next paragraph.

This prevents a common weakness in undergraduate dissertations: paragraphs that summarise several sources but never explain their significance.

2. Dissertation architecture

Universities vary in formatting requirements, but a conventional HRM dissertation includes the following components.

Front matter

The title page, declaration, acknowledgements, abstract, contents list, lists of tables and figures, and abbreviations guide the reader. The abstract should briefly state the problem, purpose, design, participants or data source, principal findings, conclusion and implications. It should not introduce claims absent from the dissertation.

Chapter 1: Introduction

The introduction establishes context, identifies the research problem, states the aim and objectives, presents the research questions, gives a brief rationale, defines the scope, and previews the dissertation. Background information should be selective. A long history of HRM is not a substitute for a clearly specified problem.

A well-formed aim is broad enough to encompass the study but narrow enough to complete. Objectives should be observable, for example:

• to examine employee perceptions of procedural fairness during performance appraisal;
• to identify factors associated with perceived appraisal fairness; and
• to develop evidence-based recommendations for improving appraisal communication.

Chapter 2: Literature review

A literature review is a critical conversation with scholarship. It should compare perspectives, identify patterns and contradictions, evaluate methods, and establish the gap addressed by the study. Organise it thematically, conceptually or chronologically only when that arrangement improves the argument.

Useful questions include:

• Which concepts are central, and how are they defined differently?
• Which theories explain the proposed relationships?
• What populations and contexts have been studied?
• Which findings are consistent or disputed?
• What methodological limitations remain?
• Why is the present study necessary?

A conceptual framework can show proposed relationships among variables or concepts. Arrows should represent hypothesised influence, association or interpretation—not automatic causation. For instance, perceived supervisor support may be related to engagement through psychological meaningfulness, but a cross-sectional survey cannot by itself establish temporal causality.

Chapter 3: Methodology

The methodology explains the research philosophy or assumptions, design, setting, sampling, instruments, data collection, analysis, quality procedures and ethics. The emphasis should be justification rather than procedural narration.

A quantitative study might explain construct operationalisation, scale reliability, sampling, missing data and statistical tests. A qualitative study should explain participant selection, interview design, reflexivity, coding, development of themes and credibility procedures. A mixed-methods study must explain not only both components but also how they are integrated.

Researchers should distinguish reliability from validity. Reliability concerns consistency; validity concerns whether the study supports the intended interpretation. In qualitative research, credibility, dependability, confirmability and transferability may be more appropriate quality concepts (Lincoln and Guba, 1985).

Chapter 4: Findings

The findings chapter reports evidence without repeating the entire method or turning every result into a recommendation. Organise it by research question or analytical theme. Tables and figures should reduce cognitive load, not decorate the text.

A table should have a number, informative title, clear units, source or note, and an explanation in the prose. A figure should be interpretable without excessive reference to surrounding paragraphs. Never present a table or chart without stating the important pattern it demonstrates.

Quantitative findings should report appropriate descriptive information and, where relevant, effect estimates and uncertainty. A statistically significant result is not automatically practically important. Qualitative findings should provide sufficiently rich excerpts to substantiate themes while protecting confidentiality.

Chapter 5: Discussion

The discussion answers the question, “What do these findings mean?” It should:

1. interpret the principal findings;
2. compare them with prior research;
3. explain convergence or divergence;
4. consider theoretical significance;
5. acknowledge alternative explanations; and
6. identify implications and limitations.

Do not introduce new empirical results in the discussion. Do not claim that a finding “proves” a theory when it merely supports, qualifies or challenges an aspect of it.

Chapter 6: Conclusion, recommendations and impact

The conclusion should directly answer the research questions and restate the contribution in a concise form. Recommendations should identify the action, responsible actor, required resources, timescale and evidence supporting the action. “Improve communication” is weak; “require quarterly two-way appraisal conversations, supported by manager training and an anonymous six-month pulse survey” is more actionable, though its feasibility must still be assessed.

3. Evidence, citations and academic integrity

3.1 Using sources critically

Sources have different purposes. Foundational theory may require a seminal article; current employment practice may require recent empirical research, official statistics or professional guidance. Evaluate authority, relevance, methodological quality, currency and potential bias.

Avoid citation dumping: a sentence followed by six references does not show synthesis. Instead, explain the relationship among sources. For example, one study may report a positive relationship between autonomy and engagement, while another finds that the relationship depends on managerial support and job insecurity. The writer’s task is to interpret the pattern.

Use a consistent citation style specified by the institution. Every in-text citation must correspond to a reference-list entry, and every listed source should be cited. Page numbers are especially useful for direct quotations and precise definitions.

3.2 Avoiding plagiarism

Plagiarism includes presenting another person’s words, ideas, data, structure or creative work as one’s own. It may be intentional or accidental. Safe practice includes:

• recording complete bibliographic details during reading;
• separating copied notes from paraphrased notes;
• using quotation marks for exact wording;
• paraphrasing by reconstructing the idea in one’s own sentence structure, followed by citation;
• citing datasets, instruments, images and organisational documents; and
• retaining drafts and evidence of the research process.

Changing a few words while retaining the original sentence structure is not adequate paraphrasing. Generative AI does not remove the author’s responsibility for originality, accuracy, confidentiality or compliance with institutional policy. AI-generated references must be verified against the original publication.

3.3 Tables, figures and appendices

Appendices contain supporting material that is useful but too detailed for the main argument: interview schedules, participant information sheets, consent forms, coding frameworks, survey instruments, supplementary tables and ethics approvals. An appendix should be referred to in the main text and should not be used to hide essential analysis.

4. From analysis to defensible claims

The strength of a claim depends on the design, evidence quality and limitations. A cross-sectional association does not establish cause; a small purposive sample may provide depth but not statistical generalisation; self-report data may be affected by recall and social-desirability bias.

A useful distinction is:

• Description: what was observed?
• Analysis: what pattern or relationship is present?
• Interpretation: what might explain it?
• Evaluation: how strong and credible is the explanation?
• Recommendation: what action is justified?

For example, if 68% of surveyed employees report that feedback is infrequent, this is descriptive. Concluding that poor feedback causes turnover would require stronger design and evidence. A defensible conclusion might be that infrequent feedback is a significant employee concern and warrants further investigation alongside objective turnover data.

When quantitative analysis is used, report measures in context. For a sample mean, the arithmetic average is:

where x̄ is the sample mean, xᵢ is observation i, and n is the number of observations. This formula applies to numerical observations measured on a scale for which averaging is meaningful. It does not, by itself, establish representativeness or causality. A mean should normally be accompanied by a measure of dispersion and a clear description of the sample.

5. Preparing the dissertation defence

A defence tests whether the researcher understands and can justify the work. It is not a request to claim perfection. Examiners commonly explore the rationale, conceptual framework, methodological choices, interpretation, limitations and implications.

5.1 Oral defence structure

A ten-minute presentation may use this sequence:

1. problem and significance;
2. aim and research questions;
3. conceptual or theoretical framework;
4. methodology and ethical safeguards;
5. two or three principal findings;
6. contribution and recommendations; and
7. limitations and future research.

Slides should support speech rather than reproduce the dissertation. Use one central message per slide, readable text, labelled visuals and consistent terminology. Present numbers selectively and explain their meaning.

When answering questions, listen fully, pause, answer directly, provide evidence, and acknowledge uncertainty. A useful response pattern is: “The design supports X. It does not establish Y because… A stronger test would require…” This demonstrates judgement rather than weakness.

5.2 Poster defence

A poster must communicate the research at a distance and invite discussion. Its visual hierarchy should move from problem to method, findings and implications. Use short sections, informative headings, a simple conceptual figure and a small number of findings. A QR code may link to supplementary material only if permitted and if the poster remains understandable without it.

The presenter should prepare a thirty-second overview, a two-minute explanation and answers to likely questions. Confidential organisational information must not be displayed.

5.3 Typical defence questions

• Why is this problem important to HRM?
• Why was this theory selected instead of a competing theory?
• How were participants recruited, and who may be missing?
• What steps improved the credibility or validity of the findings?
• Which finding was most unexpected?
• What is the strongest limitation?
• What would you change if repeating the study?
• Which recommendation should an organisation implement first, and why?

6. Communicating with academic and managerial audiences

Academic communication emphasises conceptual precision, evidence, limitations and contribution. Managerial communication emphasises decisions, feasibility, costs, risks, employee consequences and implementation. The evidence must remain consistent, but the framing can change.

A professional executive brief usually contains the problem, key evidence, implications, recommended actions, implementation risks and evaluation measures. It should not overstate return on investment or imply that employee data are merely operational assets.

7. Research impact and future research

Research impact is broader than publication or a high mark. It may include conceptual influence, improved HR policy, better managerial practice, employee benefit, professional learning, public discussion or subsequent research. Impact claims should distinguish outputs (a dissertation, presentation or report), outcomes (a policy change, training adoption or new collaboration) and impact (a demonstrable change attributable partly to the research).

An impact pathway can be planned as follows:

Finding → audience → communication product → decision or action → indicator of change

For example, a finding that onboarding information is inconsistent may lead to an HR briefing, a standardised onboarding checklist, implementation by department heads, and evaluation through time-to-productivity and new-starter experience measures. The indicators do not prove that the dissertation alone caused improvement, but they provide a basis for assessing contribution.

Future research should arise from limitations and unanswered questions, not from a generic statement that “more research is needed.” Specify the population, setting, design, variables or concepts, and reason for priority. A future longitudinal study might examine whether changes in manager feedback precede changes in engagement; a comparative study might test whether the pattern differs between public and private organisations.

8. Applied case: defending an HRM dissertation

This is an illustrative case.

Background

Maya completed a dissertation on employee perceptions of fairness in hybrid-work scheduling at a medium-sized professional-services organisation. She used a survey of 146 employees and follow-up interviews with 12 volunteers.

Problem

Managers believed that scheduling decisions were fair, but informal complaints suggested that employees with caring responsibilities and those in client-facing roles experienced the policy differently.

Analysis

The survey indicated that perceived transparency was more strongly associated with fairness ratings than the number of days worked remotely. Interview participants explained that explanations for exceptions were inconsistent. Maya therefore avoided claiming that hybrid work caused dissatisfaction. She interpreted the evidence as indicating that transparent and consistently explained decision processes were important conditions of perceived fairness.

Defence approach

In her presentation, Maya displayed the research question, a mixed-methods design diagram, one concise table of survey results, two anonymised interview excerpts and a recommendation pathway. When asked whether the findings generalised to all professional-services firms, she stated that the sample was drawn from one organisation and was not statistically representative. She argued that transferability might be possible to organisations with similar work structures, but this would require contextual judgement.

Solution and outcome

Maya recommended a documented scheduling framework, manager guidance for explaining exceptions, and a six-month review using employee feedback and scheduling records. The organisation adopted the framework for a pilot in two departments. The dissertation itself did not establish that the pilot would improve retention; it provided a reasoned basis for testing a practical intervention.

Learning points

• A mixed-methods design is valuable only when the components are integrated.
• A limitation can increase credibility when it is acknowledged precisely.
• Recommendations should be testable and assigned to responsible actors.
• Impact should be monitored rather than assumed.
• Confidentiality is essential when organisational examples are identifiable.

9. Quality-control checklist

Before submission or defence, check that:

• the title, aim, objectives and research questions match;
• each objective is answered in the findings and conclusion;
• all major claims have appropriate evidence;
• citations and references are complete and consistent;
• tables and figures are numbered, labelled and discussed;
• ethical approval, consent and data protection are documented;
• limitations are specific rather than formulaic;
• recommendations follow from findings and identify feasibility issues;
• appendices are referenced and contain no unnecessary personal data;
• the abstract accurately reflects the completed study; and
• the document has been proofread for clarity, formatting and accessibility.

Chapter Summary

A dissertation is a structured scholarly argument connecting a research problem, literature, method, evidence, interpretation and contribution. Quality depends on alignment: questions must guide methods, methods must generate suitable evidence, and conclusions must remain proportionate to that evidence. Critical literature review, transparent methodology, ethical research practice and careful citation protect academic integrity. Tables, figures and appendices should clarify rather than replace the argument. A defence requires concise communication, methodological self-awareness and the ability to answer questions honestly. Research impact is developed through planned communication and observable pathways from findings to decisions or change. Future research should be specific, justified and connected to limitations or unresolved questions.

Key Terms

Academic integrity: Honest and responsible conduct in producing and presenting scholarly work.

Abstract: A concise summary of a dissertation’s purpose, method, findings and implications.

Alignment: Logical consistency among the research problem, questions, methods, findings and conclusions.

Appendix: Supplementary material placed after the main text.

Claim: A proposition advanced by the researcher and supported by reasoning and evidence.

Conceptual framework: A structured representation of concepts and their proposed relationships.

Contribution: The knowledge, explanation, method or practical insight added by a study.

Defence: An oral or poster examination in which the researcher explains and justifies the dissertation.

Discussion: The section that interprets findings in relation to theory and previous research.

Impact: A demonstrable influence of research on knowledge, practice, policy, people or future research.

Literature synthesis: Integration and critical comparison of relevant sources.

Methodology: The rationale and principles guiding research design and evidence production.

Paraphrase: Restatement of another source’s idea in original wording and structure with citation.

Plagiarism: Presenting another person’s work or ideas as one’s own.

Research question: A focused question that the study is designed to answer.

Reliability: Consistency of a measure or research process.

Validity: The extent to which evidence supports the intended interpretation.

Transferability: The reasoned applicability of findings to comparable contexts.

Review Questions

Short Answer Questions

1. What is meant by alignment in dissertation design?
2. Distinguish reliability from validity.
3. State two functions of a literature review.
4. Why should recommendations be linked to responsible actors?
5. What is the difference between a research output and research impact?

Long Answer Questions

1. Explain how the introduction, literature review, methodology, findings and discussion should operate as an integrated argument.
2. Discuss the ethical and practical responsibilities involved in using citations, appendices and organisational data.
3. Explain how a student can prepare an effective oral defence, including strategies for answering difficult questions.
4. Compare academic and managerial communication of the same HRM research findings.
5. Evaluate the role of limitations in establishing a dissertation’s credibility.

Analytical / Application Questions

1. A student concludes that a cross-sectional survey proves that flexible work causes higher engagement. Analyse the claim and propose a more defensible conclusion.
2. Design an impact pathway for a dissertation finding that employees experience inconsistent performance feedback. Include an audience, communication product, action and indicator.
3. Using the illustrative hybrid-work case, propose one additional research question and justify an appropriate research design for answering it.

Multiple Choice Questions with Answers

Question 1

Which sequence best represents alignment in a dissertation?

A. Question, suitable method, evidence, proportionate conclusion

B. Theory, recommendation, question, unrelated evidence

C. Finding, title, sampling, unsupported conclusion

D. Appendix, abstract, method, research problem

Correct Answer: A. Question, suitable method, evidence, proportionate conclusion

Question 2

What is the primary purpose of a critical literature review?

A. To list as many sources as possible

B. To synthesise evidence and establish the research gap

C. To report the study’s original findings

D. To replace the methodology chapter

Correct Answer: B. To synthesise evidence and establish the research gap

Question 3

Which statement is most defensible for a cross-sectional survey?

A. It proves that X causes Y

B. It establishes temporal precedence

C. It can identify associations but may not establish causality

D. It eliminates self-report bias

Correct Answer: C. It can identify associations but may not establish causality

Question 4

Which practice is plagiarism?

A. Summarising a source with a citation

B. Using a direct quotation with quotation marks and citation

C. Recording a source in a reference list

D. Changing a few words while retaining the source’s structure without citation

Correct Answer: D. Changing a few words while retaining the source’s structure without citation

Question 5

What should a discussion chapter primarily do?

A. Introduce entirely new empirical results

B. Interpret findings in relation to theory and prior research

C. Repeat the table of contents

D. Provide raw interview transcripts

Correct Answer: B. Interpret findings in relation to theory and prior research

Question 6

Which item is most appropriate in an appendix?

A. The central conclusion

B. The main research question

C. A detailed interview schedule

D. The principal contribution

Correct Answer: C. A detailed interview schedule

Question 7

Which response best demonstrates methodological maturity during a defence?

A. “My study has no limitations.”

B. “The design supports this interpretation, but it cannot establish causality.”

C. “The examiner’s question is irrelevant.”

D. “The result is true because it is statistically significant.”

Correct Answer: B. “The design supports this interpretation, but it cannot establish causality.”

Question 8

What is an example of a research outcome rather than an output?

A. A dissertation submitted to a university

B. A conference poster

C. Adoption of a revised HR policy

D. A reference list

Correct Answer: C. Adoption of a revised HR policy

Question 9

Which recommendation is most actionable?

A. Improve communication

B. Managers should do better

C. Investigate HR issues sometime

D. Introduce quarterly two-way appraisal meetings and review participation after six months

Correct Answer: D. Introduce quarterly two-way appraisal meetings and review participation after six months

Question 10

In the chapter’s case, why did Maya avoid claiming that hybrid work caused dissatisfaction?

A. She had no research question

B. Her evidence and design did not establish causality

C. Qualitative data can never be useful

D. Organisations cannot be studied ethically

Correct Answer: B. Her evidence and design did not establish causality