Chapter 7 — Dissertation writing and limitations

Lesson map

Lesson 07 published. Map, example, EX039.

flowchart TD
  N0["Learning outcomes"]
  N1["7.1 Structure"]
  N2["7.2 Limitations"]
  N3["7.3 Findings versus recommendations"]
  N4["7.4 Writing for two audiences"]
  N0 --> N1 --> N2 --> N3 --> N4
  N4 --> EX["Worked example"]
  EX --> Q["Exercise EX039"]

Logical drawing

flowchart LR
  N0["Learning outcomes"]
  N1["7.1 Structure"]
  N2["7.2 Limitations"]
  N3["7.3 Findings versus recommendations"]
  N4["7.4 Writing for two audiences"]
  N0 --> N1 --> N2 --> N3 --> N4 --> Q["EX039"]

[Learning outcomes]
    |
    v
[7.1 Structure]
    |
    v
[7.2 Limitations]
    |
    v
[7.3 Findings versus recommendations]
    |
    v
[7.4 Writing for two audiences]
    |
    v
[Worked example] --> [Exercise EX039]

Learning outcomes

You will (1) structure a dissertation around a question; (2) write specific limitations; (3) separate findings from recommendations.

7.1 Structure

Priya structures her dissertation around the question: introduction and question, literature, standards and design principles, design and ethics, pilot and measurement, results, rivals, discussion and limitations. Each chapter ends with a short statement of what it contributes to answering the question. Dr. Bello reads the plan and asks her to move the scanner-click finding into its own section, because it is useful to other researchers even apart from the main result about feedback. Priya agrees, and the new section becomes one of the clearest parts of the dissertation.

7.2 Limitations

Priya's limitations section names five issues: three treated departments chosen by the security team rather than randomly; possible extra manager attention; a timing rule for scanner clicks instead of address matching; simulation clicks as a proxy for real phishing; and only two quarters of follow-up. For each, she states how it could bias the estimate and in which direction. Dr. Bello says this specificity is what distinguishes a useful limitations section from a list of disclaimers. Priya revises the section twice before submission.

7.3 Findings versus recommendations

Priya separates what the evidence shows from what she recommends. The finding is that, in this setting, feedback departments reduced human clicking more than comparison departments, without reducing reports. Her recommendations are to extend feedback to further departments as a staged rollout so comparison remains possible, to filter scanner clicks before reporting rates, and to measure reporting alongside clicks. She labels recommendations as judgements that go beyond the data, so readers can weigh them separately. Dr. Bello checks that every recommendation is marked.

7.4 Writing for two audiences

The dissertation serves examiners and Harrowgate's security team. For the team, Priya writes a two-page summary with the main estimate, the scanner-click correction and the recommendations, avoiding statistical terms where possible and never naming departments negatively. For examiners, the full text gives methods, checks and code. She asks a security team member to read the summary for clarity and revises two sentences that implied staff were careless, replacing them with neutral descriptions of behaviour. The security team member approves the revised summary, and Priya keeps both versions.

Worked example

Priya structures the dissertation by question, writes five specific limitations with bias direction, separates findings from labelled recommendations and produces a neutral summary for the security team.

Synthetic case. The dissertation is synthetic.

Errors and limits

Generic disclaimers do not help readers judge a result.

Exercises

• EX039 — Separate section. Which finding gets its own section?
  Show marking criteria

  Rubric: 3 = the scanner-click correction;; 2 = partly correct;; 1 = other;; 0 = none.
• EX040 — Five limitations. Name one of the five limitations.
  Show marking criteria

  Rubric: 3 = non-random treated departments;; 2 = one;; 1 = vague;; 0 = none.
• EX041 — Limitation detail. What does Priya state for each limitation?
  Show marking criteria

  Rubric: 3 = how it biases and in which direction;; 2 = partly correct;; 1 = vague;; 0 = none.
• EX042 — Staged rollout. Why recommend a staged rollout?
  Show marking criteria

  Rubric: 3 = comparison remains possible;; 2 = partly correct;; 1 = vague;; 0 = none.
• EX043 — Summary length. How long is the team summary?
  Show marking criteria

  Rubric: 3 = two pages;; 2 = partly correct;; 1 = wrong;; 0 = none.
• EX044 — Revised wording. What wording was revised?
  Show marking criteria

  Rubric: 3 = sentences implying carelessness;; 2 = partly correct;; 1 = vague;; 0 = none.