Appendices and Research Instruments
From 140 pages of raw field data to a quarter of a thesis given over to questionnaires — what belongs in an appendix, and the replication test that decides it.
How much of a thesis appendices actually are
What each thesis actually put there
APPENDIX CONTENTS IN THREE EXAMINED THESES
| Exemplar | Numbering | What was included |
|---|---|---|
| A — action research (1999) | 23 items numbered by chapter, e.g. 4-1 to 7-6 | Raw field data including a list of 211 problems from a search conference; researcher-built instruments and their scored results; organisational artefacts such as process charts, report forms and review forms; summarised correspondence; event chronologies; training slides; per-team feedback |
| B — mixed methods (2016) | 6 items numbered by chapter | Market data tables; the participant invitation letter; the full survey instrument as 30 pages of page-by-page screenshots including the consent question and a language screening question; the data coding scheme; full descriptive analysis with charts; a re-categorisation of one variable proxy with its analysis tables |
| C — portfolio (2004) | 2 items, A and B | The questionnaire for paper two and the interview questionnaire for paper three, reproduced in full |
Contents of three specific documents.
The test that decides what belongs
APPLYING THE TEST
| Item | Passes? | Why |
|---|---|---|
| Questionnaire or survey instrument | Yes | Nobody can repeat a survey they cannot see. Two of three examined theses reproduced theirs in full |
| Interview guide or question set | Yes | Same logic. None of the examined works supplied one — this is the most common omission |
| Coding scheme or classification frame | Yes | Without it, no reader can check how material was categorised. The 2016 thesis included one |
| Participant information and consent wording | Yes | Makes the ethics apparatus inspectable rather than asserted |
| Raw or summarised field data | Usually | The 1999 thesis included a list of 211 problems from a search conference — a reader can see what the analysis was built from |
| Full descriptive statistics not shown in the body | Yes | Lets a reader check that headline figures are consistent with the underlying distribution |
| An alternative specification you tested | Yes | The 2016 thesis included a re-categorised variable with its analysis. This is a robustness claim made inspectable |
| A figure already printed in the body | No | Duplication, not evidence |
| Interview transcripts | Judgement | None of the examined works included any. Volume and confidentiality both argue against; check your institution's requirements |
The verdict column is a synthesis for this page; no supplied source states an appendix policy.
The screenshot decision
The 2016 thesis reproduced its internet questionnaire as thirty pages of page-by-page screenshots rather than as retyped text. That is worth pausing on, because it is a deliberate choice with a real trade-off.
What screenshots preserve
- The exact wording every respondent saw
- Question order and page breaks, which affect responses
- Scale labels and their visual arrangement
- Routing and screening questions in context
- Consent wording as it was actually presented
What they cost
- No extractable text, so the instrument is not searchable
- Thirty pages for material that might retype into eight
- Screen rendering may differ from what some respondents saw
- Accessibility — a screen reader gets nothing from an image
Numbering and navigation
Both examined theses numbered appendices by chapter — appendix 4-1 belongs to chapter four. That is more useful than a flat sequence because it tells a reader which part of the argument the material supports without consulting a list.
Appendix hygiene
- Number by chapter so the reader knows what each item supports
- Use one numbering format throughout, not two
- Give appendices their own list in the front matter
- Page-number them like the rest of the document
- Check every body cross-reference against the finished appendix set
- Confirm that anything needed for replication is actually present
The instrument documentation gap
Across all five examined works, the pattern is consistent: quantitative instruments are reproduced, qualitative instruments are not. Two theses printed their questionnaires in full. None printed an interview guide, despite four of the five conducting interviews — ten in one thesis, eight in another, nine in the journal paper, and an extended programme across four phases in the action research thesis.
What to carry forward
- Appendices were a quarter to a third of the examined theses. Plan them, do not accumulate them.
- The test: could a competent researcher repeat your study, or check your claim, from what you supplied?
- Reproduce every instrument — questionnaire and interview guide. The examined works consistently supplied the first and omitted the second.
- Screenshots preserve exact wording and layout at the cost of searchability. A deliberate trade-off, worth making for instruments.
- Number appendices by chapter, list them, page-number them, and check every cross-reference against the finished document.
Frequently asked questions
How long should appendices be?
No supplied source states a length. In the examined theses they ran to roughly a third and a quarter of the document, and in one portfolio thesis two questionnaires alone occupied 39 of 146 pages. Volume follows from what replication requires, not from a target.
Should I include interview transcripts?
None of the five examined works did. Volume and participant confidentiality both argue against, and no supplied source requires it. Check your own institution's rules, and supply the interview guide regardless.
Is it acceptable to reproduce an instrument as screenshots?
One examined thesis did exactly that, across thirty pages. It preserves the exact wording, ordering and layout respondents saw, at the cost of searchability and accessibility. For an online instrument the trade is usually worth making.
How should appendices be numbered?
Both examined theses numbered by chapter, so appendix 4-1 supports chapter four. That is more informative than a flat sequence. Use one format consistently — one of the examined theses used two and made its own material harder to find.
What is the most common appendix omission?
The interview guide. Four of the five examined works conducted interviews and none reproduced its question set, while two of them reproduced their questionnaires in full.
References and source attribution
- Five examined research works supplied as exemplars: two doctoral theses (1999, 2016), a doctoral portfolio thesis (2004), a peer-reviewed journal paper (2014) and a conference paper (2017/2018). Structural observations only; chapter bodies were not reproduced.
- Supplied teaching source, Weeks 2-4: Introduction to Research Methods, Developing a Research Topic, Reviewing the Literature.
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
Suggested questions for Ask KEVOS
- Which of my research materials belong in an appendix?
- Draft an interview guide I can reproduce as an appendix.
- How should I number and list my appendices?
- What does a reader need in order to replicate my study?
- Should I include raw data or only summarised data?