Key points at a glance
- Tie the manipulation to a research question.
- Keep scenarios comparable.
- Pretest comprehension and analysis.
Define the judgement you want to study
A vignette gives participants a short hypothetical situation and asks for a judgement or decision. It can isolate a situational factor when real cases cannot be manipulated. Specify the outcome and theoretically justified factor before writing an engaging story. Too many background details introduce uncontrolled cues.
Example: You ask whether a transparent salary statement changes the appeal of an internship. Two scenarios differ in transparency but keep duration, tasks and location constant. If the pay level also changes, you cannot attribute a response difference to transparency alone.
Plan variation and allocation
Decide whether each respondent will see one scenario or several. Multiple vignettes can create order effects and fatigue, so plan randomisation or balanced order when feasible. Count factor combinations before fielding; several factors with multiple levels can overwhelm a small thesis sample.
Keep the wording equally neutral and the length similar. A warm, detailed description may influence judgement independently of your intended factor. Choose a response scale aligned with the hypothesis. Consider asking whether the case feels realistic, but avoid revealing the manipulation before the main outcome is recorded.
Pretest understanding and ethics
Ask a few people from the target population to explain the scenarios in their own words. Look for ambiguous terms, missing context and clues that make the manipulation too obvious. A manipulation check may test whether respondents noticed the planned difference; place it so it does not prime the main answer.
Hypothetical stories can still concern distressing or sensitive issues. Explain foreseeable burden and obtain required approval. Avoid unnecessary identifying details. If deception is involved, discuss ethical requirements with your supervisor before data collection.
Analyse the observations you actually made
Store which version each person saw. If respondents judge several scenarios, their responses are dependent observations and should not be counted as separate people. Explain allocation, exclusions and scale construction in the methods section. Check whether sample size supports the number of conditions before promising many interactions.
Report responses by condition before a suitable comparison. Do not equate a hypothetical intention with real behaviour. Controlled scenarios can clarify a mechanism, while generalisability still depends on plausibility, sampling and setting.
Final submission context
Formal details can feel like a separate writing task, but they matter just as much for a printed submission. Anything missing, misplaced or inconsistently formatted in the document will also appear in the bound copy.
Use this guide together with your cover page, table of contents, page numbers, source notes and appendices. Prepare the final PDF for printing and binding only after the complete file has been checked.
Practical check before PDF export
- Do headings, chapter structure and page numbers match?
- Are sources, figures and tables included completely?
- Are required elements such as declarations, appendices or the cover page included where required?
- Did you open and check the final PDF after exporting it?
Checklist
- Defined outcome and factor.
- Matched scenario wording.
- Planned allocation and pretest.
- Accounted for repeated responses.
Common mistakes
- Changing several cues unintentionally.
- Skipping the pretest.
- Treating hypothetical intent as observed action.
Prepare the finished thesis for submission
Once the content and university requirements are checked, review your PDF and the available printing and binding options.
Review printing optionsRelated content
Explore related guidance on thesis structure, sources, formatting and final submission.
Frequently asked questions
How many vignettes do I need?
Only enough to represent the theoretically justified conditions.
Must everyone see the same cases?
No, but allocation must fit the analysis.
What does a pretest reveal?
Whether scenarios are clear, comparable and plausible.