A pilot study is a small-scale rehearsal of your main research, run to expose problems with your design, instruments and procedures before you commit time, funding and participants to the full project. For UK undergraduate and postgraduate researchers, a well-run pilot is one of the most cost-effective ways to protect the quality of a dissertation or thesis. This guide walks you through planning, executing and reporting a pilot study, with worked examples and a clear template you can adapt to your own discipline.
★ Key takeaways
- A pilot study tests feasibility and process, not your hypothesis: its job is to find flaws in recruitment, instruments and analysis before the main study begins.
- Plan with explicit feasibility criteria (recruitment rate, retention, completion time and data quality) and a clear go, amend or stop rule for each.
- Pilots are typically small, often around 10 to 12 participants per arm, and are not powered to detect statistical effects, so resist reporting p-values as findings.
- Always report what you changed as a result of the pilot: amended questionnaires, revised protocols, adjusted timelines and refined inclusion criteria.
- Treat the pilot as a methods chapter in miniature, documenting decisions transparently so examiners can see your research design maturing.
What a Pilot Study Is (and Is Not)
A pilot study is a deliberately small, low-stakes version of your planned research, conducted to check that the larger study will actually work. Its purpose is feasibility and process, not discovery. You are not trying to confirm a hypothesis or measure an effect; you are stress-testing the machinery of your research before you switch it on at full scale.
It helps to separate two related ideas. A feasibility study asks whether the research can be done at all, given your resources, access and timeline. A pilot study goes one step further and rehearses the actual procedures in miniature, often using the same instruments and recruitment routes you intend to use later. In practice, many dissertation pilots do both at once.
Common pitfalls flow from misunderstanding this purpose. Students frequently treat the pilot as a dress rehearsal that also produces real results, then report significance tests on a handful of participants. With ten or twelve people, your study is almost never powered to detect anything, so a non-significant result tells you nothing about your hypothesis. Equally, a pilot is not an excuse to recycle the same participants in your main study, which would bias your final sample. The deliverable of a pilot is a set of decisions and refinements, not p-values.
The pilot study workflow from plan to refined protocol
1. Define objectives
Write specific feasibility questions and progression criteria with green, amber and red thresholds.
2. Plan tasks and ethics
Break work into owned, deadlined tasks; secure ethics approval and prepare instruments.
3. Execute small-scale
Run the rehearsal with a small sample, recording timings, queries and any procedural snags.
4. Analyse descriptively
Summarise recruitment, retention, completion times and data quality against your criteria.
5. Report and refine
Document decisions and the exact changes made to the protocol before the main study begins.
Why Pilots Matter for UK Dissertations
For UK students working to a fixed submission deadline, the pilot study is a form of insurance. Discovering that a questionnaire item is ambiguous, that your survey takes forty minutes rather than fifteen, or that your recruitment channel yields two responses a week rather than twenty, is painful at any stage. Discovering it after you have collected half your data is potentially fatal to the project.
Examiners and supervisors increasingly look for evidence that you have de-risked your methodology. A transparent pilot demonstrates exactly the methodological awareness that distinguishes a strong dissertation from a merely adequate one: you anticipated problems, tested for them, and adapted. This is the practical face of research rigour, and it is also why professional dissertation writing services routinely build a piloting phase into the timelines they propose to students.
There is an ethics dimension too. UK university ethics committees expect researchers to minimise burden and avoid wasting participants' time on poorly designed instruments. A pilot that catches a confusing consent form or a distressing question protects your participants and strengthens your ethics application. In short, piloting is not a luxury bolted on to a tidy project; it is part of doing research responsibly.
| Objective | What you measure | Example criterion | Decision if not met |
|---|---|---|---|
| Recruitment | Eligible participants enrolled per week | 8 or more per week | Add a second recruitment channel |
| Retention | Proportion completing the study | At least 80% retained | Reduce burden, add reminders |
| Instrument fit | Average completion time and item queries | Under 20 minutes, no recurring queries | Reword or remove items |
| Data quality | Missing or implausible values | Below 5% missing | Revise data-entry and validation steps |
| Analysis pipeline | Export and analysis run without error | Full run completes end to end | Fix import scripts before main study |
Planning Your Pilot Study
Good planning begins by writing down the specific questions the pilot must answer. Vague aims such as "see if it works" produce vague pilots. Instead, frame measurable feasibility objectives. Typical ones include recruitment rate (how many eligible participants you can enrol per week), retention (how many complete the study), instrument performance (how long completion takes and which items are skipped or queried), and data quality (whether your data entry and analysis pipeline runs end to end).
For each objective, set an explicit progression criterion with a traffic-light logic: green means proceed as planned, amber means proceed with modifications, and red means stop and rethink. For example, you might decide that a recruitment rate above eight per week is green, four to eight is amber, and below four is red. Defining these thresholds in advance stops you rationalising poor results after the fact.
Then break the work into tasks with owners and deadlines. List everything: drafting the instrument, securing ethics approval, recruiting, running sessions, entering data, and running a trial analysis. Estimate realistic durations, remembering that gathering materials and obtaining approvals almost always take longer than expected. A simple responsibility chart, even for a solo dissertation, keeps the moving parts visible. Finally, decide your monitoring approach: what you will record during the pilot, and how you will capture participant feedback, so that nothing useful is lost.
The deliverable of a pilot study is not a result, but a sharper, de-risked design: a set of evidence-based decisions that make your main study far more likely to succeed.The 123Essays Review Team
Executing the Pilot: A Worked Example
Consider Aisha, an MSc Health Psychology student investigating whether a four-week mindfulness app reduces exam anxiety in undergraduates. Her main study aims for around 120 participants, but first she runs a pilot with 12.
Her feasibility objectives are concrete. Recruitment: can she enrol 12 eligible students within two weeks via the university participant pool? Instruments: does her anxiety questionnaire take under fifteen minutes, and are any items misread? Adherence: do participants actually use the app, as measured by its built-in usage log? Analysis: can she export the data and run her intended repeated-measures procedure without errors?
The pilot surfaces several issues. Recruitment is slower than hoped (six in two weeks), nudging her into the amber zone and prompting a second recruitment channel. Two participants ask what "keyed" items mean on a reverse-scored question, so she rewords them. The app's usage export uses a date format her software misreads, which she fixes before it can corrupt the real dataset. Crucially, she does not report whether anxiety scores fell; with twelve people the comparison is meaningless, and treating it as a result would mislead her examiners.
The outcome of Aisha's pilot is a revised protocol: two recruitment channels, three reworded items, a corrected data-import script, and a more realistic timeline. That is precisely what a successful pilot looks like, a set of evidence-based refinements that make the main study more likely to succeed.
Analysing and Interpreting Pilot Results
Analysis at the pilot stage is mostly descriptive and diagnostic. You are checking that data behaves as expected, not chasing statistical significance. Useful outputs include counts and percentages (how many were eligible, enrolled and retained), simple summaries of completion times, the proportion of missing or implausible values, and a log of every query participants raised.
Where you do run inferential procedures, do so only to confirm the analysis pipeline functions, and report them as a process check rather than a finding. It can be legitimate to inspect the variability of an outcome to inform a later sample-size calculation, but treat any effect estimate from a tiny pilot with great caution, as confidence intervals will be very wide.
The real interpretive task is turning observations into decisions. Map each result against your pre-set progression criteria, then state plainly what you will change. A short table linking observation, criterion, decision and action is an excellent way to demonstrate disciplined reasoning, and it transfers directly into your methods chapter. Resist the temptation to over-interpret promising trends; the pilot's authority comes from honesty about its limits, not from inflated claims.
Reporting Your Pilot Study
A pilot deserves a clear, structured write-up, whether it sits as a section within your methodology chapter or as a short standalone report. Reporting guidance for randomised feasibility and pilot trials emphasises transparency about objectives, criteria, what happened and what changed, and the same principles serve qualitative and survey pilots well.
Structure your report around six elements. State the aim and feasibility objectives; describe the design and sample (size, recruitment route, inclusion criteria); explain procedures and instruments tested; present results descriptively against your progression criteria; set out the changes made as a result; and acknowledge limitations, including the small sample and the fact that the pilot is not powered for hypothesis testing.
The most valuable and most frequently omitted element is the explicit list of changes. Examiners want to see the line from problem to fix: "completion averaged 38 minutes, exceeding our 20-minute criterion, so we removed two redundant scales." This narrative of refinement is what converts a pilot from a box-ticking exercise into a genuine contribution to your research quality, and it is the part that reviewers and supervisors reward most.