Nursing master's thesis: from research questions to discussion

Have a topic for your nursing master's thesis, but are unsure what to ask patients or how to write up their answers? Start with what you want to find out. Then decide whether a questionnaire, interviews or records can help you answer that question. This guide follows a question about patients' preparation for self-care through to analysis, discussion and conclusions that address it.

You may already have an approved topic, several articles and access to potential participants, yet still be unsure where to begin. In that situation, I suggest first writing down what you actually want to find out and which data could answer that question. This is easier than adapting your aim to a questionnaire that is already finished. Below, I explain the process through a project on patients' preparation for self-care after discharge.

My name is Wioletta Boćkowska, and I have run StudentSupport since 2018. I help plan master's theses, explain research methods and develop chapters. You can also contact me for a model academic text in the field of Nursing. I first read your topic, university requirements and existing work. We then agree what I will prepare, the price and the deadline.

Where should you start with a nursing master's thesis?

First decide exactly what you want to examine, and whose experience interests you. “Patient health education” is still a very broad topic. You could focus on how adults who have had planned surgery rate their preparation for self-care. Decide when to ask them: on the day of discharge, or some time after they return home? Clarify what preparation means too. A patient may know the recommendations yet feel unsure about following them. Knowledge, confidence and experiences of receiving information call for different questions.

Do not assume that every master's thesis requires the same design or a particular number of participants. First, check your programme's current guidance and what you have agreed with your supervisor. Separate formal requirements from methodological decisions: the number of chapters does not determine whether you need a questionnaire, interviews or document analysis. Feasibility also depends on access to material, time, recruitment conditions and any approvals required.

Before meeting your supervisor, put your working topic, aim, main question, data collection plan and biggest expected difficulty on one page. That difficulty might be having no way to contact patients after discharge. Try to explain in a few sentences how your plan will answer the question. If you cannot, discuss that point before adding more items to the questionnaire.

The nursing thesis guidelines from the John Paul II Catholic University of Lublin provide one example of a distinction between bachelor's and master's work. They distinguish a case-based bachelor's thesis from an empirical master's thesis and list sample description and statistical methods among the latter's methodological elements. These are one university's requirements, rather than a rule for every programme. This is why simply expanding a bachelor's patient account does not automatically produce a suitable master's project. First, compare the type of work you plan with your institution's current requirements.

For a first-cycle thesis, see the guide Nursing case study for a bachelor's thesis: patient information, nursing problems and conclusions.

How do you connect your aim with research questions?

The aim explains what you want to learn; the research questions break that task into manageable parts. In this example, you might want to describe how patients rate their preparation for self-care and what they find difficult after discharge. Your questions would therefore concern understanding recommendations, receiving information or getting help. A general question about satisfaction with the hospital would not answer that aim.

Choose the question you genuinely want to focus on. Asking which recommendations are hardest to understand calls for a description of the answers. Examining the relationship between support and preparation means deciding how to measure both and assess their association. Exploring life after returning home requires room for patients to tell their stories. You do not have to combine all three tasks in one thesis. Each additional method brings decisions you need to justify.

Be careful with the word “effect”. A one-off survey about support and perceived preparation will generally not establish that one caused the other. The question and title can refer to associations, differences or assessment. If your project includes hypotheses, formulate them before analysis and align them with the question. Do not add a hypothesis after seeing the results and present it as an earlier assumption.

What is the theoretical chapter for?

Choose theory that helps explain what you will study. For a self-care thesis, define self-care, explain the role of patient education and consider what may make recommendations difficult to follow. Readers can then see why you ask about those issues. Avoid extending the account of illnesses simply to make the chapter longer. Keep the information needed to understand your patient group.

For each source, note how you will use it: for a definition, a rationale, an instrument description or a comparison of findings. Distinguish methodological publications from studies involving a particular patient group. Check whether the studies you compare actually measured the same thing. Knowledge of recommendations, satisfaction with information and readiness for self-care are not interchangeable concepts.

The bibliography below includes the STROBE authors' publication. Its use here is specific: it helps assess the transparency of an observational study report. The authors explain that it is neither a recipe for designing a study nor a quality assessment scale. It therefore does not replace nursing literature relevant to your topic.

A questionnaire, interviews or patient records?

Consider what your chosen method will actually tell you. “I understand the recommendations” reports a patient's assessment. It does not show whether they can perform the activity described. If you want to test knowledge, justify the questions and scoring. If you want to know what patients do at home, decide how you will obtain that information. Asking whether they were satisfied with the instructions will not answer it.

Interviews allow participants to explain difficulties in more detail: an unclear phrase, conflicting information or the challenge of fitting recommendations around everyday responsibilities. However, the number of such accounts should not be presented as the prevalence of a problem across the whole population. Document analysis, meanwhile, answers questions about what was recorded and how, rather than everything that happened during contact with a patient.

Decide what you will count as well. One person may have several hospital admissions, and one admission may generate several documents. Ten records do not necessarily mean ten patients. Explain whether you are analysing people, admissions or documents. Do not treat records relating to the same person as responses from independent participants.

How should you prepare the instrument and methodology?

For each research question, write down what you need to ask participants. “Access to support” might mean being able to telephone staff, receiving help from a relative or feeling that someone is available to help. These are different things. Combine them in one item and, once the questionnaires are back, you will not know which one a patient meant when choosing an answer.

If you choose an existing scale, check its purpose, language version, conditions of use, scoring instructions and basis for interpretation. Do not independently change its items or response options and then describe it as an unmodified standardised instrument. A questionnaire you develop yourself needs a rationale for its scope, an account of its construction and checks of comprehensibility. Including closed questions does not, by itself, create a validated scale.

The methodology should make your procedures understandable: who was eligible, how people were invited, when data were collected, what the exclusion criteria were and how incomplete responses were handled. Piloting can reveal unclear questions, but it does not replace full validation. Describe what you actually did, including changes made after the pilot.

What should you establish before collecting data?

Distinguish permission to enter a ward from permission to use material in a thesis. Establish with your supervisor and the relevant institution which approvals, participant information and data handling arrangements apply to your project. Do not copy another university's procedure as though it applied everywhere. If a required decision is still pending, do not describe it in the methodology as already obtained.

Plan to limit the data you collect. An unusual occupation, an exact surgery date and a small town could identify someone even after their name has been removed. When requesting feedback, send material without identifying details. A blank form, a variable description and a short, appropriately anonymised extract are often enough to assess a chapter's construction.

Recruitment affects interpretation. Patients who have the energy and time to respond may differ from those who do not participate. Document the route from invitation to analysis: how many people were invited, how many responded and which data were ultimately used. Missing information also matters when describing the material.

How do you plan the analysis before reporting results?

Before calculating results, decide how to enter the responses in your dataset. If patients can select several difficulties, the percentages may add up to more than 100%. For a single-choice question, state how many people answered it: that is the number used to calculate percentages. Do not turn an empty field into a “no”. Record missing answers separately and apply the same rule throughout.

Describing the sample and response distributions is the starting point. Test selection depends, among other things, on variable types, group structure, independence of observations and the assumptions of the chosen method. The name of a popular test is not sufficient justification. For a more complex plan, consider Statistical analysis for bachelor’s and master’s theses, sharing your questions, instrument and data structure before calculations begin.

For the question about support and preparation for self-care, first explain how you assessed each. You can then examine their association. Consider other possible explanations: a patient who has followed similar recommendations before may feel better prepared. If you did not ask about previous experience, you cannot include it in the calculations afterwards. Also report the estimated strength of the association and its uncertainty. Calling a result “significant” or “not significant” alone says little about its meaning.

How do you write the results chapter?

Arrange the chapter so that each section answers a research question. Explain what each result concerns, how many people or records it includes and what the number means. Avoid repeating every value in turn. Point out the main similarities, differences or difficulties in the responses. Follow your university's presentation requirements and avoid explaining the same result several times.

Instead of writing “patients were well prepared”, state what was assessed, by whom and on what basis. A high self-rating does not demonstrate correct performance of every activity. Similarly, the absence of a statistically significant difference does not prove that groups are identical. If a result differs from your expectations, report it accurately and check data quality rather than removing an inconvenient section.

In qualitative analysis, distinguish a participant's account from your interpretation. A quotation should support a specific argument, have an anonymised identifier and preserve the speaker's meaning. Look also for experiences that do not fit the dominant account. This allows the chapter to show variation in the material rather than selected confirmations alone. One approach is the Framework Method described by Gale and colleagues. Using it requires assessing its suitability for the question and material, rather than simply naming it.

How do you connect results, discussion and conclusions?

In the discussion, return to your main question and explain how your answer relates to earlier research. Before describing findings as similar, check who took part, which instrument was used and when they were asked. Someone answering on discharge day has not yet had the same experience as someone who has tried following the recommendations at home. Even similar numbers may describe different situations.

Explain what may have shaped the answers too. A confident person might rate both their preparation and the support received highly. Another might avoid criticism because they worry about staff reactions. You do not need to resolve every possibility, but explain why it matters for your conclusions. If you mention a small sample, say what it makes difficult, such as comparing groups or applying the findings to a wider patient group.

Use the conclusions to answer your opening questions briefly. If you collected accounts from patients at one institution, write about that group rather than all patients in Poland. You might suggest further investigation of a particular aspect of patient education. Do not present that suggestion as an effective solution when it was not evaluated in your study. Conclusions should follow from what you were able to establish.

How do you check the thesis as a whole?

Finally, try reading the thesis from the conclusions backwards. Find the result supporting each conclusion, then check which question and data produced that result. If a connection is missing, locate the problem. Sometimes a more cautious conclusion or a clearer explanation of the analysis is enough. In other cases, you need to revisit the research question.

• Does the title reflect the problem actually studied?
• Do the same concepts retain the same meaning throughout?
• Is the origin and coverage of the data clear?
• Does the analysis plan fit the questions and instrument?
• Does the discussion compare comparable studies?
• Do the conclusions respect limitations and avoid claiming an untested effect?

When you change the main question, revisit the methodology, results and conclusion too. Editing the introduction alone is not enough. Reading this way helps identify which passages need work and whether the whole thesis needs restructuring. For help with the next step, consider Help with your thesis methodology chapter or Thesis research section: methodology, results and discussion, depending on whether you are still planning the study or already have findings and need to develop the remaining chapters.

Keep brief notes of important decisions with your supervisor: the date, what changes and why. If you are asked to narrow the group, clarify whether recruitment is still being planned or responses have already been collected. In the latter case, explain which data you included in the analysis. Do not present the later removal of inconvenient responses as a rule set before the study. Keep the original data file and clean and code a copy. In the methodology, describe the final rules and important departures from the plan. Readers can then understand why counts differ between sections and which questionnaire version the findings concern.

An example of moving from a question to a chapter

Return to the question: which recommendations do patients find hardest to understand? A general satisfaction rating cannot answer it. You need separate questions about types of recommendations relevant to your patient group and thesis topic. Check the literature before writing those items. Remember that not every patient should have received every recommendation on your list.

Allow a “not applicable” response where appropriate. Someone who did not receive a particular recommendation should not be forced to rate its clarity. At the same time, not receiving information could be a separate research issue. In the results, therefore, distinguish the availability of information from its assessment. This prevents two different experiences being combined into one apparently simple measure.

You could therefore start the results by showing which recommendations patients received. Next, explain how they rated the clarity of those recommendations and what difficulties they reported. Follow a similar sequence in the discussion: consider possible reasons for those difficulties and compare studies that asked similar questions. Do not attribute problems to a lack of staff commitment if you did not examine how education was provided. Patients' answers alone do not establish that accusation.

Another option is to leave numerical ratings aside and focus on accounts of experience in interviews. You would then need questions that invite participants to describe receiving and using information, followed by analysis using a selected, justified qualitative method. The result would not rank the whole population. It could show how participants understand recommendations in the context of their lives. This change must be reflected in the aim, methodology, results structure and conclusions.

This plan gives you more to discuss with your supervisor than a chapter list alone. Show what you want to ask, how you will collect answers and what you can learn from them. If the feedback says “clarify the methodology”, ask whether that means concepts, patient selection, the questionnaire or calculations. Record the agreed change and check what it affects in other chapters.

If you need help with several chapters or a complete text, consider Bachelor’s and master’s thesis writing. We will agree whether I should prepare a model master's thesis, selected chapters or revisions to an existing draft. Before quoting a price, I will review the topic, requirements and data you can use.

Methodological sources

von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P., STROBE Initiative (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies. PLOS Medicine, 4(10), e296. DOI: 10.1371/journal.pmed.0040296.

Gale N.K., Heath G., Cameron E., Rashid S., Redwood S. (2013). Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Medical Research Methodology, 13, 117. DOI: 10.1186/1471-2288-13-117. This publication shows that choosing qualitative analysis also requires consideration of its application and limitations.

John Paul II Catholic University of Lublin (n.d.). Wytyczne przygotowania pracy dyplomowej (licencjackiej i magisterskiej) na kierunku PIELĘGNIARSTWO [Guidelines for preparing bachelor's and master's nursing theses], section I. University document, accessed 4 November 2026. The document does not state a publication year.