# How to choose a research method for your dissertation

**How do I choose a research method for my dissertation?** To choose a research method for your dissertation, start from the research question. Questions about how much, how many or whether one thing affects another suit quantitative methods; questions about how and why people experience something suit qualitative ones; questions that need both suit mixed methods. Then test the method against your time, access, ethics approval and skills.

Published 2026-09-26 by EdCitation. https://edcitation.com/newsletter/choosing-a-research-method-for-your-dissertation

Choosing a research method for your dissertation is a matching problem: the question decides the kind of evidence, the evidence decides the method, and your time, access and ethics approval decide which version of that method is possible.

This guide, published by EdCitation, sits under [How to write a dissertation](https://edcitation.com/newsletter/how-to-write-a-dissertation). We read the guidance on 26 September 2026 at nine universities' own pages and handbooks: Sheffield Hallam, UCL, Sunderland, the West of Scotland, Lincoln, Birmingham, Leeds and Manchester in the UK, and National University in the US. None sets one method or one sample size for every dissertation.

EdCitation never writes any part of a dissertation or chooses its method. It does the looking up: [Find sources](https://edcitation.com/) finds published studies that used a method on your topic, and the methods literature, and [Cite a source](https://edcitation.com/cite) builds each methods book into a reference from its ISBN or DOI.

## How do I choose a research method for my dissertation, step by step?

Start from the question, then let practical limits narrow the options. National University Library (2026) puts the first rule plainly: choose the method that best aligns with your research question, after seeing what studies on your topic have used.

1. Write the research question first; [how to write research questions and aims](https://edcitation.com/newsletter/how-to-write-research-questions-and-aims) covers that step.
2. Decide what evidence would answer it: numbers, people's words, documents, or published studies.
3. See how others answered similar questions: search your topic in [Find sources](https://edcitation.com/) and note which [research methods](https://edcitation.com/glossary/research-design#research-method) the studies and earlier theses used. UCL Library Services (2026) suggests earlier dissertations for exactly this.
4. Shortlist two or three designs from the table below.
5. Test each against time, access, ethics and your skills, and drop the ones that fail.
6. Read the standard methods text for the survivor, and decide how many participants, cases or documents you need, with a reason.
7. Take the choice to your supervisor with your dissertation handbook, whose requirements [Check your paper](https://edcitation.com/check) reads into a checklist, free.

## Which research method suits which research question?

Each common design answers one kind of question well and others badly.

| Method | The question it suits | Data | Typical analysis | What it needs |
| --- | --- | --- | --- | --- |
| [Survey](https://edcitation.com/glossary/quantitative-designs#survey-research) | How common is something, and what goes with what? | Numbers from questionnaires | [Descriptive](https://edcitation.com/glossary/statistics#descriptive-statistics) and [inferential statistics](https://edcitation.com/glossary/statistics#inferential-statistics) | A reachable population, a tested instrument, enough responses |
| [Experiment](https://edcitation.com/glossary/quantitative-designs#experimental-study) or [quasi-experiment](https://edcitation.com/glossary/quantitative-designs#quasi-experimental-design) | Does one thing change another? | Measurements under conditions you set | Comparison of groups, [regression](https://edcitation.com/glossary/statistics#regression-analysis) | Control over the conditions, a [power analysis](https://edcitation.com/glossary/sampling#power-analysis) |
| [Secondary data analysis](https://edcitation.com/glossary/data-collection#secondary-data-analysis) | What do existing records show? | A published dataset | Statistics on data someone else collected | Access to the dataset and its documentation |
| [Semi-structured interviews](https://edcitation.com/glossary/data-collection#semi-structured-interview) | How do people experience or explain something? | Transcripts | [Thematic analysis](https://edcitation.com/glossary/qualitative-methods#thematic-analysis) and similar | Participants who agree, time to transcribe |
| [Focus groups](https://edcitation.com/glossary/data-collection#focus-group) | How does a group talk about a shared issue? | Group discussion transcripts | Thematic analysis | Several people free at once, a moderator |
| [Case study](https://edcitation.com/glossary/research-design#case-study-research) | How or why did something happen in this setting? | Interviews, documents, observation of one or a few cases | Within-case and cross-case analysis | Access to the case, a reason for choosing it |
| [Ethnography](https://edcitation.com/glossary/qualitative-methods#ethnography) | How does a group live or work, seen from inside? | Field notes, observation, conversations | Interpretive analysis | Long access to a setting, more ethics scrutiny |
| [Content](https://edcitation.com/glossary/qualitative-methods#qualitative-content-analysis) or [document analysis](https://edcitation.com/glossary/data-collection#documentary-sources) | What do texts, policies or media say, and how? | Documents already in existence | Coding, counting or interpretation | A defined body of documents |
| [Systematised](https://edcitation.com/glossary/evidence-synthesis#systematised-review) or [scoping review](https://edcitation.com/glossary/evidence-synthesis#scoping-review) | What does the published research show? | Published studies | Screening, extraction, synthesis | A search strategy and databases; no participants |

These are typical matches, not rules. The University of the West of Scotland Library (2026) advises checking which methods are prominent in your subject first.

### Qualitative, quantitative or mixed: the first fork

The first decision is the kind of evidence. Sheffield Hallam University Library (n.d.) separates [quantitative](https://edcitation.com/glossary/research-design#quantitative-research) work, which uses numbers and statistical tests to establish relationships, from [qualitative](https://edcitation.com/glossary/qualitative-methods#qualitative-research) work, which seeks in-depth understanding from non-numerical data. Creswell and Creswell (2022) build a whole textbook on choosing among the three approaches by fit with the question.

[Mixed methods](https://edcitation.com/glossary/mixed-methods#mixed-methods-research) earns its extra work only when one kind of evidence cannot answer the question. Creswell and Plano Clark (2017) name three core designs: [convergent](https://edcitation.com/glossary/mixed-methods#convergent-design), [explanatory sequential](https://edcitation.com/glossary/mixed-methods#explanatory-sequential-design) and [exploratory sequential](https://edcitation.com/glossary/mixed-methods#exploratory-sequential-design). In a short dissertation, that is two datasets in the same weeks.

### Case study: when it is the right choice

A case study suits a question about something happening in a real setting that you cannot control. SAGE's description of Yin (2017), the standard text, stresses real-world contexts, several sources of evidence and events the researcher cannot control. The University of Manchester Global Development Institute (2023) handbook asks a student using one to explain why the approach fits, why one case rather than several, why this case, and how the information was gathered; merely saying you will do a case study is not enough.

## What practical limits should decide the method?

Time, access, ethics approval and skill decide which matching method you can actually carry out. Sheffield Hallam calls the choice a matter of "balancing realism and ambition": interviews give very rich data but take long to transcribe and analyse, while surveys reach more people with less detail.

### Time and access

Count the calendar, not only the word limit. The Manchester handbook describes a taught master's dissertation completed in three months, and warns that the primary data a hypothesis needs may not be collectable. Its marking criteria ask whether the writer shows "an appreciation of data access difficulties".

### Ethics approval

Ethics can rule a method out. On the University of Leeds Libraries Blog (Skills@Library, 2022), a student describes planning interviews with charity workers on sensitive topics, learning that approval could take up to eight weeks, and switching to a desk-based dissertation. The advice: raise ethics early, and try the ethics form against several methods. [Research ethics approval for a dissertation](https://edcitation.com/newsletter/research-ethics-approval-for-a-dissertation) covers the committees' rules.

### Skills

Pick a method you can learn in the time. A regression needs statistics, interviews need interviewing and coding, a review needs database searching. The West of Scotland and UCL both point students to the SAGE Research Methods collection.

## How many participants does a dissertation need?

Enough to answer the question, argued rather than guessed. No university page we read set a number, so it comes from your handbook, supervisor and the methods literature, and your [sample size justification](https://edcitation.com/glossary/sampling#sample-size-justification) says how.

For interviews, Hennink and Kaiser (2022) reviewed 23 studies that tested [saturation](https://edcitation.com/glossary/qualitative-methods#data-saturation): those using empirical data reached it within 9 to 17 interviews or 4 to 8 focus group discussions, mostly with fairly similar participants and narrow aims. Studies across countries or after deeper meaning needed more.

Not everyone accepts saturation as the test. Braun and Clarke (2021b) argue that it fits coding-reliability forms of thematic analysis but not [reflexive thematic analysis](https://edcitation.com/glossary/qualitative-methods#reflexive-thematic-analysis), where meaning is made in interpretation and "how many" cannot be settled wholly in advance. For reflexive work, [information power](https://edcitation.com/glossary/sampling#information-power) or a plain account of what the data had to do fits better.

For a quantitative design, a [power analysis](https://edcitation.com/glossary/sampling#power-analysis) sets the number from the expected effect, the test and the error rates. Faul et al. (2007) describe G*Power, a free program that runs these calculations for the common tests. If the sample was set by what you could reach, say so.

## Can a literature review or secondary data be the dissertation's method?

Yes, where your programme allows it, and it removes recruitment and often the ethics wait. The Manchester handbook allows desk-based analysis of published data, case studies and policy documents. University of Lincoln Library (2025) calls the [systematised review](https://edcitation.com/glossary/evidence-synthesis#systematised-review), which uses parts of the systematic review process without the full team and time, a typical postgraduate assignment that spares a student data collection and ethics delays.

It is not the easy option. University of Birmingham Library Services (2026) says to confirm the exact requirements with the programme lead, and recommends Boland, Cherry and Dickson's book for master's students doing reviews. [How to do a systematic review](https://edcitation.com/newsletter/how-to-do-a-systematic-review) and [how to do a scoping review](https://edcitation.com/newsletter/how-to-do-a-scoping-review) cover the procedures.

## How do I justify the method I chose?

By showing that it answers this question better than the alternatives, within the limits you had. The Manchester handbook frames it as questions: why interviews suit the topic, who was interviewed and why, what was asked, and how the answers were used. Its marking criteria ask whether the methods follow logically from the literature review.

Each reason that rests on someone else's work cites it, so the reader can judge, as University of Sunderland Library (2026) puts it, whether your decisions were suitable. [How to write a methodology chapter](https://edcitation.com/newsletter/methodology-chapter-how-to-write-it) covers the chapter itself; this guide stops at the choice.

## Which methods textbooks should I cite, and how?

The standard text for your method, in the edition you read. Each edition below was confirmed on the publisher's own page.

| Method | Standard text | Edition at the publisher |
| --- | --- | --- |
| Choosing among approaches | Creswell and Creswell, *Research Design* | 6th, SAGE, October 2022 |
| Mixed methods | Creswell and Plano Clark, *Designing and Conducting Mixed Methods Research* | 3rd, SAGE, August 2017 |
| Case study | Yin, *Case Study Research and Applications* | 6th, SAGE, September 2017 |
| Thematic analysis | Braun and Clarke, *Thematic Analysis: A Practical Guide* | SAGE, October 2021 |
| Business and management | Saunders, Lewis and Thornhill, *Research Methods for Business Students* | 9th, Pearson, March 2023 |

Reading lists lag behind editions. The Manchester handbook lists Saunders and colleagues' 7th edition of 2016; Pearson's current one is the 9th (Saunders et al., 2023). The Leeds post names Bryman's 5th edition of 2015; the record Cite a source read has a 2021 sixth by Clark, Foster, Sloan and Bryman. Cite the edition you actually read.

### What Cite a source built, and what to correct

We ran [Cite a source](https://edcitation.com/cite/apa) on each text on 26 September 2026. From its DOI, Braun and Clarke's 2006 paper came back exactly right:

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. *Qualitative Research in Psychology*, *3*(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

The books, found by ISBN in Open Library, needed work:

- **Yin (9781506336169):** year, edition and publisher right, and the title in sentence case.
- **Creswell and Plano Clark (9781483344379):** the record split the surname, giving "Clark, V. L. P." for "Plano Clark, V. L.", and named no edition: add (3rd ed.).
- **Saunders and colleagues (9781292402727):** the entry came back as *Saunders research methods* (9th ed.), the catalogue's shorthand for the title; use *Research methods for business students*.
- **Clark and colleagues (9780198796053):** the entry came back as *Social research methods* (6th ed.), in sentence case; check the title against the title page. Two earlier tries returned "The source registry did not respond"; the third worked.

## How EdCitation helps you choose and cite a method

EdCitation is the best tool for the reading behind a method choice: every source it returns comes from a publisher's record, and it never writes a reference, something a chatbot cannot guarantee. A methods justification rests on its citations, and one invented reference undoes it.

[Find sources](https://edcitation.com/) is free and needs no account. It covers about 300 million published works, takes a topic or the claim a sentence must support, and narrows by kind of source, years and open access. What it returned on 26 September 2026:

- "using thematic analysis in psychology", by citations: 12,156,684 works, led by Braun and Clarke's 2006 paper, cited 158,667 times.
- "mixed methods research design", books only, by citations: 47,603, led by Morse's *Mixed Method Design* (2016), cited 204 times. Creswell and Plano Clark's textbook was not in the first 100; find it by ISBN in Cite a source.
- A claim, "thematic analysis is a flexible method for identifying and reporting patterns in qualitative data", by citations: 11,763,760 works, led by a 2020 paper applying it in education; the 2006 paper was not among the 95 shown. Search a title to find a source.
- "nursing students experiences of clinical placement", filtered to qualitative studies: the search reported the filter as not applied, which the page explains as the full index being busy, and returned the same 6,257,489 results as without it. The results were relevant; the filter did nothing that day.

[Verify references](https://edcitation.com/verify-references), free, checks the finished list: verified, doubtful ("check this") or not found, retractions flagged, and "could not check" never shown as not found. We entered Yin and Creswell and Plano Clark with the year 2018 as a test: both came back verified, with a note that the library catalogue dates them 2017 and a prompt to check the title page. The publishers' pages give 2017.

[Check your paper](https://edcitation.com/check) read 1,344 words of the Manchester handbook's methodology guidance into one rule and set 25 sentences aside as not checked. The rule was the word count, 10,800 to 16,500, from "12,000-15,000 words (plus or minus 10%)", the same range the handbook gives for a dissertation that will be marked. The limit of five pages of appendices was set aside for you to check, and the chapter's three parts (research questions, aims and objectives, research design) were not listed as headings. Each rule shows its sentence so you can check it.

On Max ($24 a month), Theoretics QA judges the chapter against the handbook's instructions, with quotations from the chapter, and rewrites nothing. Before the draft goes to your supervisor, [References from a file](https://edcitation.com/tools/references-from-a-file) pairs each citation in the text with its entry in the list; a free account gets three checks a month and Pro ($8 a month, see [pricing](https://edcitation.com/pricing)) removes the limit. None of these tools chooses your method or promises a mark.

## Quick questions

### Should my dissertation be qualitative or quantitative?

Whichever kind of evidence answers your question. Questions about how many or whether one thing affects another need numbers; questions about how people experience something need their words.

### How many interviews do I need for a dissertation?

Your handbook and supervisor decide, and you justify the number. Hennink and Kaiser (2022) found saturation within 9 to 17 interviews among fairly similar participants; Braun and Clarke (2021b) argue saturation does not suit reflexive thematic analysis.

### Can I do a dissertation without collecting data?

Often, if your programme allows it. Manchester's development handbook allows desk-based analysis, and Lincoln's library calls the systematised review a typical postgraduate assignment.

### Which methods textbook should I cite?

The standard text for your method, in the edition you read. EdCitation's [Cite a source](https://edcitation.com/cite) builds each from its ISBN and sets the title in sentence case; check the edition, authors and title.

### Can EdCitation choose my research method?

No. EdCitation never writes or decides any part of a dissertation. [Find sources](https://edcitation.com/) shows which methods published studies on your topic used; your supervisor helps you choose.

## References

- Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. *Qualitative Research in Psychology, 3*(2), 77–101. [https://doi.org/10.1191/1478088706qp063oa](https://doi.org/10.1191/1478088706qp063oa)
- Braun, V., & Clarke, V. (2021a). *Thematic analysis: A practical guide*. SAGE Publications. [https://www.sagepub.com/en-gb/eur/thematic-analysis/book248481](https://www.sagepub.com/en-gb/eur/thematic-analysis/book248481)
- Braun, V., & Clarke, V. (2021b). To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. *Qualitative Research in Sport, Exercise and Health, 13*(2), 201–216. [https://doi.org/10.1080/2159676X.2019.1704846](https://doi.org/10.1080/2159676X.2019.1704846)
- Creswell, J. W., & Creswell, J. D. (2022). *Research design: Qualitative, quantitative, and mixed methods approaches* (6th ed.). SAGE Publications. [https://collegepublishing.sagepub.com/products/research-design-6-270550](https://collegepublishing.sagepub.com/products/research-design-6-270550)
- Creswell, J. W., & Plano Clark, V. L. (2017). *Designing and conducting mixed methods research* (3rd ed.). SAGE Publications. [https://www.sagepub.com/en-us/nam/designing-and-conducting-mixed-methods-research/book241842](https://www.sagepub.com/en-us/nam/designing-and-conducting-mixed-methods-research/book241842)
- Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. *Behavior Research Methods, 39*(2), 175–191. [https://doi.org/10.3758/BF03193146](https://doi.org/10.3758/BF03193146)
- Hennink, M., & Kaiser, B. N. (2022). Sample sizes for saturation in qualitative research: A systematic review of empirical tests. *Social Science & Medicine, 292*, Article 114523. [https://doi.org/10.1016/j.socscimed.2021.114523](https://doi.org/10.1016/j.socscimed.2021.114523)
- National University Library. (2026, August 24). *Research methodology*. Research Methodology and Design. [https://resources.nu.edu/methods](https://resources.nu.edu/methods)
- Saunders, M. N. K., Lewis, P., & Thornhill, A. (2023). *Research methods for business students* (9th ed.). Pearson. [https://www.pearson.com/nl/en_NL/higher-education/subject-catalogue/business-and-management/saunders-research-methods-for-business-students-9ed.html](https://www.pearson.com/nl/en_NL/higher-education/subject-catalogue/business-and-management/saunders-research-methods-for-business-students-9ed.html)
- Sheffield Hallam University Library. (n.d.). *Methodology and methods*. Dissertations and research projects. [https://libguides.shu.ac.uk/researchprojects/methodolgy-methods](https://libguides.shu.ac.uk/researchprojects/methodolgy-methods)
- Skills@Library. (2022, December 8). *Research methods and approaches*. University of Leeds Libraries Blog. [https://leedsunilibrary.wordpress.com/2022/12/08/research-methods-and-approaches/](https://leedsunilibrary.wordpress.com/2022/12/08/research-methods-and-approaches/)
- UCL Library Services. (2026, August 25). *Research methods*. Support for dissertations and research projects. [https://library-guides.ucl.ac.uk/dissertations/research](https://library-guides.ucl.ac.uk/dissertations/research)
- University of Birmingham Library Services. (2026, January 28). *Systematic and other reviews*. [https://libguides.bham.ac.uk/asc/systematicreviews](https://libguides.bham.ac.uk/asc/systematicreviews)
- University of Lincoln Library. (2025, October 9). *Introduction*. Writing a systematised review in business. [https://guides.library.lincoln.ac.uk/c.php?g=716013](https://guides.library.lincoln.ac.uk/c.php?g=716013)
- University of Manchester Global Development Institute. (2023). *GDI dissertation handbook: Taught masters programmes in 2023-24* [PDF]. School of Environment, Education and Development. [https://documents.manchester.ac.uk/display.aspx?DocID=27174](https://documents.manchester.ac.uk/display.aspx?DocID=27174)
- University of Sunderland Library. (2026, September 22). *Writing your dissertation: Methods/methodology*. Study skills: Dissertations. [https://libraryguides.sunderland.ac.uk/Dissertations/Method](https://libraryguides.sunderland.ac.uk/Dissertations/Method)
- University of the West of Scotland Library. (2026, September 16). *Research methods*. Dissertation support. [https://uws-uk.libguides.com/c.php?g=730589&p=5327108](https://uws-uk.libguides.com/c.php?g=730589&p=5327108)
- Yin, R. K. (2017). *Case study research and applications: Design and methods* (6th ed.). SAGE Publications. [https://www.sagepub.com/en-us/nam/case-study-research-and-applications/book250150](https://www.sagepub.com/en-us/nam/case-study-research-and-applications/book250150)