A research paper methodology explains how you answered the question well enough that another researcher could rerun the study and expect the same kind of result. Write it as seven labeled parts in order — design, participants, instruments, procedure, analysis, limitations, ethics — and give every choice a "because" sentence that ties it to your research question.
What the methodology section of a research paper must do
Run the rerun test: a stranger with your paper, your instruments and one free week should be able to repeat your study and get a comparable result. If they would have to email you three questions before starting, the methodology section of a research paper is missing something. Marks live in that test, because the section is where a grader checks whether your findings were earned or merely announced.
The recipe never changes order. Each part feeds the next:
- Design — the approach, named and defended.
- Participants — who took part, how many, how chosen.
- Instruments — the tools the data passes through.
- Procedure — the steps, in the order they happened.
- Analysis — the plan that turns raw data into findings.
- Limitations — the weak spots you own before anyone else finds them.
- Ethics — consent, anonymity and the right to stop.
This page stays on the methods only. The whole-paper walkthrough lives in how to write a research paper, the opening moves in the research paper introduction guide, and the compressed one-paragraph version (which mentions your methods in two sentences) in the research paper abstract. Here we slow down, because the methodology is where most marks are quietly lost.
The methodology skeleton builder
Fill the slots below. The builder composes six labeled paragraphs — Design, Participants, Instruments, Procedure, Analysis, Limitations — and writes a different justification sentence depending on the design you pick. Treat the output as a skeleton to edit, not a finished section: your course details always win.
Methodology skeleton builder
Six slots, one design choice, a rerun-proof draft.
Switching the design rewrites the "because" sentence, so rebuild after any change.
Draft: 0 words — fill a slot or load the example.
Draft outgrew its section? The research paper outline guide sets word budgets for every part of the paper, including the methodology.
How to write a methodology section: one job per paragraph
Graders rarely hand out marks for length; they hand them out for coverage. Every part below earns its own paragraph, and every part contains one sentence of justification of method choice — a "because" that connects the choice to your question. A sharp question makes that sentence easy to write, which is why formats like the PICO(T) question exist: question and method are two ends of the same stick.
- Design. Name the approach and defend it in one sentence. "A survey fits this question because the question asks about habits only participants can report, and identical items keep the answers comparable." That "because" is the paragraph. The rest is naming the design accurately — experiment, survey, interviews, case study or mixed — and it should match what you found in your literature review.
- Participants. Who, how many, how chosen, and — the part students forget — who was left out. "Fifteen final-year nursing students who had completed two clinical placements" already tells a reader more than "students were surveyed" ever will.
- Instruments. Everything the data passes through: the survey and its number of items, the interview guide and its main themes, the observation sheet and its categories, the quiz and its mark scheme. Borrowed instruments get a citation; homemade ones get described item by item.
- Procedure. The steps, in order, so a stranger could follow them. Write them the way a recipe would: booked room, one condition at a time, quiz the next morning. If two readers could reconstruct two different studies from your procedure, it needs another pass.
- Analysis. Numbers get named tests; words get named coding moves. "Independent-samples t-test comparing group means" and "open coding of transcripts into themes" are both data analysis plans. Vague ones — "results were examined" — are where replicability dies.
- Limitations. One to three honest weak spots, each tied to what it touches. More on this below, because it deserves its own section.
- Ethics. Consent, anonymity, the right to stop. Short paragraph, zero drama, always present.
Qualitative vs quantitative methods, and where mixed methods fit
The methodology section changes shape with the design. Same seven parts, different centers of gravity — pick the tab that matches your study:
What changes: the section leans on numbers, groups and measurement. You must say how many, how the groups formed, how each variable was measured, and which statistical test answers the question. Validity (does the instrument measure what it claims) and reliability (would it score the same twice) carry the weight here.
- Design and variables — which condition is tested, which outcome is measured
- Population and sample — size, sampling method, who is excluded
- Instruments — surveys, tests, scales: item counts and scoring
- Procedure — repeatable steps in order
- Analysis — named statistical test, significance level
- Limitations and ethics
What changes: you become the instrument. The section defends the setting, explains how participants were chosen (usually purposively — chosen because they lived the experience), and describes how transcripts or field notes became themes. Trustworthiness replaces the statistics: triangulation, where two data sources or two coders point at the same theme, is your credibility move.
- Approach and rationale — why words beat numbers for this question
- Participants and setting — purposive selection, criteria stated
- Data collection — interviews, observation: length, recording, note format
- Analysis — coding steps, how themes were named and checked
- Trustworthiness — triangulation, member checking where used
- Limitations and ethics
What changes: you now have two strands and one story. Mixed methods sections must say which strand ran first, how the results connect, and why one alone would fail. Methods-heavy fields lean this way — nursing research papers often pair a survey with follow-up interviews for exactly that reason.
- Design type — exploratory (numbers later) or explanatory (numbers first)
- Quantitative strand — compressed version of the quant list
- Qualitative strand — compressed version of the qual list
- Integration point — where the two sets of results meet and what each adds
- Limitations and ethics for both strands
Sampling and data collection: who, how many, how chosen
Sampling is where the rerun test bites hardest, because "students were surveyed" hides every decision that mattered. Name the sampling method and the trade becomes visible:
| Sampling method | How it works | Fits when | Watch out |
|---|---|---|---|
| Random | Every member of the population has an equal chance of being picked. | Surveys and experiments where you can list the whole population. | You need that full list, and classrooms rarely give you one. |
| Stratified | Split the population into groups, sample each group. | Subgroups matter — first-years vs final-years, for example. | You must know the group sizes before you sample. |
| Convenience | Whoever is reachable: your seminar, your dorm, your timeline. | Pilots and course projects. | Say so in limitations — the results may not travel. |
| Purposive | Choose people who lived the exact experience you study. | Qualitative interviews and case studies. | State the selection criteria, or the choice looks arbitrary. |
| Snowball | Early participants recruit the next ones. | Hard-to-reach groups. | The sample grows through friendships, so it clusters. |
Data collection deserves the same precision as sampling. Surveys need item counts and answer scales; interviews need length, recording method and transcription; observation needs the sheet or category list the observer used. One quiet paragraph on each — what the instrument was, who completed it, when — covers validity and reliability without turning the section into a methods textbook.
Research methodology example: the note-taking quiz study
Here is what the builder produces when it is filled with one illustrative study — note-taking and quiz scores. Nothing below is a real experiment or a real dataset; it exists to show the shape:
Design
This study uses a quantitative experimental design. The research question — "Do students who take longhand notes score higher on a follow-up quiz than students who follow the slides only?" — is the reason: an experiment can isolate the effect of one condition by holding everything else steady, and the two groups can be compared on the same measure. Because groups are compared, the procedure must say how participants ended up in each one.
Participants
Sixty undergraduates from two first-year psychology sections, recruited through a class sign-up sheet and assigned by coin flip to the longhand group or the slides-only group. Volunteers, so the sample may lean toward students who enjoy experiments.
Instruments
A ten-minute recorded lecture, a one-page background form, and a twenty-question multiple-choice quiz with a shared answer key. Both groups sat the identical quiz.
Procedure
The study ran in three steps. First, the two conditions met in separate sittings so note-takers could not see the slides-only group. Next, the lecture played once: the longhand group took handwritten notes while the slides-only group followed without writing. Finally, everyone sat the quiz the next morning, with notes handed in the day before.
Analysis
Compare the mean quiz scores of the two groups with an independent-samples t-test, and report the size of the gap rather than only whether it clears the significance line.
Limitations
One lecture and one quiz measure a narrow slice of learning. Volunteers may differ from students at large. The next-morning quiz tests short-term recall only, so nothing here says the effect lasts a semester.
Notice what the example does and does not do. It never claims a finding, never reports a score, and never names a real course or campus — those belong in the results section of a real study. Swap in your own question and the paragraph order still holds.
Methodology limitations: own them before the grader does
Students treat the limitations paragraph as a confession. Treat it as an argument instead: naming a weak spot and saying what it touches proves you understand your design better than the reader does. Three honest limitations beat one vague apology every time.
- Sample limits. "One campus, volunteers only, sixty people" — and therefore the results describe that slice, not every student everywhere.
- Instrument limits. "A single quiz measures recall once" — and therefore the numbers say nothing about learning a month later.
- Design limits. "No follow-up condition" — and therefore the study shows a gap exists, not why.
What does not belong: "the time frame was short" with no consequence attached, or a list of five limitations that includes imaginary ones to look thorough. Each limitation earns its place by naming what it puts out of reach. Graders read this paragraph as a map of your judgment — leaving it blank hands them the pen.
Ethics: consent, anonymity and the right to stop
The ethics paragraph says, in plain language, how the people in your study were treated: that they gave informed consent — they knew what the study was about and what would happen to their data — that responses were kept anonymous or confidential, and that they could stop at any point without consequence. Course rules differ on how much formal process a student project needs, so follow your brief; some modules want the approval route named, others do not. What never changes is the paragraph itself: a methodology section that describes people but skips ethics reads as unfinished, and it is the cheapest paragraph in the section to get right.
If your instruments borrow published scales or the paper quotes sources, cite them the way the rest of the paper does — the APA citation generator handles the formatting so the ethics paragraph can stay about people, not punctuation.
Four methodology mistakes (flip to see the fix)
Method that ignores the question
The question asks "how" or "why", but the data is a rating scale.
The fix
Match the shape: "whether" questions need a comparison, "how" and "why" questions need accounts in words. Re-read the question and name the exact fit in the design paragraph.
Sampling detail missing
"Students were surveyed" — who, how many, chosen how?
The fix
Four facts, no excuses: who, how many, how chosen, who was left out. That fourth one is where the marks hide.
No limitations paragraph
The draft ends at analysis and hopes nobody notices.
The fix
One to three limits, each tied to what it touches. Owning a weak spot earns trust; a grader finding it costs marks.
Ethics skipped
Consent and anonymity treated as somebody else's paragraph.
The fix
Three sentences: informed consent, anonymity or confidentiality, right to stop. Add your course's process only if the brief asks for it.
Methodology checklist: is the section done?
Tap each line as it clears. Six out of six is a submit.
0 of 6 checks
- Design named in one sentence, with a "because" that ties it to the question
- Participants cover who, how many, how chosen — and who was left out
- Every instrument described: item counts, guides, sheets, mark scheme
- Procedure steps written so a stranger could repeat the study
- Analysis plan matches the design — named tests or named coding moves
- Limitations and ethics paragraphs written, not skipped
Quiz: does your method fit your question?
1. The question is "How do first-year students describe their exam-week stress?" Which design answers it?
2. Your question asks whether a flashcard app changes quiz scores. What is the smallest design that can answer it?
3. A study of commuter students surveys only the 20 students in the researcher's own seminar. What is the sampling problem?
4. Which sentence belongs in the limitations paragraph?
5. Which detail belongs in the ethics paragraph of the methodology?
Method section stalling while the deadline closes in?
If the draft keeps missing the rerun test, a quoted, deadline-bound writer can take the methodology — or the whole paper — off your hands and hand back a draft you can defend.
Frequently asked questions
How long should the methodology section be?
Proportional, not padded. In an empirical paper the methodology commonly takes roughly one part in five to one in four of the word count, but your rubric overrides every rule of thumb. What matters is coverage: all seven parts present, each with its justification sentence, and not a word spent defending choices the assignment made for you.
What is the difference between method and methodology?
Methodology is the strategy — the design and the reasoning for it. Methods are the concrete moves: the survey, the sample, the t-test. A methodology section covers both, but the grading eye lands on the reasoning: a "because" sentence per choice is what separates the two on paper.
Can I write "I" in a methodology section?
Style guides differ, so check yours. APA accepts first person; some fields prefer the passive ("participants were assigned"). Pick one register and hold it across the whole section — mixing "I recruited" with "it was decided" in the same paragraph reads as two people writing.
Do I have to name an ethics approval?
Only if your course or field requires one — and then name the actual process your brief describes, never an invented one. For most coursework a short paragraph on informed consent, anonymity and the right to stop is what the grader looks for. Say what you actually did; the paragraph's honesty is the point.
My paper is literature-based — do I still need a methodology?
Many essays have none, and forcing one adds fluff. When the assignment asks how you gathered material — some literature reviews and dissertations do — write a short paragraph on source selection: which databases, which inclusion rules, which date range. Same rerun test, smaller kitchen.
Was this guide useful?