To write the research methodology chapter, describe your design, sampling, instruments, analysis and ethics so another researcher could repeat the study. Then give the reason for each choice from your research questions. Examiners use this chapter to decide whether your findings can be trusted, and most viva questions also come from here. A weak chapter only lists what was done. A good one explains why each choice suited your questions, and it is honest about what the design cannot show.
In this guide we go through what goes inside the chapter and the main research designs, and when each one is suitable. You will also see how to present sampling, instruments and analysis, and how to give reasons for every choice. It is mainly meant for empirical theses in management, social sciences, education, health and engineering. For the methods section of a paper the same ideas apply, only in fewer words. Last reviewed September 2026.
What the methodology chapter has to do
The chapter has two jobs. One is to describe what you did, clearly enough that some other researcher could repeat the same study. The other is to convince the examiner that your method was really capable of answering your research questions.
Most chapters are weak in the second job. Many Indian theses spend pages on textbook definitions, like what is research and what is sampling, followed by types of sampling and types of research design. Examiners don’t give much importance for these pages. What they are searching for is the design you used and the reason it fits your questions. So cut those definition pages and spend the same pages on your reasons.
Usually the chapter is arranged like this.
- Research philosophy or approach (keep it short, and include it only if your field expects it)
- Research design
- Population, sampling frame, sampling method and sample size
- Data sources and instruments
- Pilot study, reliability and validity
- Data collection procedure
- Methods of analysis and tools
- Ethical considerations
- Limitations of the method
In engineering and science theses the chapter is many times called “materials and methods” or “proposed methodology”. In place of sampling and instruments, they have materials, equipment, the experimental set-up, datasets and evaluation metrics. The thinking behind it is still the same.
Choosing and justifying the research design
First comes the research question, and the design comes from it. Suppose your question is asking “how much” or “how many”. Then mostly a descriptive survey will do. When it asks “does X affect Y”, you are looking at a correlational, experimental or quasi-experimental design. And if you want to know “how” or “why” people go through something, qualitative work is suitable.
| Design | Use it when you want to | Typical data | Example |
|---|---|---|---|
| Descriptive (survey) | Describe the level or pattern of something in a population | Questionnaire, secondary data | Awareness of crop insurance among farmers in Thanjavur |
| Correlational / explanatory | Test relationships between variables, often with a model | Questionnaire with validated scales; panel data | Effect of work–life balance on nurses’ intention to quit |
| Experimental | Establish cause and effect by manipulating a variable with random assignment | Measurements from treatment and control groups | Effect of a new teaching method on test scores, with randomised classes |
| Quasi-experimental | Estimate an effect where random assignment is not possible | Pre/post measures, comparison groups | Learning outcomes in schools before and after a state programme |
| Qualitative (case study, grounded theory, phenomenology, ethnography) | Understand meanings, processes or experiences in depth | Interviews, focus groups, observation, documents | How first-generation women entrepreneurs obtain credit |
| Mixed methods | Combine breadth and depth, or use one to build or explain the other | Both of the above, in a planned sequence | Interviews to build a scale, then a survey to test it |
| Design and development (engineering) | Build and evaluate a system, algorithm or material | Simulation, experiments, benchmark datasets | A deep learning model for paddy disease detection, tested on field images |
The complete list of designs, along with the question each one answers, is in our page on types of research design.
Philosophy and approach
In management and social science, examiners many times expect a short statement of your research philosophy (positivist, interpretivist, pragmatist). They also look for your approach, whether deductive or inductive. Many scholars structure it with the “research onion” from Saunders, Lewis and Thornhill’s textbook. One page is enough. No need to list every philosophy there is. Just show the link between how you look at the problem and the design you took. Engineering and natural science theses usually leave this section out.
Justify against the alternatives
Your justification is strongest when you name the obvious alternative and say why you did not go with it. For example, “A cross-sectional survey was chosen because the objective was to estimate prevalence across four districts. With a longitudinal design the claims about change over time would be stronger. It was not feasible within the registration period.” From this the examiner can see you actually weighed the choice. And the viva question on it is already answered before anybody asks.
Population, sampling and sample size
Give exact informations at every step.
- Population: who or what your findings are meant to apply to. Just “women” will not do as a population. “Women members of self-help groups registered under the Tamil Nadu State Rural Livelihoods Mission in Villupuram and Cuddalore districts” is.
- Sampling frame: the real list from which you drew the sample, and how complete that list was.
- Sampling method: either probability (simple random, stratified, cluster, systematic) or non-probability (convenience, purposive, snowball, quota). To choose one and defend it, read our sampling techniques guide. In case you used a non-probability method, write that openly and explain what it means for generalisation. Many scholars are writing “random” for a convenience sample, so don’t do that.
- Sample size: the final number and the way you arrived at it. It may come from a formula, a power analysis, a rule of thumb for your analysis method, or saturation if the work is qualitative. Cochran’s formula, G*Power and the rules followed for SEM are all explained in our guide to sample size calculation.
- Response rate: the number of people you approached, the number who responded, and how many responses were still usable after screening.
In qualitative studies, explain how participants were selected and why they were in a position to speak on your question. Also write how you decided that the number of interviews was enough.
Instruments, reliability and validity
Each instrument needs a description. By instrument we mean the questionnaire, interview guide, test, observation schedule, lab equipment or dataset. With a questionnaire, mention where each scale came from, how many items it has, the response format and whatever changes you made. Maybe you translated a scale into Tamil, Hindi or some other language. Then explain the method (forward and back translation is what most people use) and name who checked it. Put full instrument in an appendix.
Pilot study
Write what the pilot was testing, how big its sample was, and what you changed after it. If the pilot showed nothing to change, write that and tell why.
Reliability and validity
The evidence has to come from your data, and the original author’s figures alone are not enough. With multi-item scales, mostly this is internal consistency, reported as Cronbach’s alpha or as composite reliability in SEM, and 0.70 is the threshold commonly used. SEM studies also need convergent and discriminant validity taken from the measurement model. For content validity, mostly a few experts go through the questionnaire. Describe them broadly and write which items they asked you to change.
Qualitative research uses matching ideas called credibility, dependability and transferability. Write the steps you followed, such as member checking, an audit trail, a second coder or thick description.
For experimental and engineering studies, explain the calibration, replication and controls. The datasets and the baselines you compared against should also be given.
Methods of analysis, ethics and limitations
Methods of analysis
Show which analysis answers which objective. Examiners like a small table for this. They keep it open and check each later chapter against it.
| Objective | Data | Analysis |
|---|---|---|
| 1. To assess the level of digital literacy among rural women SHG members | Survey, 20-item literacy scale | Descriptive statistics; comparison across districts (ANOVA) |
| 2. To examine the effect of digital literacy on use of mobile banking | Same survey, usage items | Logistic regression, controlling for age, education and income |
| 3. To understand barriers to adoption | Interviews with 25 members | Thematic analysis |
Give the software name with its version (SPSS 29, AMOS, SmartPLS 4, R, NVivo, ATLAS.ti, MATLAB, Python with named libraries). Against every test, write why it fits your data and which assumptions you tested. In case you are unsure which test is right, our guide to choosing a statistical test takes you through the decision. If the study has hypotheses, you can list them here itself, or just point to the chapter where they were first stated. How to word them is covered in our guide on writing a research hypothesis.
Ethics
If your research involves human participants, you need approval from your institution’s ethics committee before data collection begins (see research ethics approval). For health research the ICMR National Ethical Guidelines are applicable. In the chapter, give the committee name, the reference number and the date of approval. Write how informed consent was taken and how the data were anonymised and stored. Some studies include vulnerable people, say children or patients. Write what extra protection you gave them. Check the dates once. If approval is dated after your data collection, committees and examiners will surely ask about it.
Limitations
No design is without limits. Write yours in plain words, for example a non-probability sample, self-reported data, only one district, or a cross-sectional design. For each one, say what you did to reduce it and how it restricts what your findings can claim. Many viva questions are about these weak points only. If they are already written honestly, those questions are answered in the chapter itself.
Common mistakes in the methodology chapter, and a final check
- Textbook background. Pages defining research, types of design and types of sampling, never connected to your own study.
- Copying the synopsis. Scholars many times paste the synopsis plan, still in future tense. In the thesis you write what you actually did, and the things that changed after the synopsis also go in.
- Mismatched analysis. For example a parametric test run on ordinal data without any justification. Or a regression with 15 predictors when there are just 60 responses.
- Calling everything “random” though you just met whoever was free. That is convenience sampling and the chapter should call it that.
- Missing details. No dates for the data collection, no response rate, a scale with no source given.
Before giving the draft to your guide, go through this list once.
- Every choice is defended using your research questions. The reason is not that other scholars used it or that you know the software.
- You have stated the population, sampling frame, sampling method and sample size, and also the calculation or rule that gave the number.
- Each instrument has a source. For an adapted scale, you have said what was changed and the reason.
- Reliability and validity figures are from your own data. Quoting only the original author is not enough.
- For each objective there is specific data and one specific analysis matched to it.
- You have described ethics approval, consent and data protection, and given approval numbers wherever they exist.
- Limits of the design are written clearly, along with the steps you took to reduce them.
- The whole chapter is in past tense and reports what you actually did. The synopsis plan is not repeated.
How this chapter connects to the rest of the thesis is in our guide to PhD thesis structure. If you want a methodologist in your field to look at your design and analysis plan, our research methodology consulting is for that. The decisions still remain yours, and you have to defend them in the viva.
Want a specialist to help with this in your own work? Research Methodology Consulting: a free first consultation, then a written scope and fixed quote.
Sources
FAQ
Questions scholars ask
How long should the methodology chapter be?
In a social science or management thesis it is commonly 20 to 40 pages. Engineering theses many times have a shorter chapter, since the methods may be spread over the technical chapters. Detail and reasons should decide the pages. Long definitions only make it look thick.
Do I need a research philosophy section?
Depends on the field. Examiners in management, education and many social sciences often expect a short one. In engineering, natural sciences and medicine it is rarely included. Look at recent theses from your department to see what your examiners are used to.
What if my method changed from what the synopsis proposed?
That is normal. Report what you really did and give a short reason for the change. For a big change, make sure your research committee approved it and that the approval is on record.
Is a convenience sample acceptable in a PhD?
It can be, mainly when there is no sampling frame available. Only thing, you have to name it as convenience sampling and not claim beyond that. Examiners mostly object to two things. One is a non-probability sample written as random, and other is findings stretched to the whole population.
Should I write the methodology in past or future tense?
Past tense in the thesis, as you are reporting work already done. The synopsis is in future tense because at that stage the study is yet to be done.
