How to Write Research Methodology Chapter 3 for a Business Research Proposal
Quick answer: To write research methodology Chapter 3, choose one research paradigm and explain its ontology, epistemology and axiology. Then justify a design that answers your research questions, describe the context and access, define your respondents and sample size, and show how your instrument will be piloted and tested for validity and reliability. Every choice must follow from the one before it.
That last sentence is where most marks are won or lost. Learning how to write research methodology Chapter 3 is less about definitions and more about consistency. This guide uses the five-part structure that many Malaysian Masters programmes now set (paradigm, design, context, respondents, instrument), including Sunway University’s RSM5224 Assessment 3 research methodology brief.
Short on time before your deadline? A research methods expert can check your paradigm and sampling plan before you write. Send your brief for a free review and quote.
What a research methodology Chapter 3 must show
A methodology chapter in a proposal is a plan, not a report, so it is written in the future tense: “data will be collected”. Markers look for three things: justification, alignment and feasibility. Justification means each choice is backed by a methods source. Alignment means the paradigm, design, sample and instrument fit together and fit the research questions from Chapter 1. Feasibility means one student can actually carry out the plan.
| Section | Core question it answers | Must include | Typical length (2,000-word chapter) |
|---|---|---|---|
| 3.1 Research paradigm | What do I assume about reality and knowledge? | Named paradigm, ontology, epistemology, axiology, why not the alternatives | About 400 words |
| 3.2 Research design | How will I answer my questions? | Approach, method choice, strategy, time horizon, link to RQs and analysis | About 400 words |
| 3.3 Research context | Where, and why there? | Industry, location, problem relevance, feasibility, access | About 400 words |
| 3.4 Choice of respondents | Who, how many and how? | Population, inclusion criteria, sampling, recruitment, sample size, ethics | About 400 words |
| 3.5 Research instrument | With what, and is it any good? | Instrument source, administration, pilot, validity, reliability | About 400 words |
In my experience of reviewing these chapters, the strongest ones are not the longest. They are the ones where a reader could guess section 3.5 just from reading section 3.1.
Step 1: Choose and justify your research paradigm
A research paradigm is the set of beliefs about reality, knowledge and values that shapes how a study is designed. In business research, most proposals choose positivism, interpretivism or pragmatism. Pick the one your research questions already imply, and do not choose a paradigm first and bend the questions to fit it.
| Positivism | Interpretivism | Pragmatism | |
|---|---|---|---|
| Ontology (what is real?) | One measurable reality | Multiple, socially constructed realities | Reality judged by practical consequences |
| Epistemology (what counts as knowledge?) | Observable, measurable data; generalisation | Participants’ meanings and interpretations | Whatever best answers the question |
| Axiology (role of values) | Researcher stays detached | Researcher’s values are acknowledged | Values drive the choice of question |
| Usual RQ wording | “Does X affect Y?” | “How do managers experience X?” | “What is happening and why?” |
| Usual methods | Surveys, experiments | Interviews, case studies | Mixed methods |
Saunders, Lewis and Thornhill’s Research Methods for Business Students (9th edition, 2023) also covers critical realism and postmodernism. You can mention them, but most 400-word sections only need to compare two or three paradigms.
How to write ontology, epistemology and axiology without sounding like a textbook
Write one definition sentence and one application sentence for each. For example: “Ontologically, this study treats customer trust as a real attitude that can be measured with established scales.” That second sentence is the one the marker is looking for. Students often define all three terms well and never mention their own study, which earns very little credit under Analysis.
Step 2: Build a research design that follows from the paradigm
Research design is the overall plan linking your questions to data and analysis. Describe it in layers: approach (deductive or inductive), method choice (quantitative, qualitative or mixed), strategy (survey, experiment, case study, interviews) and time horizon (cross-sectional or longitudinal).
A positivist study testing hypotheses from Chapter 2 is usually deductive, quantitative and survey-based, and cross-sectional because of the time available. An interpretivist study is usually inductive and uses semi-structured interviews or a case study. One mistake loses more marks than any other: a positivist paradigm paired with “exploring lived experiences”. Markers spot it straight away.
Add a small table that maps each research question to the data needed and the analysis planned, such as regression or PLS-SEM for surveys, or thematic analysis for interviews. It takes 60 words of table space and answers the “data collection and data analysis” requirement that most briefs include. If SPSS or SmartPLS is unfamiliar, plan the analysis now. Our SPSS homework help for survey analysis covers this stage when you reach it.
Step 3: Describe the research context and access
The context section answers “why here?” Name the industry, the location in Malaysia (for example the Klang Valley, Penang or Johor) and the type of organisation, and link them to the problem from Chapter 1. Then deal with feasibility. How will you reach respondents? Do you need gatekeeper permission from an HR manager? What will you do if a company says no?
Honest limits help here. Saying that the findings may not transfer to manufacturing or to the public sector shows judgement. It does not count against you.
Step 4: Define respondents and justify the sample size
This is the section where students lose the most marks, because sample size is usually stated without any reasoning. Cover population, inclusion criteria, sampling technique, recruitment, sample size and ethics, in that order.
G*Power or Krejcie and Morgan: which should you use?
For a survey that tests relationships, use a G*Power a priori power analysis as your main justification. Krejcie and Morgan’s 1970 table estimates a proportion for a known population size. It does not tell you how many responses you need to detect an effect, so treat it as a benchmark, not the main argument.
| Tool | Best for | Worked example | Watch out for |
|---|---|---|---|
| G*Power (Faul et al., 2007) | Hypothesis testing (regression, t-tests, ANOVA) | Multiple regression, 2 predictors, f² = 0.15, α = .05: n = 68 at power .80; n = 107 at power .95 | Report every input; the number of predictors changes the result |
| Krejcie and Morgan (1970) table | Descriptive surveys with a known population size | Very large population, 95% confidence, ±5% margin: n = 384 | Assumes random sampling; often misused for non-probability samples |
| Saturation | Qualitative interviews | Stop when new interviews add no new themes; plan a range and a stopping rule | Say how you will judge saturation, not only that you will reach it |
For non-probability samples, which are common because Malaysian firms rarely share staff lists, Memon et al. (2020) is a useful, freely available review of sample size rules for survey research. Then add response-rate planning: if you need 200 usable responses, you will probably need to invite more than 200 people.
Ethics belongs in this section
Cover informed consent, voluntary participation, anonymity, data storage and Malaysia’s Personal Data Protection Act 2010 (Act 709). Also state that you will get approval from your university’s ethics committee before collecting data. Where a brief maps a learning outcome to ethical conduct, a single sentence on ethics is a visible gap.
Not sure your sample size will hold up? We can run your G*Power inputs with you and check your inclusion criteria. Get Chapter 3 support from a business research expert.
Step 5: Explain the research instrument, pilot test and data quality
For surveys, adapt published scales rather than writing your own items. A tested scale brings its validity evidence with it, which is far stronger than items you wrote the night before. Give a table of construct, source, number of items and response format. Keep the original response format where you can. Turning a 7-point scale into a 5-point one without a reason weakens the link to the original validation.
- Content validity: ask two or three academics or practitioners to review the items before the pilot.
- Pilot test: about 30 eligible respondents is a common, defensible size. Remove pilot responses from the main data set.
- Reliability: Cronbach’s alpha and composite reliability of .70 or above are usually treated as acceptable.
- Construct validity: for PLS-SEM, plan to report outer loadings, AVE and HTMT (Hair et al., 2022).
- Common method bias: keep predictors and outcome in separate sections, guarantee anonymity, and plan a statistical check.
For interviews, describe the interview guide, how it maps to your research questions, a pilot interview, and how you will show trustworthiness (credibility, dependability, transferability, confirmability) in place of reliability and validity.
Research methodology Chapter 3 checklist for maximum marks
- Paradigm, approach and design all point the same way
- Each research question is mapped to data and analysis
- Ontology, epistemology and axiology are each applied to your own study
- Inclusion criteria are specific and checkable
- Sample size is shown with its inputs (G*Power) or a saturation rule
- Ethics covers consent, anonymity, storage, PDPA 2010 and approval
- Instrument sources are cited, and the pilot, validity and reliability are planned
- At least 8–12 methodology sources cited in APA 7
- Each section is close to its word allocation; tables count towards the total if the brief says so
Want to see the checklist applied? Our model Chapter 3 methodology extract (for reference only) shows a full paradigm table, a G*Power calculation and an instrument table with APA 7 references. If your proposal is part of a longer final-year project, our dissertation helpers in Malaysia and thesis writing support for Masters and PhD students cover the later chapters too.
Before you submit your research methodology Chapter 3
Read your Chapter 3 backwards: start at the instrument and ask whether it matches the paradigm you claimed in 3.1. If it does not, fix 3.1, because changing the instrument at this stage costs far more time. That one check is the quickest way to improve how you write research methodology Chapter 3 under deadline pressure. If you would like an expert to check that alignment, or need guidance on a specific section, our team works from your own brief and chapters. We provide model answers and structured guidance for your own writing, and your files stay confidential.
Your Chapter 3, checked by a research methods expert. Paradigm alignment, G*Power inputs, ethics and instrument design reviewed against your brief. Request help now.
About the expert
Amirul Azam holds an MBA from the University of Malaya. His profile lists marketing, management, organizational behaviour and international business, over a decade in managerial roles in marketing and advertising, six years of academic assistance and 2,341 completed orders.
Frequently asked questions
Can I use pragmatism if I’m only doing a survey?
You can, but it is harder to defend. Pragmatism is usually chosen to justify mixing methods. If your design is a single quantitative survey testing hypotheses, positivism fits more naturally and is easier to align with your design and analysis. Choose pragmatism only if your questions genuinely need both numbers and explanations.
Is 384 respondents always the right sample size?
No. The figure 384 comes from the Krejcie and Morgan table for a very large population at 95% confidence and a 5% margin of error. It suits descriptive surveys with random sampling. For hypothesis testing, a G*Power analysis based on your number of predictors usually gives a smaller, better-justified minimum.
Do I write Chapter 3 in the past or future tense?
In a proposal, use the future tense, for example “Data will be collected through an online questionnaire”. Change to the past tense only in a completed dissertation, after the data have actually been collected. Mixing tenses suggests the chapter was copied from someone else’s finished thesis.
How many people do I need for a pilot test?
About 30 eligible respondents is common and defensible for a Masters survey pilot, as it gives a usable first estimate of Cronbach’s alpha and shows up confusing items. Say who the pilot respondents are, and state that their answers will not be included in the main data set.
Can I copy questionnaire items from published scales?
You can adapt published scales with citation, and that is good practice. Cite the original authors and explain any wording changes. Put full item lists in an appendix where allowed, and describe them in your own words in the main text. This keeps your Turnitin similarity down.
What if my research is qualitative?
Use interpretivism, an inductive approach and semi-structured interviews or a case study. Choose information-rich participants through purposive sampling, and justify numbers with a saturation stopping rule. Replace reliability and validity with trustworthiness criteria, and pilot your interview guide before the main fieldwork.
References
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
Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2022). A primer on partial least squares structural equation modeling (PLS-SEM) (3rd ed.). SAGE.
Krejcie, R. V., & Morgan, D. W. (1970). Determining sample size for research activities. Educational and Psychological Measurement, 30(3), 607–610. https://doi.org/10.1177/001316447003000308
Memon, M. A., Ting, H., Cheah, J.-H., Thurasamy, R., Chuah, F., & Cham, T. H. (2020). Sample size for survey research: Review and recommendations. Journal of Applied Structural Equation Modeling, 4(2), i–xx. https://doi.org/10.47263/jasem.4(2)01
Saunders, M. N. K., Lewis, P., & Thornhill, A. (2023). Research methods for business students (9th ed.). Pearson.
