Research Methodology Exam Questions & Answers
Published: Jan 09, 2024 · Last updated: 06 November 2026
Choosing the method is the first part. If the methodology chapter itself needs writing, see our research methodology writing support — written to your university's format, with a free written quote in 1 working day.
Questions on research methodology are where many scholars drop marks they did not expect to lose. Since 2020, our team at Zonduo in Nagercoil has worked with research scholars and postgraduate students through methodology papers, chapter-three reviews and viva preparation. The pattern has been consistent: students rarely lose marks because they do not know a definition. They lose marks because they cannot apply that definition to a situation the examiner has invented.
This guide is built from what we see in that work: the questions that keep coming back, model answers, the mistakes we correct most often in scholars’ drafts, and a simple structure for answering under exam pressure. If you are writing your own methodology chapter, our methodology support is described at the end of the page.
Exploratory research investigates a problem that is not yet clearly defined, usually to develop hypotheses rather than test them. Descriptive research documents the characteristics of a population or phenomenon without establishing cause. Explanatory research tests cause-and-effect relationships between variables.
A research design is the overall plan for conducting a study — specifying what data will be collected, from whom, by what method, and how it will be analysed to answer the research question. It is decided before data collection begins.
A research method is the specific technique used to collect or analyse data — a survey, an interview, a regression analysis. Methodology is the broader justification for why those methods were chosen and how they fit the research question.
In probability sampling, every member of the population has a known, non-zero chance of being selected — random, stratified, cluster and systematic sampling all fall here. In non-probability sampling, selection depends on judgement or convenience, so results cannot be statistically generalised to the wider population.
Stratified sampling divides the population into subgroups (strata) that share a characteristic, then samples randomly within each subgroup. This ensures each group is represented proportionally, which simple random sampling cannot guarantee.
Sampling error is the difference between a sample statistic and the true population value, arising because a sample is only part of the population. It decreases as sample size increases and is distinct from non-sampling errors such as measurement or response bias.
Primary data is collected first-hand by the researcher for the specific study — through surveys, interviews, observation or experiments. Secondary data already exists, having been collected by someone else for a different purpose, such as census records or published studies.
A Likert scale measures attitudes or opinions using ordered response options, typically five or seven points ranging from strong disagreement to strong agreement. It produces ordinal data, which affects which statistical tests can legitimately be applied.
Closed-ended questions offer fixed response options and produce quantifiable data that is easy to analyse. Open-ended questions allow respondents to answer in their own words, producing richer qualitative data that takes longer to code and interpret.
The null hypothesis states that there is no effect or no relationship between the variables being studied. The alternative hypothesis states that there is one. Statistical testing determines whether there is sufficient evidence to reject the null — it never proves the alternative.
A Type I error is rejecting a null hypothesis that is actually true — a false positive. A Type II error is failing to reject a null hypothesis that is actually false — a false negative. Reducing one generally increases the risk of the other.
A p-value of 0.05 means there is a 5% probability of observing results at least as extreme as those found, assuming the null hypothesis is true. It is a conventional threshold for significance, not a measure of how important or how large an effect is.
Validity is whether an instrument measures what it claims to measure. Reliability is whether it produces consistent results when repeated under the same conditions. An instrument can be reliable without being valid — consistently measuring the wrong thing.
Internal validity concerns whether the study genuinely establishes the causal relationship it claims, free from confounding variables. External validity concerns whether those findings can be generalised beyond the specific sample, setting and time period studied.
When we go through past question papers with scholars, nearly every question falls into one of four types. Identifying the type tells you how much to write and what the examiner is looking for.
Example: “What is stratified sampling?” These usually carry a few marks. Give one precise sentence and a one-line example. A longer answer rarely earns more.
Example: “Differentiate between validity and reliability.” These are the pairs that blur together in answers. State each concept in one contrasting sentence, then give a single example that shows both at work.
Example: “A researcher wants to study job satisfaction among nurses in three hospitals. Which sampling technique should she use?” This is where scholars drop the most marks in our experience. Name your choice, justify it with the details given in the question, and state one limitation.
Example: “Evaluate the limitations of questionnaire-based research.” Examiners want both sides: what the method does well, where it fails, and how a careful researcher reduces the weakness.
These come from the answers and methodology chapters scholars bring to us. Almost none of them is about forgetting a definition. They are about reasoning.
A scholar writes that “the methodology is a questionnaire”. A questionnaire is a method. Methodology is the reasoning for why that method suits the research question. Examiners notice this quickly.
A significance test can only reject the null hypothesis or fail to reject it. Writing that the data proves the alternative hypothesis shows that the logic of testing has not been understood.
Likert responses are ordinal. Many studies still report means and run parametric tests on them. That is not automatically wrong, especially for composite scales, but the answer must acknowledge the issue and justify the choice, or use a non-parametric test instead.
“We collected 100 responses” is a number, not a justification. Say whether it comes from a sample-size formula, a power analysis or a published guideline for the technique, and state the size of the population.
If an instrument is described as reliable, the answer should say how that was checked, for example Cronbach’s alpha for internal consistency, with 0.70 as the conventional minimum.
More respondents reduce sampling error. They do nothing about bias caused by who was asked or how the question was worded.
The structure we use has four steps: name the choice, justify it from the scenario, state the limitation, and say what you would do if the ideal option were not available. The example below was written for this guide to show the structure.
A researcher wants to study job satisfaction among staff nurses in three hospitals of one district. Which sampling technique is appropriate? Justify your answer.
Proportionate stratified random sampling, with each hospital treated as a stratum. The population divides naturally into three hospitals of different sizes, and job satisfaction may differ between them, so simple random sampling across all nurses could under-represent the smallest hospital. Within each hospital, nurses would be selected randomly from the staff list in proportion to its nursing strength. The main limitation is that it needs an up-to-date staff list for every hospital. If access is refused, purposive sampling is the fallback, but the results could then not be statistically generalised.
Why this answer scores: it names a technique, uses the details in the scenario (three hospitals, different sizes), states a limitation honestly, and gives a fallback.
A viva is different from a written paper. Examiners stop asking what a concept is and start asking why you made a particular choice in your own study. These are the five we prepare scholars for most often.
Answer from your research question, not from preference. Say what a different design would have given up.
State the basis, the population size and the response you actually received. If you fell short of the target, say so and explain the effect.
Describe the pilot study, who reviewed the items, and the reliability coefficient you obtained.
Link it to the type of data and the assumptions the test requires, and say how you checked them.
Name two or three honestly. Examiners trust a candidate who names limitations before being asked.
For a deeper look at justifying your choices, read our guide on choosing the right research methodology for the thesis, or see how we support thesis writing.
Reading these five is not the same as answering them about your own study. We build a mock viva from your actual methodology chapter — your design, your sample size, your instrument, your statistical test — and put the questions your examiner is most likely to ask, so the first time you hear them is not in the room.
Send us your chapter and we reply with a free written quote within one working day. See how our methodology support works →
For a deeper look at justifying your choices, read our guide on choosing the right research methodology for the thesis, or see how we support thesis writing.
Most fall into four types: definitions, distinctions between paired concepts such as validity and reliability, application questions that give a scenario and ask for a justified choice, and critique questions that ask for the strengths and limitations of a method. Application and critique questions carry more marks and need reasoning, not recall.
It depends on the marks. As a working guide, a two-mark question needs one or two precise sentences, a five-mark question needs a definition, an explanation and an example, and a ten-mark question needs a structured answer with justification and at least one limitation.
No. A written exam tests whether you know the concepts. A viva asks why you made specific choices in your own study, such as your design, sample size, instrument and statistical test, and expects you to defend them and name their limitations.
Reading model answers helps for the exam. Writing and defending your own methodology is a different task, and it is where most scholars come to us. Since 2020 we have supported PhD, MPhil and postgraduate scholars with:
Send us your topic or chapter on WhatsApp. We reply with a free written quote within one working day, and you pay nothing in advance.
WhatsApp Us for a Free Quote Call +91 87547 14627Choosing an appropriate research methodology for research paper is the most importation for academic writings. To know more about choosing of methodology for your research