Research Methods: Types, Examples and How to Choose the Right One
A clear guide to research methods for students and engineers: qualitative, quantitative and mixed methods, research designs, data collection, and how to write the methods section of a research paper.
Quick summary: Research methods are the techniques you use to collect and analyse data to answer a research question. The three main types of research methods are quantitative (numbers and measurement), qualitative (words, meanings and experiences) and mixed methods (both together). This guide explains each type, the most common research designs, how data is collected, how research works in engineering, the steps of the scientific method, and how to write a strong methods section for a research paper or thesis.
Every research project, from a final year project to a PhD thesis, eventually meets the same question from a supervisor or reviewer: how exactly did you get these results? If the answer is unclear, even good results lose their value.
I have published more than twenty peer-reviewed papers, reviewed for international journals, and supervised student research for over twenty years. The most common reason I see papers rejected is not weak results. It is a weak or poorly explained research method.
This guide explains research methods in plain language, with examples from engineering and science, so you can choose the right method and describe it with confidence.
01 What Are Research Methods?
Research methods are the specific techniques and procedures used to collect and analyse data in order to answer a research question. They include how you gather data, such as experiments, surveys, interviews or simulations, and how you analyse it, such as statistical tests or thematic analysis.
To understand research methods, it helps to start with the word research itself. Research is a systematic investigation to find new facts, test ideas or solve a problem, using evidence that others can check. The key word is systematic. Research follows a planned method, so that someone else could repeat the work and reach the same result.
Most research is also empirical. The empirical method of research means building knowledge from direct observation, measurement or experiment, rather than from opinion or theory alone. When an engineer tests a new controller on a real motor and records its response, that is empirical research.
02 Research Methods vs Research Methodology
These two terms are often used as if they mean the same thing. They do not, and examiners notice the difference.
| Point | Research methods | Research methodology |
|---|---|---|
| What it is | The tools and techniques you use | The overall plan and the reasoning behind your choice of methods |
| Question it answers | How did you collect and analyse data? | Why is this the right way to answer your research question? |
| Example | A survey of 200 engineers; a regression analysis | A quantitative approach, because the aim is to measure and compare |
| Where it appears | The methods section of a paper | The methodology chapter of a thesis |
In short, methodology is the strategy and methods are the tools. A good research methodology explains why the chosen methods are the right ones for the question.
03 Types of Research Methods
There are three main types of research methods. The right one depends on what kind of answer your research question needs.
Quantitative research methods
Quantitative research methods collect numerical data and analyse it with statistics. They answer questions such as how much, how many, how often or how strongly two things are related. Examples include laboratory experiments, structured surveys with rating scales, sensor measurements and simulations. Results are usually shown as tables, graphs and statistical tests.
Qualitative research methods
Qualitative research methods collect non-numerical data, such as words, observations and documents, to understand meanings, reasons and experiences. They answer questions such as why people make certain choices or how a process works in practice. Common types of qualitative methods of research include interviews, focus groups, observation, case studies and document analysis.
Mixed methods research
Mixed methods research combines quantitative and qualitative methods in one study. For example, you might survey 300 students to measure how often they use an online lab, then interview 12 of them to understand why. The numbers show what is happening, and the interviews explain why. Mixed methods take more time, but they often give the most complete answer.
| Point | Quantitative | Qualitative | Mixed methods |
|---|---|---|---|
| Data | Numbers | Words, images, observations | Both |
| Typical sample | Large | Small and focused | Large and small parts |
| Analysis | Statistics | Themes and patterns | Both, then combined |
| Strength | Measurable, comparable, repeatable | Depth and context | Breadth and depth together |
| Limitation | Can miss the reasons behind results | Harder to generalise | More time and skills needed |
04 Types of Research Design
A research design is the overall structure of your study: what you will compare, when you will measure, and how you will control other influences. These are the types of research design you will meet most often.
Experimental
You change one variable and measure its effect on another, while controlling everything else. It is the strongest design for showing cause and effect.
Quasi-experimental
Like an experiment, but without random assignment, for example comparing two existing class sections taught in different ways.
Survey
You collect answers from a sample of people using questionnaires. Survey research methods are useful for measuring opinions, habits and practices across many people.
Correlational
You measure two or more variables to see whether they are related, without changing anything. It shows association, not cause.
Descriptive
You describe a situation or population as it is, such as the current state of automation in local factories.
Case study
You study one case, or a few, in depth, such as a single plant, project or organisation.
Action research
You improve a real practice in cycles of plan, act, observe and reflect. Common in education and workplaces.
Simulation and modelling
You build a mathematical or computer model of a system and test it under different conditions. Widely used in engineering.
Question: Does a fuzzy logic controller keep room temperature more stable than a standard on/off thermostat?
Design: Experimental. The same test room is controlled by each method for ten days each, under similar outside temperatures. Room temperature is logged every minute, and the variation around the set point is compared with a statistical test.
05 Data Collection Methods
Data comes from two broad sources.
- Primary data: data you collect yourself for this study, through experiments, measurements, surveys, interviews, observation or simulations.
- Secondary data: data that already exists, such as published datasets, government statistics, company records or previous studies. A literature review is a form of secondary research.
Whatever the source, record exactly how the data was collected: the instrument, its settings and accuracy, the sample size, the dates and the conditions. That record is what makes your research repeatable.
Ethics come first with human participants. If your research involves surveys, interviews or experiments with people, you will usually need approval from your university's research ethics committee before collecting any data, plus informed consent from every participant.
06 Research Methods in Engineering
Engineering research uses all of the methods above, but some are especially common:
- Experimental testing: building a prototype or test rig and measuring its performance, such as efficiency, accuracy or response time.
- Modelling and simulation: using tools such as MATLAB, Simulink, ANSYS or COMSOL to test designs before building them. Digital twins take this further by linking a model to live data, as explained in the free Digital Twin book series.
- Design, build and test: designing a new system, building it, and evaluating it against requirements. Most final year projects follow this design.
- Statistical and data-driven methods: design of experiments, regression and machine learning to find patterns in measured data. See our guide to artificial intelligence for how machine learning is used in engineering.
- Field studies and surveys: collecting data from real plants, users or engineers, for example to study how industry uses automation or safety practices.
A good engineering study often combines several. In my own SABIC-funded research on an electrohydrodynamic spray system for agriculture, we designed and built the system, ran controlled experiments to measure spray performance, and then optimised the design using the measured data. Each method answered a different part of the question.
07 The Scientific Research Method Steps
Most research, whatever the field, follows the same basic steps of the scientific method.
Identify the problem
Start from a real gap, need or observation. A clear problem leads to a clear study.
Review the literature
Find out what is already known, which methods others used, and where the gap is.
Write the research question or hypothesis
Make it specific and testable. "Is method A more accurate than method B under condition C?" is better than "Study method A".
Choose the research design and methods
Decide on quantitative, qualitative or mixed methods, the design, the sample and the instruments.
Collect the data
Follow your plan carefully and record everything, including problems and changes.
Analyse the data
Use the analysis you planned in advance, such as statistical tests or thematic coding.
Draw conclusions and report
Answer the question honestly, state the limitations, and share the work in a report, thesis or paper.
08 How to Choose the Right Research Method
Answer these five questions before you choose a method:
- What does your question ask? "How much" and "does it work better" point to quantitative methods. "Why" and "how do people experience" point to qualitative methods.
- What data can you realistically get? Access to equipment, participants or datasets often decides the method.
- How much time and money do you have? Large experiments and mixed methods take longer.
- What do others in your field use? Reviewers expect methods accepted in your discipline, unless you explain why you chose differently.
- What skills do you have, or can you learn in time? A method you cannot carry out well is the wrong method.
Start from the question, never from the method. Students often decide "I will do a survey" before they know what they want to find out. Write the research question first, then pick the method that answers it most directly.
09 How to Write the Methods Section of a Research Paper
The methods section of a research paper, or the methodology chapter of a thesis, should give enough detail for another researcher to repeat your study. Include these parts, in this order:
- Research design: the type of study and why you chose it.
- Participants, samples or materials: who or what you studied, how many, and how they were selected.
- Equipment and instruments: devices, software, versions, settings and accuracy, or survey and interview instruments.
- Procedure: step by step, what you did and under what conditions.
- Data analysis: the statistical tests, software or coding approach you used.
- Ethics and limitations: approvals obtained and any known weaknesses of the method.
"An experimental design was used to compare the speed control accuracy of a PID controller and a fuzzy logic controller on a 24 V DC motor. Each controller was tested at five set-point speeds between 500 and 2,500 rpm, with ten trials per speed. Motor speed was measured with an optical encoder (1,000 pulses per revolution) and logged at 1 kHz using MATLAB. Steady-state error and settling time were calculated for each trial and compared using a paired t-test at a 5% significance level."
Notice what makes this methodology example work: the design is named, the conditions and numbers are specific, the instrument is described, and the analysis is stated. A reader could repeat it.
Write the methods section first. It is the easiest part to write while the work is fresh, and it often shows gaps in your data before it is too late to fix them.
10 Common Mistakes to Avoid
Choosing the method before the question
The question decides the method. Picking a favourite method first leads to data that does not answer anything.
Samples that are too small
A quantitative study with very few data points cannot support strong conclusions. Plan the sample size before you start.
Claiming cause from correlation
Two variables moving together does not prove one causes the other. Only a controlled experiment can show cause and effect.
Vague methods sections
"Data was collected and analysed" tells the reader nothing. Give instruments, numbers, settings and tests.
Changing the analysis after seeing the results
Plan your analysis in advance. Trying many tests until one looks significant produces unreliable findings.
Hiding limitations
Every method has weaknesses. Stating them honestly makes your work more credible, not less.
Working on a Thesis or Research Paper?
Dr. Ahsan Rahman offers academic research guidance for undergraduate, postgraduate and doctoral researchers, from choosing a topic and method to preparing publication-ready work.
See Research Guidance →11 Frequently Asked Questions
What are research methods?
Research methods are the techniques used to collect and analyse data to answer a research question. They include data collection methods, such as experiments, surveys, interviews and simulations, and analysis methods, such as statistical tests or thematic analysis.
What are the three main types of research methods?
The three main types are quantitative research methods, which use numbers and statistics; qualitative research methods, which use words, observations and meanings; and mixed methods research, which combines both in one study.
What is the difference between research methods and research methodology?
Research methods are the specific tools and techniques used to collect and analyse data. Research methodology is the overall strategy and the reasoning that explains why those methods are the right way to answer the research question. Methods are the tools; methodology is the plan behind them.
What is mixed methods research?
Mixed methods research combines quantitative and qualitative methods in one study. For example, a researcher might use a survey to measure how common a problem is, then use interviews to understand why it happens. It gives both breadth and depth.
What are the types of research design?
Common types of research design include experimental, quasi-experimental, survey, correlational, descriptive, case study, action research, and simulation or modelling designs. The right design depends on whether you need to show cause and effect, describe a situation, or explore a case in depth.
What are the steps of the scientific research method?
The main steps are: identify the problem, review the literature, write a research question or hypothesis, choose the research design and methods, collect the data, analyse the data, and draw conclusions and report the results.
What should the methods section of a research paper include?
It should include the research design, the participants or samples, the equipment and instruments, the step-by-step procedure, the data analysis methods, and any ethics approvals and limitations. It should be detailed enough for another researcher to repeat the study.
★ Final Thoughts
Research methods are not paperwork that comes after the real work. They are the real work. A clear question, the right method, careful data collection and an honest analysis are what turn effort into knowledge that others can trust and build on.
If you remember one thing, make it this: start with the question, choose the method that answers it most directly, and write down exactly what you did. Reviewers, examiners and future readers will thank you.
One thing to do this week: write your research question in one sentence, then write one paragraph explaining which research method you will use and why. If you cannot explain the "why", that is the part to work on next.