What Is Research? Definition, Types, and Process

Research is careful, systematic work done to add to what we know, or to find new uses for what we already know. It is not the same as looking things up.

The same word covers a physicist probing the nature of matter, a nurse testing a new routine on a ward, and an engineer improving a battery. What ties them together is not the topic or the equipment. Each frames questions so that evidence can answer them, and each shares the answers so others can check them.

The Official Definition of Research

Governments that measure research spending use a standard definition: creative and systematic work done to increase the stock of knowledge.1NCSES. Definitions of Research and Development: An Annotated Compilation of Official Sources That knowledge explicitly includes knowledge of humankind, culture and society, not only the natural sciences.2NCSES. Definitions of Research and Development: An Annotated Compilation of Official Sources The same definition also counts work that devises new applications of knowledge already available.3NCSES. Definitions of Research and Development: An Annotated Compilation of Official Sources So history, sociology and economics count as research alongside chemistry and biology.

Both qualities, creative and systematic, are required. Creative means the work aims at something new rather than repeating a known result. Systematic means it follows a planned method that someone else could follow and check.

Comparing phone reviews online is gathering existing information; nothing new comes out of it. A classroom “research paper” that summarizes sources is a literature review, a real and useful skill but a different one. Formal research adds something checkable: new data, a new analysis, or a new application. A literature review crosses that line when it systematically combines studies to reach a conclusion no single study had reached.

Basic, Applied, and Development Research

Research is commonly sorted by what it is for. The standard categories are:

  • Basic research: experimental or theoretical work to understand the foundations of how things work, without any particular use in view.4Wikipedia. Frascati Manual
  • Applied research: also produces new knowledge, but is aimed mainly at a specific practical goal.5Wikipedia. Frascati Manual
  • Experimental development: systematic work that builds on research and experience to produce new products or processes, or to improve existing ones.6Wikipedia. Frascati Manual

Take one subject, rust on bridges. Studying how metal corrodes just to understand it is basic research. Testing which coating keeps bridges from rusting is applied research. Turning a lab-proven coating into a factory product is experimental development.

The categories overlap more than a list suggests. One well-known model maps research on two separate scales: how much it seeks understanding, and how much it considers use.7PubMed Central. At the nexus of science, engineering, and medicine: Pasteur’s quadrant reconsidered Pure understanding is nicknamed Bohr’s quadrant, after the physicist Niels Bohr, and pure use is Edison’s quadrant.8PubMed Central. At the nexus of science, engineering, and medicine: Pasteur’s quadrant reconsidered Research that seeks deep understanding while driven by a practical problem is called Pasteur’s quadrant, or use-inspired basic research.9PubMed Central. At the nexus of science, engineering, and medicine: Pasteur’s quadrant reconsidered The name fits, because Louis Pasteur uncovered basic biology of microbes while trying to solve problems like spoiled food and disease.

Basic research is often dismissed as useless because it has no goal at the start. Yet many practical technologies later grew from it, so its value often appears decades afterward.

Exploratory, Descriptive, and Explanatory Research

Another way to sort research is by the kind of question it asks.

Because each type has a different aim, choosing one decides much of how the project is designed.13Research Methods for the Social Sciences: An Introduction. 3.2 Exploration, Description, Explanation Exploratory projects often check whether a bigger study is feasible and get the lay of the land.14Scientific Inquiry in Social Work. What Is Research? Definition, Types, and Process

Skipping breakfast shows how that looks. An exploratory study might interview a few students about why they skip it. A descriptive study would then survey a whole school to record how many skip it and when. An explanatory study would test whether skipping breakfast lowers test scores, and why.

The same logic helps you pick a type for your own project. If you can’t yet list the likely factors, start exploratory. If you need a count or a profile, choose descriptive. If you have a specific cause in mind to test, choose explanatory. Mixing up the purpose leads to questions your data cannot answer, such as trying to explain “why” with a survey built only to count.

The Steps of a Research Project

Most projects begin with a question, often drawn from gaps in what earlier studies have found. The question is often sharpened into a hypothesis, a prediction that evidence could support or contradict, though exploratory and many qualitative studies start without one. A testable prediction can claim two things are linked, or that changing one thing causes a change in another.15Journal of Korean Medical Science. A Practical Guide to Writing Quantitative and Qualitative Research Questions and Hypotheses in Scholarly Articles Researchers often state a “null hypothesis” that there is no relationship, then see whether the data rule it out.16Journal of Korean Medical Science. A Practical Guide to Writing Quantitative and Qualitative Research Questions and Hypotheses in Scholarly Articles

Before measuring anything, researchers define each concept in clear, precise terms, a step called conceptualization.17Research Methods in Criminology. Research Design Part 1 – Measurement, Conceptualization and Operationalization Then comes operationalization, which means deciding exactly how a concept will be measured or changed in the study.18bookdown.org. Chapter 7 Formulating Research Questions and Hypotheses | Quantitative Research in Mass Communications An abstract idea becomes concrete indicators that can actually be observed.19Research Methods in Criminology. Research Design Part 1 – Measurement, Conceptualization and Operationalization

Take “stress.” First you define it as the body’s and mind’s response to pressure. Then you choose indicators, such as a stress questionnaire score or heart rate. Different choices measure slightly different things, so results depend on them. That is why readers of a study should check how its key words were measured.

Next comes collecting data. In an experiment, researchers assign people or things to different conditions, ideally at random. Observational studies record what happens without assigning conditions, and many are designed like experiments in every way except randomization.20PubMed Central. The Relative Merits of Observational and Experimental Research: Four Key Principles for Optimising Observational Research Designs In interview-based studies, data collection often continues until new interviews stop revealing new themes, a point called saturation.21Field Methods. Quantifying Thematic Saturation in Qualitative Data Analysis

Results are then analyzed, written up, and shared so others can check and build on them. When many studies exist, a systematic review gathers and combines them using a reporting checklist called PRISMA, updated in 2020.22BMJ. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews

The process is a loop more than a line. Findings usually raise new questions that start the next project.

Qualitative, Quantitative, and Mixed Methods

Quantitative research measures things as numbers and uses statistics to test predictions.23Journal of Korean Medical Science. A Practical Guide to Writing Quantitative and Qualitative Research Questions and Hypotheses in Scholarly Articles Qualitative research works with interviews, observation and texts to understand experiences and meanings. Common qualitative approaches include narrative research, phenomenology, grounded theory, ethnography and case study.24SAGE Publications. Introducing Qualitative Designs

  • Grounded theory: builds a new theory from the data itself rather than testing one decided in advance.
  • Phenomenology: describes what an experience is like for the people living it.
  • Ethnography: studies a group’s culture through long immersion in it.

Choosing between the two starts with what you need to know: “how many” or “what is it like.” Numbers show how common something is; interviews show why people act as they do. Qualitative studies use fewer participants but get much more detail from each.

Mixed methods projects can run numbers first then interviews, interviews first then numbers, or both at once side by side.25PubMed Central. Achieving Integration in Mixed Methods Designs—Principles and Practices Interviews can explain a surprising survey result, or shape the questions a later survey asks. Results are often brought together in a “joint display,” a table placing both kinds of findings side by side.26PubMed. The concept of integration in mixed methods research: a step-by-step guide using an example study in physiotherapy

How to Tell Whether Research Is Good

Reliability means a measuring tool gives consistent results. Validity means it measures what it claims to, in other words whether the findings hold up.27PubMed Central. Internal, External, and Ecological Validity in Research Design, Conduct, and Evaluation Internal validity asks whether a study’s design and analysis answer its question without bias.28PubMed Central. Internal, External, and Ecological Validity in Research Design, Conduct, and Evaluation External validity asks whether the findings apply to other people and settings.29PubMed Central. Internal, External, and Ecological Validity in Research Design, Conduct, and Evaluation Both are often overlooked; external validity suffers, for example, when a sample is too small to represent a larger population.30PubMed. Research Reliability and Validity: Why do they matter?

A tool can have one quality without the other. A bathroom scale that always reads 2 kg heavy is reliable but not valid.

Money matters too. Industry-funded research tends to favour topics tied to products that can be sold.31PubMed Central. The Influence of Industry Sponsorship on the Research Agenda: A Scoping Review Corporate interests can steer research away from questions that matter most for public health.32PubMed Central. The Influence of Industry Sponsorship on the Research Agenda: A Scoping Review Checking who funded a study is a reasonable first step when judging it.

When you meet a study in the news, a few quick checks help. Look at how many people were studied and who they were. Ask whether it was an experiment or just observed a link. Check whether it was reviewed by experts and who paid for it. And be wary when one study is reported as settling a question.

Ethical Rules for Studying People

The Belmont Report sets out the basic ethical principles for medical and behavioural research involving human participants.33PubMed. The Belmont Report. Ethical principles and guidelines for the protection of human subjects of research Its three principles are respect for persons, beneficence (minimizing harm and maximizing benefit), and justice in who bears the burdens and benefits. Institutional Review Boards use these principles to approve or reject studies before they begin.34PubMed. The Belmont Report. Ethical principles and guidelines for the protection of human subjects of research

For a participant, this means giving informed consent and being free to withdraw at any time. Personal data must be kept confidential. Respect for persons also gives extra protection to children and others less able to consent. Student projects involving interviews or surveys usually need ethics approval first.

How Research Gets Checked and Shared

Before publication in a journal, other experts in the field review a study for flaws, a process called peer review. Open peer review can mean naming reviewers, publishing their reports, or letting more people take part.35PubMed Central. What is open peer review? A systematic review

Expert checking catches many errors and weak reasoning before publication. But reviewers usually don’t redo the experiment, so some mistakes slip through. A preprint is shared before that checking happens and should be treated as preliminary.

Not all knowledge comes through formal science. Indigenous knowledge built over centuries offers deep understanding of local ecosystems, and combining it with scientific methods has helped in conservation, farming and health. Fair collaboration requires respecting communities’ ownership of their knowledge.