Understanding the different types of research is one of the most important steps in planning a successful project. Whether you are writing a research proposal, preparing for your upgrade or designing your methodology chapter, knowing which approach fits your study will shape every decision you make.
The challenge is that research is classified in many different ways. You will encounter terms like qualitative, quantitative, experimental, descriptive, exploratory and longitudinal, and it is not always obvious how they relate to each other or which ones apply to your work.
This guide breaks down every major type of research, organised by the dimension used to classify it. Each type includes a clear definition, when to use it and a real-world example. By the end, you will have a solid understanding of the research landscape and be better equipped to choose the right approach for your own study.
What is research?
Research is the systematic process of investigating questions, testing hypotheses and generating new knowledge. It involves collecting, analysing and interpreting data to answer a specific question or solve a defined problem.
In an academic context, research forms the foundation of every PhD thesis. Your choice of research aims and objectives will directly influence which types of research you use, how you collect data and how your findings are presented.
If you are new to the concept, our guide on what is research and its purpose covers the fundamentals in more detail.
How types of research are classified
There is no single way to categorise types of research methods. Instead, researchers classify research across several dimensions, and most projects combine multiple types of research methodology simultaneously. For example, a study might be applied, quantitative, experimental and cross-sectional all at once.
The main classification dimensions are:
- By purpose – theoretical or applied
- By depth of scope – exploratory, descriptive, explanatory or correlational
- By data type – qualitative, quantitative or mixed methods
- By variable manipulation – experimental, non-experimental or quasi-experimental
- By inference method – deductive, inductive or hypothetical-deductive
- By time period – longitudinal or cross-sectional
- By information source – primary or secondary
- By data collection method – documentary, field, laboratory or mixed
The table below provides a quick-reference summary of the main types of research across all dimensions.
Comparison of types of research
| Type | Category | Data Used | Best For | Example |
|---|---|---|---|---|
| Theoretical | Purpose | Secondary/documentary | Generating new theories | Philosophical analysis of educational frameworks |
| Applied | Purpose | Primary and secondary | Solving practical problems | Market research informing product development |
| Exploratory | Depth | Qualitative or mixed | Investigating new or poorly understood topics | Pilot study on social media and self-image |
| Descriptive | Depth | Quantitative or qualitative | Documenting characteristics of a phenomenon | Census data analysis of demographic patterns |
| Explanatory | Depth | Quantitative | Establishing cause-and-effect relationships | Testing material behaviour under load |
| Correlational | Depth | Quantitative | Identifying relationships between variables | Examining links between sleep and academic performance |
| Qualitative | Data type | Non-numerical | Understanding meaning and experience | In-depth interviews with PhD students |
| Quantitative | Data type | Numerical | Measuring and generalising | Statistical analysis of survey responses |
| Mixed methods | Data type | Both | Combining breadth and depth | Survey followed by focus group interviews |
| Experimental | Variables | Controlled | Testing hypotheses under controlled conditions | Randomised controlled drug trial |
| Non-experimental | Variables | Observed | Studying phenomena without intervention | Observing teaching methods in classrooms |
| Quasi-experimental | Variables | Partially controlled | Testing with existing groups | Evaluating a training programme across departments |
| Longitudinal | Time | Repeated measures | Tracking changes over time | 10-year cohort study on career outcomes |
| Cross-sectional | Time | Snapshot | Capturing data at one point | National student satisfaction survey |
Types of research by purpose
Research classified by purpose falls into two broad categories: theoretical and applied.
Theoretical research
Theoretical research (also called basic, fundamental or pure research) focuses on generating knowledge for its own sake, without immediate concern for practical application. The goal is to expand understanding, develop new theories or refine existing ones.
This type of research relies heavily on documentary analysis, mathematical models and critical reflection. It is particularly common in philosophy, theoretical physics and pure mathematics.
Example: A PhD student in philosophy might conduct theoretical research exploring new frameworks for understanding consciousness, drawing on existing literature and logical analysis rather than empirical data collection.
Theoretical research often lays the groundwork that applied research later builds upon.
Applied research
Applied research aims to solve specific, practical problems. It uses theory-driven knowledge to address real-world challenges and is especially common in STEM fields, business and healthcare.
Applied research has two main subdivisions:
- Technological applied research improves efficiency in productive sectors by enhancing processes, tools or machinery
- Scientific applied research measures variables to predict behaviours, often used in commercial contexts such as consumer research and project viability assessments
Example: A biomedical engineering PhD might develop and test a new prosthetic device, applying principles from materials science and biomechanics to solve a specific clinical problem.
Types of research by depth of scope
The depth of scope refers to how deeply a study investigates its subject. There are four main types.
Exploratory research
Exploratory research is a preliminary investigation into a topic that is not yet well understood. It establishes reference frames, identifies key variables and generates hypotheses for deeper study. Exploratory research tends to rely more on observing patterns in data than on testing existing theory.
This type is common at the early stages of a PhD, particularly when defining the scope and delimitations of your project.
Example: A psychology PhD student might conduct exploratory research into how short-form video content affects adolescent self-perception, using focus groups and open-ended interviews to identify themes before designing a larger study.
Descriptive research
Descriptive research documents the characteristics of a phenomenon without investigating why it occurs. The researcher observes and records without intervening, which is critical for preventing behavioural changes that could distort the findings.
Descriptive studies often use surveys, observational protocols and existing datasets.
Example: A social sciences PhD might analyse public census data to compare the demographic profiles of elected officials in urban and rural constituencies, describing patterns without attempting to explain their causes.
Explanatory research
Explanatory research (also called causal research) is one of the most common types of research in doctoral study. It establishes cause-and-effect relationships, allowing findings to be generalised to similar realities and contexts.
Explanatory research builds on descriptive research by adding controlled variables and testing specific hypotheses. According to the SAGE Research Methods encyclopaedia, causal research designs are among the most widely used types of research design in the social and behavioural sciences.
Example: An engineering PhD might examine how brittle materials behave under varying compressive loads, manipulating load conditions to establish causal relationships between force and fracture patterns.
Correlational research
Correlational research identifies relationships between two or more variables, examining how changes in one element are associated with changes in another. It is important to note that correlation does not imply causation; this type of research reveals associations, not causes.
Example: A PhD in education might investigate the relationship between hours of independent study and final examination scores across a cohort of undergraduate students. Understanding how to choose a PhD research topic will help you decide which of these research methodology types best suits your study.
Types of research by data type
The data you collect determines whether your research is qualitative, quantitative or mixed methods.
Qualitative research
Qualitative research explores meaning, experience and social phenomena through non-numerical data. Common methods include interviews, focus groups, ethnographic observation and discourse analysis.
Qualitative research is subjective by nature, as not all variables can be fully controlled. Its strength lies in answering ‘why’ and ‘how’ questions rather than ‘how many’ or ‘how much’. It is widely used in the social sciences, humanities and health research. For a detailed comparison of approaches, Scribbr’s guide to qualitative vs quantitative research provides a useful overview.
Example: A PhD student researching the experiences of first-generation university students might conduct semi-structured interviews with 20 participants, analysing transcripts thematically to identify shared challenges and coping strategies.
Quantitative research
Quantitative research studies phenomena through numerical data collection, using mathematical, statistical and computational tools to analyse results. It enables generalised conclusions that can be projected across populations.
Quantitative methods include experiments, surveys with closed-ended questions, statistical modelling and computational simulations.
Example: A PhD in public health might analyse hospital admission data from 50 NHS trusts over five years, using regression analysis to identify predictors of readmission rates.
Mixed methods research
Mixed methods research combines qualitative and quantitative approaches within a single study. This increasingly popular approach allows researchers to capture both the breadth of quantitative data and the depth of qualitative insight.
Common designs include sequential explanatory (quantitative first, then qualitative to explain findings) and concurrent triangulation (both collected simultaneously for comparison).
Mixed methods is now recognised as a distinct research paradigm alongside qualitative and quantitative traditions, and is growing rapidly across disciplines including education, health sciences and social policy.
Example: A PhD in nursing might survey 500 patients about their discharge experience (quantitative), then conduct in-depth interviews with 15 participants who reported low satisfaction scores (qualitative) to understand the reasons behind the numbers.
Types of research by variable manipulation
This classification concerns whether the researcher actively manipulates variables or simply observes them.
Experimental research
Experimental research designs or replicates phenomena under strictly controlled conditions. The researcher manipulates one or more independent variables and measures the effect on dependent variables, following the scientific method with study and control groups.
This is the gold standard for establishing causation and is central to laboratory-based sciences, clinical research and psychology.
Example: A pharmaceutical sciences PhD might conduct a randomised controlled trial comparing a new drug’s effectiveness against a placebo, with participants randomly assigned to treatment and control groups.
Non-experimental research
Non-experimental (observational) research analyses phenomena in their natural context without any direct researcher intervention. The focus is on measuring and describing variables as they occur naturally.
This approach is often used when manipulating variables would be unethical or impractical. It is common in descriptive and correlational research.
Example: An epidemiology PhD might study the relationship between air pollution levels and respiratory disease rates across different regions, using existing environmental and health data without any experimental manipulation.
Quasi-experimental research
Quasi-experimental research sits between experimental and non-experimental approaches. The researcher controls some variables but cannot randomly assign participants to groups. Instead, existing groups or populations are used, which ensures the findings reflect real-world conditions.
Example: A PhD in education might evaluate the effectiveness of a new teaching intervention by comparing outcomes in schools that adopted the programme (treatment group) against schools that did not (comparison group), without random assignment.
Types of research by inference method
The inference method describes how a study moves between theory and observation.
Deductive research
Deductive research starts with a general theory or hypothesis and works towards specific observations to confirm or refute it. The conclusions are logically valid if the premises are correct and the reasoning is sound.
This is the classic ‘top-down’ approach, common in the natural sciences and quantitative social science.
Inductive research
Inductive research works in the opposite direction, starting with specific observations and building towards broader generalisations or new theory. It is a ‘bottom-up’ approach commonly associated with qualitative research and grounded theory methodology.
Hypothetical-deductive research
The hypothetical-deductive method combines elements of both approaches. The researcher observes a phenomenon, forms a hypothesis based on those observations, uses deduction to derive testable predictions, and then verifies or rejects the hypothesis through further observation or experimentation.
This is arguably the most widely used approach in modern scientific research.
Types of research by time period
The timeframe of a study determines whether it is longitudinal or cross-sectional.
Longitudinal studies
Longitudinal research (also called diachronic research) tracks the same subjects, events or groups over an extended period. Data is collected at multiple time points, allowing the researcher to observe changes, trends and developments.
Longitudinal studies are common in medical research, psychology, sociology and education. They are powerful for identifying causal patterns but require significant time and resources. A systematic review published in BMC Medical Research Methodology provides detailed guidance on best practices for longitudinal study design.
Example: A psychology PhD might follow a cohort of 200 university students over four years, measuring wellbeing, academic performance and career outcomes at six-month intervals.
Cross-sectional studies
Cross-sectional research (also called synchronous research) collects data from a population at a single point in time. It provides a snapshot of a phenomenon and is faster and less resource-intensive than longitudinal research.
Example: A sociology PhD might survey 1,000 adults in a single month to examine the relationship between social media use and political engagement at that moment in time.
Types of research by information source
Primary research
Primary research collects original, first-hand data directly from sources. The researcher designs and conducts data collection, whether through experiments, surveys, interviews, observations or fieldwork.
Primary research is the backbone of most doctoral projects and forms the basis of your PhD thesis.
Secondary research
Secondary research analyses data and information that has already been collected by other researchers. Sources include published studies, government datasets, historical records and systematic reviews.
Secondary research is central to the literature review stage of any PhD, and some doctoral projects are built entirely on secondary data analysis.
Types of research by data collection method
The final classification concerns how data is physically gathered.
Documentary research
Documentary (or desk) research involves the systematic review of existing information sources. It is the foundation of literature reviews, historical research and case study analysis. The researcher works with published texts, archives, databases and digital records.
Field research
Field research collects data directly at the location where the phenomenon occurs. This includes ethnographic observation, site visits, environmental sampling and community-based data collection.
Field research is common in geography, anthropology, ecology and public health, and demands careful planning around logistics, ethics and safety.
Laboratory research
Laboratory research takes place in controlled environments where the researcher can isolate dependent variables and establish causal relationships through the scientific method. It is the standard approach in chemistry, physics, biology and experimental psychology.
Mixed-method data collection
Many PhD projects combine documentary, field and laboratory approaches. A biomedical researcher, for instance, might review existing literature (documentary), collect tissue samples from clinical sites (field) and analyse them under controlled conditions (laboratory).
How to choose the right type of research
Selecting the right research approach and research design depends on several factors:
- Your research question: Is it asking ‘what’, ‘why’, ‘how’ or ‘how much’? Descriptive questions suit different methods from explanatory ones
- Your discipline: STEM subjects lean towards quantitative and experimental methods, while humanities and social sciences more frequently use qualitative and interpretive approaches
- Available resources: Longitudinal and experimental studies require more time and funding than cross-sectional or documentary research
- Ethical considerations: Some research questions cannot ethically be studied experimentally, making observational or quasi-experimental designs more appropriate
- Your supervisor’s expertise: Your methodology should align with the support your supervisor can provide. Our guide on working with your PhD supervisor covers this relationship in more detail
Most PhD projects use a combination of research types. A social sciences PhD, for example, might be applied, mixed methods, quasi-experimental and cross-sectional. The important thing is that your choices are justified and clearly explained in your research design.
Frequently asked questions
What are the four main types of research?
The four types most commonly referenced are qualitative, quantitative, basic (theoretical) and applied research. However, research is classified across many dimensions, and there are far more than four types in total. The most useful classification for PhD students considers type by purpose, data, depth, variable control and timeframe.
What is the difference between qualitative and quantitative research?
Qualitative research collects non-numerical data (interviews, observations, texts) to explore meaning and experience. Quantitative research collects numerical data (measurements, surveys, statistics) to test hypotheses and identify patterns. Many modern studies use mixed methods to combine both approaches.
How do I choose a research methodology for my PhD?
Start with your research question. Consider what type of data you need to answer it, whether you can or should manipulate variables, and how much time and resources you have. Discuss options with your supervisor early, and review how similar studies in your discipline have approached their methodology. Your university’s research methods training programme can also help you make an informed decision.
Summary
Choosing the right types of research for your project is a decision that shapes your entire PhD journey. From deciding between qualitative and quantitative data to choosing between experimental and observational approaches, each choice influences how you collect, analyse and present your findings.
The key takeaways from this guide:
- Research is classified across multiple dimensions, not just one
- Most PhD projects combine several types of research simultaneously
- Your research question, discipline and available resources should guide your choice
- Mixed methods research is growing rapidly and offers the benefits of both qualitative and quantitative approaches
- There is no single ‘correct’ type of research; the right choice depends on what you are trying to discover
If you are ready to start planning your research, explore the significance of the study section of your proposal, or search for funded PhD opportunities on DiscoverPhDs to find a project that matches your interests.
Updated: 7th March 2026
