What is Qualitative Research? Types, Methods, Examples

Contents

• What is qualitative research?
• Key characteristics of qualitative research
• Qualitative research approaches
• Qualitative research methods (data collection)
• Qualitative data analysis
• When to use qualitative research
• Advantages and disadvantages of qualitative research
• Qualitative vs quantitative research
• Is qualitative research right for your study?
• Sampling strategies in qualitative research
• Ensuring rigor in qualitative research
• Ethical standards in qualitative research
• Addressing researcher bias
• Frequently Asked Questions

Qualitative research helps researchers understand the ‘why’ and ‘how’ behind human behavior: the motivations, experiences, and meanings that numbers alone cannot capture. This article explains what qualitative research is, describes its major approaches and data collection methods, and outlines when and how to use them.

What is qualitative research?

Qualitative research is the systematic process of collecting, analyzing, and interpreting non-numerical data. Rather than measuring how much or how many, it explores how and why, examining people’s subjective experiences, perceptions, motivations, and social contexts.

Qualitative data typically takes the form of words, narratives, images, audio recordings, and observations. It is widely used in the social sciences, healthcare, education, psychology, and anthropology to generate hypotheses and build theory from real-world evidence.^1,2,3

Common examples of qualitative research include:

• Interviewing patients to understand their experience of a new treatment
• Conducting focus groups with students to explore attitudes towards online learning
• Observing workplace dynamics to investigate organizational culture
• Analyzing policy documents to identify patterns in government communication

Key characteristics of qualitative research

Qualitative research is distinguished by several defining features:

• Naturalistic inquiry: Data are collected in real-world settings where participants behave naturally, without laboratory constraints.
• Non-numerical data: Findings are expressed in words, themes, and narratives rather than statistics.
• Inductive reasoning: Theories and patterns emerge from the data rather than being tested against a pre-set hypothesis.
• Researcher as instrument: The researcher’s observation, communication, and interpretation are central to data collection.
• Flexible design: The research design can be adjusted as new insights emerge during the study.
• Holistic perspective: Analysis considers social, cultural, emotional, and contextual dimensions together.
• Small, purposive samples: Participants are selected for the richness of their experience, not for statistical representativeness.

Qualitative research approaches

A qualitative research approach is the overarching framework that guides how a study is designed and interpreted. The five main approaches are described below.

Qualitative research methods (data collection)

While an approach sets the philosophical direction, a method is the practical technique used to collect data. Several methods are often combined within a single study.

Interviews

Interviews are one of the most widely used qualitative methods. They allow researchers to access individual narratives, opinions, and detailed personal accounts. Interviews may be:

• Structured: A fixed set of questions in a predetermined order; this type is useful for consistency across many participants.
• Semi-structured: A flexible guide of open-ended questions, allowing the conversation to follow relevant threads.
• Unstructured: A conversational approach with minimal pre-set questions, allowing the participant to direct the discussion.

Face-to-face interviews also allow the researcher to observe non-verbal cues such as body language, which can enrich interpretation.

Focus group discussions

Focus groups bring together a small number of participants (typically 6–10) to discuss a topic under the guidance of a moderator. The group dynamic generates insights that individual interviews cannot. Debates, peer challenges, and collective sense-making reveal shared and divergent perspectives. Focus groups are particularly useful for:

• Exploring social norms and community attitudes
• Testing responses to a new product, policy, or idea
• Understanding disagreements or variations in experience within a group

Participant observation

In participant observation, the researcher embeds themselves in a setting and observes interactions, behaviors, and events over time. Field notes record nuanced descriptions of the environment, individuals, group dynamics, and informal conversations. This method is central to ethnographic research.

Content analysis

Content analysis involves the systematic examination of existing texts, documents, images, or media. Researchers look at how words, symbols, and themes are used to construct meaning. It can be applied to journal articles, social media posts, policy papers, news coverage, or audio-visual material. The process yields coded categories that are then interpreted for patterns.

Document study and record keeping

Researchers review written or digital materials such as organizational reports, historical archives, academic literature, policy documents, or personal diaries. This method requires no direct interaction with participants and can be used to triangulate findings from other methods.

Qualitative observation (non-participant)

Unlike participant observation, in non-participant observation the researcher remains an outside observer. Data are gathered through the five senses (noting what is seen, heard, or otherwise perceived) without becoming part of the setting. This approach minimizes the risk of the researcher influencing what they are studying.

Qualitative data analysis

Once data are collected, the analysis transforms raw material like transcripts, field notes, images, or documents, into meaningful insights. The general steps in qualitative data analysis are:

1. Prepare and organize the data: transcribe recordings, compile field notes, and store materials securely.
2. Review and familiarize: read through all data thoroughly before beginning to code.
3. Develop a coding system: assign labels (codes) to segments of data that reflect a concept, theme, or pattern.
4. Identify themes: group related codes into broader categories or themes.
5. Interpret and validate: examine what the themes mean in context, and check interpretations against the data and, where possible, against other sources (triangulation).
6. Present findings: write up themes narratively, supported by representative quotes or observations.

Common qualitative data analysis methods include:

When to use qualitative research

Qualitative research is the most appropriate choice in the following situations:

• Little is known about the topic: It is ideal for exploratory work where no established theory or hypothesis yet exists.
• The research question asks ‘how’ or ‘why’: If you need to understand a process, a decision, or an experience rather than measure a quantity, qualitative methods are best suited.1
• Human experience and meaning are central: Whenever participant perspectives, emotions, or interpretations are the subject of inquiry.
• Complex social phenomena need unpacking: Qualitative research can capture the interplay of factors in ways that quantitative data cannot.
• Hypotheses need to be generated before testing: Findings from qualitative studies often inform the design of subsequent quantitative research.

Typical use cases include:

• New product development: understanding user needs before building a survey
• Healthcare research: exploring patient experiences of a condition or treatment2
• Education research: investigating classroom dynamics or teaching practices
• Policy evaluation: understanding how a program is experienced by its beneficiaries
• Organizational research: studying culture, leadership, and communication in the workplace

Advantages and disadvantages of qualitative research

Advantages

• Depth of understanding: Produces rich, detailed insight into human behavior, beliefs, and motivations that quantitative data cannot capture.
• Contextual sensitivity: Findings are grounded in the real-world environments where phenomena actually occur.
• Flexibility: The research design can evolve in response to emerging findings.
• Hypothesis generation: Well suited to exploring new areas and building theory from the ground up.
• Participant engagement: Open-ended methods allow participants to speak in their own terms, often yielding more authentic data.
• Multiple data sources: Combining methods (interviews, observation, documents) enables triangulation and a more complete picture.

Disadvantages

• Limited generalizability: Small, purposive samples cannot be statistically extrapolated to wider populations.
• Subjectivity: Interpretation is influenced by the researcher’s perspective; bias must be actively managed.
• Time and resource intensive: Data collection, transcription, and analysis are all time-consuming processes.
• Difficult to replicate: Context-specific studies may not yield identical results if repeated.
• Validity and reliability challenges: Demonstrating rigor requires explicit documentation of methods and reflexive practice.3

Qualitative vs quantitative research

The two main paradigms in research are complementary. Choosing between them depends on the research question, the state of existing knowledge, and the kind of insight sought.

Many studies use a mixed-methods design, combining both approaches to gain the benefits of depth and breadth. For example, qualitative interviews may be used to design a survey instrument, or quantitative findings may prompt qualitative investigation of an unexpected result.

Is qualitative research right for your study?

Choosing the right research method is one of the first and most important decisions in any study. The table below maps common research situations to the most appropriate method, to help you decide.

Use this checklist as a quick self-check before committing to a qualitative design:

• My research question asks how, why, or what something means.
• There is limited existing theory or research on this topic.
• Participants’ perceptions, feelings, or lived experiences are central to the inquiry.
• I need to generate hypotheses rather than test ones that already exist.
• The context and meaning behind behavior matters as much as the behavior itself.

If most of these apply, qualitative research is likely the right approach. If your priority is statistical generalization, consider quantitative or mixed-methods designs instead.

Sampling strategies in qualitative research

Unlike quantitative research, which relies on large random samples to support statistical generalization, qualitative research uses non-probability sampling: selecting participants for the richness and relevance of their experience. Sample sizes are typically small, and data collection continues until theoretical saturation is reached (the point at which new data no longer generates new themes or insights).

A note on sample size

There is no universal rule for sample size in qualitative research. A rough guide:

• In-depth interviews: 12–30 participants is common for most studies; smaller for phenomenology (6–12), larger for grounded theory.
• Focus groups: 3–6 groups of 6–10 participants each.
• Case studies: 1–4 cases, depending on depth required.
• Ethnography: Determined by access and immersion time, not a numeric target.

Ensuring rigor in qualitative research

Rigor in qualitative research is not assessed by the same criteria as quantitative research. Validity and reliability are replaced by four parallel concepts, first proposed by Lincoln and Guba (1985) and widely used since.¹

Practical techniques for strengthening rigor

• Triangulation: Cross-check findings using multiple data sources (e.g. interviews + observation + documents), multiple methods, or multiple researchers analyzing the same data independently.
• Member checking: Share draft themes or interpretations with participants and invite corrections or clarifications.
• Peer debriefing: Discuss your interpretations regularly with a colleague or supervisor who is not involved in the study.
• Negative case analysis: Actively look for data that contradicts your emerging themes. Revising an interpretation to account for disconfirming evidence strengthens its credibility.
• Reflexive journaling: Keep a running log of your assumptions, reactions, and interpretive decisions throughout fieldwork and analysis.

Ethical standards in qualitative research

Qualitative research often involves close, personal interaction with participants, which makes ethical conduct especially important. The core principles below apply to all qualitative studies, and most institutions require formal ethics approval before data collection begins.³

Working with vulnerable populations

Additional safeguards are needed when research involves:

• Children or young people (under 18): parental or guardian consent is required alongside the child’s own assent.)
• People with cognitive impairment or mental illness: capacity to consent must be carefully assessed.
• Individuals in dependent relationships (e.g. patients, students, employees): voluntary participation must be clearly protected from any perceived pressure.
• Survivors of trauma or abuse: interview protocols should be designed with a trauma-informed approach, and researcher wellbeing should also be considered.

Addressing researcher bias

In qualitative research, the researcher is both the primary data-collection instrument and the interpreter of findings. This makes bias an inherent risk rather than an aberration. The goal is not to eliminate bias (which is impossible) but to acknowledge, examine, and manage it transparently.

Common sources of bias

Reflexivity in practice

Reflexivity is the ongoing process of critically examining your own influence on the research. Practical steps include:

• Write a positionality statement before data collection, describing your personal and professional relationship to the research topic.
• Keep a reflexive journal throughout fieldwork and analysis, noting moments where your assumptions were challenged or your reactions may have shaped interpretation.
• Use member checking to verify whether your interpretations resonate with participants’ own understanding.
• Present disconfirming evidence alongside your main findings to show that alternative interpretations were considered.
• Seek external audit: ask a peer or supervisor to review your analytical process and challenge your conclusions.

Frequently Asked Questions

Can I conduct qualitative interviews online rather than in person?

Yes, and online interviews have become standard practice rather than a compromise. Research comparing video, audio, chat, and email modes found little meaningful difference in data quality across formats. Participants in online settings often report feeling more comfortable discussing sensitive topics due to perceived anonymity and the absence of face-to-face social pressure.

The main practical considerations are:

• Video (Zoom, Teams): Closest to in-person; allows some non-verbal observation. The most common format for semi-structured and in-depth interviews. Test equipment in advance and have a backup recorder.
• Audio-only: Lower technical burden; useful when participants have limited bandwidth or prefer not to be on camera. Slightly reduces rapport but not data quality.
• Asynchronous (email, online forms): Useful for participants in different time zones or those who need more time to reflect. Responses tend to be more considered but may lack spontaneous depth.
• Online focus groups: Work well for participants who already know each other or share an online community; harder to manage simultaneous contributions than in a physical room.

One emerging concern with online recruitment is the rise of “imposter participants”, i.e., individuals who misrepresent their eligibility to access incentives. Verification strategies (screening questions, follow-up checks, referral-based recruitment) are increasingly recommended, particularly for studies with cash incentives.

What software can I use to manage and analyse qualitative data?

Qualitative data analysis software (often called QDAS or CAQDAS) does not analyse your data for you. It organises and manages it so you can analyse more systematically. The choice of tool depends on your budget, institution, and the scale of the project.

For most student and early-career researchers, starting with manual coding in Word or a spreadsheet is a reasonable approach before committing to learning a full QDAS package. Software adds value mainly when datasets are large (20+ transcripts), teams need to work collaboratively, or you need to manage multiple data types alongside each other.

Can I use AI tools to help with qualitative coding and analysis?

AI tools can support certain tasks in qualitative research, but using them for core analytical work carries significant risks that are not yet fully resolved.

Where AI can legitimately help:

• Transcribing audio recordings (tools like Otter.ai, Whisper, and Teams transcription are now widely used and reliable for most accents and audio quality)
• Summarising lengthy documents before reading in full
• Generating initial codebook suggestions to react to, not to adopt wholesale
• Checking for consistency across a large number of codes

Where caution is essential:

• Theme generation: Studies comparing AI-assisted and human-only thematic analysis have found significant variation in outputs. Theme counts range widely and AI outputs can reflect training data biases rather than the actual data
• Fabricated quotes: Multiple researchers have reported AI tools generating plausible-sounding but entirely invented quotes attributed to participants. This is a serious validity risk.
• Interpretive depth: AI tools currently struggle with the kind of reflexive, contextually sensitive interpretation that distinguishes rigorous qualitative analysis from surface-level pattern matching

The emerging consensus is that AI is a useful assistant for logistics and early-stage organisation, but should not replace human coding, theme development, or interpretation. Any AI-assisted steps should be disclosed transparently in your methods section.

How do I handle participants who go off-topic or say something unexpected during an interview?

Unexpected responses are one of the most valuable features of qualitative research, not a problem to be corrected. The design of qualitative methods specifically allows for this and your research questions may evolve as a result.

Practical guidance:

• Follow the thread briefly before redirecting. If a participant raises something unprompted, it often signals that the topic is more salient to them than your scheduled questions. Explore it with a neutral probe (“Can you say more about that?”) before returning to your guide.
• Distinguish off-topic from adjacent-topic. A participant who starts discussing their manager when you asked about team culture is not off-topic. They are telling you where the relevant boundaries actually lie. A participant describing their weekend plans is genuinely off-track.
• Redirect without dismissing. A phrase such as “That’s really helpful context—I’d like to come back to that. Can I ask you about X first?” preserves rapport while keeping the interview on track.
• Debrief thoroughly afterwards. Note unexpected themes in your field notes or reflexive journal immediately after the interview. These deviations often point to the most analytically interesting findings.
• Revise your interview guide if needed. Qualitative research design is iterative. If multiple participants go to the same unexpected place, your guide should reflect that.

Unexpected responses that contradict your emerging themes should be treated as especially valuable. They are the raw material for negative case analysis and often strengthen the credibility of the final findings.

What does “transferability” mean, and can qualitative research ever be generalised?

Qualitative research does not generalise statistically, but it can transfer analytically — and that is often more useful. This is one of the most common points of confusion for researchers new to qualitative methods.

Statistical generalisation (the kind used in quantitative research) requires a large, randomly selected sample that is representative of a defined population. Qualitative research does not meet these criteria, and it is not designed to.

Analytical or theoretical generalisation works differently. A well-conducted qualitative study produces concepts, frameworks, or theoretical propositions that readers can apply to their own contexts, not because the sample was representative, but because the analysis was rigorous, the context was described in sufficient detail, and the reasoning was transparent.

This is what Lincoln and Guba called transferability: rather than the researcher claiming findings apply elsewhere, it is the reader who judges whether the described context is similar enough to their own situation for the findings to be relevant. This places an obligation on the researcher to provide thick description (detailed accounts of the setting, participants, and context) so readers have enough information to make that judgement themselves.

A single in-depth case study can have profound generalizability in the analytical sense: if it generates a new theoretical framework or disrupts an existing assumption, it adds to knowledge regardless of sample size.

References

1. Busetto L, Wick W, Gumbinger C. How to use and assess qualitative research methods. Neurological Research and Practice. 2020;2(1):14.
2. Ziebland S, Wyke S. Health and illness in a connected world: how might sharing experiences on the internet affect people’s health? Milbank Quarterly. 2012;90(2):219–249.
3. Tenny S, Brannan JM, Brannan GD. Qualitative Study. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2023. Available at https://www.ncbi.nlm.nih.gov/books/NBK470395/
4. Weng ML. What Is Qualitative Research? An Overview and Guidelines. https://doi.org/10.1177/14413582241264619
5. Hammarberg K, Kirkman M, de Lacey S. Qualitative research methods: when to use them and how to judge them. https://doi.org/10.1093/humrep/dev334
6. Bazen A, Barg FK, Takeshita J. Research Techniques Made Simple: An Introduction to Qualitative Research. https://doi.org/10.1016/j.jid.2020.11.029

This article was originally published on September 25, 2018, and updated on June 12, 2026.