03 · What You Need to Know
Understanding the Relationship Between Qualitative Sample Size and Analytical Depth
What Is Analytical Depth in Qualitative Research?
Analytical depth concerns the extent to which an interpretation develops meaningful understanding of participants' experiences, social processes, contextual conditions, or conceptual relationships.
It goes beyond identifying recurring topics. A researcher may observe that many teachers describe assessment integrity as a concern, but a deeper analysis examines how they define integrity, negotiate competing expectations, and respond to particular assessment situations.
Depth may involve attention to contradictions, temporal developments, differences between cases, and the researcher's role in constructing interpretations.
Its meaning varies across qualitative methodologies. An idiographic phenomenological analysis, a qualitative descriptive study, and an ethnography do not necessarily pursue identical forms of analytical depth.
Consequently, depth should be evaluated according to the study's analytical purpose rather than the length of the final report or the number of quotations included.
Why Can a Larger Sample Become Difficult to Analyze?
Each additional participant can generate material requiring familiarization, interpretation, documentation, and integration into the developing analysis.
In interview research, this may include reviewing recordings, checking transcripts, writing analytical notes, examining contextual information, and comparing accounts.
The workload does not necessarily increase in a simple linear relationship with participant count. A comparative analysis may require additional work to examine relationships across participants and groups.
When recruitment expands without corresponding analytical resources, researchers may have less opportunity to revisit individual accounts, investigate contradictions, or refine interpretations.
This is a risk rather than an inevitable consequence of large samples. Experienced teams with appropriate analytical procedures may work rigorously with extensive qualitative datasets.
Does More Data Automatically Produce Better Qualitative Evidence?
No. Additional data are valuable when they contribute information relevant to the research question.
A new participant might introduce a previously unexamined experience, challenge an emerging interpretation, or clarify an important contextual difference.
Alternatively, an additional interview might provide little information beyond what the existing material already supports.
The value of recruitment therefore depends on its contribution to the intended analysis rather than the numerical increase in the dataset.
This distinction is consistent with the information power framework proposed by Malterud, Siersma, and Guassora (2016), which relates sample adequacy to the relevance and richness of information.
How Does the Methodology Affect the Risk of Excessive Recruitment?
Qualitative methodologies differ in how they balance the number of cases against the depth and scope of interpretation.
| Methodological approach |
Analytical emphasis |
Possible consequence of unnecessary recruitment |
| Interpretative phenomenological analysis |
Detailed idiographic examination of individual experiences. |
Less opportunity for intensive case-level interpretation. |
| Grounded theory |
Conceptual category development through iterative analysis and sampling. |
Accumulation of interviews that do not advance theoretical development. |
| Reflexive thematic analysis |
Interpretive development of patterns of meaning across a dataset. |
Insufficient time for recursive engagement and theme refinement. |
| Qualitative descriptive research |
Detailed description of relevant experiences or practices. |
Extensive descriptive material without sufficient synthesis. |
| Case study |
Understanding a bounded case using relevant contextual evidence. |
Excessive interview collection at the expense of documents, observations, or contextual analysis. |
| Ethnography |
Understanding social practices through contextual field engagement. |
More interviews without corresponding attention to observation and field relationships. |
These are possible risks, not claims that any methodology requires a universally small sample. The appropriateness of recruitment depends on the particular design and intended analytical contribution.
Why Is the Issue Especially Important in Idiographic Research?
Idiographic approaches emphasize detailed examination of particular cases before developing broader interpretations.
In interpretative phenomenological analysis, for example, researchers commonly examine individual accounts closely, attending to experiential meanings and differences among participants.
Smith, Flowers, and Larkin (2022) discuss the methodological foundations and analytical commitments of interpretative phenomenological analysis.
Recruiting substantially more participants than can be examined idiographically may create a mismatch between the declared methodology and the analysis actually conducted.
The problem is not that a large sample is prohibited. It is that the analytical commitments of the methodology must remain feasible.
Can a Large Sample Encourage Superficial Coding?
It can when researchers respond to data volume by prioritizing rapid categorization over the analytical work required by the study.
For example, a researcher may assign broad labels such as workload, ethics, and student engagement to hundreds of interview excerpts without examining how these issues relate to one another.
Such coding may be suitable for an initial descriptive task, but it does not necessarily provide a developed interpretation of participants' experiences.
In some approaches, extensive coding is an appropriate analytical stage. The concern arises when coding replaces rather than supports the intended interpretation.
A large dataset should not be mistaken for a deep analysis simply because it contains many coded segments.
Does Qualitative Analysis Software Solve the Problem?
Qualitative data analysis software can assist with organizing transcripts, retrieving coded material, managing annotations, and examining relationships across a dataset.
These functions can make extensive qualitative material more manageable.
However, software does not independently establish the meaning of participants' accounts or determine whether an interpretation is sufficiently developed.
Automated coding and AI-assisted summarization may also omit contextual detail or impose categories that require careful human examination.
Software can support analytical work, but it does not eliminate the need for methodological judgment, reflexivity, and adequate researcher engagement with the data.
How Does Interview Length Affect the Analytical Workload?
Participant count is only one component of data volume.
Twenty interviews lasting ninety minutes each may generate substantially more material than forty interviews lasting twenty minutes each.
However, duration is not a direct measure of information richness. A short interview may contain highly relevant material, while a lengthy interview may include substantial repetition.
Researchers should consider the expected volume and complexity of the data alongside the quality of the interviews.
Planning should allow sufficient time for analysis, not merely for completing the scheduled interviews.
How Can You Estimate the Analytical Workload Before Recruitment?
A practical planning exercise can estimate the volume of recordings, transcripts, and analytical work associated with alternative recruitment targets.
For example, thirty interviews lasting approximately sixty minutes each would produce around thirty hours of recordings.
Transcription, checking, familiarization, coding, memo writing, and interpretation would require additional time. The amount varies with recording quality, transcription methods, analytical approach, and researcher experience.
These estimates should not be converted into a universal number of hours required per interview.
The purpose is to determine whether the proposed sample is realistically compatible with the analytical work the study requires.
Can Too Many Participants Reduce Attention to Contradictory Accounts?
Potentially. When researchers face a large volume of material, they may emphasize the most frequent patterns because those are easier to summarize.
Yet unusual or contradictory accounts can be important for understanding the boundaries of an interpretation.
For example, most teachers might report that AI-assisted assessment design reduces workload, while a smaller group describes increased verification responsibilities.
The less common experience may reveal important contextual conditions that a frequency-oriented summary would overlook.
Analytical depth requires attention to the significance of evidence, not merely the number of participants expressing a particular view.
Is a Smaller Sample Always Better for Interpretive Research?
No. A small sample may permit intensive analysis but still lack information needed to address the research question.
For example, a study comparing teachers' assessment practices across several disciplines may require enough participants to examine meaningful differences.
A researcher cannot resolve insufficient participant variation simply by analyzing a narrow set of interviews more deeply.
Likewise, a study involving complex social processes may require extended recruitment or multiple data sources.
The appropriate balance depends on what the study intends to understand and the evidence needed to support that understanding.
How Does Information Power Help Evaluate Excessive Recruitment?
Information power provides a framework for considering whether the sample offers sufficient relevant information for the research aim.
Malterud and colleagues (2016) identify study aim, sample specificity, established theory, dialogue quality, and analysis strategy as interacting considerations.
If the sample is highly relevant, interviews are rich, and the analytical aim is focused, additional participants may contribute limited value.
However, a broad aim or substantial participant variation may justify further recruitment.
The information power framework does not prescribe a maximum sample size. It helps researchers reason about the relationship between information and analytical purpose.
Can Data Saturation Justify Stopping Before the Maximum Sample Is Reached?
Sometimes, when saturation is consistent with the methodology and has been assessed through an appropriate analytical procedure.
For coding-oriented studies, researchers may examine whether additional interviews contribute new relevant codes or meanings.
However, code repetition does not necessarily establish sufficient interpretive depth. Hennink, Kaiser, and Marconi (2017) distinguish code saturation from meaning saturation.
Moreover, saturation may be inappropriate for approaches such as reflexive thematic analysis.
Researchers should therefore evaluate whether saturation fits the methodology before using it to justify stopping recruitment.
What Is the Difference Between Excessive Recruitment and Necessary Variation?
Excessive recruitment occurs when additional participants create burdens or analytical demands that are not justified by their expected contribution to the research aim.
Necessary variation refers to differences among participants or contexts that must be examined to support the intended analysis.
For example, interviewing more teachers with nearly identical experiences may add little information to a focused study. Recruiting teachers with substantially different assessment responsibilities may address an important analytical gap.
Researchers should distinguish numerical expansion from purposeful expansion.
The specificity of the sample and the variation it contains are therefore more informative than participant count alone.
Does a Large Qualitative Sample Improve Generalizability?
Not automatically. Qualitative studies often use purposive rather than probability sampling.
Increasing the number of purposively recruited participants does not by itself produce statistically representative population estimates.
A larger sample may broaden the range of experiences examined and strengthen certain comparative claims, but its contribution depends on the sampling strategy and analysis.
Transferability requires sufficient contextual information for readers to evaluate the relevance of findings to other settings.
Researchers should not justify excessive recruitment solely by claiming that more participants automatically make qualitative findings statistically generalizable.
When Is a Large Qualitative Sample Appropriate?
Large qualitative samples may be appropriate for studies involving extensive institutional comparisons, diverse participant populations, or broad descriptive objectives.
They may also be feasible in collaborative projects with trained analytical teams and carefully designed procedures.
For example, a multi-institutional study investigating differences in online assessment policies may require interviews across several professional roles and institutional settings.
In such a design, broader recruitment serves an identifiable analytical purpose.
The key question is whether the researchers can analyze the resulting material adequately and whether the sample supports the intended claims.
Should Researchers Stop Recruiting Once They Have Enough Data?
When the methodology and approved protocol permit it, recruitment should be guided by whether additional participants are needed to address the analytical aim.
Continuing recruitment solely to reach a round number may create unnecessary work and participant burden.
However, stopping early requires an appropriate justification rather than a vague assertion that the data seem repetitive.
The decision about when participant recruitment should stop should consider analytical adequacy, relevant variation, and the research design.
Researchers must also respect existing recruitment commitments and applicable ethics requirements.
How Can Researchers Preserve Depth in a Large Qualitative Study?
Large studies benefit from an explicit analytical strategy that addresses the volume and diversity of the data.
Researchers may use case summaries, analytical memos, structured comparisons, and carefully coordinated team discussions.
For interpretive approaches, the analysis should preserve the context and meaning of participant accounts rather than reduce them to isolated quotations.
Team-based analysis also requires attention to methodological consistency. Agreement among coders is not a universal criterion for rigor across qualitative traditions.
Ultimately, the analytical procedures should be appropriate to the methodology and sufficiently resourced to support the intended interpretation.
Watch Out
Do not increase qualitative recruitment simply because more participants are available. Additional interviews involve participant time, confidentiality obligations, data-management responsibilities, and analytical work. Their expected contribution should justify these costs.
