03 · What You Need to Know
How to Plan Grounded Theory Sample Size Without Undermining Theoretical Sampling
Why Is Grounded Theory Sample Size Difficult to Determine Beforehand?
Grounded theory differs from many research designs because data collection and analysis are intended to inform one another.
Researchers begin with evidence relevant to the research problem, develop provisional analytical categories, and use subsequent comparisons to identify what additional evidence is needed.
Consequently, the final sampling requirements depend partly on analytical questions that do not yet exist when the study begins.
Suppose initial interviews suggest that university instructors negotiate AI-related responsibility differently depending on whether institutional policies are clear or ambiguous.
The researcher may need additional evidence from settings with different policy arrangements to investigate that relationship.
Later analysis might reveal that professional status or disciplinary expectations also matter. These developments could require further theoretically directed sampling.
The researcher cannot know the full extent of these analytical needs before examining the initial material.
This is why theoretical sampling is central to grounded theory research.
What Is the Difference Between an Initial Target and a Final Sample Size?
An initial target is a planning estimate. It helps researchers establish recruitment procedures, allocate resources, and prepare ethics applications.
The final sample size is the number of participants actually included in the study after recruitment and data collection have ended.
Neither number independently establishes the theoretical adequacy of the resulting analysis.
Initial recruitment target
A provisional number or range used to organize recruitment and estimate the resources required for the study.
Final achieved sample
The participants actually recruited and included, which may differ from the original target and must be reported accurately.
Researchers should also distinguish an approved recruitment ceiling from an analytical stopping criterion. An ethics committee may approve recruitment of up to 30 participants, for example, without determining whether 30 interviews will be necessary or sufficient for theoretical development.
These distinctions should be explicit in the research proposal.
Is There a Minimum Number of Participants for Grounded Theory?
There is no universally accepted minimum participant number applicable to every grounded theory study.
Grounded theory traditions differ in their analytical procedures, and studies vary in the complexity of the processes being investigated.
Published grounded theory projects may involve different sample sizes, but those numbers should not be converted into universal methodological requirements.
For example, a narrowly focused study involving detailed accounts of one professional process may have different sampling needs from a project investigating interactions across several institutions and participant groups.
The adequacy of the evidence depends on the conceptual development of the categories, not simply on the number of people interviewed.
Researchers should therefore avoid claims such as "grounded theory requires at least 20 participants" unless they are referring to a clearly identified institutional rule or context-specific recommendation rather than a universal methodological standard.
How Does Grounded Theory Sample Size Differ From Statistical Sample Size?
Statistical sample size calculations commonly depend on considerations such as the intended estimand, expected variability, precision, statistical power, and analytical model.
Grounded theory generally pursues a different form of knowledge. It seeks to develop a conceptual explanation of a process through systematically analyzed empirical material.
Accordingly, grounded theory sample size is not ordinarily established through statistical power analysis.
Recruiting more participants does not automatically make the emerging theory more explanatory, just as recruiting fewer participants does not automatically make the study inadequate.
Grounded theory researchers may make carefully bounded claims about the process investigated, but their purposive and theoretical sampling strategies do not ordinarily support population prevalence estimates.
The number of interviews should therefore be justified through the analytical purpose rather than statistical representativeness.
What Does Theoretical Sampling Mean for the Final Number?
Theoretical sampling makes sample size responsive to category development.
Initial participants may be selected purposively because they have experience relevant to the research problem. As analysis proceeds, additional evidence is sought to clarify categories, examine their relationships, and investigate alternative explanations.
The next theoretically useful source may be another participant, an observation, a document, or a follow-up account from someone already interviewed.
Consequently, the number of participants is not necessarily the same as the number of theoretically significant sampling decisions.
For example, a researcher may interview 12 instructors and subsequently examine institutional policy documents to investigate an emerging category concerning authorization.
The documents may contribute to theory development without increasing the participant count.
Conversely, recruiting five additional participants may contribute little if their accounts merely repeat already examined incidents without addressing unresolved theoretical questions.
The central concern is the contribution of the additional evidence to the emerging explanation.
How Do Grounded Theory Traditions Approach Sample Size?
Classic, Straussian, and constructivist grounded theory all emphasize theoretical development, although their philosophical assumptions and analytical procedures differ.
| Grounded theory tradition |
Sampling orientation |
Implication for sample size |
| Classic grounded theory |
Theoretical sampling guided by emerging conceptual categories and their integration. |
Final sampling needs follow the development of the emerging theory rather than a universal numerical rule. |
| Straussian grounded theory |
Sampling directed toward category properties, dimensions, conditions, and relationships. |
Additional evidence may be required to elaborate and relate categories sufficiently. |
| Constructivist grounded theory |
Theoretically directed sampling and interpretive category development. |
Sample adequacy depends on the development and grounding of the theoretical interpretation, not simply interview totals. |
Researchers should consult the methodological literature of their chosen tradition when establishing recruitment plans and stopping criteria.
It would be misleading to present one tradition's procedural guidance as a universal grounded theory requirement.
What Is Theoretical Saturation?
Theoretical saturation is commonly associated with the point at which further evidence is no longer contributing significant development to relevant theoretical categories and their relationships.
Its precise meaning and application vary across grounded theory traditions.
Theoretical saturation is not equivalent to reaching a particular participant number. Nor is it necessarily established when no new descriptive codes appear in the next interview.
For example, a researcher may have identified the category "seeking authorization" across many interviews but still lack sufficient understanding of how authorization-seeking changes under different institutional conditions.
The category may therefore remain theoretically underdeveloped despite repeated references to it.
Researchers should examine whether category properties, dimensions, relationships, and relevant variations have been adequately explored.
The stopping decision should follow a defensible account of theoretical saturation rather than a numerical shortcut.
Can You Use Information Power to Plan Grounded Theory Sampling?
Malterud, Siersma, and Guassora proposed information power as a framework for reasoning about sample adequacy in qualitative interview studies.
The framework considers the study aim, sample specificity, use of established theory, quality of dialogue, and analytical strategy.
These considerations may help grounded theory researchers estimate the resources and initial recruitment needed for a project.
However, information power is not a substitute for theoretical sampling or the methodology-specific evaluation of category development.
For example, a highly specific initial sample may provide detailed evidence about one aspect of a process while leaving important theoretical relationships unexplored.
The researcher may still need additional theoretically directed evidence.
Information power should therefore be treated as a supplementary planning perspective rather than a formula that determines the final grounded theory sample size.
What About Published Numerical Recommendations?
Methodological literature sometimes discusses numerical ranges or typical participant counts for qualitative research.
Such recommendations may be useful for estimating feasibility, but their meaning depends on the methodological assumptions, research context, and analytical purpose from which they were developed.
For example, empirical studies of saturation in qualitative interviews may investigate how many interviews are needed to identify recurring codes or meanings.
Those findings should not automatically be treated as evidence that a grounded theory can be theoretically saturated after the same number of interviews.
Hennink, Kaiser, and Marconi distinguish code saturation from meaning saturation. Their distinction illustrates why identifying recurring issues and developing a sufficiently detailed understanding are not equivalent.
Grounded theory adds the further requirement of developing explanatory categories and relationships.
Researchers should therefore evaluate the relevance of numerical guidance before applying it to grounded theory.
Can You Specify a Sample Size Range in Your Proposal?
Yes. A provisional range may be more transparent than claiming an exact final number when theoretical sampling is anticipated.
For example, a researcher might propose an initial recruitment target of 12 participants with provision for additional theoretically directed recruitment, subject to an approved maximum and applicable amendments.
The numbers should be justified through the study's scope, likely recruitment opportunities, analytical resources, and institutional requirements.
A range does not eliminate the need for theoretical sampling. Researchers should explain how analysis will determine which additional evidence is sought and why.
Similarly, a recruitment range should not be presented as a guarantee that theoretical saturation will occur within its boundaries.
It is a planning instrument, not a substitute for analytical judgment.
What If an Ethics Committee Requires an Exact Number?
Ethics committees and institutional research offices may require numerical recruitment estimates for administrative review.
Researchers should distinguish those estimates from the methodological determination of theoretical adequacy.
Where the application permits it, the researcher may specify an anticipated target and maximum recruitment number while explaining the iterative sampling procedure.
If an exact number is required, the researcher should clarify whether it represents an initial target, a recruitment ceiling, or a binding limit.
The study should not exceed approved recruitment limits or introduce substantially different participant groups without obtaining required authorization.
When emerging analysis indicates a need for additional evidence beyond the approved plan, the researcher may need to request an amendment.
Grounded theory's flexibility does not override research ethics requirements.
What If Your University Requires a Fixed Final Sample?
Some academic programs require students to specify an exact number of participants before data collection.
Such requirements may reflect administrative expectations, project timelines, or supervisory conventions rather than universal grounded theory principles.
Researchers should clarify whether the number is genuinely fixed and whether theoretically directed sampling can occur within the approved recruitment arrangement.
For example, a study may be able to recruit participants in successive stages while remaining within a maximum approved number.
However, if all participants must be selected before analysis and no further evidence can be collected, the design may have limited capacity to support theoretical sampling.
The researcher should consider whether the proposed study can still fulfill the commitments of the selected grounded theory tradition.
If not, another qualitative design may be more appropriate than claiming grounded theory while removing its iterative logic.
What Happens If You Reach the Planned Number Before the Theory Is Adequately Developed?
Reaching a numerical target does not automatically establish theoretical saturation.
If important categories remain underdeveloped, the researcher should identify what additional evidence would be needed and determine whether further collection is feasible and ethically permitted.
Possible responses include seeking an approved recruitment extension, collecting other relevant evidence, or conducting further theoretically directed analysis where the existing material permits it.
However, additional coding of the same transcripts should not automatically be presented as equivalent to obtaining new evidence through theoretical sampling.
If further data collection is impossible, the researcher should report the limitations of the theoretical account.
A preliminary explanation may still be valuable, but it should not be presented as a fully developed theory when the analytical evidence does not support that claim.
What If the Categories Appear Adequately Developed Before the Target Is Reached?
The researcher should examine whether the apparent adequacy is defensible within the chosen methodology.
A lack of new interview topics is not necessarily sufficient evidence that category relationships have been fully examined.
Researchers should review analytical memos, comparisons, contrasting cases, and unresolved theoretical questions.
If the categories are adequately developed and the approved protocol permits recruitment to stop early, continuing solely to reach an arbitrary number may offer little analytical benefit.
However, institutional requirements or approved procedures may still govern the recruitment decision.
The researcher should document the basis for stopping and explain any difference between the planned and achieved sample.
How Should You Document Sample Size Decisions?
Grounded theory researchers should maintain a transparent record of how recruitment and evidence collection developed.
Useful documentation includes the initial sampling rationale, anticipated participant target, recruitment dates, emerging categories, analytical memos, theoretical sampling decisions, and reasons for ending data collection.
For example, a memo might explain that early interviews raised questions about the relationship between institutional policy clarity and instructors' authorization-seeking practices.
The researcher could then document why additional participants from contrasting institutional contexts were recruited and how their accounts affected the developing category.
Such records demonstrate that changes to the sample followed analytical reasoning rather than convenience alone.
They also help distinguish a legitimate stopping decision from a retrospective claim that the planned number happened to produce saturation.
How Should Sample Size Be Reported in the Final Study?
The final methodology should distinguish planned recruitment from actual recruitment and explain how sampling evolved.
Researchers should report the achieved participant number, relevant participant characteristics, the initial sampling strategy, subsequent theoretical sampling, and the basis for ending data collection.
Where additional documents, observations, or incidents contributed to theoretical sampling, those sources should also be described.
If recruitment ended because of time, funding, access, or institutional restrictions, the report should acknowledge those constraints rather than automatically claim theoretical saturation.
The final account should demonstrate how sampling contributed to the development of the theory.
This reporting is part of the broader methodological requirement that data collection and analysis remain meaningfully iterative.
