Qualitative Sample Size: How Many Participants Do You Need?

"How many participants do I need?" is the question every qualitative researcher dreads, because the honest answer — "it depends" — never satisfies a dissertation committee. Unlike quantitative research, where power analysis gives you a precise number, qualitative sample size is guided by principles rather than formulas. This post gives you the practical knowledge you need to choose and defend your sample size.

Why There Is No Magic Number

In quantitative research, sample size is about statistical power — you need enough participants to detect an effect if one exists. In qualitative research, the goal is different. You are not trying to generalize to a population. You are trying to develop a deep understanding of a phenomenon. That requires a different logic for determining how many participants are enough.

The short answer is: you need enough participants to achieve saturation — the point at which new data stops generating new insights. But what that looks like in practice varies enormously depending on your methodology, research questions, and the diversity of your participant pool.

Sample Size Guidelines by Method

While no rule is absolute, the methodological literature offers some guidance:

Phenomenology

Creswell and Poth recommend 5 to 25 participants for a phenomenological study. Dukes (1984) suggests 3 to 10. The idea is that you need enough participants who have experienced the phenomenon to identify its essential structure, but not so many that you lose depth.

Grounded Theory

Charmaz (2006) suggests 20 to 30 interviews for a grounded theory study, though some studies achieve theoretical saturation with fewer. Strauss and Corbin (1998) argue that the number depends on the complexity of the theory being developed.

Case Study

A case study might involve a single participant or a single site, with multiple data sources (interviews, observations, documents). The sample size question here is less about how many participants and more about how many cases. Yin (2018) recommends 4 to 10 cases for a multiple-case study.

Narrative Inquiry

Narrative studies often focus on a very small number of participants — sometimes just one or two — because the goal is to understand the depth and complexity of individual stories.

Generic Qualitative Study

For a basic qualitative study without a specific methodological tradition, Guest, Bunce, and Johnson (2006) found that saturation often occurred within the first twelve interviews, with basic themes emerging as early as six interviews.

The Saturation Argument

Most qualitative researchers justify their sample size through the concept of saturation. There are two main types:

Data saturation means you have heard the same stories and experiences repeated across participants, and new interviews are not producing new codes or categories.

Theoretical saturation is specific to grounded theory and means your emerging theory is fully developed — all categories are well-defined, relationships between them are established, and new data fits within the existing framework.

The challenge is that saturation is hard to plan for in advance. You cannot know when you will reach it until you are in the midst of data collection and analysis. This is why qualitative researchers often collect and analyze data simultaneously rather than finishing all interviews before beginning analysis.

How to Defend Your Sample Size to Your Committee

Start with the Literature

Cite published studies that used similar methods and had similar sample sizes. If you are proposing fifteen participants for a phenomenological study, point to well-regarded phenomenological studies in your field that used twelve to twenty participants.

Invoke Saturation

Explain that you will continue collecting data until saturation is reached, and describe how you will assess saturation. For example: "I will analyze each transcript before conducting the next interview. I will track the emergence of new codes and consider saturation achieved when three consecutive interviews produce no new codes."

Be Practical

It is acceptable to acknowledge practical constraints. Studying a rare population with limited access is a legitimate reason for a smaller sample. Just be transparent about it.

Propose a Range

Instead of committing to an exact number, propose a range. "I anticipate conducting 12 to 18 interviews, with the final number determined by data saturation." This gives you flexibility while showing your committee that you have thought about the issue.

Common Mistakes

Using quantitative logic. Do not try to calculate statistical power for a qualitative study. It is a category error that signals to your committee that you do not understand qualitative methodology.

Choosing a number arbitrarily. Saying "I'll interview twenty people" without explaining why twenty is the right number is a red flag. Always connect your sample size to your methodology and saturation.

Equating more with better. A study with fifty shallow interviews is not stronger than a study with twelve deep ones. In qualitative research, depth almost always matters more than breadth.

Ignoring within-case data. Remember that qualitative studies often generate enormous amounts of data from each participant. A single sixty-minute interview produces roughly twenty to thirty pages of transcript. Fifteen interviews give you three hundred to four hundred fifty pages of text to analyze. That is a substantial dataset.

A Practical Approach

Here is what I recommend for most dissertation studies:

1. Review the methodological literature for your chosen approach and note the recommended range.
2. Review published studies in your field that use the same method and note their sample sizes.
3. Propose a sample size within the intersection of those two ranges.
4. Build in a saturation check by analyzing data concurrently with collection.
5. Be prepared to justify your final number in your findings chapter by showing evidence of saturation.

The sample size question is never fully resolved before data collection begins. But if you approach it thoughtfully, ground your decision in the literature, and remain attentive to saturation during your study, you will have a defensible answer for your committee.