Qualitative Saturation: What It Can and Cannot Tell You

What Is Saturation?

Saturation is the point at which further data collection or analysis is judged unlikely to add anything important. It began in grounded theory as theoretical saturation: the developing categories stop changing. It has since spread to other approaches as a way to justify how many interviews or groups are enough. Saunders and colleagues found that it is understood in several different ways and that the logic of predicting what is unobserved from what has been observed is uncertain.

So the same word covers different claims. Code saturation means no new codes are appearing. Thematic saturation means no new themes. Theoretical saturation means a theory's categories are developed. "Meaning" saturation, where understanding of an issue is judged complete, is a further form that Hennink and Kaiser note needed larger samples. A report should say which one it means and how it checked.

Decision This Supports

The practical decision is how many interviews or groups to plan and when to stop. A commissioner often wants a number. A defensible answer gives a reasoned range tied to the study, not a universal figure.

Before You Start

• Name the analytic tradition. See choosing a qualitative analytic tradition.
• Describe how varied the people are and how narrow the question is.
• Decide whether you will assess saturation at all, and how.

What Studies Suggest

Guest, Bunce and Johnson analyzed sixty interviews from one study and found that saturation occurred within the first twelve, although basic elements for metathemes were present as early as six. Hennink and Kaiser reviewed 23 studies and reported that empirical studies reached saturation within a narrow range of interviews or focus groups, particularly where the population was fairly homogeneous and the aims narrow. They stress that these results apply to certain types of study, and note that multi-country work, meta-themes and meaning saturation needed larger samples.

These findings help plan. They do not set a minimum or guarantee that a given sample is enough. Malterud and colleagues offer another logic, information power: the more relevant information the sample holds, the fewer participants are needed, depending on the study aim, sample specificity, use of established theory, quality of dialogue and analysis strategy.

Where Saturation Does Not Fit

Braun and Clarke argue that in reflexive thematic analysis, themes are developed by the researcher through interpretation, so there is no fixed set of themes waiting to be exhausted. Judging when to stop is situated and subjective and cannot be fully decided in advance. Applying a saturation test to that approach contradicts its assumptions. Grounded theory, by contrast, ties saturation to theoretical sampling and the state of the emerging categories.

A Small Worked Example

This example is fictional.

An evaluator plans interviews with participants of a hypothetical learning program. The population is varied by district, sex and age, and the question has three parts. She does not cite a number from another study. She writes: "We plan 12 to 18 interviews, purposively varied by district, sex and age, and will review after every four whether new circumstances are still appearing, such as a reason for leaving not yet heard. If new circumstances keep appearing, we will add interviews if the timeline permits; if not, we will report them as unexplored." The log of each review, with its reasoning, becomes part of the methods.

Limits and Cautions

• Saturation is a judgment, not a statistic. Avoid reporting it as proven.
• A saturated sample can still be biased if the wrong people were reached. See purposive sampling.
• Do not carry a figure such as "12 interviews" from one study into a different population or question.
• Time and budget often decide the sample. Say so, and state what the sample cannot cover.

Frequently Asked Questions

How Many Interviews Do I Need?

There is no single answer. Base the plan on the aim, the variety in the population, the approach, and the decision. Studies suggest narrow ranges only for homogeneous samples with narrow aims.

Can I Say We Reached Saturation?

Only if you say what kind, how you checked, and in which approach. In reflexive thematic analysis the claim does not fit.