Article Summary: Sample sizes in qualitative research can be much lower than sample sizes in quantitative research. The key is having the right participant segmentation and study design. Data saturation is also a key principle to understand.
Qualitative research is a bit of a puzzle for new practitioners: since it is done via interviewing participants, observation, or studying people’s patterns and movements (in the case of user experience design), one can’t obviously have a huge sample size that is statistically significant. Interviewing 200+ people is not only incredibly time-consuming, it’s also quite expensive.
And, moreover, the goal of qualitative research is not to understand how much or how many. The goal is to collect themes and see patterns. It’s to uncover the “why” versus the amount.
So in this post, we’re going to explore the question every qualitative researcher asks, at one point or another: How do you justify the sample size in qualitative research?
Here are some guidelines.
Qualitative sample size guideline #1: Segmentation of participants
In qualitative research, because the goal is to understand themes and patterns of a particular subset (versus a broad population), the first step is segmentation. You may also know of this as “persona” development, but regardless of what you call it, the idea is to first bucket your various buyer/customer types into like-categories. For example, if you’re selling sales software, your target isn’t every single company who sells products. It’s likely much more specific: like mid-market sized VP-level sales execs who have a technology product and use a cloud-based CRM. If that’s your main buyer, that’s your segment who you would focus on in qualitative research.
Generally, most companies have multiple targets, so the trick is to think about all the various buyers/consumers and identify which underlying traits they have in common, as well as which traits differentiate them from other targets. Typically, this is where quantitative data comes into play: either through internal data analysis or surveys. Whatever your process, this is step 1 to figure out the segments you will be bucketing participants into so you can move into the qualitative phase, where you’ll ask in-depth questions, via interviews, to each segment category. At this stage, it’s time to bring in your recruiting company to find your participants.
Qualitative sample size guideline #2: Figure out the appropriate study design
After you’ve tackled your segmentation exercise and know how to divide up your participants, you’ll need to think through the qualitative methodology that is most appropriate for answering your research questions. At InterQ Research, we always design studies through the lens of contextual research. This means that you want to set up your studies to be as close to real life as possible. Is your product sale done through a group discussion or individual decision? Often, when teams decide on software or technology stacks, they’ll want to test it and talk amongst themselves. If this is the case, you would need to interview the team or a team of like-minded professionals to see how they come to a decision. In this case, focus groups would be a great methodology.
Conversely, if your product is thought through on an individual-basis, like, perhaps, a person navigating a website when purchasing a plane ticket, then you’d want to interview the individual, alone. In this case, you’d want to choose a hybrid approach, of a user experience/journey mapping exercise, along with an in-depth interview.
In qualitative research, there are numerous methodologies, and frequently, mixed-methodologies work best, in order to see the context of how people behave, as well as to understand how they think.
But I digress. Let’s get back to sample sizes in qualitative research.
Qualitative sample size guideline #3: Your sample size is completed when you reach saturation
So far we’ve covered how to first segment your audiences, and then we’ve talked about the methodology to choose, based on context. The third principle in qualitative research is to understand the theory of data saturation.
Saturation in qualitative research means that, when interviewing a distinct segment of participants, you are able to explore all of the common themes the sample set has in common. In other words, after doing, let’s say, 15 interviews about a specific topic, you start to hear the participants all say similar things. Since you have a fairly homogenous sample, these themes will start to come out after 10-20 interviews, if you’ve done your recruiting well (and sometimes as soon as 6 interviews). Once you hear the same themes, with no new information, this is data saturation.
The beauty of qualitative research is that if you:
- Segment your audiences carefully, into distinct groups, and,
- Choose the right methodology
You’ll start to hit saturation, and you will get diminishing returns with more interviews. In this manner, qualitative research can have smaller sample sizes than quantitative, since it’s thematic, versus statistical.
Let’s wrap it up: So what is the ideal sample size in qualitative research?
To bring this one home, let’s answer the question we sought out to investigate: the sample size in qualitative research.
Typically, sample sizes will range from 6-20, per segment. (So if you have 5 segments, 6 is your multiplier for the total number you’ll need, so you would have a total sample size of 30.) For very specific tasks, such as in user experience research, moderators will see the same themes after as few as 5-6 interviews. In most studies, though, researchers will reach saturation after 10-20 interviews. The variable here depends on how homogenous the sample is, as well as the type of questions being asked. Some researchers aim for a bakers dozen (13), and see if they’ve reached saturation after 13. If not, the study can be expanded to find more participants so that all the themes can be explored. But 13 is a good place to start.
Interested in running a qualitative research study? Request a proposal >
Author Bio: Joanna Jones is the founder and CEO of InterQ Research. At InterQ, she oversees study design, manages clients, and moderators studies.
