During the research, it is necessary to determine how many respondents to involve to find sufficient information. At the same time, the research needs to be cost and time effective. Before determining the number of respondents, it is important to specify what the in-depth interview actually has to accomplish to be considered valid.
Research evaluation criteria
There are many criteria for evaluating research. Most of them rely on the following:
- Credibility – the aim is to prove that we have sufficiently identified the research subject. Adequate study duration, consultation with team members on established hypotheses, involvement of different respondents within the target group, and feedback on research results increase credibility. Thanks to this, we contribute to the objectivity of the entire research and eliminate the researcher's subjectivity.
- Transferability – the results of our research can be used for a similar type of research in another environment or with a different group of people.
- Credibility – we use multiple sources of information, different groups of respondents, and different data collection methods. Inviting an independent expert to review the research process and results is ideal.
- Verifiability – is related to the review of the whole research and states that it is necessary to have access to the whole process. It is not enough to see only the results. It is recommended to review everything from raw data collection methods through analysis, synthesis, methodological notes, and other follow-up steps.
So how many respondents do we need?
The concept of saturation is often mentioned in research methods. It sets a threshold if incoming data contains little or no new information. Based on meta-saturation studies in qualitative research, we now know that 6 to 8 interviews can identify most new ideas and data. With 10 to 12 interviews, we can achieve up to 92% saturation. On average, it is recommended to conduct 6 to 12 interviews – the exact number will depend on whether you have data from previous surveys, your time and financial capacity, and whether you will combine the in-depth interview method with the focus group, observation, other methods or your previous experience and knowledge. I would like to emphasize that the number of interviews always relates to one persona or one target group due to the size of the sample and the depth of the expected knowledge. If you have more than one persona or target group, you should focus on each of them separately and pay close attention to them.
From our experience, we most often conduct in-depth interviews with 6 to 8 respondents per persona or target group. Due to that, we are able to penetrate into the research problem more than sufficiently, and it also makes sense from an economic point of view. If you are interested in in-depth interviews, you can go through my article, in which I get into more detail. The in-depth interview is just one of the methods we use. If you are interested in how many respondents you need for usability testing, you can read about that in the article from Michal.
- Jan Hendl – Kvalitativní výzkum: Základní metody a aplikace, ISBN: 80-7367-040-2