When only two participants are involved, I have argued this should be treated as a "dyadic interview" because the interaction is more like a conversation between two people rather than a group discussion. My personal preference is to think of group discussions as involving four or more people, which leaves three participants as a "gray area."

If you look at my publications here on RG, you will find at least two related to dyadic interviews.

Matthias - the 'numbers' needed for focus groups varies according to a number of factors i.e. topic, availability, sampling method etc. I would argue that you should still stick with the focus group methodology (and it is a methodology in its own right) if that's how they were originally structured. You can't classify them as 'individual' interviews, case studies etc. It will be up to the reviewers and editors, if you are submitting your work, to comment/decide as to the acceptability of low group numbers - and that would be an unknown until you do.

The attached chapter may assist a little further.

Dear researcher

The ideal size of a focus group for most noncommercial topics is five to eight participants. Don’t plan

focus groups with more than 10 participants because large groups are difficult to

control and they limit each person’s opportunity to share insights and observations.

In addition, group dynamics change when participants want to but aren’t able to

describe their experiences. For example, if people do not have an opportunity to

share experiences in the total group, they may lean over to the next person to whis-

per. This phenomenon is a signal that the group is too large. Small focus groups, or

mini-focus groups, with four to six participants are becoming increasingly popular because the smaller groups are easier to recruit and host and are more comfortable for participants.

Good luck

When only two participants are involved, I have argued this should be treated as a "dyadic interview" because the interaction is more like a conversation between two people rather than a group discussion. My personal preference is to think of group discussions as involving four or more people, which leaves three participants as a "gray area."

If you look at my publications here on RG, you will find at least two related to dyadic interviews.

For the record, I think that individual interviews often provide better data than focus groups do. In a focus group, an individual may be more reluctant to speak up, particularly if their opinion differs from that of the rest of the group. In one-on-one interviews, this is less likely.

David L Morgan thank you very much for your explanation. I tried to make some reasearch for this but it seems to be that dyadic interview are for couples or persons who know each other ? Because that would not be right in our case.

Kind regards

I don't think that the number of participants is important to define a group. Rather than a conceptual viewpoint (which will vary a lot depending on the discipline you are in), I suggest adopting a pragmatic approach based on your objectives. As a method, focus group is most useful to explore in depth different views on the topic you are researching. If you are working with only 2 or 3 participants and their answers are sophisticated and complex, than you may still be able to meet your goals. You may want to give them a list of questions in advance, so they will have time to reflect on their answers.

Usually, we have focused groups, with s- Not one group. The number of participants in each group varies between 6-10. If the groups are too small, the discussion will be slow and usually led with one person. Again if the participants are too many, most likely a good number of them will be just listening.

Matthias Kernig, you are right that most of the earlier literature on dyadic interviews is limited to pairs that have pre-existing relationships. That is not the case in what I have written, however. In addition to my book on dyadic interviewing (from Routledge), here are two papers illustrating our approach.

Also note that we found no problems with one person dominating the interviews in this setting, because the format encouraged a conversational dynamic.

Article Dyadic Interviews as a Tool for Qualitative Evaluation

Article Introducing Dyadic Interviews as a Method for Collecting Qua...

From a group perspective, a small group can comprise of 2 people (dyad) or three (triad). However, for purposes of FGDs, these small groups may not generate the rich diversity that may be desired by a qualitative researcher. In practice, the most commonly preferred group sizes are 5-8, 6-8, 8-10, 8-12 and the size may be larger than 12 depending on the context but my best experience has been working with 6-10. Still others talk of mini FGDs of 4 discussants. Normally, a group size may increase depending on the discussants' level of experience with the topic. Little experience would mean you need more people for purposes of dynamism for better results. Check if your data could be unduly affected by this.

I agree with Michael W. Marek on the likely intimidation of some members of FGD making them not to share some sensitive ideas or contrary opinion publicly. This makes individual interviews ideal. However, if a researcher has to use FGD, then there must be a careful plan to avoid intimidation of others from freely and openly giving their opinions.

Our research found that a comparison of dyadic interviews and focus groups did not produce any difference on the topics that were discussed or the depth in which they were discussed.

As for differences in the "diversity" of points of view, there is nothing known about this. On the one hand, a 6-person focus group lasting 90 minutes will produce an average of 15 minutes of speech per person, while a dyadic interview lasting an hour will produce an average of 30 minutes person. But whether it would take 3 dyadic interviews to equal the diversity in one 6-person focus group is an open question.

Article A System for Coding the Interaction in Focus Groups and Dyad...

David L Morgan thank you, really appreciate your answers. I was already reading the second article you mentioned here when I saw your message.

Kind regards

Matthias Kernig. In writing your paper, I suggest that you mention what other researchers used, and why did you decide to use this number of groups and numbers in each group, and what other considerations did you consider in planning for these sessions, and how these interviews were carried out. Peer-reviewers and editors are interested in these details with evidence for decisions made and explanations.

Dear Matthias Kernig

I suggest you to review these studies. According to Johnson and Christensen (2004); 6-12 people; Langford et al. According to (2002) and Morgan (1997); 6-10 people; According to Krueger (2000); It is enough to interview 6-9 people. At the same time, it is appropriate to ask 6-8 people 12-20 questions in the homogeneous sample to be selected in qualitative research (Kuzel, 1992).

References.

* Johnson, B., & Christensen L. (2004). Educational research: Quantitative, qualitative, and mixed approaches. Needham Heights, MA: Allyn & Bacon.

* Langford, B. E., Schoenfeld, G., & Izzo, G. (2002). Nominal grouping sessions etc. Focus groups. Qualitative Market Research: An International Journal, 5 (1), 58-70.

* Kuzel, A. J. (1992). Sampling in qualitative inquiry. In B. F. Crabtrree, W. L. Miller (Eds.), Doing qualitative research (pp. 33-45). Thousand Oaks, CA: Sage Publications.

* Krueger, R. (2000). A focus groups: A practical guide for applied research. Thousand Oakz, CA: Sage Publications.

* Morgan, D. L. (1997). Focus groups as qualitative research. Qualitative research methods series. 16. Thousand Oaks. CA: Sage Publications.

Five or more, in my opinion, but whether it can be regarded as a traditional focus group depends more on form. If the participants are being interviewed rather than allowed to also discuss in between themselves, then it is not, as far as I'm concerned, a focus group.

Group interviews with more than a dozen candidates do not always do this method