The most common use of individual interviews before focus groups is to hear from key informants who can help you do a better job with things such as recruiting participants, writing interview guides, and moderating effectively. In your case, you might want to talk to 1 or 2 people from each of your key categories to help maximize the effectiveness of your full set of interviews.

This approach is particularly appropriate if you are less familiar with the interview topic or a set of participants, or if you are more familiar with some categories of participants and stakeholders than others. You typically would not include these key informants in the subsequent focus groups.

Alternatively, when starting with focus groups and then moving on to do individual interviews, the typical goal is to get more depth and detail on the topics that appear in the group interviews. That might involve identifying people who seem to be "typical cases" or you might pick people with more unusual experiences to expand your coverage of the topic. Either way, the idea is to follow-up by interviewing the people who can do the most to advance your understanding.

I discuss both of these design strategies in my book, Focus Groups as Qualitative Research, SAGE 1996.

In addition, Jana has raised the possibility of something like "member checking" where you go back to participants at a later stage in the research to hear their responses to your results. I also mention this in my book, with the idea that you can use either a focus group to comment on results from individual interviews, or vice versa.

In a different context (teen relational violence), I found it very fruitful to begin with focus groups and then to use the  broad themes that were generated in groups to narrow in and delve more deeply through one-on-one interviews. You may want to look at

Mishler, E. (1986). Research interviewing: Context and narrative. Cambridge, MA: Harvard University Press.   It is old, but a bit of a classic

I have found it very helpful to start with individual (in-depth) interviews and then explore further the issues that come out of the individual interviews in focus groups. I have used this approach in social science and health systems research using qualitative methods.

Nataskja - I would personally agree with Francis. I have tried both approaches but have found individual interviews first, then followed by focus group/s to be the most productive - particularly if the focus groups are comprised of the original individual interview participants. They've been through the less intimidating process of individual interview and, if they've enjoyed the experience, are more likely to be willing to participate further in a group setting to maintain the momentum of the study. It's never worked the other way around for me. That said, in defence of Judith, it can be 'horses for courses' depending on the context and nature of the study

I personally did them simultaneously, mainly due to time restraints, but actually found this very fruitful as I could add themes from both data collection tools as I went. It also improved my recruitment rate as some participants couldn't attend the focus groups but were keen to contribute in some way, and so were interviewed. Other's did both, either before or after the focus group, as had an interest in the discussion points but wanted to give more in-depth information via a semi-structured interview that they didn't feel they could say in a focus group. This worked well but again may vary in different contexts (e.g. my sample was with a professional body in the UK). It may be that for more sensitive topics, you might want to separate the two, as you may find some topics are not discussed in focus groups.

There is plenty of literature out there (e.g. Barbour et al 1999 Combining focus groups and interviews: Telling how it is; telling how it feels) sorry I can't think of anything more up to date off the top of my head. Great discussion point though.

In pursuing research, I try to use all possible methods in a specified time. Right now I lead FGI in conjunction with a auditorium web survey . The results lead to the deepening of the group interview. Meeting finish a detailed of the individual interview. Thus for 4 hours will test fullness of problem.

I used both in parallel and left it to the participants to choose whether they wanted to participate in an individual interview or a focus group. Some participants were willing to participate in either setting, while others preferred one format over the other. Due to time constraints the "double approach" was useful in that respect too. I got rich data through both methods, with the groups contributing to highlighting similiarities and differences in participants' experiences. Good luck with your project.

This depends on whether you are in position 1 - where wish to develop your techniques for collecting information or qualitative data and build towards themes, or position 2 - where you may have collected all or most of your information or qualitative data and wish to undertake analysis for emerging themes. The first position really means that there is a need for analysis as you proceed (identifying issues and themes as you go along) - a situation such as this may be the case if inductive reasoning predominates. Early findings from a single evidence source (interviews, observation or focus groups) for position 1 could highlight gaps and show a need for fine tuning your methodology. Therefore a focus group from early evaluation may show a need to move to individual interviews of subjects with particular attributes, or vice-versa. One or the other of these methods may indicate some observation is required? This developmental analysis can serve to build and inform next steps and models. The approach may involve lots of pre-coding, theme building and development. The second position is typical of a researcher who may be more pragmatic in terms of identifying that perhaps in order to address the need to identify and evidence themes there may be a need to undertake interviews, focus groups, observations or historical written evidence from reliable sources. This qualitative set of sources can be examined as individual methodological information or qualitative data and pre-coded, or the entire set of information or qualitative data could be examined holistically and subjected to axial coding either inductively to evidence or establish emerging themes, or semi-inductively (where there are pre-determined themes), for the former the themes would truly emerge, for latter the researcher would search for thematic evidence.

Use of software such as NVivo can greatly support those researchers who find themselves in either position 1 or position 2 

There is one further pragmatic point that you may want to consider.  In another context, wine consumption in the UK, I became aware that 10 people were responsible for 90% of all the wine imported into the UK in the off-trade. I also knew that they would all know or know of each other though many informal channels I would not have access to and that they would only agree to individual interviews because of confidentiality.  If I didn't get the first interview right the rest would almost certainly not see me. So in this research I developed the themes to be investigated with this small essential group (possibly your CCs) by holding a series of focus groups with consumers, of whom there are many, testing theories from secondary research and identifying new themes that they brought forward.  Focus groups was an appropriate setting for this group of stakeholders as wine consumption is usually a shared activity..  You might want to look at the following paper for more detail but my point is that the interview method and order used should be that which best replicates natural situations (if possible) and or respects the individual needs of the group you want to extract data from in order to maximise the richness and depth potential of the data collection.

Sobaih, A.E., Ritchie, C., Jones, E. 2012. Consulting the Oracle? Applications of Modified Delphi Technique to Qualitative Research in the Hospitality Industry International Journal of Contemporary Hospitality Management. Vol 24 [6] 886 – 906

In my opinion there is no 'right' order it depends upon the aim of the research, populations to be sampled, finance available and the time that you have to undertake the project in

Good luck

Everybody- thank you so much for your instructive comments and valuable inputs. Also for recommending literature, Judith and Samantha. This is highly appreciated!

I find it thought- stimulating to note that there appear to be two main considerations that have been taken into account when combining methods: the impact of the research design on the researchers focus on themes and further refinement, and its impact on the interview participants' readiness to (further) participate in the respective interview on the other hand. The latter is something I look forward to delve more into, also considering the remarks you made regarding the interview participants'  ease to participate in a given interview situation, Dean.

The idea of doing it simultaneously, Sigrid and Samantha, appears to me as interesting approach as well.

Interestingly, a great deal of research referred to in the discussion above, appears to to apply two or more qualitative methods to the same interview participants or target group.

While this is the case in project as well, in addition, there will be three focus groups that consist of further health care stakeholders.

That makes me give thought to the question: Would above mentioned  considerations and choices taken in combining methods differ if each interview situation would comprise different (groups of) participants?

No, in my research there were in total 4 populations sampled and both the first, the initiating group, and the last were consumers since the project was about wine consumers but the themes discussed in the last round, focus groups in this case, were significantly different from those in the first, as would be expected. 

As De Certeau suggested the discussion is always circulating forwards but there is likely to be a logic in how influence on the process under investigation is exerted and that may suggest the order of the interviews/ focus groups so that after analysis the themes identified in one iteration can be built upon and or tested in the next

Nataskja, if your research design/schedule allows it may also be fruitful to decide "on the spot", depending for instance, on which participant was or was not especially vocal during the focus group discussion, in particular if certain politics are involved. Some individuals are easily silenced in the group because of their particular socio-political or professional standing, and you may be able to collect more interesting/objective data from these participants in individual interviews.

Hello Nataskja, based on the project you are describing, I would recommend starting with the focus groups. Focus groups allow you to get a good idea for what the broad topics and issues are in inter-disciplinary co-operation in cancer care. You can then follow up on this in the in-depth interviews with the advantage of already knowing some of the areas you want to better understand and learn more about. However, what might be a useful idea to think about is to have a final focus group / workshop where you present results of your research and invite feedback. This might be particularly relevant if one of your research aims is the suggestions of best practices to improve cooperation.

Caroline, Stephanie, Jana, thanks a lot for your very instructive remarks! Jana, have you tried the approach yourself? If so, what is your experience with follow-up/ feedback focus groups, inhowfar did they impact the results?

Caroline, my apologies for missing out your answer earlier- it is first now that your comment is visible to me. Anyhow, the point you made adds a very interesting perpective to the discussion, considering the indeed small population of CCs. I've read your paper with great interest.

Going through all comments once again, it appears to me that one of the mot crucial experiences made in combining methods is to bear in mind the needs and characteristics of the groups/interview respondents and create interview situations which supports the most natural flow during the interview. It is very interesting to read which different kinds of solutions those of you came up with, to whom a neither FG/Ind.I or Ind.I/FG seemed to be the perfect fit. Be it parallel approaches, circulating methods or a follow up focus groups - it inspires me to think a little "out-of-the-box" and give a thought to possible adaptions of methodological combinations to get the best fit for the project at hand. 

Anyhow, this might bring along some practical concerns to consider. Has anyone experienced challenges or difficultis in introducing new facettes of established methodological procedures?

For the moment I have no time to answer to your critical question. For a long time I trioed to use with go and by these two methods because it is a way towards new knowledge

Nataskja, we are trying the approach I've described in a current project but it is much too soon to let you know how it went. Maybe in Spring next year  ;-)?!

All the best with your research!

The most common use of individual interviews before focus groups is to hear from key informants who can help you do a better job with things such as recruiting participants, writing interview guides, and moderating effectively. In your case, you might want to talk to 1 or 2 people from each of your key categories to help maximize the effectiveness of your full set of interviews.

This approach is particularly appropriate if you are less familiar with the interview topic or a set of participants, or if you are more familiar with some categories of participants and stakeholders than others. You typically would not include these key informants in the subsequent focus groups.

Alternatively, when starting with focus groups and then moving on to do individual interviews, the typical goal is to get more depth and detail on the topics that appear in the group interviews. That might involve identifying people who seem to be "typical cases" or you might pick people with more unusual experiences to expand your coverage of the topic. Either way, the idea is to follow-up by interviewing the people who can do the most to advance your understanding.

I discuss both of these design strategies in my book, Focus Groups as Qualitative Research, SAGE 1996.

In addition, Jana has raised the possibility of something like "member checking" where you go back to participants at a later stage in the research to hear their responses to your results. I also mention this in my book, with the idea that you can use either a focus group to comment on results from individual interviews, or vice versa.

Hi Nataskja.

i think it does depend on your research question and how inductive you want to be.  I  did find this article in the journal of advanced nursing which might help?  The article raises some interesting points. 

Lambert SD & Loiselle CG (2008). Combining Individual interviews and focus groups to enhance data richness J.Adv Nurs 62(2) pp228-237.

http://www.ncbi.nlm.nih.gov/pubmed/18394035

http://www.ncbi.nlm.nih.gov/pubmed/18394035

Another way to use both interviews and focus groups is to see how participants answer in different contexts. For example, I did focus groups with family members as well as individual interviews with each family member to see how answers changed when family members were away from each other: http://ijoc.org/index.php/ijoc/article/view/2250

I also have done research where I have dialogued with focus groups about friends-with-benefits relationships. I then interviewed members of those focus groups separately, and it allows me to consider what people share individually that they do not say when part of a group discussion.

I imagine interviews and focus groups could be used in many other intriguing ways to get a sense of what people will share in one context versus what they share in another.

yes you may use them both in one study, if in your study you have two strata of sample and the other sample comprise fewer participants that may not be accommodated in a focus group interview. this just for practical reasons.

you may also use individual interviews as follow up of focus group discussions. this is normally for validation of data collected in a FGD.  

Nataskja, your quest is great, but hard to answer without more information. As others have indicated, the answer depends on the nature of the research questions you want to answer, and how much you already know about the topics you are exploring. David Morgan knows more about FGDs than nearly everyone else. I have used both methods many times, but the information collected from group discussions are most valuable for learning about terms and concepts, and not about behaviour. Individual interviews provide narratives that show how respondents make logical connections for themselves. So if you are seeking to understand individual experiences, then use interviews. If you want to compare knowledge of different groups about a topic then FGDs will give you that.

Waal, very informative discussion. I am conducting one-on-one interviewing lecturers, followed by observation and then use students focus group interviews on validating what the lecturers claimed to be doing.

Please help if this is academically sound.

It might also depend on the study population or sample. at times you find that in other settings where you want to conduct interviews that is enough individuals to fit in a focus group and when you visit another setting you actually encounter few people to fit in a focus group and that is when you may decide to go in-depth individual interviews. Yes, at times the other strategy might be used to verify data collected through another strategy e.g. focus groups followed by in-depth individual interviews.

This is an interative process. I usually fins that the most pragmatic approach is to approach the group where there are the most potential participants first and gather that data set, then I move with more focussed and validated questions to the second sample set, if I feel that the results still need validating I may return to the first, more numerous population again.  In some ways this is like a modified version of the Delphi Technique which you might want to look into.

Thank you, Caroline and Thandisizwe for your contribution to my question

Nyarai

Hello, i want to find out if you can use the same set of questions for indepth interviews and focus group discussion  and then compare to see if group responses differ or are samilar to one on one interview responses.

Can this be done in one study area?

Hi Esi! I would think this is possible. It is an interesting question and I would recommend to publish it on your profile, to make sure that other researchers can see and discuss it! Best wishes, Nataskja

Dear ALL: Does it mean that it is academically correct to use both Focused Group and Interview together in a qualitative research?

While researchers often use individual interviews and focus group discussions in the same study, it would not be appropriate to use the same questions in both exercises. The context of individual interviews allows you to vary and modify questions according to the answers and comments provided. Such answers may provide information about actions undertaken. In a group discussion, you want the group to consider certain general themes or general questions,and the discussion may go off in various directions. But one does not get information about individual actions in a group discussion. One gets ideas and opinions.

I concur with previous contributors, especially with Stan Yoder. In sensitive or controversial fields such as family planning, HIV and abortion, for example, Focus Groups have been used extensively to provide the "context", the "language" (including euphemisms) and overall perceptions (e.g., stigma, shame) of people about the subject. Results of these FGs may give rich insight to design individual interview questions, and even questionnaire items for quantitative surveys. All this is good input and contribution to the current trend of "mixed methods" approach in social science research. Good luck!

I used semi-structured interviews followed by sets of 2 focus groups in research with women who had mothered babies while enduring domestic violence. The research yielded deep and rich findings while allowing the women to find common ground, discuss differences and support each other.

I always begin my research by deep individual and recurrent interviews through which I try to build what could be a focus group. It is a rhyzomatic way of understanding the relation between statistic and quantitative societal data and the subterranean desires and objectives of individuals and families