Site Admin|Contact|Home 


research ‘terminology’ and about ‘data saturation’: How many interviews are enough?

Wednesday, May 23rd, 2007 @ 4:43 pm
Leave a  comment
Posted in  Methodology

 talking about methodology and method, i shall once again quote my greek ancestors and give the etymology of the word: Method (from Greek methodos or met hodos meaning “way across”). so, this is again a path, a journey, a ‘mobile’ force….including some sort of travelling…

today, when in tutorial with William, i learned a few new terms of ‘research terminology’ that is useful to know,also for my review panel coming up (for example: saturation, snowballing etc). I also wrestle with the question: How many participants are enough?

I found two good links that give explanations in easy way. Click here and also here

Also, around the question:How many interviews are enough in qualitative research? I found this article here that argues that in a study with 60 participants who volunteered for in-depth interviews, saturation occured in the first 12 ones…

I asked Chris about how did he justify his 16 participants in his heuristic study and he responded by saying:

“I wrote the following in my thesis (p49):
Moustakas recommends ‘perhaps as many as 10-15 co-researchers’ (ibid p.47). One of the problems of the legacy of Positivism is the belief that, in some way, larger numbers of participants = better data. In fact, as Kvale (1996) argues it is easy to end up with ‘too much to handle in a meaningful way’ (ibid p.178). Originally I imagined twenty to thirty participants- a view that did not survive seeing the first transcript of over thirty pages of rich text. Finally I gave up having a fixed target. I do not believe that the sixteen participants, whose stories are here, along with many other interactions, represent a ‘saturation’ of the topic. Many things were repeated and yet each person added something new. What I do claim is that their stories represent a significant insight into the research question.
The panel had wanted me to talk about ‘saturation’ hence my explicit rejection of it! I was asked about it at Viva but they seemed happy enough with my answer!”

 also, Steph send an email (27.5.2007) where she talked about the ‘how many participants are enough’ by writing:

“I am ending up with only 5 participants. These are the same people from my 
earlier focus group discussion. I started out with 8 for the focus group and 2 
dropped out. Recently I had a dual role boundary and had to reluctantly drop
one  member from the study.
I have a difficulty with the concept of saturation. As I am heavily  involved
in the analysis at the moment, everytime I look at what I have  understood
about the data (including process) more meaning emerges. So the  idea that if
you research x people you will reach saturation point where nothing  new will
happen doesn’t really grab me. I dont think the more the merrier,  what I think
is small is beautiful. I guess its if you want breadth or  depth data. What I
am seeking is a possible truth about a phenomenon,  not the truth. Clearly
there has to be a cut off point and it may be the  job  of some research is to
pave the way for further study.  especially if it is a relatively unknown
phenomenon. The thought of analysing  more than a small number of participants with
my topic of dreams,to the depth I  want to go would I think be unrealistic”

Print this post

Leave a Comment:

Your comment: