I am currently conducting a study on the motivations of older adults in the use of online communities. I have interviewed several informants and conduct a thematic analysis in the interview transcript. I would like to know what will be the minimum sample size so that I can check if I have theoretical saturation. Any reference material you can recommend?
Versions of this question have been asked several times on RG, so I suggest you use the search functions to read those discussions.
The short answer is that there is no known number at which you can expect theoretical saturation to occur. You have to determine that through your own data collection.
You may find the following materials important:
- Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). sage.
-Creswell, J. W. (2nd ed). (2007). Qualitative Inquiry and Research Design: Choosing Among Five Approaches. USA: SAGE Publication.
-Mason, M. (2010). Sample Size and Saturation in PhD Studies Using Qualitative Interviews. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 11(3). Available from http://www.qualitative-research.net/index.php/fqs/article/view/1428/3028
Is it not rather a question of approach? You could, for example, recunstructing on cases and define only the sample of your selcetion.
Like Flick (1999):
Social constructions of change: qualitative methods for analysing developmental processes
Uwe Flick https://doi.org/10.1177/053901899038004007
Versions of this question have been asked several times on RG, so I suggest you use the search functions to read those discussions.
The short answer is that there is no known number at which you can expect theoretical saturation to occur. You have to determine that through your own data collection.
1. Guest, Bunce, and Johnson (2006) propose that saturation often occurs around 12 participants in homogeneous groups. ( Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and variability. Field Methods, 18(1), 24. doi: 10.1177/1525822X05279903 )
2. Crouch & McKenzie (2006) propose that less than 20 participants in a qualitative study helps a researcher build and maintain a close relationship and thus improve the “open” and “frank” exchange of information. ( Crouch, M., & McKenzie, H. (2006). The logic of small samples in interview-based qualitative research. Social Science Information, 45(4), 18. doi: 10.1177/0539018406069584 )
Now, its your choice. Thank you.
Thank you all for your valuable insights. Currently reading your recommended citations.
Note that the Guest et al. (2006) study is about code saturation (i.e., the point where no new codes appear) rather than theoretical saturation (i.e. the point where no new insights appear). For the second of these, see Hennink, Kaiser, & Marconi (2017)
David L Morgan Muhammad Abdullah Avais Thank your for your advise. All along I was inclined to go for theoretical saturation until I revisited my research questions and read Guest et. al (2006). My direction is clear and will be looking for code saturation during my analysis.
I personally think that code saturation is a weak standard. In particular, if you stop when you no longer find a new code in an interview, you may not have enough data about how those final, rarer, codes are related to the other aspects of your data.
To put it another way, if all you want to know are what is in your data, then it would be enough to stop when you no longer get any codes. But if you want to understand how and why your codes are related, then you may well need more data than that.
In qualitative research the sample size adequacy based on the information saturation that you have got from the respondents.
You may get more information from these articles.
https://pubmed.ncbi.nlm.nih.gov/22803288/
Article Characterising and justifying sample size sufficiency in int...
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size