Hi,
I am conducting a thematic analysis, and could use an advice what comes to the power and theme prevalence calculation. I have 11 participants with 80% power. I have completed the familiarizing, coding, grouping codes into themes and revision. However, I am struggling to find information how to present the strength of each theme.
Should I calculate the theme prevalence by comparing the number of codes in one theme with the full dataset? Or with the number of overall codes selected among all themes? Somewhere I also saw a suggestion to calculate the theme prevalence by comparison with the lest prevalent theme: "(a) the expected population theme prevalence of the least prevalent theme, derived either from prior knowledge or based on the prevalence of the rarest themes considered worth uncovering, e.g. 1 in 10, 1 in 100".
A practical example: I have 45 codes and 5 themes. 19 codes support the most popular theme (Happiness) and 5 codes the least popular one (Anxiety). Therefore, is it correct to say that the theme prevalence of 'Happiness' is 42.22% and of 'Anxiety' 11.11% ?
With the desired instances of 2, power of 80% and sample size 11 the population theme prevalence should be 25%:
Article Supporting thinking on sample sizes for thematic analyses: a...
In respect to this, the theme 'Anxiety' is not significant enough as a theme, right? Shall I still mention it (it is significant to my research question.)
Is it correct to adjust the desired number of theme instances afterwards? The power analysis for qualitative and especially thematic analysis doesn't seem like a strict requirement, more of a supporting tool, but I would like to include this into my report.
Would really appreciate some advice and clarification. Thank you!
We should not lose sight of the fact that this is a qualitative study and the sample is non-random and usually small. It is not aimed for statistical significance and/or generalizability. If you have coded, categorized and come up with themes. I think it is enough to report your themes. I find the use of terms like "prevalence" and "power" in qualitative studies, misleading as they imply statistical significance which the study is not designed for.
The prevalence is calculated by the formula - the absolute number of phenomena in the aggregate of observation units multiplied by a thousand and divided by the total number of observation units in the aggregate.