In the case of the respondent's perception analysis, the sample size was 30 for making the perception of two factors with 4 levels and 3 levels.
Abdur Rahman it depends: the sample n=30 is for how many population units N=? It was a probabilistic sample design or non-probabilistic sample design? If probabilistic, how was the sample calculated?
Hello Abdur Rahman :
In order to know if 30 is enough, you need to know the population size, how accurate you believe your samples are, the margin of error that you are willing to allow...
Without more information regarding your study, it is difficult to calculate the smallest sample size required in an accurate way.
However, I understand that sometimes is difficult to know the population size for perception analysis. In this context, most authors suggest a data range of 5 to 10 times the number of factors used (you can find more information about sample size in this book: "Multivariate Data Analysis - A global perspective", from Hair et al., 2010).
I hope it helps.
in addition you must take care of the minimum 'effect size' you consider as relevant, clinical significance not statistical one. The decision on clinical significance is up to the expert in the field and influences the sample size.
(see attached)
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size