it seems you will use independent sample ttest to do the comparision. Depend on the effect size, the power and alpha in your study variables. if the effect size is moderate, power .80, and alpha .05, then you need at least 34 participants per group.
it seems you will use independent sample ttest to do the comparision. Depend on the effect size, the power and alpha in your study variables. if the effect size is moderate, power .80, and alpha .05, then you need at least 34 participants per group.
I agree with prof. Muayyad Ahmad. In literature, the are many formulas and calculators available to calculate sample size and the researcher is confused about which formula to use.
I suggest this paper https://www.karger.com/Article/FullText/322830
You should first know your hypothesis, afterwards know the population of the groups either from literature or pilot study.
This will enable you to have an idea of the kind of research you want and will help you to formulating your questionnaires, this will allow you to choose the kind of statistical analysis you want to consider.
It is advisable to use the gpower software to estimate your sample size.
p: The prevalence of the condition/ health state. I
q: i. When p is in percentage terms: (100-p)
d (or l): The precision of the estimate.
Za [Z alpha]: The value of z from the probability tables. If the values are normally distributed, then 95% of the values will fall within 2 standard errors of the mean.
If your samples are independents and equal in size and the two populations have the same variance, alpha is 0.05, power 0.80, and your effect size is medium, I would suggest an approximative but simple formulat: 16/d², which gives 64 for each sample.