Expecting a sample size ratio between group 1 and 2 of 0.54, a required sample sizes of 128 patients per group (alpha = 0.05, Fisher's exact test, 2-sided) for a power of 80% will be needed.
At the first stage, the list of wards from all cities and towns of the state/UT formed the urban sampling frame from which a required sample of wards, that is, primary sampling units (PSUs), was selected using probability proportional to size (PPS) sampling.
With a type I error probability of 5% and a power of 80%, we estimated a required sample size required of 500 subjects.
Such a situation does not apply to typical sample size calculations, but we expect a required sample size of about 190 patients (expected effect sizes of f=0.15 (medium effect as usually expected for criterion of clinical important amount of change) for multiple regression analyses under the condition of α=0.05 and 1−β=0.80 (as commonly defined)).
The specifications above lead to a required sample size of 104 children per group, which we rounded to 105 per group so that 35 therapists would be required per group.
This indicated a required sample size of n = 60 (20 per group) to detect a large effect size (f = 0.4), with power = 0.95, α = 0.05.
Sample size calculations based on an alpha of 0.05, 10 predictor variables, expected effect size of 0.20 (moderate effect size), and power of 0.80 indicated a required sample size of 91 subjects.
In general statistics in Cohen's paper [6], a required sample size n (the number of cases) for the paired t test is estimated as n = 90 when the effect size, the statistical significance, and the power were 0.3, 0.05, and 0.8, respectively, from an a priori analysis.
These assumptions yield a required sample size of 260 patients.
After calculations, a required sample of 985 nurses was determined.
This resulted in a required sample size of 242 herds.
