where Z= Z value in normal distribution tables (1.96 for 95% confidence level), p= the estimated proportion of the population that presents the characteristics (0.5 is used as a conservative value, higher or lower values yield a smaller required sample size), and H= the precision level or margin of error, expressed as decimal (10%= 0.1).
This results in more statistical power and smaller required sample sizes than in a parallel group design.
Thus, the inclusion of severity of impairment on the FAQ may account for the smaller required sample size calculation as a result of increased statistical power.
Due to their cross-over design, aggregated SPTs have a smaller required sample size than their conventional parallel arm RCT counterparts for equivalent levels of statistical power, and are better at controlling for confounding [ 20].
When organizations' target populations were smaller than the required sample size needed to estimate stratum-specific mortality rates (n = 1,650 households in 55 clusters; see below for details), geographically contiguous organizations' populations were pooled.
For detection/exclusion of infections at a pre-specified prevalence threshold and confidence level (e.g. 99%), the required sample is rather small, but the information obtained from the data is limited.
We estimated a required sample size based on a small between conditions effect of Cohen's f = .10.10
Simulation results indicate that a few small risk-free intervals only marginally affect the required sample size, whereas the presence of longer and more frequent risk-free intervals increases the sample size requirements to N=184 individuals.
Smallest detectable difference was calculated as 3.23, 32 giving a required sample size of six patients per group (α = 0.05, power = 0.80).
Most instruments in our study had a SD between 10-15% of the instruments' range, resulting in a required sample size of N = 400 to detect small differences between baseline and after four weeks [ 21].
