Calculates power directly for the case of a clustered prevalence survey where the aim is to detect the presence of *any* deletions over all clusters. This design can be useful as a pilot study to identify priority regions where high deletion prevalence is likely. Note that we need to take account of intra-cluster correlation here, as a high ICC will make it more likely that we see zero deletions even when the prevalence is non-zero.
get_power_presence(N, prevalence = 0.01, ICC = 0.05)vector of the number of samples obtained from each cluster.
assumed true prevalence of pfhrp2/3 deletions as a proportion between 0 and 1.
assumed true intra-cluster correlation (ICC) between 0 and 1.
get_power_presence(N = c(120, 90, 150), prevalence = 0.01, ICC = 0.1)
#> [1] 51.78108