Standardized allelic richness was estimated by resampling 1000 times to standardize to the smallest sample size (n = 30) to account for differences between sampled locations, using StandArich R package [ 28].
The lower bound is the smallest sample size that would yield the desired level of precision for some model, while the upper bound is the smallest sample size that would yield the desired level of precision for all models.
For comparison, the reported descriptive statistics are for the smallest sample size regression of 2,465 observations that are used in analysis.
Our example illustrates that minimizing cost to achieve 80% power is not as simple as finding the smallest sample size capable of achieving 80% power, since smaller sample sizes require each subject to be scanned longer.
Two tendencies are discernible across all designs: The largest deviations can be found for the smallest sample size (n=100) combined with the fewest items (k=5, i.e. d f=4).
As all datasets have different country samples, the analysis is further restricted to a comparable sample, again taking IMD as a reference, because it has the smallest sample size, and because IMD country rankings are sample-dependent.
The estimates of allelic richness were standardised to the smallest sample size in our dataset, N = 17.
For dataset B, k- and a-LoCoH have nearly identical accuracy except for the smallest sample size where a-LoCoH obtains a smaller error.
We chose 25 individuals as the smallest sample size because smaller samples would yield too few individuals per stage class to resemble what a real study might sample.
Allelic richness was calculated using CONTRIB [62] with rarefaction to the smallest sample size obtained minus 1 to produce estimates that are unbiased by differences in sample sizes.
Almost all models showed significant predictive ability based on jackknife manipulations within time periods (all P<0.05, except for H4 and GI8 Neanderthals, the periods with smallest sample size and most restricted distributions).
