The kinship coefficients and sample size reduction with respect to allele frequency estimation of common relative pairs are listed in Table 1, and those for sibships with 1,2, … siblings are listed in Table 2. For more complicated relationships or pedigrees with loop, one can consult (Maruyama and Yasuda, 1970; Lange 1997). Several rules-of-thumb can be stated: two siblings contribute 1.333 samples, uncle-nephew pair contributes 1.6 samples, three siblings are equivalent to 1.5 samples, etc. If the relationship between two pedigree members is distant, the correlation is close to zero and they can be treated as two independent samples (e.g., second cousins contribute 1.94 samples). For larger sibship, there is a diminishing return in adding extra sibs: adding the second, the third, the fourth, and the fifth sibs only adds 0.333, 0.167, 0.1, 0.067 samples. Even in the limit of infinite number of sibs, the effective sample size can not be larger than 2, as the extra sibs merely resample the finite pool of four parental alleles.
