Analytical and statistical stochastic approaches are used to model the dispersion of monogenic variants through large populations. These approaches are used to quantify the magnitude of the selective advantage of a monogenic heterozygous variant in the presence of a homozygous disadvantage. Dunbar’s results regarding the cognitive upper limit of the number of stable social relationships that humans can maintain are used to determine a realistic effective community size from which an individual can select mates. By envisaging human community structure as a network where social proximity rather than physical geography predominates, a significant simplification is achieved, implicitly accounting for the effects of migration and consanguinity, and with population structure and genetic drift becoming emergent features of the model. Effective community size has a dramatic effect on the probability of establishing beneficial alleles. It also affects the eventual equilibrium values that are reached in the case of variants conferring a heterozygous selective advantage, but a homozygous disadvantage, as in the case of cystic fibrosis and sickle cell disease. The magnitude of this selective advantage can then be estimated based on observed occurrence levels of a specific allele in a population, without requiring prior information regarding its phenotypic manifestation.