Stansfield and Carlton [Human Biology 79:255–260 (2007)] reported that the distributions of the combinations of the sexes in human sibships are binomial. They inferred that the probabilities of male births are equal and independent within and across all sibships. Here I argue that their argument is both invalid and false. Contrary to their inference, a binomial distribution may result when equal and counterbalancing measures of Poisson and Lexis variation are simultaneously present. These conditions are approximately met with respect to human births.
James, William H.
"Variation of the Probability of a Male Birth Within and Between Sibships,"
1, Article 2.
Available at: http://digitalcommons.wayne.edu/humbiol/vol81/iss1/2