Abstract

OBJECTIVE: To investigate whether the sex ratio of children varies between fathers of different occupations. METHODS: The sex ratio (the ratio of the number of boys to the number of girls at birth) was calculated in relation to paternal occupation in the cohort of all 253,433 live births in Cumbria, north west England, from 1950-89. Exact binomial confidence intervals were used to estimate whether the sex ratio in each occupational category was significantly different from that for the rest of the cohort. RESULTS: There were fewer occupational categories with significantly different sex ratios at the 5% level than expected by chance alone, assuming the same binomial distribution of sexes at birth within each paternal occupation. CONCLUSIONS: Significant variation of the sex ratio with fathers' occupations was not found. There is some evidence that the sex ratio shows less variance than expected under a binomial model which assumes independence of the sex of each child; a possible explanation of this may be parental preference for limiting family size after children of both sexes have been born or some other factor which results in children within a family being more likely to be of both sexes rather than the same sex.