Note on the choice of statistic for testing hypotheses regarding seasonality

Season of birth in infantile autism and other types of childhood psychoses

Season of birth in 328 children with infantile autism and other types of childhood psychoses born between 1945 and 1980 was compared with that of a control population born in the same period. For some parts of the period (the years 1951-56, 1963-68 and 1975-80 combined) an excess of March born boys with infantile autism was found. For boys with autistic-like disorders, born in the same period a maximum occurred in November, while no seasonality was observed for a borderline sample.

The choice of a statistic for testing hypotheses regarding seasonality

This paper discusses statistical methods for testing hypotheses of seasonality in births across one year, using monthly frequencies. It argues that the Freedman test, a variant of the Kolmogorov-Smirnov one-sample test, and the Edwards test and its modifications are most appropriate for this purpose. These tests take into account the order of the monthly frequencies and have a higher power than the ordinary chi-square goodness-of-fit test, regarding seasonality. The paper addresses some points mentioned in earlier papers by McCullough, and O'Brien and Holbert.

Seasonality of birth in schizophrenia. An insufficient stratification of control population?

The seasonality of birth in 142 schizophrenics from Barcelona (Spain) has been studied, showing a pattern similar to that described by other authors, with a maximum in winter. If sample is compared with the local general population there are significant differences, but these disappear when the comparison is made with a control population properly stratified by age, sex, place of birth, place of residence and social status. The same results are obtained using statistical methods suggested by other authors. The seasonality pattern usually obtained in schizophrenia could be due to a selective stratification of the ill population based on variables that might or might not be related to illness.