We agree with Joober and colleagues about the importance of careful statistical analysis in all research but do not agree with their criticisms of our article.1 We would like to point out that the statistical issues raised about regression analysis are not specific to testing for gene × environment interaction. They apply to testing for statistical interactions regardless of whether a given interaction term involves social × social variables (eg, family conflict × sex), gene variant × gene variant, or interaction between any other set of quantitative variables.
Thapar A, Harold G, Holmans P, Rice F, Langley K, Fowler T, Aggleton J, van den Bree M, Owen M, O’Donovan M. Promoting Measured Genes and Measured Environments: On the Importance of Careful Statistical Analyses and Biological Relevance—Reply. Arch Gen Psychiatry. 2007;64(3):378-379. doi:10.1001/archpsyc.64.3.378-b