high SES). The approach used for testing G × E interactions can vary across study design, such that adoption designs or studies with measured genes are generally required for the mean levels approach, while the moderated variance components approach may be used in both twin and adoption studies in the absence of measured genes.
