Confounding fuels misinterpretation in human genetics.

The scientific literature has seen a resurgence of interest in genetic influences on human behaviour and socioeconomic outcomes. Such studies face the central difficulty of distinguishing possible causal influences, in particular genetic and non-genetic ones. When confounding between possible influences is not rigorously addressed, it invites over- and misinterpretation of data. We illustrate the breadth of this problem through a discussion of the literature and a reanalysis of two examples. The first paper we discuss suggested that patterns of similarity in social status between relatives indicate that social status is largely determined by one's DNA. Our reanalysis shows that the paper's conclusions are based on the conflation of genetic and non-genetic transmission (for example, of wealth) within families. The second paper we discuss posited that genetic variants underlying bisexual behaviour are maintained in the population because they also affect risk-taking behaviour and thereby confer an evolutionary fitness advantage through increased sexual promiscuity. In this case, too, our reanalysis shows that, though possible explanations cannot be distinguished, only one is chosen and presented as a conclusion. We discuss how issues of confounding apply more broadly to studies that claim to establish genetic underpinnings to human behaviour and societal outcomes.