One of the few consistent findings originating from the research in schizophrenia is the high familiality in the transmission of the disease. Data from family studies have been used in the attempt to disentangle the genetic from the environmental components of this familiality. The analysis of this body of research has led several groups to the conclusion that there is a higher heritability of schizophrenia through the female rather than the male lineage and that the concordance for the illness is higher among DZ twins than siblings. These assumptions have generated a flurry of hypotheses on the aetiology of schizophrenia. This article demonstrates that the above findings can instead be explained as the result of an artefact originating from the uncertainty in the paternity of the probands. None of the studies has in fact considered that, in some populations, up to 30% of children are not genetically related to their putative fathers. This issue is potentially important for the genetic studies in all pathologies, but has a particular relevance for schizophrenia, due to the high profile acquired by family studies in the research for the aetiology in this illness. The conclusions reached in most family studies in schizophrenia should therefore be reappraised.