For the observation of a higher risk of Alzheimer's disease (AD) in women, causes like a general longer life expectancy of women, a longer lifespan of women with AD, a later age at onset and a lower level of education compared with men have been discussed. However, methodological artefacts could also contribute to the observed gender difference. Data from a comprehensive family study was re-evaluated, in which a gender difference in the risk of AD had been observed. Available relatives had been directly interviewed, information on unavailable ones was provided by thirds. A differential, gender specific participation led to the under-representation of older and demented men, creating a selection bias, that may be the cause for the observed gender difference in the risk of AD. Study methods must be taken into account when interpreting information on gender differences in the risk of AD.