Increasing within-country migration from rural to urban areas is setting new challenges for survey organization. For example, the educational level of population in urban and rural areas differ, resulting in differences in health behaviours and health outcomes between areas. Data from the national cross-sectional surveys of the FINRISK Study conducted in Finland in 1997-2012 among the adult population were used. Women living in the capital region were more likely to be survey non-participants than women living in rural areas. The differences between rural and urban areas have diminished over time among men, but not among women. Even though participation rates between different levels of urbanizations are becoming similar, observed differences in population profiles - that is, educational level - between areas may bias survey results unless results are adequately adjusted.