This cross-sectional study was carried out in October, 2001, with the aim of measuring formaldehyde levels in 399 homes representative of those in the central districts of Ankara province. Measurements were carried out with a Formaldemeter 400 in the living rooms and kitchens of these homes. The average formaldehyde levels in living rooms and kitchens were 0.064 and 0.060 ppm, respectively. The correlations between formaldehyde levels and type of house, construction materials, and method of ventilation were assessed. Formaldehyde levels were found to be significantly higher in apartments than in detached houses (living room: chi(2) = 38.7, p < 0.001; kitchen: chi(2) = 43.2, p < 0.001), significantly higher in reinforced concrete homes than in brick and mortar homes (living room: chi(2) = 43.1, p < 0.001; kitchen: chi(2) = 34.9, p < 0.001), and significantly higher in homes using mechanical methods of ventilation than in homes with natural ventilation only (chi(2) = 6.2, p < 0.05). People living in homes with formaldehyde levels higher than 0.10 ppm had a significantly higher incidence of watering eyes, dry throats, and running noses than people living in homes with lower levels. The correlation between kitchen formaldehyde levels and the type of fuel used in the kitchen was assessed, and homes using natural gas in the kitchen (53.4%) were determined to have significantly higher levels than homes using bottled propane (46.6%) (chi(2) = 48.8, p < 0.001). The results of the study show a significant correlation between formaldehyde levels and the type of home and construction materials, and between formaldehyde levels and residents' complaints of watering eyes, dry throats, and running noses. This study was carried out during the warm season. The results suggest that a similar study should be conducted during the winter, when heating is used and ventilation is less adequate.