Questionnaire surveys in several countries have consistently detected an association between symptoms and residential mould growth. Confirmation by objective measures would strengthen the argument for causality. To address this issue, quantitative and qualitative fungal measures (airborne ergosterol and viable fungi in dust) were compared to respiratory symptoms (n = 403) and nocturnal cough recordings (n = 145) in Canadian elementary schoolchildren during the winter of 1993-1994. There was a 25 percent to 50 percent relative increase in symptom prevalence when mould was reported to be present (p < 0.05). However, neither symptoms nor recorded cough was related to objective measures of mould. In conclusion, the inability to find an association between objective measures of fungus and health suggest that either these objective measures, or the traditionally used questionnaire data are inaccurate. This discrepancy limits the acceptance of a causal relation between indoor fungal growth and illness.