We used two methods to collect data on indoor smoking exposure of 3-month-old infants. First, parents of approximately 100 children completed a questionnaire. We then measured nicotine in the air of the living rooms in smoking and non-smoking households with a passive sampler for a period of 2 weeks, several months after the questionnaire had been completed. Smoking habits reported in the questionnaire generally with reported number of cigarettes smoked during the measurement weeks, and with nicotine concentrations in the air. These results suggest that exposure classification based on questionnaire data is likely to be reasonably valid.