The relationship among the three most frequently used environmental tobacco smoke measuring techniques was examined in this study: (1) indoor nicotine concentrations, (2) time-activity data, and (3) urine cotinine levels. Three families that included members who smoked and one family whose members did not smoke were selected from central Taiwan to participate in the study. The children's urine cotinine-creatinine ratios and the nicotine concentrations in the living rooms varied significantly between smokers' and non-smokers' families. The multiple linear regression model indicated that both a child's cotinine-creatinine ratio, established by daily first-urine samples, and the day-averaged urine samples correlated well with the daily butt counts multiplied by the number of hours spent in the living room. It was concluded that daily first urine cotinine-creatinine ratios, as well as day-averaged urine cotinine-creatinine ratios, can predict the environmental tobacco smoke exposure of the previous day.