The purpose of present study was to examine validity of pedometer with accelerometer (Lifecorder-EX ; LC) and to assess the daily physical activity (PA) of Japanese junior high school students by LC method. Firstly, to examine validity of LC, ten students (13.6±0.5 years old) wearing LC on their waists performed the walking and running tests. Energy expenditure was measured by the expired gas analysis method during the tests. The results showed that the activity level of LC had a high correlation with the exercise intensity (VO₂ : r=0.948, METs : r=0.931), and it was from 7 to 9 (LC7－9) in the more intense activity than running. Secondly, PA of 206 boys and 220 girls aged 12－15 years were measured using LC. Step counts and activity time of LC7－9 each were 9500－15000 counts and 10－19 minutes on weekdays, and 6500－10000 counts and 5－12 minutes on weekends. PA analyzed by MANOVA were lower on weekends than on weekdays, and decreased in proportion as grade, and that of boys were significantly higher than that of girls. Present study suggests that LC method is a good way of measuring PA for junior high school students in free-living condition.

The purpose of present study was to examine validity of accelerometer (Lifecorder ; LC) and to assess the free-living physical activity (PA) in Japanese elementary school students used by LC method. Firstly, to examine validity of LC, twelve children, aged 11.2±1.0 years, were measured total energy expenditure (TEE) by doubly labeled water method and resting energy expenditure (REE) by the expired gas analysis. Physical activity energy expenditure (PAEE) is calculated as 0.9 TEE minus REE. At the same time, PA was evaluated by both daily walk count (WC) and vigorous activity time (LC7-9) using LC method. PAEE per 1 kg body weight significantly correlated with WC (r=0.785, p<0.005) and LC7-9 (r=0.828, p<0.001). Secondly, 140 boys and 167 girls, aged 8-11 years were measured PA by LC method. WC and LC7-9 each were 13000-18000 counts and 16-32 minutes in week-day, and 8000-12500 counts and 8-18 minutes in week-end. PA analyzed by MANOVA were significantly higher in boys than in girls, and lower in week-end than in week-day, and decreased in proportion as grade. Present study suggests that LC method is a good measure of PA in free-living condition for elementary school students.