This study examined the reliability and validity of the National College Health Risk Behavior Survey (NCHRBS) vigorous physical activity (VPA), moderate physical activity (MPA), flexibility (FLEX), and muscular strength and/or endurance (MSE) questions. Twenty college students completed the four items twice during the same day. During the next 7 days they wore an accelerometer and a pedometer during all waking hours. They also recorded their daily activity in a log. At the end of the week, subjects completed the NCHRBS physical activity items again. The intraclass correlation coefficients ranged from 0.94-0.99 for the four items. The vigorous item was highly correlated with log VPA (r=0.82) and number of days with [greater than or equal to] 29 minutes of VPA from the accelerometer (r=0.60). The flexibility item was moderately correlated (r=0.57) and the muscular strength and/or endurance item was highly correlated (r=0.89) with corresponding log activities. The moderate item was highly correlated with log MPA (r=0.66) and number of days with [greater than or equal to] 30 minutes of MPA from the accelerometer (r=0.61). In conclusion, the NCHRBS physical activity items have excellent test-retest reliability and validity indices similar to other self-report physical activity questions. (Contains 2 tables and 1 figure.)