There is a well-known tendency to delay and prolong our sleep during weekends (Saturday and Sunday), with an advance and reduction of sleep during workdays (Monday to Friday). The objective of this work was to determine if the changes of sleep during weekends are produced by a partial sleep deprivation or a lack of entraining of circadian rhythms to an advanced phase, during workdays. The subjects were 52 undergraduate female students, mean age = 17.5 years, SD = 1.32. All students attended school following a regular schedule, from Monday to Friday. Two groups of students were studied: one attended school from 07:00 to 12:00 h (morning group, n = 30); the other attended school from 14:00 to 18:00 (afternoon group, n = 22). None of the students worked or was engaged in other activity with a fixed schedule. All kept a sleep-wake diary for 2 weeks, in which they recorded their bedtimes, wakeup times, and sleep-onset latencies. The morning group delayed 47.4 min [t(29) = 4.72, p < 0.0001] and prolonged their sleep 118.2 min [t(29) = 9.4, p < 0.0001] during weekends. Although the afternoon group had the opportunity to maintain a delayed phase and a long sleep time throughout the week, they delayed their bedtime by 24 min [t(21) = 2.99, p < 0.01] during weekends, without changing their sleep duration. The findings suggest that the prolonged sleep during weekends is due to reduction of sleep during workdays, whereas the delay of bedtime seems to be associated with a tendency of the human circadian system to maintain a delayed phase