To examine the validity of self-reported survey estimates of sleep patterns in adolescents through a comparison of retrospective survey descriptions of usual school- and weekend-night sleep habits with diary-reported sleep patterns and actigraphically estimated sleep behaviors over a subsequent week.

High school students completed a Sleep Habits Survey about the previous 2 weeks and then wore an actigraph (AMI, Ard-sley, NY) for 8 days while keeping a daily sleep diary. Matched-pair t tests assessed average differences between survey and diary reports and between survey and actigraph estimates. Pearson correlations assessed the extent to which survey reports were in agreement with diary reports and actigraphy estimates.

302 high school students (196 girls, 106 boys) in grades 9-12 from five high schools.

School-night survey total sleep times and wake times did not differ from sleep amounts reported in the diary or estimated by actigraphy; survey bedtimes were slightly earlier. On weekends, survey total sleep times and wake times were longer and later, respectively, than estimated with actigraphy and reported on diaries. Moreover, school- and weekend-night survey variables were significantly correlated both with diary and actigraphy variables. Strengths of the associations were consistently greater for school-night variables than the corresponding weekend-night variables.

The findings support the validity of the Sleep Habits Survey estimates in comparison with diary and actigraphy. Strengths and limitations for survey measures of high school students’ usual sleep/wake patterns are discussed.