An Experimental Study of Bifurcated (Weekend and Weekday) and Unitary (Past Week) Retrospective Assessments of Sleep
Description: Discordance between weekday and weekend sleep schedules is common (Bonnet & Arand, 1995; Breslau, Roth, Rosenthal, & Andreski, 1997; Machado, Varella, & Andrade, 1998; Strauch & Meier, 1988; Tsai & Li, 2004). Brief retrospective self-report measures are essential for epidemiological research studies (Moul, Hall, Pilkonis, & Buysse, 2004), but self-reports are prone to error in recall, and the greater the variability in nightly sleep, the less reliable are retrospective reports (Babkoff, Weller, & Lavidor, 1996). More accurate self-report responses may be possible if measures prompt participants to consider variations in sleep schedules that are consistent (i.e., weekday and weekend sleep schedules). The current study experimentally examined whether Bifurcated (Weekday and Weekend) retrospective assessments of sleep are more accurate than Unitary (Past Week) assessments. Participants were randomly assigned to complete one of the two versions (Bifurcated vs. Unitary) of the Sleep Questionnaire. One hundred and thirty-one participants were included in the analyses. Results of a a series of analyses demonstrated that the Bifurcated version of the Sleep Questionnaire provided more accurate and less variable estimates of total sleep time than the Unitary version of the Sleep Questionnaire. Differences between the versions of the Sleep Questionnaire for other sleep variables were less consistent, and the increased length of the Bifurcated version of the Sleep Questionnaire may have contributed to increased missing and unusable data in this group. Overall, the findings suggest that in both research and clinical work, retrospective measures that examine weekday and weekend sleep separately may offer advantages over retrospective measures that do not differentiate between weekday and weekend sleep.
Date: August 2016
Creator: Sethi, Kevin J.
Partner: UNT Libraries