Table 6.
Sleep duration and timing question design considerations
| Component of Interest | Example(s) | Design Considerations for these examples |
|---|---|---|
| Event definitions (sleep duration) | a) Nocturnal sleep “How many hours do you usually sleep at night?” b) Time in bed attempting to sleep (“How many hours of sleep do you get at night, not including time spent awake in bed?”) c) Major sleep episode (“How much sleep do you usually get at night (or your main sleep episode)?”) |
• Reporting nighttime sleep may be challenging for shift workers who commonly sleep during the day. Requesting information about “main sleep episode” (as opposed to “at night”) would enable shift workers and others on non-traditional schedules to response with their primary sleep episode. • Instead of asking for sleep duration directly, researchers can also ask for sleep start and end times in order to calculate sleep duration. • Researchers interested in the time spent in bed leading up to sleep (herein proposed as “time in bed attempting to sleep”) may administer questions assessing the first attempt to sleep and the time out of bed in the morning. On the other hand, using the time a person reports “going to bed” and “getting out of bed” (common questions administered to measure time in bed) may be extremely protracted for individuals who spend extended periods of time in bed before sleep. • Researchers interested in measuring the major sleep episode duration may administer questions assessing the times a person “first attempts to sleep,” how long it takes to “fall asleep,” and when the person “wakes up.” |
| Context (all domains) | “How much sleep do you get at night on weekdays or workdays?” “How much sleep do you get at night on weekends or free days?” |
• Requesting information about week versus work days allows participants to respond specifically if the sleep timing or duration differs between the two conditions. • Asking participants to distinguish between sleep on week/weekend nights may not capture accurate timing details for people who work on weekends. • Surveys intended for retired individuals may not need to distinguish between weekend and work days. |
| Time Frame (all domains) | “On average how many hours did you sleep each night during the past 4 weeks?” “During the past month, what time have you usually gone to bed at night?” |
• Asking participants to report their “usual” sleep duration and/or timing may affect the quality of responses because individuals are limited in their ability to report their behavior in an abstract time frame. • Providing a timeframe (e.g., “During the past month” or “In the past 2 weeks”) guides participant responses. |
| Response option (all domains) | “HH:MM” “[<5 hours; hrs; 6 hrs; 7 hrs; 8 hrs; 9 hrs; 10+ hours]” |
• The HH:MM response option enables more precise recording of actual duration. • Individuals may differ in how precise their responses are (e.g., every one minute or rounding to quarter hour). • Closed-ended scales may be beneficial for reducing the survey completion time, but may reduce accuracy |
| Event definition (bedtime) | “At what time do you first attempt to sleep?” At what time do you fall asleep?” At what time do you go to bed?” |
• “First attempt to sleep” is beneficial in that it asks precisely when the person first attempts sleep, which, in conjunction with sleep latency may be used to calculate sleep duration. • “Fall asleep” offers more precise guidance to the participant to report the time they fall asleep. • “Go to bed” is ambiguous; it is unclear whether the person should respond with the time they get in bed or first attempt sleep. |
| Event definition (wake time) | “At what time do you usually get out of bed?” “At what time do you typically get up?” “At what time do you usually wake up?” |
• “Get out of bed” is useful in capturing time in bed. • “Gets up” is ambiguous; this could mean when a person gets out of bed or wakes up. • “Wake up” is clear and assesses the time a person wakes from sleep. |
| Event definition | “During the past week, on how many days did you nap for an hour or two?” “During the past week, on how many days did you nap for 5 minutes or more?” |
• A question that asks for participants to report napping for one or two hours may capture longer naps but not shorter naps. • A question asking for naps “5 minutes or longer” will capture both short naps as well as long naps (without distinguishing between the two). |
Note.
Bold indicates the wording approaches a survey designer may consider.