The Influence of Sleep on Subjective Well-Being: An Experience Sampling Study

Abstract

Previous research has associated sleep with subjective well-being (SWB), but less is known about the underlying within-person processes. In the current study, we investigated how self-reported and actigraphy-measured sleep parameters (sleep onset latency, sleep duration, sleep satisfaction, social jetlag, and sleep efficiency) influence SWB (positive and negative affect, life satisfaction) at the within- and between-person level. Multilevel analyses of data from 112 university students who completed a two-week experience sampling study revealed that higher within-person sleep satisfaction was a significant predictor of all three components of next day’s SWB (ps < .005). Higher between-person sleep satisfaction was also related to higher levels of positive affect and life satisfaction (ps < .005) whereas shorter self-reported between-person sleep onset latency was associated with higher positive affect and life satisfaction, and lower negative affect (ps < .05). However, longer actigraphy-measured within-person sleep onset latency was associated with higher next day’s life satisfaction (p = .028). When including within- and between-person sleep parameters into the same models predicting SWB, only within-and between-person sleep satisfaction remained a significant predictor of all components of SWB. Additionally, we found an effect of higher self-reported within-person sleep onset latency on positive affect and of higher self-reported within-person sleep duration on life satisfaction (ps < .05). Our results indicate that the evaluative component of sleep—sleep satisfaction—is most consistently linked with SWB. Thus, sleep interventions that are successful in not only altering sleep patterns but also enhancing sleep satisfaction may stand a better chance at improving students’ SWB.

The influence of sleep on subjective well-being: An experience sampling study.

“…This is different from many previous studies that used cross-sectional designs to examine the relationships between sleep and SWB (see, e.g., Diaz-Morales & Escribano, 2015;Gaina et al, 2005;Rossa et al, 2014). Even though there are individual differences in human sleep indicators, meaning that humans differ among each other in the amount of daily sleep they require or their sleep quality (van Dongen et al, 2005), sleep parameters also vary within humans, that is, from night to night (Buysse et al, 2010;Lenneis et al, 2021). By assessing multiple observations in the same participants over a period of 2 weeks, we were able to examine (a) if and to what extent deviations from one's personal average levels are related to SWB and (b) how interindividual differences in sleep parameters influence SWB.…”

“…Of these, 104 (95.41%) had actigraphy data available. The dataset has been used in other studies (Das-Friebel et al, 2020;Lenneis et al, 2021) but it has not been used for the present purpose. The sample size was determined by other experience sampling studies that used similar number of participants, as well as time and resources available for data collection (Dimotakis et al, 2013).…”

“…Results showed that in PA, NA, and LS, 54.68%, 51.39%, and 69.29% of the variances were explained by between-participant effects, respectively. However, as the ICC is only defined for random-intercept models, but not for more complex models (e.g., those involving random slopes; Lenneis et al, 2021), we did not calculate it for the other models.…”

“…Van Dongen et al (2005) found interindividual variability in human sleep parameters, indicating that people differ among each other in the number of hours they sleep, in their sleep quality, or in their midpoint of sleep (chronotype). Furthermore, both the amount and quality of sleep also fluctuate within people; for example, there is a substantive amount of night-to-night variability in various sleep parameters (Buysse et al, 2010;Lenneis et al, 2021) that might affect SWB. Components of SWB also show a substantive amount of intraindividual variability (Mill et al, 2016;Willroth et al, 2020).…”

“…There is also a problem with the numbering from Hypothesis 9 onwards, but the contents of the hypotheses remain the same.3 We excluded 22 instances due to several reasons in the following order: Six instances because participants had indicated the same wake-up and going-to-bed times, one because they went to bed before trying to fall asleep, one because they needed more than 5 hr to fall asleep, six because their sleep duration was less than or equal one hour, one because their sleep duration was more than 15 hr, three because their mid-sleep score was more than 15, and lastly four because there was no information available on whether it was a work or free day. The instances were also excluded in a previous paper using the same dataset(Lenneis et al, 2021).…”