Impact of pandemic lockdown on learning behaviour and sleep quality in German students

Data collection

This study was carried out by the Department of Biology, Eberhard Karls University, Tübingen, Germany. The study sample comprises students of the University of Tübingen. The questionnaire was provided in German language only and distributed online during two different time periods in 2020 (February 27–March 21) and 2021 (February 27–March 27). The first data collection phase was directly before the first strict lockdown to contain the COVID-19 pandemic in Germany, and the second data collection phase was during the second lockdown. Participants were made aware of the study via mailing lists and official university platforms. The background of the study was clarified, and the voluntary and anonymous nature of participation was pointed out up front. In addition, the participants were made aware that they would not suffer any disadvantages if they terminated the survey prematurely and that participation would not be remunerated. The survey was hosted on the SoSciSurvey online platform, which fulfils the European Union’s data privacy standards. Data were collected from a total of N = 637 (male: 191; female: 445; other: 8) participants; first data collection N = 312, second data collection N = 325.

Questionnaires

In addition to collecting data on demographics and the highest educational qualification (type, year and state in which the qualification was obtained), the survey was compiled from validated questionnaires (Morningness–Eveningness Stability Scale improved, MESSi; Pittsburgh Sleep Quality Index, PSQI; Sleep timing questionnaire, STQ; Pediatric Daytime Sleepiness Scale, PDSS). Furthermore, the participants provided information on the current status of lectures “type of teaching currently offered” twice, but coded slightly differently in each case, to form a scale (1: If you worked remotely from home in the past 2 months, how often did this follow a fixed schedule with set times? Exclusively/a lot/medium/little/not at all; 2: If you worked remotely from home in the past 2 months, how often was this based to on-demand events? Exclusively/a lot/medium/little/not at all). Also, participants were asked whether they had completed the survey at both survey times. Participants for whom this circumstance applied were excluded from the analysis (N = 398). The reasoning was that the survey took place anonymously and the datasets of this linked sample could not have been combined as dependent datasets. Therefore, independent datasets were analysed. The questionnaire took an average of 11.29 min to complete with a standard deviation of 4.76 min.

Morningness–Eveningness Stability Scale improved

In this study we used the MESSi to assess participants’ morningness–eveningness. The questionnaire is separated into three subscales—the morning affect subscale (MA), the eveningness subscale (EV) and the distinctness subscale (DI). Each subscale consists of five questions in a 1–5 Likert format. The MA queries the affective facet (e.g., “How alert do you feel during the first half hour after awakening in the morning?”), the EV collects data on the overall physical and mental situation in the evening (e.g., “In general, how is your energy level in the evening?”), while DI is looking into the amplitude of active phases (e.g., “There are moments during the day where I feel unable to do anything.”). Higher values on the subscales indicate participants being more prone to those facets (MA → prone to morningness/EV → prone to eveningness/DI → higher daytime fluctuations of active phases). The MESSi questionnaire has been validated in various studies (factorial invariance, structure, reliability in different languages, e.g., [8,9,10,11,12]; and by actigraphy [13]). Cronbach’s α in the current sample was 0.891 for MA, 0.870 for EV and 0.759 for DI.

Pittsburgh Sleep Quality Index

The PSQI is a self-reported inventory that measures various sleep variables in a retrospective design considering the past 4 weeks. It is used in clinical and non-clinical settings and includes 19 self-rated and 5 externally rated questions. The latter are not included in the quantitative evaluation. The PSQI is divided into seven sections: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medications and daytime dysfunction over the past month (e.g., “During the past month, how often have you had trouble sleeping because you cannot get to sleep within 30 min?”). The sum of the 1–4 Likert scale leads to a classification from “good” sleepers to “poor” sleepers [14]. The current study investigated the frequency of sleep disturbances (eight items) and performed an assessment of overall sleep quality (1 item). Cronbach’s α in the current sample was 0.642.

Sleep timing questionnaire

The STQ consists of 18 items asking for the individual bed and awakening times on weekdays and at weekends, the stability of these schedules, and the frequency and length of night awakenings. It was developed to provide a precise picture of a person’s typical sleep rhythm [15]. Sleep technicians can thus adjust the timing of polysomnographic examinations to the specific needs and habits of the patient [16]. It is more time efficient than a sleep diary format. As we had already collected sleep quality aspects, we only asked for variation individual bed and awakening times on weekdays or at weekends and, in addition, calculated the length of sleep. Cronbach’s α in the current sample was 0.831.

Paediatric Daytime Sleepiness Scale

The PDSS is a questionnaire originally introduced in English [17] and translated into German by Schneider and Randler (2009) [18]. It queries the daytime sleepiness of students. The questionnaire contains eight items of which two are clearly related to a school or learning environment. All questions are asked in a 1–5 Likert format, with one question being reverse coded. The total score of the questionnaire is obtained by adding up the answer scores. Cronbach’s α of this scale was 0.808 in this sample.

Habitual sleep–wake variables

We collected bedtimes and rise times on weekdays and free days to calculate sleep duration, midpoint of sleep and social jetlag. Sleep duration (SD) results from the difference between sleeping and waking times and is summed up to average sleep duration (SDaverage) as follows:

The term midpoint of sleep (MS) refers to the clock time-based midpoint between sleep onset and awakening. Since bedtime and waking times differ measurably between weekdays and weekend days/days off, the term social jetlag was coined [7]. Social jetlag quantifies the difference between the midpoint of sleep on weekdays compared to the midpoint of sleep on days off. Accordingly, we use an algorithm to correct the midpoint of sleep (MSfreedays corrected) and to include the weekend oversleep in the analysis [19]:

$$MS_\,\text}=MS_}-0.5x\left(\fracSD_}-(5xSD_}\\+2xSD_})\end}\right)$$

Statistics

Data analysis was performed using SPSS 27.0 (IBM, Armonk, NY, USA). Examined characteristics were “status of lectures”, MESSi, STQ, PDSS, PSQI and self-reported sleep variables. As not all participants filled in all questions, the exact sample size is always given for the calculations; thus, N and degrees of freedom may differ.

留言 (0)

沒有登入
gif