Extended Data Fig. 3: Method of deriving Learning Management System (LMS) login parameters.
From: Early morning university classes are associated with impaired sleep and academic performance
a,b, The diurnal distribution of LMS logins is shown on days that students had (a) classes starting at 08:00, and (b) no scheduled classes. Login data in both panels were compiled from the same group of students (n = 10,950) in the second semester of the 2018/19 academic year. Students’ LMS logins were summed by time of day in 5-min bins. The time series was normalized by dividing the number of logins in each 5-min bin by the total number of logins across all bins. LMS login parameters were derived using a threshold crossing method in which the LMS activity threshold was calculated as 50% of the average normalized number of logins per bin (1 bin / 288 bins × 0.5 = 0.001736 or 0.17%). The LMS login offset was defined as the clock time when the normalized number of logins dropped and remained below the threshold, and the LMS login onset was defined as the clock time when logins exceeded and stayed above the threshold. The blue line in each plot shows the LMS activity threshold, and the ‘X’ symbols show the downward and upward threshold crossings. The grey box indicates the LMS inactive period calculated from the LMS login offset to the LMS login onset.
