A new paper from the School of Emergent Priorities finds a desk's arrival-time regularity forecasts next week's tidiness-sweep failure better than Facilities' existing hours-logged risk score.
A fourteen-week study finds how consistently a desk is used forecasts a failed tidiness sweep better than how much it's used
A desk that logs barely any hours can still be the one likely to fail Facilities’ Evening Tidiness Sweep — so long as it keeps the hours it does log at roughly the same time every day. A new paper from the School of Emergent Priorities finds a desk’s arrival-time regularity forecasts a failed tidiness check better than the total hours a desk racks up, the very number Facilities’ own risk score has ranked desks by since it was introduced.
The Adaptive Metrics Lab’s Desk Utilisation Risk Score has, since it was introduced, flagged the University’s flexible-tenancy hot-desks most likely to fail the nightly sweep by how many hours each was logged occupied that week. Researchers from the School’s Trajectory Analytics Group set that score against an alternative signal adapted from circadian rest-activity-rhythm research, more usually applied to sleep and health outcomes, scoring instead how consistently a desk’s occupant arrives at the same time of day.
A desk that turns up at the same time every day has time to develop a routine around tidying it. One that turns up whenever still gets its hours in, just never in a shape a habit can settle into.
— Dr Fenna Okoro, Senior Lecturer and Convenor, Horizon Register, School of Emergent Priorities
The comparison ran across a full teaching term on two of the University’s hot-desk floors, tracking each desk’s badge-swipe pattern and its outcome at the nightly sweep week over week. The regularity-based forecast outperformed the hours-based score on both floors, and combining the two added nothing regularity alone did not already predict — a result the researchers say leaves the existing score answering a question the sweep was never really asking.
“It’s an unusually clean case of an institution already holding the better signal and ranking by the worse one instead,” said Associate Professor Kwame Lindqvist.
The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.h4dvr2.