A School of Emergent Priorities poster imports a food-delivery app's regression-discontinuity threshold into the casual-tutor shift roster.
A regression-discontinuity poster from the School of Emergent Priorities finds a food-delivery platform's threshold portable, and the Horizon Register agrees
A food-delivery platform decides how much work a courier sees next the moment a rolling rating clears an unannounced bar. The School of Emergent Priorities’ Horizon Register has spent a fortnight confirming the bar is real, sharp, and, it argues, exportable, in a new poster released today, Towards a Portable Threshold.
The poster estimates that crossing the cutoff lifts weekly job offers by roughly seven shifts, a jump placebo tests elsewhere on the rating scale don’t reproduce. The Register reads the clean result as license to act rather than as a finding to sit with.
A discontinuity this sharp doesn’t ask where it came from before it starts explaining our own roster.
— Associate Professor Kwame Lindqvist, Director, Master of Priority Studies
On the strength of the poster, casual-tutor shift offers will be revised to hold flat below a survey-derived equivalent of the platform’s threshold and step up above it, with no phased introduction — the same design, unmodified, applied to what a tutor’s feedback score should be allowed to change. “The Register wasn’t looking for a cutoff to adopt,” said Dr Marek Solheim, Lecturer and Deputy Convenor of the Horizon Register. “It was looking for one it could trust.”
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.2618xe.