Same Roster, Four Different Verdicts

Test--retest instability in a casual-shift rostering platform's fairness-compliance score, and the managers who stop reading the roster once they have one

A paper from the School of Continuous Improvement resubmits one week's unchanged casual-shift roster data to a rostering platform's fairness-compliance engine four times over twelve weeks, finding the score moves by an average of 9.7 points per re-run and 20 of 34 stores cross into a different compliance band at least once. Interviews with 21 store managers find 81% do not reopen the roster once a score is returned, and removing the instrument's heaviest-weighted criterion leaves the drift statistically unchanged.

@misc{slop_o8swbm,
  author       = {Renke Sabel and Marit Osayande and Joost Nwosu},
  title        = {Same Roster, Four Different Verdicts: Test--retest instability in a casual-shift rostering platform's fairness-compliance score, and the managers who stop reading the roster once they have one},
  year         = {2026},
  publisher    = {Slop University},
  doi          = {10.5555/slop.o8swbm},
  url          = {https://slop.university/outputs/slop-paper-re-running-a-casual-shift-o8swbm/},
  version      = {1.0},
  note         = {Research paper},
}