A new Slop University poster compares two campuses on a shared swipe-card engagement-risk platform, finding staff override access, not the score itself, decides which flagged students actually get called home.
Comparing two campuses on a shared swipe-card platform finds override access, not the score, decides who's actually contacted
At one feeder-network campus, whether a flagged student is actually called home has almost nothing to do with the number that flagged them. At its sister campus, running the identical platform, it has almost everything to do with it. A new Slop University poster traces the whole difference to a single setting: whether staff can override the list.
A score becomes the decision the moment nobody is allowed to second-guess it. Give someone the standing to disagree, and you find out how much judgement the list was quietly asking them to skip.
— Dr Mirela Hanke, Postdoctoral Fellow, School of Continuous Improvement
Researchers matched engagement-risk percentiles for 214 flagged students against wellbeing-team contact logs at two campuses on a shared swipe-card platform over a full term — one where staff retain standing to override the platform’s daily outreach list, one where the list is enacted as issued. Contact rates came out close to identical between the two sites. Who was actually contacted did not.
“The list is the easy part to audit,” said Associate Professor Casimir Beng, Lead of the Adaptive Metrics Lab. “What’s harder to see is everything a name on it was standing in for.”
The University reads the comparison as a clean instance of the question its evaluation-ecosystem agenda keeps returning to: not whether a measurement is wrong, but what happens once somebody is allowed to act on something the measurement missed.
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.zo2x5o.