A new paper from the School of Continuous Improvement finds that removing the one input tied to whether a defibrillator would actually work changes the University's published AED readiness figure by well under a percentage point.
A new paper traces what the campus's 96 per cent defibrillator readiness figure actually certifies when its one functional-check term is dropped
Every wall-mounted defibrillator on campus carries a small QR tag, and once a month someone from Facilities walks past with a handheld reader and scans it. That scan is most of what stands behind the number in the University’s annual safety report: AED Fleet Readiness, currently 96.2 per cent. A new paper from the School of Continuous Improvement asks what else, if anything, the figure is counting.
Ninety-six per cent is a good headline number precisely because it doesn’t move. That’s not nothing, but it’s not the same thing as knowing every cabinet would work.
— Senior Lecturer Renke Sabel, Convenor, Indicator Commons
The paper, co-authored with Associate Professor Casimir Beng and Research Fellow Osei Vandermeer, continues the School’s practice of turning its own instruments back on themselves. The team matched a semester of the ledger’s monthly scan records against the separate procurement table that tracks each cabinet’s pad-expiry status, then recalculated the published readiness figure under two alternative formulas: one that drops the pad-expiry term altogether, and one that doubles its weight. Both alternatives landed within a percentage point of the figure the University currently reports.
Associate Professor Beng said the finding “sits alongside a run of similar results from across the School’s instrumentation work, and the consistency itself is starting to become the more interesting pattern.” Facilities has indicated it has no plans to revisit the current weighting, noting the figure has held steady across several reporting cycles.
The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.wfyo68.