A new poster from the School of Emergent Priorities covertly tags 214 hired graduation gowns and cross-references ceremony photographs against the hire desk's damage-fee ledger, finding billing tracked the docket's named borrower even after the desk was shown an accurate record of who wore the gown.
Covertly tagged graduation gowns and a term of ceremony photographs show the hire desk kept billing the name on the docket even once shown who actually wore the gown
A hired graduation gown changes hands more than the hire desk’s paperwork admits. Numbered fibre tags, stitched into the collar lining of 214 gowns across the University’s Autumn and Spring 2025 conferrals, let a new poster from the School of Emergent Priorities check the desk’s damage-fee ledger against the ceremony’s own photographs. The tags found that more than a third of tracked gowns were worn, at the moment the shutter clicked, by someone other than the name on the hire docket. The desk’s damage fees followed the docket name regardless — a gap the School of Emergent Priorities has been quietly cataloguing wherever a record and the thing it records stop lining up.
A docket says who signed. A photograph says who was standing there. We just wanted to know which one the fee actually listens to.
— Dr Anouk Mensah, Postdoctoral Fellow, School of Emergent Priorities
A second phase of the study put the same tag-to-photograph mismatch report in front of hire-desk staff before they billed a fresh batch of cases, to see whether the paperwork would follow the evidence once the evidence was sitting on the counter. “Handing someone a better record and watching what they do with it is its own kind of instrument,” said Dr Dagny Okafor, who convenes the School of Continuous Improvement’s Evaluation of Evaluation program. “This one just happened to be a gown.”
The method, in short: covert fibre tags gave each gown an identity independent of its hire docket, letting the ceremony’s own proof-sheet photographs stand as a second, unrelated record of who actually wore it. Damage-fee attribution was then coded against both records across a before and an after batch.
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.fe4usp.