Skip to main content
Slop University tests its own promise of an anonymous tip line

Slop University tests its own promise of an anonymous tip line

A new paper from the School of Continuous Improvement finds the staff integrity tip line's own timestamp and phrasing data are often enough to narrow an anonymous submission to a name, and audits existing case notes for the same inference happening informally.

A new paper finds two routine ticketing fields narrow most team-scoped tips to a name, and audits whether triage already infers the same thing informally

A staff member who files an anonymous integrity tip is told, correctly, that the intake form asks for no name. A new paper from the School of Continuous Improvement reports that the tip line’s own ticketing system, entirely by accident, often supplies one anyway.

Lecturer Dagny Okafor, Postdoctoral Fellow Mirela Hanke, and Senior Lecturer Renke Sabel built a lightweight tool that scores an anonymous tip against a pool of likely staff using only two fields the ticketing system already keeps for ordinary operational reasons: when a tip arrived, and what it said. Tested against a sample of closed tickets, the tool placed the true submitter at the top of a short, team-scoped list more often than not, with its reach narrowing as the candidate pool widens to a lab, programme, or school.

A field kept for scheduling and a field kept for record-keeping turn out, together, to do quite a lot of work a form’s blank name field was supposed to make unnecessary.

— Dr Dagny Okafor, Lecturer and Convenor, Evaluation of Evaluation

The team paired the tool with an independent audit of existing triage case notes, finding language consistent with the same kind of inference already present at a broadly similar rate, without any tool involved. “The capability isn’t hypothetical,” said Dr Mirela Hanke, a co-author. “In a lot of these notes, something like it already seems to be happening.”

The School situates the paper within its wider practice of auditing the University’s own assurance instruments before turning them on anything else. A triage-confidence indicator drawing on the tool is now under discussion for inclusion on the Living Dashboard.

The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.go3vki.