A School of Emergent Priorities poster tests the staff car-park Priority Score against independent commuting-need measures across 214 renewals, and finds the rank tracks how completely a form was filled in rather than the need it asks about.
A new poster checks the staff permit ranking against commuting-need measures drawn from records no applicant ever sees, and finds the two rarely meet
Two hundred and fourteen staff car-park renewal applications, and a parallel set of commuting-need measures assembled from records no applicant ever touched: this is the material of a new poster from the School of Emergent Priorities, presented this week by the Office of Research Outputs.
The Priority Score is the University’s instrument for ordering more permit applications than it has bays, gathering commute distance, caregiving responsibilities, and shift timing into a single renewal rank. Dr Fenna Okoro and Dr Petra Umbile set that rank beside independently drawn need records, and modelled it against a second family of measures the form was never meant to capture: optional fields completed, free text written, resubmissions made.
Rank showed no reliable association with the combined need measures. It tracked the completeness of the form closely, and agreed with a blind three-member need panel on 61 of the 214 renewals. Across eight renewal cycles the share of top-tier permits going to the most thoroughly completed applications rose steadily, while the share going to the highest-need quartile held flat.
We are grateful to the staff who filled in these forms, twice a year, without knowing that the care they took was the thing being measured.
— Dr Okoro, Senior Lecturer and Convenor of the Horizon Register
The publication continues a line of work the School has made its own: how a stated priority becomes a rank, and a rank becomes a decision.
“Whether a renewal rank ought to reward a careful applicant is not a question this data settles,” said Dr Umbile, Postdoctoral Fellow in the School of Emergent Priorities. “It is a question the data makes harder to leave unasked.”
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.ojetat.