Counted Once, Funded Less

A cluster-randomised trial of a distinct-person count across 62 funded community groups

A research poster from the School of Continuous Improvement reports a cluster-randomised trial across one metropolitan council's twelve-month community-grants round, in which 62 funded groups were allocated 1:1 at award either to a hand tally counter issued with a one-page distinct-person definition or to the standard acquittal form, and 248 acquitted events were returned under the council's single self-reported engagement figure. Trial-arm groups report a median 152 people per event against 236 in the control arm (adjusted incidence rate ratio 0.63, 95% CI 0.52 to 0.77, ICC 0.11) and sit a median 6 per cent above an independent concurrent observer where control-arm returns sit 68 per cent above; re-derived on the program's own cost per participant, 12 of the 31 counted groups fall a renewal band against 2 of 31 controls.

@misc{slop_sxwqok,
  author       = {Solveig Adeyemi and Sten Okwuosa and Joost Nwosu},
  title        = {Counted Once, Funded Less: A cluster-randomised trial of a distinct-person count across 62 funded community groups},
  year         = {2026},
  publisher    = {Slop University},
  doi          = {10.5555/slop.sxwqok},
  url          = {https://slop.university/outputs/slop-poster-the-folding-table-sxwqok/},
  version      = {1.0},
  note         = {Research poster},
}