A new Slop University research poster models two years of sign-in flows at eleven community sheds and finds that the recruitment open days lifting a shed's funding return are associated with fewer men in the room on an ordinary week.
A four-state model of two years of sign-in flows at eleven community sheds puts each additional open day at 12.4 more counted attendances and 1.9 fewer men present on a median Tuesday
The School of Continuous Improvement has spent much of the year on formulas that pay for arrivals. A research poster deposited today through the Office of Research Outputs takes that interest to the community shed, where the arrival is a man at an open day and the formula is the one that keeps the roller door up.
Working from twenty-four months of sign-in sheets across eleven sheds, covering 4,318 attendance events and 1,062 distinct men, the team fitted a four-state model of how attendance is taken up, held, and let go. Read against each shed’s own open-day cadence, the counted figure and the room move apart: an additional open day a year is associated with 12.4 more counted attendances and 1.9 fewer men present on a median Tuesday. In steady state the model puts the regulars at 0.21 of everyone who has ever signed in.
Nobody at these sheds is doing anything other than what the scheme asks of them, and what the scheme asks for is a door that opens. This is the second study in a year where the School has found itself measuring the distance between a return and a room, and I doubt it is the last.
— Dr Ingeborg Nwachukwu, Lecturer and Convenor, Capability Sprints
Dr Bram Ntuli, Lecturer and Convenor of Improvement Grand Rounds, noted that the participating sheds “handed over two years of paper and asked for nothing back”, and said the question the work puts to a scheme’s administrators is “which of the two numbers they would rather be right about”.
Slop University counts it a good use of the School’s method when the object of study is something a reader could walk into on a weekday morning. The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.ngssxj.