A new Slop University research poster conceals the maker tag on 288 provenance-verified craft-market items and asks four panels to judge which were made by the stallholder, finding calls correct 53.4 per cent of the time against stated confidence of 77.4 per cent.
Twenty-four adjudicators judged 288 provenance-verified market items with the swing tag concealed, landing on 53.4 per cent correct while reporting 77.4 per cent confidence
Twenty-four people spent nine market days deciding which goods on a stall had been made by the person selling them. They were right 53.4 per cent of the time. They were sure of it 77.4 per cent of the time.
A research poster released today by the School of Emergent Priorities sets those two numbers beside each other. Four panels — practising makers, market coordinators, regular shoppers and first-time visitors — judged 288 items whose provenance the researchers had established separately, with the maker tag concealed. Against an even baseline, no panel cleared 59 per cent correct. Every panel reported confidence in the seventies or above.
We set out to learn something about the tags, and came away with something about ourselves: this School has been treating a mark on an object as though it were a measurement of the object, and has been doing it for longer than the tags have been around.
— Dr Iben Chikere, Lecturer, School of Emergent Priorities
The items came from 41 stalls across four suburban markets, catalogued over nine monthly trading days; provenance was established from supply invoices and a workshop visit with eighteen consenting stallholders. Restoring the tag lifted stated confidence by a further 9.1 points and left accuracy where it was.
The association running the markets publishes a local-maker share, and across the same nine days that share rose from 61 to 88 per cent. The audited composition of the stalls did not measurably shift.
“A tag is cheap to apply and expensive to check,” said Dr Ingeborg Nwachukwu, Lecturer and Convenor of Capability Sprints. “Any indicator built on the cheap half of that will keep going up.”
For the School, the value of the study is the setting: somewhere the difference between an assurance and a piece of information could be measured rather than argued about.
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.95ai6g.