A Slop University research poster finds that once a chat platform's fifteen-minute edit deadline passes, correcting a mistake publicly rather than silently keeps disputes running significantly longer, and the University is proposing an identical deadline for staff email.
A regression-discontinuity study finds public corrections keep group-chat arguments running 6.1 minutes longer once the edit window closes, prompting an identical deadline for staff email
The fifteen-minute edit window on a group chat is a setting almost nobody chose and almost everyone lives inside. A research poster published today by the School of Emergent Priorities finds that a correction sent a minute after that window closes behaves nothing like one sent a minute before it, and that the gap is wide enough for the University to consider binding a staff policy to it.
Associate Professor Kwame Lindqvist, Director of the Master of Priority Studies, led the work with doctoral candidates Tove Ilesanmi and Sunniva Balogun, treating the platform’s fixed edit deadline as a natural experiment. Comparing 214 disputed messages either side of the fifteen-minute mark, the team finds that a correction sent as a visible reply, rather than folded quietly into an edit, keeps the surrounding argument running six minutes longer on average, a gap that does not appear at a placebo cutoff of ten minutes.
A deadline nobody designed for its social consequences is still setting them, and once you can measure that, it is hard to leave the measuring there.
— Associate Professor Kwame Lindqvist, Director, Master of Priority Studies
The University points to the finding as an example of how far its instruments now travel beyond the classroom. The Office of Digital Conduct is proposing that staff email adopt the same fifteen-minute silent-correction window the chat platform enforces by default, after which any amendment would need to go out as a visible reply-all. A two-division pilot is scheduled for the new academic year, with the research team continuing to track dispute-thread length as the measure of whether the deadline is doing what the policy now assumes it does.
The full research poster is available from the University’s research repository under an open licence, doi:10.5555/slop.ojc9q8.