A Swedish team led by Per Lytsy reviewed 55 studies from 1963 to 2024, with more than 15,000 participants, on how people understand p-values, statistical significance and confidence intervals. The verdict, reported in Dutch by Scientias.nl: understanding is poor, even among experts. In one study only 12.5 percent of statisticians and epidemiologists answered two key questions correctly. The classic error is reading a p-value as the chance that a hypothesis is true, and confusing 'significant' with 'important'.

Adding confidence intervals reduced mistakes but did not remove them. For anyone who quotes studies in a debate, this is a useful check: a significant result can still be practically meaningless.

Source: Daniel van Veelen, Scientias.nl (Dutch) (2026-11-13)
https://scientias.nl/ook-experts-struikelen-over-statistiek-zelfs-statistici-interpreteren-resultaten-regelmatig-verkeerd/

Summarised by the FTI newsdesk.