Animal health · Proc Natl Acad Sci U S A · 2024
Summary: 1 of 6 virus-fed calves still carried the bacterium. Its own results: 2 of 8.
The summary’s counts against the paper’s own, from the same end-of-study test
The paper says
To test whether viruses that attack a harmful gut bacterium protect calves, researchers gave some calves both, others the bacterium alone. The paper’s summary at the top says 1 of 6 virus-fed calves still carried it at the end, against 6 of 6 bacterium-only calves.
Its own data show
The paper’s account of how the study was run gives the virus-fed and bacterium-only groups 8 calves each. Its results, from the same test 16 weeks after the calves got the bacterium, count 2 of 8 virus-fed and 8 of 8 bacterium-only calves still carrying it. It reports no calf left out.
A count like “1 of 6” has two parts: how many calves carried the bacterium, and how many were tested. The summary gets both wrong. The viruses helped either way, but summary readers see a smaller study (12 calves given the bacterium, not 16) and a bigger effect (1 in 6 virus-fed calves still carrying it, not 1 in 4).