Nothing can go wrong go wrong go wrong

[supergee]

One in five genetics papers contains errors thanks to Microsoft Excel [Science]

Thanx to andrewducker

Comments

[weofodthignen]

Interesting, and of course it makes sense that scientists would use spreadsheets and also that the people who coded them would be thinking in terms of business/economics uses. It makes me wonder how this fits in with the co-writing and peer reviewing processes. Is the wrong coding getting introduced as co-authors share drafts of the paper, or as the journal passes it to reviewers and gets it back, or even during publication? For some of these I wonder whether the reviewers are even looking at the tabulated results. However, it's science, so what do I know.

What you link to has a big scary message about a paywall, and says the story was originally published by the WaPo, with a link to this blog.

M