The extent and consequences of p-hacking in science.
Head ML, Holman L, Lanfear R, Kahn AT, Jennions MD. · 2015
Read The extent and consequences of p-hacking in science. by Head ML, Holman L, Lanfear R, Kahn AT, Jennions MD. (2015) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4359000.txt.
Description
Megan Head, Luke Holman, Rob Lanfear, Andrew Kahn, and Michael Jennions measured a quiet bias in science publishing, in PLoS Biology in 2015. Journals favor new, clear, positive results. One response, called p-hacking, is when researchers keep testing or trimming data until a weak result crosses the line for statistical significance. The team used text mining of published papers to gauge how common that pattern is across fields.
Signs of p-hacking appeared widely across science, the team found. Yet when they showed how meta-analyses, studies that pool many studies, can test for the bias, its pull looked weak next to the real effects being measured. So p-hacking is likely common, they conclude, but it probably does not badly distort the broad verdicts drawn from pooled evidence. Spotting it still matters for single studies.
Catalog record
| CREATOR | Head ML, Holman L, Lanfear R, Kahn AT, Jennions MD. |
|---|---|
| YEAR | 2015 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |