Figure 8. | Publication bias and the canonization of false facts

Open accessCopyright infoDownload PDF

Research: Publication bias and the canonization of false facts

Figure 8.

Affiliation details

University of Copenhagen, Denmark; University of Washington, United States; North Carolina State University, United States
Figure 8.
Download figureOpen in new tabFigure 8. p-hacking dramatically increases the chances of canonizing false claims.

Probability that a false claim is canonized as fact vs. fraction of negative outcomes. Throughout, all positive outcomes are published (p1=1), and the nominal false positive rate is αnom=0.05, the false negative rate is β=0.2, and evidentiary standards are strong (τ0=0.001 and τ1=0.999). Curves show actual false positive rates αact=0.05 (yellow), 0.10,,0.25 (red). Compared with Figure 4C, in which the nominal rates are equal to the actual rates, the probability of canonizing a false claim as fact is substantially higher.

DOI: http://dx.doi.org/10.7554/eLife.21451.009