Title |
Inappropriate Fiddling with Statistical Analyses to Obtain a Desirable P-value: Tests to Detect its Presence in Published Literature
|
---|---|
Published in |
PLOS ONE, October 2012
|
DOI | 10.1371/journal.pone.0046363 |
Pubmed ID | |
Authors |
Gary L. Gadbury, David B. Allison |
Abstract |
Much has been written regarding p-values below certain thresholds (most notably 0.05) denoting statistical significance and the tendency of such p-values to be more readily publishable in peer-reviewed journals. Intuition suggests that there may be a tendency to manipulate statistical analyses to push a "near significant p-value" to a level that is considered significant. This article presents a method for detecting the presence of such manipulation (herein called "fiddling") in a distribution of p-values from independent studies. Simulations are used to illustrate the properties of the method. The results suggest that the method has low type I error and that power approaches acceptable levels as the number of p-values being studied approaches 1000. |
X Demographics
Geographical breakdown
Country | Count | As % |
---|---|---|
United Kingdom | 18 | 22% |
United States | 18 | 22% |
Canada | 5 | 6% |
Spain | 4 | 5% |
Germany | 2 | 2% |
France | 2 | 2% |
Sweden | 2 | 2% |
Comoros | 1 | 1% |
Netherlands | 1 | 1% |
Other | 6 | 7% |
Unknown | 24 | 29% |
Demographic breakdown
Type | Count | As % |
---|---|---|
Members of the public | 48 | 58% |
Scientists | 30 | 36% |
Science communicators (journalists, bloggers, editors) | 4 | 5% |
Practitioners (doctors, other healthcare professionals) | 1 | 1% |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
United States | 7 | 6% |
Netherlands | 2 | 2% |
Germany | 1 | <1% |
Switzerland | 1 | <1% |
Spain | 1 | <1% |
Russia | 1 | <1% |
Unknown | 103 | 89% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Student > Ph. D. Student | 25 | 22% |
Researcher | 23 | 20% |
Professor | 11 | 9% |
Professor > Associate Professor | 10 | 9% |
Student > Master | 9 | 8% |
Other | 30 | 26% |
Unknown | 8 | 7% |
Readers by discipline | Count | As % |
---|---|---|
Psychology | 24 | 21% |
Medicine and Dentistry | 18 | 16% |
Agricultural and Biological Sciences | 16 | 14% |
Computer Science | 9 | 8% |
Biochemistry, Genetics and Molecular Biology | 5 | 4% |
Other | 31 | 27% |
Unknown | 13 | 11% |