The p‐value and model specification in statistics
Open Access
- 9 July 2019
- journal article
- research article
- Published by Wiley in General Psychiatry
- Vol. 32 (3) , e100081
Abstract
The p value has been widely used as a way to summarise the significance in data analysis. However, misuse and misinterpretation of the p value is common in practice. Our result shows that if the model specification is wrong, the distribution of the p value may be inappropriate, which makes the decision based on the p value invalid.Keywords
This publication has 6 references indexed in Scilit:
- Scientists rise up against statistical significanceNature, 2019
- Association of Mechanism of Injury With Risk for Venous Thromboembolism After TraumaJAMA Surgery, 2017
- The ASA Statement on p -Values: Context, Process, and PurposeThe American Statistician, 2016
- Probability2010
- Categorical Data Analysis2002
- Algorithm AS 304: Fisher's Randomization Test for Two Small Independent SamplesJournal of the Royal Statistical Society Series C: Applied Statistics, 1996