1. Statistics in publishing: the (mis)use of the p -value (Part 2).
- Author
-
Stunt, Jonáh J., Broekstra, Dieuwke C., and de Boer, Michiel R.
- Subjects
DISTAL radius fractures ,STATISTICS ,DISTRIBUTION (Probability theory) - Abstract
This is the second of a two-part article, in which we discuss problems associated with null hypothesis significance testing (NHST) and the use of the I p i -value. To start with misuse, many researchers just use CIs to conduct NHST: when the CI does not include the value of no effect, the null hypothesis is rejected, and the effect is significant. The area under the curve represents possible I p i -values; more specifically, the I p i -value is the probability where a datapoint or more extreme data can be observed under this curve. When the probability of an observed, or more extreme observation under the assumption that H SB 0 sb is true (i.e. the I p i -value), is very small, we state that this value does not support our null hypothesis and consequently reject it. [Extracted from the article]
- Published
- 2022
- Full Text
- View/download PDF