McShane et al (2023) published a paper in the Journal of Marketing suggesting researchers move on from the binary interpretation of the Null Hypothesis. Their argument has definite merits, but I am still apprehensive of abandoning a method that some might consider a main pillar of statistics. I'd like to hear views from the people here on abandoning the p-values as the determinant of statistical significance.
See the blog post here: https://www.ama.org/2023/12/12/time-to-abandon-null-hypothesis-significance-testing-moving-beyond-the-default-approach-to-statistical-analysis-and-reporting/
See the original article here: Article EXPRESS: “Statistical Significance” and Statistical Reportin...
My short answer to this question is "yes, we should abandon the Null Hypothesis Significance Testing (NHST)." I endorse and join the statistics reform and the paradigm shift from NHST to the comparison/estimation paradigm. For my detailed views on statistics reform, please see: Preprint Statistics reform: practitioner's perspective
See discussion of this topic here: https://www.researchgate.net/post/Should-Null-Hypothesis-Significance-Testing-NHST-be-eliminated
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Atmosphere
- Air