Read on Twitter
Hello friends. Here's a list of things that you should actually worry about instead of p-values:
- Poorly-conceived research questions
- Poor data stewardship
- Sloppy research procedures
- Poor measurement
- Avoidable errors in the analysis
- Wrong model altogether
- Poorly-conceived research questions
- Poor data stewardship
- Sloppy research procedures
- Poor measurement
- Avoidable errors in the analysis
- Wrong model altogether
- Shallow theorizing
- Treating single studies as definitive answers to anything
- Not enough space in the "workload model" to properly design and plan research
- Misaligned career incentives
- Ridiculous funding models
- Treating single studies as definitive answers to anything
- Not enough space in the "workload model" to properly design and plan research
- Misaligned career incentives
- Ridiculous funding models
Though some statisticians/scientists like to debate about different statistical schools of thought, they usually give the same answer in most contexts, and any practical differences between them are probably smaller than those caused by the things I just listed.
Some papers that have influenced by opinions: https://www.thelancet.com/series/research
And my own 2 cents on part of the solution here: https://medium.com/@darren_dahly/statistical-reform-be2d4f742a4
