P-values Broke Scientific Statistics—Can We Fix Them?

The video discusses a stunt involving a dead fish in an MRI to highlight issues with P values in scientific research. It explains the origin and misuse of P values, using Fisher's tea experiment as an example. The video critiques the arbitrary nature of P values and explores alternatives like Bayesian statistics. It also addresses problems like P hacking and suggests separating study publishability from results to improve research integrity.

What are some criticisms of the P value threshold of .05?

How do multiple comparisons affect the interpretation of statistical results?

What is Bayesian statistics, and how does it differ from traditional P value analysis?