Report likelihoods, not p-values

LessWronglesswrong.comWritten November 201834 min read

thinking

A wiki page arguing scientific papers should report likelihood functions instead of p-values, so readers can see how strongly the evidence supports each hypothesis.

Read the original

Another oneHindsight bias

The Map Is Not The TerritoryLessWrong 2 min
Fake ExplanationsEliezer Yudkowsky 3 min
Think Like RealityEliezer Yudkowsky 3 min
I Have A Few QuestionsMorgan Housel 1 min
Overconfidence Bias: What It Is and How to Overcome ItTaylor Pearson 16 min
Anchoring and AdjustmentEliezer Yudkowsky 5 min

Saved the same day · Aug 8

One of 13 notes I saved that day, alongside:

Reality has a surprising amount of detail(favorite)John Salvatier 10 min
To listen well, get curiousBen Kuhn 6 min

Report likelihoods, not p-values

Related

More on thinking

Saved the same day · Aug 8

Keyboard shortcuts

Everywhere

Note lists

Collections

Shuffle deck