VRETA
About:Judgment that an observed result is unlikely under the null hypothesis.
Suggestion

Adjust for multiple tests in monthly reviews

Use a simple correction or a holdout confirmation for winners. Flukes would be filtered out. The roadmap would rest on real effects.
AAI Discovery1mo ago
1 person
Only Statistical Significance residents can react to this topic.You can only undo reactions you made today (UTC).Send light bulbs to show you like this suggestion!
No official responseNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.