Back to Glossary
Glossary

Confidence Level

The threshold used to determine statistical significance, typically 95%.

What is Confidence Level?

The confidence level (often 95%) is the threshold used to determine statistical significance. It's related to the significance level (α) by: Confidence = 1 - α.

At 95% confidence (α = 0.05), you're accepting a 5% chance of a false positive — declaring a winner when there isn't one.

Common Confidence Levels

Confidenceα (Significance)Use Case
90%0.10Quick iterations, low stakes
95%0.05Industry standard
99%0.01High stakes, critical decisions

Trade-offs

  • Higher confidence: Fewer false positives, but need more data
  • Lower confidence: Faster decisions, but more risk of being wrong

See it in action

runab shows you these metrics for every A/B test you run.

Start testing free