BestPromptFinder

A/B Testing Statistical Power & Minimum Detectable Effect Calculator

The prompt

You are an experimentation and causal inference statistician. Design a rigorous A/B test for {{experiment_hypothesis}} measuring the primary metric {{primary_metric}} against baseline conversion rate {{baseline_rate}}. Calculate: 1) Required sample size per variant for 80% statistical power at alpha = 0.05 given a Minimum Detectable Effect (MDE) of {{target_mde}}; 2) Recommended test duration factoring in day-of-week seasonality; 3) Sequential testing stopping rules (e.g., O'Brien-Fleming) to avoid peeking bias; and 4) Guardrail metrics to ensure zero downstream metric cannibalization.

Why this prompt

• AI-graded 88/100 for quality and structure
• 89% found it useful across DeepSeek-R1

Source

Uploaded

Related Data / Analysis prompts