Back to Quality Hub
Green Belt: Hypothesis Testing
Do numbers lie?
When we see two different means, is it due to process improvement or just random coincidence? Hypothesis testing answers this with P-value.
t-Test Simulator
Is Supplier A statistically better than Supplier B?
Group A (Supplier A)
Mean (Avg)100
Sigma (StdDev)5
Group B (Supplier B)
Mean (Avg)105
Sigma (StdDev)5
Sample Size (n)30
Conclusion
Significant Difference
P-value
0.0001
Alpha = 0.05
t-stat: 3.87
Overlap Area ≈ P-value
What does this mean?
When P < 0.05, we are 95% confident the groups are different. A low P-value means the chance this difference is just "luck" is very small.