đ Example: F-Test for Two Variances
đ¯ Problem:
Two factories produce the same type of product. A quality inspector wants to know if their product weight variances are different.
The data:
- Factory A:
- Sample size
- Sample variance
- Factory B:
- Sample size
- Sample variance
At a 5% significance level, test whether the variances are significantly different.
✅ Step 1: Set Hypotheses
We are testing:
- (variances are equal)
- (variances are not equal)
✅ Step 2: Compute F-statistic
Use the formula:
F = \frac{\text{larger variance}}{\text{smaller variance}} = \frac{20}{10} = 2.00
Let:
- Numerator df =
- Denominator df =
✅ Step 3: Find Critical Values
We use a two-tailed test at Îą = 0.05 → Each tail has 0.025.
We need:
- Upper critical value
- Lower critical value
From F-distribution tables:
- → so
F_{0.025,14,11} = \frac{1}{3.21} \approx 0.311
✅ Step 4: Decision Rule
- Reject if or
- Our F = 2.00 → Lies between 0.311 and 3.29
✅ Step 5: Conclusion
Since is within the acceptance region, we do not reject the null hypothesis.
đ Conclusion: There is not enough evidence to say the variances are different. So, the product weight variation between the two factories is statistically similar.