Unit 5.5: Distributions of Differences Between Sample Means - Study Guide
KEY CONCEPTS
1. Distribution of Difference Between Sample Means
When comparing two independent groups, we analyze the difference between their sample means (x̄₁ - x̄₂).
Formula: μ(x̄₁-x̄₂) = μ₁ - μ₂
2. Standard Error of the Difference
Measures the variability of the difference between sample means.
Formula: SE(x̄₁-x̄₂) = √(s₁²/n₁ + s₂²/n₂)
Where:
- s₁, s₂ = sample standard deviations
- n₁, n₂ = sample sizes
3. Degrees of Freedom
For t-procedures with two samples, use the conservative approach:
df = smaller of (n₁ - 1) and (n₂ - 1)
4. Margin of Error
The margin of error accounts for sampling variability in estimating the difference.
Formula: E = t* × SE(x̄₁-x̄₂)
Where t* = critical value from t-distribution table based on desired confidence level and degrees of freedom
5. Confidence Interval
A confidence interval estimates the range for the true difference between population means.
Formula: (x̄₁ - x̄₂) ± E
Or: (x̄₁ - x̄₂) ± t* × SE(x̄₁-x̄₂)
6. Hypothesis Testing for Difference of Means
Test whether there is a significant difference between two population means.
Null Hypothesis (H₀): μ₁ - μ₂ = 0 (no difference)
Alternative Hypothesis (Hₐ):
- Two-sided: μ₁ - μ₂ ≠ 0
- One-sided: μ₁ - μ₂ > 0 or μ₁ - μ₂ < 0
INTERPRETING RESULTS
When interpreting confidence intervals:
- If the interval contains 0: No evidence of a significant difference
- If the interval is entirely positive: Evidence that μ₁ > μ₂
- If the interval is entirely negative: Evidence that μ₁ < μ₂
Remember: The confidence level represents the long-run proportion of intervals that would capture the true parameter if repeated many times.