Combined Standard Deviation is used to find the pooled standard deviation when two or more groups of data are combined into a single dataset.
## Core concept
When datasets from different groups are merged, the combined standard deviation reflects the overall variability of the merged data. This is not simply the arithmetic mean of individual standard deviations—it accounts for group sizes, individual means, and individual standard deviations.
Use combined SD when: - Comparing overall variability across merged datasets - Analyzing pooled results from multiple sources - Exam-style consolidation problems involving two or more groups
Key insight: Combined SD depends on three factors: 1. Individual group sizes (n₁, n₂, etc.) 2. Individual group means (x̄₁, x̄₂, etc.) 3. Individual group standard deviations (σ₁, σ₂, etc.)
## Formula / rule
For two groups:
$$\sigma_{combined} = \sqrt{\frac{n_1(\sigma_1^2 + d_1^2) + n_2(\sigma_2^2 + d_2^2)}{n_1 + n_2}}$$
Where: - n₁, n₂ = sizes of groups 1 and 2 - σ₁, σ₂ = standard deviations of groups 1 and 2 - d₁ = x̄₁ − x̄combined (deviation of group 1 mean from overall mean) - d₂ = x̄₂ − x̄combined (deviation of group 2 mean from overall mean) - x̄combined = (n₁x̄₁ + n₂x̄₂)/(n₁ + n₂)
Step-by-step procedure: 1. Calculate the combined mean using weighted average formula 2. Find deviations (d₁, d₂) of each group mean from combined mean 3. Square individual standard deviations (σ₁², σ₂²) 4. Apply the formula above 5. Take the square root of the final result
## Common exam applications
Typical scenarios: - Two departments with different productivity measures merged for organizational review - Combining test scores from two batches of students - Pooling sales data from multiple regional offices - Merging performance metrics from morning and evening shifts
Exam pattern: Usually a 4–6 mark question requiring calculation and interpretation.
## Worked example
Group 1: n₁ = 40, x̄₁ = 50, σ₁ = 5 Group 2: n₂ = 60, x̄₂ = 55, σ₂ = 6
Step 1: Combined mean $$\bar{x}_{combined} = \frac{40(50) + 60(55)}{40 + 60} = \frac{2000 + 3300}{100} = 53$$
Step 2: Deviations - d₁ = 50 − 53 = −3 - d₂ = 55 − 53 = +2
Step 3: Apply formula $$\sigma_{combined}^2 = \frac{40(25 + 9) + 60(36 + 4)}{100} = \frac{40(34) + 60(40)}{100} = \frac{1360 + 2400}{100} = \frac{3760}{100} = 37.6$$
$$\sigma_{combined} = \sqrt{37.6} ≈ 6.13$$
## Common mistakes
- Ignoring deviations: Forgetting to add d² terms; this gives incorrect results
- Wrong combined mean: Using simple average instead of weighted average
- Confusing with weighted average of SDs: Combined SD ≠ (n₁σ₁ + n₂σ₂)/(n₁ + n₂)
- Sign errors: Deviations can be negative; square them correctly
- Forgetting the square root: Final answer must be a positive value (SD cannot be negative)
## Quick revision checklist
✓ Combined mean = weighted average of group means ✓ Deviations measured from combined mean, not individual means ✓ Formula includes both variance term (σ²) and deviation term (d²) ✓ Always take positive square root at the end ✓ Combined SD > individual SDs when groups have very different means