Variance: Average of squared deviations from the mean, measuring data spread
Standard Deviation: Square root of variance, in same units as original data
- Calculate the mean (average)
- Find deviation of each value from the mean
- Square each deviation
- Calculate average of squared deviations (variance)
- Take square root to get standard deviation
Mean = (2 + 4 + 6 + 8 + 10) ÷ 5 = 30 ÷ 5 = 6
2 - 6 = -4, 4 - 6 = -2, 6 - 6 = 0, 8 - 6 = 2, 10 - 6 = 4
(-4)² = 16, (-2)² = 4, (0)² = 0, (2)² = 4, (4)² = 16
Variance = (16 + 4 + 0 + 4 + 16) ÷ 5 = 40 ÷ 5 = 8
Standard Deviation = √8 ≈ 2.83
Variance = 8, Standard Deviation ≈ 2.83
• Mean: x̄ = Σx/n
• Variance: σ² = Σ(x - x̄)²/n
• Standard Deviation: σ = √σ²
Population Variance: σ² = Σ(x - μ)²/N (uses N)
Sample Variance: s² = Σ(x - x̄)²/(n-1) (uses n-1)
Mean = (5 + 7 + 9 + 11 + 13) ÷ 5 = 45 ÷ 5 = 9
(5-9)² = 16, (7-9)² = 4, (9-9)² = 0, (11-9)² = 4, (13-9)² = 16
σ² = (16 + 4 + 0 + 4 + 16) ÷ 5 = 40 ÷ 5 = 8
s² = (16 + 4 + 0 + 4 + 16) ÷ (5-1) = 40 ÷ 4 = 10
Population Variance = 8, Sample Variance = 10
• Population: Divide by N (total count)
• Sample: Divide by (n-1) for unbiased estimate
• Bessel's Correction: Using (n-1) reduces bias in sample estimates
Score: 10, 20, 30, 40
Frequency: 2, 3, 4, 1
Grouped Data Formula: σ² = Σf(x - x̄)²/Σf where f is frequency
x̄ = (10×2 + 20×3 + 30×4 + 40×1) ÷ (2+3+4+1) = (20+60+120+40) ÷ 10 = 240 ÷ 10 = 24
(10-24)² = 196, (20-24)² = 16, (30-24)² = 36, (40-24)² = 256
196×2 = 392, 16×3 = 48, 36×4 = 144, 256×1 = 256
σ² = (392 + 48 + 144 + 256) ÷ 10 = 840 ÷ 10 = 84
σ = √84 ≈ 9.17
Mean = 24, Variance = 84, Standard Deviation ≈ 9.17
• Weighted Mean: x̄ = Σ(f×x)/Σf
• Grouped Variance: σ² = Σf(x - x̄)²/Σf
• Frequency Weighting: Each data point appears f times in calculation
Variance: Average squared deviation from the mean, measure of data spread
Standard Deviation: Square root of variance, expressed in same units as data
Population: Entire group being studied (parameter)
Sample: Subset of population (statistic)
- Identify the type: Population vs sample data
- Calculate the mean: Sum of values divided by count
- Find deviations: Subtract mean from each value
- Square deviations: Eliminate negative values
- Average squares: Divide by N or (n-1)
- Calculate SD: Take square root of variance
• Population Variance: σ² = Σ(x - μ)²/N
• Sample Variance: s² = Σ(x - x̄)²/(n-1)
• Standard Deviation: σ = √σ²
• Weighted Variance: σ² = Σf(x - x̄)²/Σf
Coefficient of Variation: CV = (σ/μ) × 100%, measures relative variability
CV_A = (10/50) × 100% = 0.2 × 100% = 20%
CV_B = (20/100) × 100% = 0.2 × 100% = 20%
Since CV_A = CV_B = 20%, both datasets have equal relative variability
Although Dataset B has larger absolute variability (std dev = 20 vs 10), the relative variability compared to its mean is the same as Dataset A
Both datasets have equal relative variability (CV = 20%)
• Coefficient of Variation: CV = (σ/μ) × 100%
• Relative Comparison: Useful for comparing variability across different scales
• Unitless Measure: Allows comparison regardless of measurement units
Linear Transformation Properties: If Y = aX + b, then E[Y] = aE[X] + b and Var(Y) = a²Var(X)
Y = 2X + 5, so a = 2 and b = 5
E[Y] = E[2X + 5] = 2E[X] + 5 = 2(15) + 5 = 30 + 5 = 35
Var(Y) = Var(2X + 5) = 2²Var(X) = 4 × 9 = 36
SD(Y) = √36 = 6
Y has mean = 35, variance = 36, standard deviation = 6
• Linear Transformation: If Y = aX + b, E[Y] = aE[X] + b
• Variance Property: Var(aX + b) = a²Var(X)
• Additive Constant: Adding constant doesn't change variance
Measures of Spread: Quantify how data values vary around the center
Dispersion: Degree to which data points differ from central tendency
Homogeneity: Low variance indicates data points cluster closely together
Heterogeneity: High variance indicates data points are spread out widely
- Data Analysis: Determine if working with population or sample
- Central Tendency: Calculate mean (most common reference point)
- Deviation Calculation: Find distance of each value from mean
- Squaring: Square deviations to eliminate negatives and emphasize larger differences
- Averaging: Calculate mean of squared deviations
- Interpretation: Convert back to original units using square root
• Translation Invariance: Adding constant doesn't change variance: Var(X + c) = Var(X)
• Scaling Property: Multiplying by constant: Var(aX) = a²Var(X)
• Linear Transformation: Var(aX + b) = a²Var(X)
• Independence: Var(X + Y) = Var(X) + Var(Y) if X and Y are independent
• Non-negativity: Variance ≥ 0, equals zero only when all values are identical
Dataset 1: σ = 5 (low variability)
Dataset 2: σ = 10 (medium variability)
Dataset 3: σ = 15 (high variability)
Analysis: The chart demonstrates how standard deviation affects the spread of data around the mean.
- Low SD: Data tightly clustered around mean (bell-shaped curve is narrow)
- Medium SD: Moderate spread around mean
- High SD: Wide spread of data points from mean (flatter curve)