Variance: Definition and Interpretation
What you'll learn: You'll understand how variance quantifies the spread of a random variable's values around its mean.
The Core Idea
After computing a random variable's expected value (mean) μ, we often want to know: how spread out are the values? Two random variables might have the same mean but wildly different behaviors—one tightly clustered near μ, another scattered far and wide.
Variance measures this spread. It's defined as:
Var(X) = σ² = E[(X - μ)²]
In words: variance is the expected value of the squared deviation from the mean.
Let's unpack this:
- (X - μ) is the deviation—how far a value is from the mean
- Squaring it (X - μ)² makes all deviations positive and emphasizes larger deviations more
- Taking the expectation E[...] gives us the average squared distance from μ
Why Square the Deviations?
If we just averaged (X - μ), positive and negative deviations would cancel out, always giving zero. Squaring ensures all contributions are positive and lets us meaningfully measure spread.
Interpretation
- Low variance: Values cluster tightly around the mean (predictable)
- High variance: Values are widely dispersed (more variability)
Think of two basketball players with the same average score (μ = 20 points). Player A scores 18-22 points every game (low variance—consistent). Player B scores anywhere from 5-35 points (high variance—unpredictable).
The units of variance are the square of the original units (points²), which is why we often use the standard deviation σ = √Var(X) for interpretation.
Key Takeaway: Variance σ² = E[(X - μ)²] quantifies how spread out a random variable's values are around its mean, with larger values indicating greater dispersion.