4.6Normal distribution and z-scores
The idea
The normal (bell curve) distribution is symmetric around its mean μ and described by its spread σ. To find probabilities you standardise to a z-score.
Think of it like this: Heights of adults: most are near the average, a few are very short or very tall, symmetrically.
Why it exists
Sums of many small effects are approximately normal (central limit theorem), so it is the default model for measurement noise.
How it works
- Z = (X − μ)/σ turns any normal into the standard normal N(0, 1).
- P(X ≤ x) = Φ(z), read from the table.
- About 68%, 95%, 99.7% of the values lie within 1, 2, 3 standard deviations of μ.
- The curve is symmetric: P(Z ≤ −z) = 1 − P(Z ≤ z).
- Sums of independent normals are normal (means add, variances add).
Worked example: X is N(10, 4) (so σ = 2); P(X ≤ 12)
z = (12 − 10)/2 = 1, so P = Φ(1) ≈ 0.8413. The z-score calculator below gives the same result and the central probability 0.683.
Now play with it
- →In the explorer, double σ and see the bell flatten and widen.
- →Use the z-calculator with x = 14 to see a z of 2 (central probability 0.954).
Dashed line = mean. Shaded/green part = the range below.
Symmetric bell. About 68%, 95% and 99.7% of the mass lies within 1, 2 and 3 standard deviations of the mean.
Standardise first, then use the standard normal table. For z = 1, 2, 3 the central probabilities are about 0.683, 0.954 and 0.997.
Check yourself
X ~ N(50, 25) (σ = 5). Which x has a z-score of 2?
Exam notes (the dense version)
The bell curve N(μ, σ²) is symmetric about μ. Standardise: Z = (X − μ)/σ ~ N(0, 1), then look up Φ(z). About 68% / 95% / 99.7% lie within 1 / 2 / 3 standard deviations. A sum of independent normals is normal (means add, variances add).
Previous year questions: Standard Distributions (8)
PYQs are grouped by topic, so this is the full set for Standard Distributions; try the ones that match this subtopic. Open all 8 on the topic page.