Standard Normal Variate (Z): Definition, Table Use and

The standard normal variate, universally written as Z, is a normal random variable rescaled to have mean 0 and standard deviation 1, written Z ~ N(0, 1). It is the common language into which every normal distribution is translated before its probabilities are looked up.

It matters because no one memorises a separate probability table for every possible μ and σ. By converting any X ~ N(μ, σ²) into Z, the entire world's normal probabilities are read from one single table, the standard normal table, which is printed in every statistics textbook.

Key Concepts

• Standardisation formula: Z = (X - μ)/σ. Subtracting μ centres the data at 0; dividing by σ scales the spread to 1.
• Z-score meaning: Z = 2 means the value is 2 standard deviations above the mean; Z = -1.5 means 1.5 standard deviations below.
• Table types: some tables give P(Z ≤ z) (cumulative from the left), others give the area from 0 to z. Know which one your exam provides.
• Symmetry: P(Z > z) = P(Z < -z), and P(Z ≤ z) = 1 - P(Z ≤ -z). The curve mirrors around 0.
• φ and Φ: φ(z) is the density height at z; Φ(z) = P(Z ≤ z) is the cumulative area, the value the table reports.

Formulas

Z = (X - μ)/σ, and back again: X = μ + Zσ. Probability transfer: P(a < X < b) = Φ((b-μ)/σ) - Φ((a-μ)/σ). Useful constants: Φ(1) ≈ 0.8413, Φ(1.645) ≈ 0.95, Φ(1.96) ≈ 0.975, Φ(2) ≈ 0.9772, Φ(2.576) ≈ 0.995. Tails: P(|Z| > 1.96) = 0.05 and P(|Z| > 2.576) = 0.01, the two most used cut-offs in hypothesis testing.

Worked Example

Heights are normal with μ = 170 cm and σ = 8 cm. Find P(a person is taller than 182 cm). Step 1: Z = (182 - 170)/8 = 12/8 = 1.5. Step 2: the table gives Φ(1.5) = 0.9332, the area below 182 cm. Step 3: P(X > 182) = 1 - 0.9332 = 0.0668, about 6.7%. Reverse question: what height marks the tallest 10%? Need Φ(z) = 0.90, so z = 1.28, and X = 170 + 1.28 × 8 = 180.24 cm. Anyone above about 180 cm is in the top tenth.

Common Misconceptions

• Standardising does not change the shape: Z is still normal and bell-shaped; only the centre (now 0) and scale (now 1) changed. Relative positions of values are preserved.
• Z can be any real number: Z = 3.2 or Z = -2.7 are perfectly valid. The 68-95-99.7 rule says values beyond ±3 are rare, not impossible.
• Read the table direction carefully: if your table gives areas from 0 to z, then P(Z ≤ 1.5) = 0.5 + 0.4332 = 0.9332, not 0.4332. Half the answers in exams go wrong on this table-reading step.

Real-World Applications

• Grade standardisation: converting raw marks from different exams onto one Z scale so classes can be compared fairly.
• Outlier detection: flagging |Z| > 3 as suspicious in fraud checks, sensor data and lab results.
• Hypothesis testing: Z-tests compare a sample mean to a claimed value using standard normal cut-offs.

The standard normal variate Z = (X - μ)/σ is the universal translator for bell curves: subtract the mean, divide by the spread, and every normal probability becomes one table lookup. Learn which table your exam uses and the key cut-offs 1.645, 1.96 and 2.576, and normal-distribution questions turn into quick arithmetic.