A z-score (or standard score) measures how many standard deviations a data point is from the mean. It converts raw scores into a standardised scale that enables comparison across different datasets.

The Z-Score Formula

z = (x − μ) ÷ σ

Where:

  • x = individual data point
  • μ (mu) = population mean
  • σ (sigma) = population standard deviation

For a sample, replace μ with x̄ (sample mean) and σ with s (sample SD).

Worked Example

A student scores 72 on an exam. The class mean is 65, and the standard deviation is 8.

z = (72 − 65) ÷ 8 = 7 ÷ 8 = 0.875

This student scored 0.875 standard deviations above the mean.

Interpreting Z-Scores

Z-scoreInterpretationPercentile (approx.)
−3Extremely below average0.1%
−2Well below average2.3%
−1Below average15.9%
0At the mean50.0%
+1Above average84.1%
+2Well above average97.7%
+3Extremely above average99.9%

The 68-95-99.7 Rule

In a normal distribution:

  • 68% of data falls within ±1 standard deviation
  • 95% within ±2 standard deviations
  • 99.7% within ±3 standard deviations

Converting Z-Score to Percentile

Once you have a z-score, look up the standard normal table (Z-table) or use:

Percentile = Φ(z) × 100

Where Φ is the cumulative normal distribution function.

Example: z = 1.5 → Φ(1.5) = 0.9332 → 93.3rd percentile

Applications of Z-Scores

Finance:

  • Altman Z-Score predicts bankruptcy risk
  • Used in risk management to identify outliers

Healthcare:

  • BMI for age z-scores for children
  • Bone density (DXA) T-scores are a form of z-score

Quality control:

  • Six Sigma uses z-scores to measure process capability
  • A "6-sigma" process has a z-score of 6 (3.4 defects per million)

Standardising test scores:

  • IQ scores: mean 100, SD 15 (a z-score of +2 → IQ 130)
  • SAT scores: mean 1000, SD 200 (scaled from z-scores)

Comparing Scores Across Different Tests

Example: Alice scored 80 on Test A (mean 70, SD 10). Bob scored 55 on Test B (mean 40, SD 8).

Alice's z = (80 − 70) ÷ 10 = 1.0
Bob's z = (55 − 40) ÷ 8 = 1.875

Despite the lower raw score, Bob performed better relative to his peers.