Normality Plot (Q–Q Plot)

Introduction: Q–Q Plot

A normality plot, often called a Q–Q (quantile–quantile) plot, is a graphical method used to assess whether a dataset follows a normal distribution. It visually compares the data’s quantiles with those of a theoretical normal distribution. If the data points fall close to a straight line, the dataset is approximately normal.

Background

The normality plot is a key tool in exploratory data analysis. Since many statistical methods — such as regression, t-tests, and ANOVA — assume that data are normally distributed, checking this assumption is an essential first step. It helps analysts verify whether the dataset meets parametric test requirements or if transformation or non-parametric alternatives are needed.

Key Elements / Features

• Quantile comparison: Data quantiles are plotted against the corresponding quantiles of a standard normal distribution.
• Line of normality: If the dataset is normal, the points align closely with a diagonal reference line.

Deviation patterns:

• Curvature indicates skewness (right or left).
• Divergence at the ends suggests heavy or light tails.

Formula (for normal quantiles):

\(
z_i = \Phi^{-1}\left(\frac{i – 0.5}{n}\right)
\)

Where:

• \(
  z_i = \text{theoretical quantile from the standard normal distribution}
  \)
• \(
  \Phi^{-1} = \text{inverse of the cumulative normal distribution function}
  \)
• \(
  i = \text{rank of the ordered data point}
  \)
• \(
  n = \text{total number of observations}
  \)

Applications / Examples

• Statistical assumptions: Confirm whether data meet the normality assumption required for parametric tests.
• Model validation: Check regression or ANOVA residuals for normality.
• Data transformation: Guide when to use log, square root, or Box–Cox transformations.

Example:
In a process capability study, a Q–Q plot of measurement data shows slight curvature at the upper end — indicating a right-skewed distribution. Applying a log transformation brings the data closer to normality, validating the capability analysis.

Relevance / Impact

Normality plots provide a visual and intuitive method for assessing data distribution. Identifying deviations from normality early ensures the correct choice of statistical tests, improves model validity, and strengthens decision-making in Lean Six Sigma projects.

See also

• Normal Distribution
• Central Limit Theorem
• Histogram
• Box–Cox Transformation