What is qq plot

Last updated: April 1, 2026

Quick Answer: A QQ plot (quantile-quantile plot) is a statistical graph that compares the distribution of a dataset against a theoretical probability distribution to test normality or assess goodness of fit.

Key Facts

What is a QQ Plot?

A QQ plot, short for quantile-quantile plot, is a graphical tool used to assess whether a dataset follows a particular probability distribution, most commonly the normal distribution. It works by plotting the quantiles of your data against the quantiles of a theoretical distribution on a two-dimensional graph.

How QQ Plots Work

The plot displays data points on the y-axis and theoretical quantiles on the x-axis. If your data follows the assumed distribution perfectly, all points would fall exactly on a straight diagonal line. In reality, some deviation is expected due to random variation, but substantial departures from this line indicate that the data doesn't follow the assumed distribution pattern.

Key Features and Interpretation

The corners and edges of a QQ plot are particularly informative. Tails (the ends of the plot) reveal heavy tails or outliers in your data. If points curve upward at both ends, the data has heavier tails than a normal distribution. If they curve downward, the tails are lighter. The center of the plot reveals skewness, with points curving left or right indicating that your data is asymmetrical.

Common Applications

QQ plots are extensively used in statistical analysis, particularly before choosing statistical tests that assume normality. Data scientists use them during exploratory data analysis to understand data characteristics. Quality control engineers employ QQ plots to verify that manufacturing processes produce normally distributed outputs. Researchers also use them to validate model assumptions in regression analysis and hypothesis testing.

Advantages and Limitations

QQ plots provide visual, intuitive assessment without formal hypothesis testing. They clearly show where and how distributions differ. However, interpretation requires some statistical knowledge, and visual assessment can be subjective. For large datasets, minor deviations become visually exaggerated. Formal normality tests like Shapiro-Wilk or Anderson-Darling provide more definitive results but don't show distribution differences as clearly.

Related Questions

What is the difference between a histogram and a QQ plot?

A histogram shows the frequency distribution of data in bins, while a QQ plot compares quantiles against a theoretical distribution. Histograms are better for visualizing overall shape, while QQ plots are superior for assessing fit to specific distributions.

What does it mean if QQ plot points deviate from the line?

Deviations indicate your data doesn't follow the assumed distribution. Curves at the top or bottom reveal heavy or light tails, while curves in the middle indicate skewness or bimodality.

How is a QQ plot different from a P-P plot?

Both compare distributions, but QQ plots use quantiles while P-P (probability-probability) plots use cumulative probabilities. QQ plots are more sensitive to tail behavior and are generally preferred for normality assessment.

Sources

  1. Wikipedia - Q-Q Plot CC-BY-SA-4.0
  2. Carnegie Mellon University Statistics Educational Use