Median
Middle value of a sorted dataset — the 50 % value.
Also known as: 50th percentile, Q2
The median splits a sorted dataset into two equal halves: 50 % of values fall below, 50 % above. For odd it is the middle value; for even the average of the two middle values.
Unlike the mean, the median is robust against outliers — a single extreme value does not shift it. For strongly skewed distributions (income, waiting times, lifetimes) the median is therefore often the more honest measure of location.
See also
Used in
- Histogram
- Boxplot
- Process Capability
- Hypothesis Test
- Regression Analysis
- Regression (Attributive)
- Correlation Analysis
- MSA Type 2
- DoE Planner
- DoE Advisor
- Sample Size
- Data Transformation
- Distribution Fit
- Calculator
- Outlier Test
- Multivariate Control Chart (Hotelling T²)
- Pairwise Comparison
- Control Chart
- Individual Value Plot
- XY Plot
- Probability Plot
- Run Chart
In the Algorithm Lab
Sources
- Montgomery, D.C.: Introduction to Statistical Quality Control, 8th ed., Wiley, 2020 — Kap. 3