Module 3 · Lesson 12 of 17
Descriptive statistics and graphical summaries
Listen to this article
Ready
100% Free
This lesson and the entire Lean Six Sigma Black Belt academy are free of charge — no fees, ever.
Learning objectives
- Compute and interpret centre, spread and shape statistics.
- Choose the correct graph before formal tests.
- Detect outliers and stratification opportunities.
Summarise before you test
- Count, missingness, mean, median, standard deviation, range, quartiles, IQR.
- Histogram, box plot, run chart and stratified views.
- Robust summaries when outliers are present.
Garment-factory example
Wash shrinkage may have a mean within tolerance but a long upper tail concentrated in one fabric lot — visible on a stratified box plot but hidden in a headline average.
Method
- Clean the data and record how.
- Display raw distribution before any summary.
- Report both mean and median; investigate their gap.
- Stratify by the most plausible contextual factor.
Common mistakes
- Deleting outliers automatically.
- Reporting precision beyond instrument resolution.
Knowledge check
Pick one answer per question. Explanations appear after you submit.
1. The median is preferred over the mean when:
2. A stratified box plot helps to:
Author: NOSK.life · Last reviewed: 2026-07-21