Quantile-quantile chart
Normal quantile-quantile chart that checks whether a numeric sample follows a normal curve.
Use a quantile-quantile chart (Q-Q plot) to check how closely a sample follows a normal distribution: points near the line are normal-like, points that bend away are not.
Most deliveries track a normal curve, but the slowest few fall well above it
Days to deliver against the normal quantile; the dashed line passes through the quartiles
This chart cannot be displayed here: it needs JavaScript and WebGL.
x is the standard-normal quantile at plotting position (rank - 0.5) / n; the dashed line runs through the first and third quartile pairs
The same chart in each style
Source
```chart
type: qq
title: Most deliveries track a normal curve, but the slowest few fall well above it
subtitle: Days to deliver against the normal quantile; the dashed line passes through the quartiles
data:
inline:
- {days: 2.1}
- {days: 2.6}
- {days: 2.9}
- {days: 3.1}
- {days: 3.3}
- {days: 3.4}
- {days: 3.6}
- {days: 3.7}
- {days: 3.9}
- {days: 4.0}
- {days: 4.2}
- {days: 4.3}
- {days: 4.5}
- {days: 4.6}
- {days: 4.9}
- {days: 5.1}
- {days: 5.4}
- {days: 5.8}
- {days: 6.3}
- {days: 7.2}
- {days: 8.9}
encoding: {y: days}
axes:
y: {label: Days to deliver}
```
Fields that matter
type: qqwithencoding.yfor the numeric sample, one value per rowencoding.colordraws one set of points and one line per group;encoding.facetsplits the chart into panels- x is the standard-normal quantile at plotting position (rank - 0.5) / n; the dashed reference line runs through the first and third quartile pairs
- Readouts give the sample value as you wrote it, its theoretical quantile and its rank; RichView computes the last two, so they are marked as derived
- Rows without a value are listed as excluded; text that is not a number is an error
axes.ysetslabel,formatandunit; the chart is square by default (style.ratiochanges it)
Interactions
- Hover or focus a point to read its value, quantile and rank.
- Click a legend entry to toggle a group when the chart has several.
- Open the Table view, Download PNG or CSV, or Expand the chart.