Relationship

Correlation ellipse plot

Encodes pairwise correlation through ellipse orientation, elongation and often colour.

Plot
Relationship
Multivariate exploration

What it answers

How do correlation signs and strengths differ?

Shape: Matrix of tilted ellipses
What it shows: Positive/negative associations and relative strength.

Example

The question

Which measures are strongly related, and in which direction?

Correlation ellipses for five weekly measures

Each cell holds an ellipse: the narrower it is, the stronger the correlation; tilting up means positive, down negative. Visits and orders are the narrowest (0.8, positive); price and discount tilt down (−0.6).

Illustrative data. It belongs to no company.

When to use it

Use it when

Compact correlation matrices for informed readers.

Avoid it when

When numeric precision or unfamiliar ellipse encodings matter.

How to read it

Read ellipse direction for sign and narrowness for magnitude.

Note

Ellipses summarise coefficients rather than reproducing the observations.

The data you need

Layout: Wide observation table
Fields: ObservationID: text; V1...Vp: numeric columns
One row is: One observation with several measured numeric variables. What is the grain?
Rules: Align variables by observation. Decide scaling, missing-data handling and which columns are features.
Optional: Optional Group. PCA, clustering and correlations require calculations before rendering their derived results.
Preparation: Compute the correlation matrix, then map coefficient sign/strength to ellipse orientation and elongation.
Units: Ellipses encode a unitless correlation coefficient with its chosen sign/strength mapping.

A wide table: one row per item, with a separate column for each measure or period.

Worked example

Correlations+0.8 and-0.2 produce a narrow upward ellipse and a rounder downward ellipse respectively.

ID	V1	V2	V3
A	10	30	5
B	12	20	8
C	18	25	6

All charts in the library