Describe › Standalone plots

Scatter plot

A scatter plot draws two numeric columns against each other, with optional grouping by category and an optional LOESS smoother.

What is Scatter plot?

The scatter plot is the workhorse bivariate visualisation. Each dot is one observation in (x, y) space; with a grouping column, colour separates the groups so you can see whether the relationship differs across them.

The LOESS overlay (locally-weighted tricube smoother) traces the conditional mean of y given x — useful for seeing whether the relationship is linear, monotonically curved, or non-monotonic. With grouping, each group gets its own dashed smoother in its colour.

When should I use Scatter plot?

  • First look at any (x, y) relationship.
  • Visualising correlation strength + shape before fitting a model.
  • Showing group differences in joint distributions.

What data does it need?

X + Y numeric columns + optional grouping column + LOESS toggle.

What does it report?

ECharts scatter (one or more series), optionally with LOESS smoothers.

How do I interpret the result?

A LOESS clearly curving away from linear ⇒ try polynomial / nonlinear models instead of linear_regression.

See also