Four Features, or It Is Not a Description

You'll learnwhich variable goes on which axis and why the choice matters later, the four features a complete scatterplot description covers — form, direction, strength, unusual points — why each one alone can be shared by very different clouds, and why an unusual point is investigated rather than deleted.

A scatterplot shows two quantitative variables measured on the same individuals, one point per individual, and describing one is a four-item checklist rather than a sentence. "It is positive" is the most common answer and it loses most of the credit — this page draws a second cloud that is also positive at r = 0.54 next to the students' r = 0.977, and one word cannot tell them apart. "Linear, because a line can be drawn through it" is the second most common, and the page answers with a cloud built from a curve that still produces r = 0.925 while its residuals arch 8.9 points from ends to middle. Form, direction, strength, unusual features — in context, every time.

8 STEPS · 6 QUICK CHECKS · AXES · FORM · DIRECTION · STRENGTH · UNUSUAL POINTS · v1

both clouds are positive: r = 0.977 and 0.54 · a curve still gives r = 0.925 · each cluster rises, the pair falls

Before you start

What you're looking at

Eight students' weekly study hours and exam scores, one point per student — and then three more clouds, each built to break one shortcut in describing the first.

The question

What has to be in a description of a scatterplot before it actually describes these data and not a thousand other plots?

Watch for

Pairs of clouds sharing one word — "positive", "strong", "linear" — while showing completely different things. Each shared word is a feature the description still owes.

Step 1 / 8