Describing Patterns in Scatter Plots

Lesson 2 of 2: Scatter Plots

In this lesson:

• Name what a scatter plot's cloud is doing
• Read clusters and outliers, and write a full description

What You Will Be Able To Do

By the end of this lesson, you should be able to:

1. Describe association as positive, negative, or none
2. Distinguish linear from nonlinear association
3. Identify and interpret clusters and outliers in context

Five Clouds, Five Different Shapes

How would you describe each so someone who can't see it knows what it shows?

Trace the Trend With Your Finger

Look at last lesson's study-time cloud.

Trace its general direction — it drifts up to the right.

That up-to-the-right drift has a name: positive association.

Positive Versus Negative Association Direction

• Positive: as one goes up, the other tends up (slopes up-right)
• Negative: as one goes up, the other tends down (slopes down-right)

Linear vs. Nonlinear: Same Direction, Different Shape

• Linear: the trend follows an approximately straight path
• Nonlinear: the trend bends — a curve, not a line

No Association At All Between Variables

Knowing one variable tells you nothing about the other.

Is Scatter the Same as No Pattern?

All three trend upward. Does loose scatter mean "no association"? Predict before advancing.

Real Data Is Messy But Readable

The three-part description:

1. Direction — positive, negative, or none
2. Shape — linear or nonlinear
3. Strength — tightly clustered or widely scattered

Say Association, Never Say It Causes

• Use the word association, not "correlation" (that's high school)
• Association means two things tend to move together
• It does not prove one causes the other

Quick Check: Classify These Three

For each: direction (up/down/none) and shape (linear/nonlinear)?

From the Whole Cloud to the Stray Points

You've named what the cloud does overall.

Now zoom in on the points that break the pattern, and the places where points bunch up.

Outliers: Points That Break the Pattern

• An outlier sits far from the overall trend
• It might be an error, an unusual case, or a different cause
• Don't delete it by reflex — investigate first

Clusters: Groups That Bunch Together

• A cluster is a group of points bunched apart from others
• Clusters often reveal subgroups in the data

Putting It All Together In Context

Negative linear; a cluster of warm southern cities; one cold-for-its-latitude outlier.

Your Turn: Describe a New Plot

Write a complete 3-4 sentence description of a new scatter plot.

Include: direction, shape, any clusters, any outliers, and a plausible explanation.

Key Takeaways From This Lesson

✓ Describe direction, shape, and strength
✓ Outliers and clusters carry real information
✓ Always interpret in context

Watch out: loose scatter is still association, not "no pattern"
Watch out: an outlier isn't automatically an error
Watch out: association is never proof of causation

Where We Go From Here Next

When the association is linear, the next standard draws a line through the cloud.

That line becomes an equation you can use to predict.