Describing and Comparing Distributions

Lesson 1 of 2: Interpret Data in Context

In this lesson:

• Describe a distribution's shape, center, and spread
• Interpret two-group differences for a real decision

Learning Objectives for This Unit

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

1. Describe shape, center, and spread in context
2. Name the direction of skew correctly
3. Interpret two-group differences in context
4. Explain how an outlier affects each statistic
5. Decide whether to keep or remove an outlier

Is the Mean Really a Description?

I tell you the mean commute is 25 minutes.

• Have I described the commutes? Not really.
• Are they bunched or varied? Mostly short with a few long?

A real description names shape, center, and spread — about the real thing.

Read the Commute-Time Histogram Together

• Most cluster around 20–30 minutes
• A long tail stretches out to 70 minutes

One number couldn't show that tail. Shape comes first.

Name Skew for the Tail

This commute distribution is right-skewed — the long tail points right.

• Rule: skew is named for the tail, not the peak
• The peak is on the left, but the tail stretches right

Trace the tail with your finger — that direction names the skew.

Center: What a Typical Commute Is

Because the shape is skewed, use the median (resistant).

• The median commute is about 25 minutes

Say it in context:

A typical worker commutes about 25 minutes.

Not "the median is 25" — a sentence about the commuters.

Spread: How Much Commutes Vary

For skewed data, use the IQR — the middle 50% width.

• The IQR is about 15 minutes

Say it in context:

The middle half of commutes span about 15 minutes.

Spread tells you how consistent or variable the commutes are.

Assemble the Full Description Sentence

Put shape, center, and spread into one sentence:

Commute times are right-skewed — most workers commute around 25 minutes, a few stretch to 70, and the middle half span about 15 minutes.

Shape, center, spread — all about the commuters. Reuse this template.

Your Turn: Describe a Distribution

Summer highs: symmetric, single peak near 85°F, most days 80–90°F.

Write a complete description — shape, center, spread, in context.

Symmetric, so use mean and SD. Try it, then advance.

Answer: Highs are roughly symmetric, centered near 85°F, most days within about 5°F — a consistent summer.

Quick Check: Name the Skew

Name the skew of each:

1. Long tail toward high values
2. Long tail toward low values
3. Roughly symmetric, both sides even

Skew points to the tail. Decide all three, then advance.

Answers: 1) right-skewed; 2) left-skewed; 3) symmetric (no skew).

From Describing to Comparing Groups

Describing one distribution is the unit you reuse twice to compare.

• But comparison adds a step: interpret the difference in each feature
• A difference in center, and a difference in spread, each carry meaning

Describe each group, then read what the differences mean.

Two Coffee Shops, Side by Side

• Store A: lower median, wider box (larger IQR)
• Store B: higher median, tight box (small IQR)

Interpret the Center Difference in Context

Store A's median wait is lower than Store B's.

• For a customer: Store A is usually faster
• On a typical visit, you'll wait less at Store A

Not just "A's median is lower" — what it means for the customer.

Interpret the Spread Difference in Context

Store A has a much larger IQR than Store B.

• Store A: unpredictable — some visits quick, some long
• Store B: reliably consistent every time

If you can't risk a long wait, Store B's reliability may beat A's speed.

Lower Is Not Always Better

Before calling a difference good or bad, check the context.

• Lower wait time → good
• Lower test score → bad
• Higher error rate → bad

The same difference can be good or bad depending on what's measured.

Your Turn: Interpret a Comparison

Two delivery services: A has a lower median but large spread; B has a higher median but small spread.

Write a center sentence and a spread sentence.

Try it, then advance.

Answer: A is usually faster but unreliable; B is slower but dependable. Choose B for a guaranteed time.

Full Task: Describe and Compare

Given two groups, each as a plot:

1. Write a complete description of each
2. Interpret the center difference in context
3. Interpret the spread difference in context

Check whether lower or higher is good here.

Answer: Two descriptions, then center and spread difference sentences, with the good/bad direction.

Key Takeaways From This Lesson

✓ Describe shape, center, and spread as a sentence

✓ Skew is named for the tail, not the peak

✓ Comparing means interpreting both differences

"Lower" is good only relative to context

A description is a sentence, not numbers

Next: when one outlier distorts the story.