Year 9 · Unit 5 Scatter graphs and correlation · Lesson 2 of 6

Describing correlation

Three words describe a scatter graph: the direction of the dots, whether they follow a straight line, and how tightly they hug it.

Statistics and ProbabilityNo calculator

Warm-up

Three quick ones from earlier work. Try each in your head or on paper before you open the answer.

  1. The dots on a scatter graph fall from left to right. Is the direction positive or negative?
    AnswerNegative.
  2. Which variable goes on the x axis, the independent or the dependent one?
    AnswerThe independent variable.
  3. Find the mean of 10, 14, 18 and 22.
    Answer(10 + 14 + 18 + 22) ÷ 4 = 16.
Words used in this lesson
correlation
a link between two variables seen in a scatter graph. For example: positive, linear, strong.
linear
following a straight line. For example: dots in a band.
strength
how closely the dots follow the pattern. For example: strong or weak.
outlier
a dot that sits far from the pattern. For example: one dot far below the band.
non-linear
following a curve, not a straight line. For example: an arch.

Three words

A reporter has ten seconds to describe a scatter graph, so she uses three words: one for the direction, one for the form and one for the strength. Before you read on, guess what the three words might be for dots in a tight band running from the bottom left to the top right.

Direction: positive (the dots rise), negative (they fall) or none. Form: linear (the dots cluster about a straight line) or non-linear (a curve). Strength: strong, moderate or weak, by how closely the dots follow the pattern.

A dot that sits far from the pattern is an outlier. Mention it separately; it does not change the three words by itself.

Pick a pattern and slide the spread. Say the three words before you read them.

1
positivedirection
linearform
strongstrength

Rises, spread 1: positive, linear, strong. The largest gap from the trend line is 10.

Show me the steps
  1. Rises, spread 0: every dot sits on one straight line going up. Positive, linear, strong.
  2. Rises, spread 4: the dots stay in a band that rises, but the band is wide. Positive, linear, weak.
  3. Falls, spread 1: the dots go down to the right in a narrow band. Negative, linear, strong.
  4. No link: the dots have no direction. Changing the spread changes nothing, because there is no pattern to spread from.
  5. Curve: the dots rise and then fall. The relationship is clear but not linear, so it has no single direction.

Two slips to watch for. Strong does not mean steep: a shallow line of tightly packed dots is a strong correlation. And “no linear correlation” is not “no relationship”: dots in a clear arch follow a pattern, it is just not a straight line.

Worked examples

Puppies' ages (weeks) against weight (kg): the dots rise in a narrow band, and one dot sits far below the band. Describe it.

  1. The dots rise from left to right: the direction is positive.
  2. They cluster about a straight line: the form is linear.
  3. The band is narrow: the strength is strong. One dot far below the band is an outlier.

Graph A: dots within 2 units of the line y = 3x + 5. Graph B: dots within 12 units of the same line. Which correlation is stronger?

  1. The direction and steepness are the same: both lines have gradient 3.
  2. A's dots hug the line more closely (within 2 units, not 12).
  3. So A has the stronger correlation: strength is about closeness, not steepness.

Your go. The first step is given. Shoe length and height for 30 pupils: the dots rise in a band of medium width with no sign of a curve. Describe the correlation.

  1. As shoe length rises, height rises, so the direction is positive.
  2. Give the form and the strength.
Check the answer

Positive, linear and moderate.

Check before you go on

1. The dots on a scatter graph are spread widely about a rising pattern. Which description fits?

Your turn

Start at Level 1 and go as far as you can. Type negative numbers with a minus sign, for example -7. A wrong answer gets a reason, not just a cross: read it, then try again.

Level 1 · Getting started

2. Five pairs (x, y): (1, 18), (2, 29), (3, 33), (4, 47), (5, 52). By how much does y rise from the first pair to the last?

3. A trend line is y = 8x + 10. At x = 7 a dot sits at y = 71. How far above the trend line is the dot?

Level 2 · Building

4. Graph A's dots lie within 3 units of its line. Graph B's dots lie within 9 units of the same line. B's band is how many times as wide as A's?

5. Dots on a scatter graph rise steadily and then fall steadily in a clear arch. Which statement is best?

6. A scatter graph has 20 dots. Five of them lie far from the band that holds the rest. What percentage of the dots are inside the band?

Level 3 · Stretch

7. The trend for x from 1 to 8 is y = 3x + 20, and no dot is more than 4 units from the line. What is the largest value of y that a dot could have?

8. Graph C: every dot lies within 3 units of a steep line that falls. Graph D: every dot lies within 3 units of a shallow line that rises. Which statement is correct?

Something to think about

Two graphs show a rising pattern. In graph A the dots are tightly packed about a very shallow line. In graph B the dots are widely scattered about a steep line. A pupil says B has the stronger correlation because it is steeper. What would you say?

One way to think about it

Steepness is the gradient of the pattern; strength is how closely the dots follow it. A is the stronger correlation, even though it is shallow, because its dots lie close to the line. B is steeper but its dots are spread out.

Exit ticket

9. The dots fall in a band of medium width about a straight line. Two dots sit far below the band. Which is the best description?