Week 9 · Unit 7: Recording, organising and representing data · 4 lessons
1
Scatter graphs and correlation
Student’s Book 8: Unit 7, pp. 63–72
GoalPlot and read a scatter graph and say whether the correlation is positive, negative or none.
New words
scatter graph · a graph of paired data in which each pair of values is plotted as one pointcorrelation · the kind of relationship between two sets of data shown by a scatter graphpositive correlation · as one value increases, the other tends to increase toonegative correlation · as one value increases, the other tends to decrease
Explanation
A scatter graph is used to find out whether two sets of data are related. Each pair of values makes one point, with one value on the horizontal axis and the other on the vertical axis. If the points go up from left to right, there is positive correlation. If they go down, there is negative correlation. If they are scattered with no pattern, there is no correlation. The closer the points lie to a straight line, the stronger the correlation. Be careful: correlation shows a link, but it does not prove that one thing causes the other.
Worked examples
A gardener records the hours of sunshine and the number of ripe tomatoes picked on 10 days. The point (5, 10) means 5 hours of sunshine and 10 tomatoes. The points rise from left to right, so there is positive correlation. It is strong, because the points lie close to a straight line.
A phone shop records the age of second-hand phones and their price. The points fall from left to right: older phones cost less. This is negative correlation. The point (4, 270) means a 4-year-old phone costs $270.
Shoe size and a spelling score for pupils of the same age. The points are scattered with no pattern, so there is no correlation. Knowing a pupil’s shoe size does not help us guess their score. Warning: if the pupils were of all ages, shoe size and score might both rise with age and show positive correlation, but a bigger shoe does not make anyone spell better. Age would be the hidden link.
Class activity
Make your own scatter graph. Ask 10 people (family or friends) for their height and the length of their right foot in centimetres. Draw axes on squared paper, with height along the bottom and foot length up the side. Plot each pair as a cross. Describe the correlation. Then plot a second graph with their height and the number of letters in their first name, and compare the two graphs.
Practice
1
What type of relationship would you expect, if any? Give a reason. (a) a person’s arm span and their height (b) hours of piano practice in a week and mistakes made in a piece (c) a person’s birth month number and their maths mark (d) the distance from home to school and the travel time
(a) Positive: tall people have longer arms. (b) Negative: more practice, fewer mistakes. (c) None: the month has no link with the mark. (d) Positive: farther usually takes longer.
2
Eight pairs of values: hours of homework per week 1, 2, 2, 3, 4, 5, 6, 7 and marks 35, 40, 48, 50, 60, 64, 75, 80. Describe the correlation you expect when you plot them.
Hours
1
2
2
3
4
5
6
7
Mark
35
40
48
50
60
64
75
80
Positive correlation, fairly strong: the marks rise as the hours increase.
3
The graph shows the outside temperature and the number of hot drinks sold. Describe the correlation and what it means.
Negative correlation, strong: when it is warmer, fewer hot drinks are sold.
4
Convince a friend. Nodira says: ‘The more fire engines there are at a fire, the more damage it causes, so fire engines cause damage.’ The data show positive correlation. Is she right?
No. Correlation does not prove cause. Big fires cause a lot of damage, and big fires also need many fire engines. The size of the fire links the two.