Frequency polygons, line graphs and scatter graphs
Week 34 · Unit 19: Interpreting and discussing results · 4 lessons
3
Scatter graphs and correlation
Coursebook 9: Unit 19, pp. 177–183
GoalDraw a scatter graph from paired data, name the type of correlation and explain what it means and does not mean.
New words
scatter graph · a graph of paired data in which each pair of values is plotted as one pointcorrelation · a relationship between two sets of datapositive and negative correlation · positive: as one value increases the other tends to increase; negative: as one increases the other tends to decrease
Explanation
A scatter graph shows two measurements for each person or object, for example height and arm span. Each pair gives one point. Look at the pattern the points make. If they run from bottom left to top right, there is positive correlation: when one value is larger, the other tends to be larger too. If they run from top left to bottom right, there is negative correlation. If the points are spread with no pattern, there is no correlation. The closer the points are to a straight line, the stronger the correlation. Correlation only tells you that two things go together. It does not prove that one causes the other, because both may depend on something else.
Type
What the points do
Example
Positive
rise from left to right
height and arm span
Negative
fall from left to right
rain and visitors to a park
None
no pattern
shoe size and a quiz score
Worked examples
Heights and arm spans (cm) of ten students are plotted, with height across and arm span up. The points are (142, 140), (148, 149), (151, 150), (155, 158), (158, 157), (160, 163), (163, 162), (167, 169), (170, 172) and (174, 175). They lie close to a straight line rising from left to right. There is strong positive correlation: taller students tend to have a larger arm span.
A park manager records the rainfall (mm) and the number of visitors on eight days: rain 0, 2, 3, 5, 8, 10, 12, 15 and visitors 220, 200, 205, 160, 130, 110, 95, 60. The points fall from left to right and are close to a line. There is strong negative correlation: on rainier days fewer people visit.
Twelve students give their shoe size and their score on a quiz out of 20. The points are scattered over the whole graph with no pattern, so there is no correlation: shoe size tells us nothing about the quiz score. Now a warning. Sales of woollen scarves and of hot chocolate both rise together in winter, so the scatter graph would show positive correlation. Buying scarves does not make people buy hot chocolate: both depend on the cold weather.
Class activity
Hand and foot. Collect the hand span (tip of thumb to tip of little finger, hand stretched) and the foot length of at least 12 people, in centimetres. Write the pairs in a table, plot them on a scatter graph with hand span across, and decide what kind of correlation you see. Keep your data for the next lesson.
Practice
1
Cottages by the sea: distance from the beach (km) and price per night ($): (0.2, 140), (0.5, 150), (1, 120), (1.5, 115), (2.5, 95), (3, 100), (4.5, 80), (6, 70). Draw a scatter graph (distance across, 0 to 7; price up, 60 to 160). Describe the correlation.
The points go down from left to right and stay fairly close to a line. There is negative correlation: the further from the beach, the cheaper the cottage.
2
Say whether you expect positive, negative or no correlation, and why: (a) a child’s age (3 to 12) and shoe size; (b) outside temperature and the amount of home heating used; (c) the number of letters in a name and the height of the person; (d) the height of a mountain village above sea level and its average temperature.
(a) positive: older children have bigger feet. (b) negative: the colder it is, the more heating. (c) none: no link. (d) negative: air is colder at greater height.
3
The scatter graph shows the age of apple trees and the mass of apples picked in a season. What type of correlation is it, and how strong? Does it show that every older tree gives more apples? Explain.
Positive correlation, fairly strong: the points are around a rising line but with some scatter. It does not show that every older tree gives more. For example the 4-year-old tree gave 15 kg but the 3-year-old tree gave 25 kg. It shows only a general tendency.
4
The number of visitors to a swimming pool and the number of bottles of sun cream sold in a town are positively correlated. Does swimming cause sun cream sales? Explain, and suggest a reason for the link.
No, one does not cause the other. Both rise on hot, sunny days: more people swim and more people buy sun cream. The weather is the hidden cause.