IB Mathematics: Applications and Interpretation · Topic 4 · Statistics and probability
4.4 Correlation and regression
Plot a scatter diagram, describe the strength and direction of correlation, and use a regression line to interpret and predict values.
1/5A scatter diagram shows how two variables are related
- A scatter diagram shows how two variables are related
- Correlation describes the direction and strength of a pattern
- Pearson’s r gives a number for the strength of linear correlation
- A regression line models the trend between two variables
- Predictions are reliable only inside the range of the data
- Check your recall
- Practice: 10 questions
Part 1 of 5: A scatter diagram shows how two variables are related
Suppose you record the hours a student revises and the score in a test. Each student gives a pair of values. Data in pairs like this is bivariate data.
To display it, draw a scatter diagram. Put the independent variable, the one you choose or control, on the horizontal axis. Put the dependent variable, the one that responds, on the vertical axis. Plot one point for each pair.
Always plot the data before calculating anything. The diagram shows whether a straight line is a sensible model.
Key terms · tap one for its definition
bivariate data
Bivariate data is a set of paired values, with two variables recorded for each individual or item.
scatter diagram
A scatter diagram is a graph that plots each pair of values as one point, with one variable on each axis.
Part 2 of 5: Correlation describes the direction and strength of a pattern
Now look at the overall pattern of the points. With positive correlation, the points slope upwards from left to right. With negative correlation, they slope downwards.
Strength depends on how closely the points follow a straight line. Tight clusters show strong correlation, and loose scatter shows weak correlation. If there is no pattern, there is no correlation. Correlation here always means a straight-line pattern.
Sometimes one point lies far from the rest. This is an outlier. Check whether it is an error before deciding to keep it.
Key terms · tap one for its definition
positive correlation
Positive correlation is a tendency for to increase as increases, giving .
negative correlation
Negative correlation is a tendency for to decrease as increases, giving .
outlier
An outlier is a data point that lies unusually far from the pattern formed by the other points.

Quick check
One question on the part you just read
Easy
Show answer and explanation
Answer: A · Worked explanation
Part 3 of 5: Pearson’s r gives a number for the strength of linear correlation
Words such as ‘strong’ are vague, so we also use a number. Pearson’s product-moment correlation coefficient, , measures linear correlation. Your calculator finds it from the data. A perfect correlation puts every point exactly on a straight line.
The value of always lies between -1 and . The sign gives the direction. The size, ignoring the sign, gives the strength, and values near either end show strong correlation.
A rough guide to interpreting r
- Value of $r$
- Description
- Perfect positive linear correlation
- Value of $r$
- near
- Description
- Strong positive correlation
- Value of $r$
- near
- Description
- Moderate positive correlation
- Value of $r$
- near 0
- Description
- Little or no linear correlation
- Value of $r$
- near -0.9
- Description
- Strong negative correlation
- Value of $r$
- Description
- Perfect negative linear correlation
Key term
Pearson’s product-moment correlation coefficient
Pearson’s product-moment correlation coefficient, , is a number from -1 to measuring the strength and direction of linear correlation.
Part 4 of 5: A regression line models the trend between two variables
When points lie near a straight line, a line of best fit can model the trend. The regression line of on is the least squares line. It makes the sum of the squared vertical distances from the points as small as possible. Your calculator gives its equation as .
The line always passes through the mean point, .
The gradient is the average change in for each increase of 1 in . The intercept is the predicted value of when . In the revision data, , and . Each extra hour of revision goes with about 4.25 extra marks.
Key terms · tap one for its definition
regression line
The regression line of on is the least squares line , used to predict from a given .
mean point
The mean point is the point whose coordinates are the means of the values and values.

Quick check
One question on the part you just read
Easy
Show answer and explanation
Answer: D · Worked explanation
Part 5 of 5: Predictions are reliable only inside the range of the data
Now use the line to make predictions. Substitute a value of into the equation to find a predicted . Predicting within the range of the data is interpolation. Predicting beyond that range is extrapolation.
Worked example
Predict and judge reliability
- (a) Substitute : .
- Evaluate: . Since 6 lies inside 1 to 8, this is interpolation.
- (b) Substitute : .
- Since 14 lies outside 1 to 8, this is extrapolation. The result exceeds the maximum mark, so it cannot be right.
Answer: (a) About 67.8 marks, a reliable interpolation. (b) The value 101.75 marks is an unreliable extrapolation.
A prediction assumes that the same pattern continues. This is more believable inside the data range than outside it. Use the line only to predict from . The line was built to predict , so it should not be rearranged to estimate .
Key terms · tap one for its definition
interpolation
Interpolation is predicting a value of for an value inside the range of the data.
extrapolation
Extrapolation is predicting a value of for an value outside the range of the data.
Check your recall[14 marks]
6 written questions. Answer each one in your head or on paper, then open the mark scheme.
- 1.Describe[2 marks]
Describe the correlation between two variables when .
Show mark schemeHide mark scheme
Model answer
There is strong negative linear correlation. The second variable tends to decrease as the first increases.
- 1 mark for stating that the correlation is negative
- 1 mark for stating that the correlation is strong
- 2.Distinguish[2 marks]
Distinguish between interpolation and extrapolation.
Show mark schemeHide mark scheme
Model answer
Interpolation predicts for an value inside the range of the data. Extrapolation predicts for an value outside that range, so it is less reliable.
- 1 mark for stating that interpolation uses an x value inside the range of the data
- 1 mark for stating that extrapolation uses an x value outside the range of the data
- 3.Calculate[3 marks]
Calculate the predicted value of when , using the regression line found from data with between 5 and 20. State whether this is interpolation or extrapolation.
Show mark schemeHide mark scheme
Model answer
. Since 12 lies between 5 and 20, this is interpolation.
- 1 mark for substituting into the equation
- 1 mark for obtaining
- 1 mark for stating that this is interpolation because 12 lies inside the data range
- 4.Explain[2 marks]
Explain why a strong correlation between two variables does not prove that one causes the other.
Show mark schemeHide mark scheme
Model answer
Correlation only shows that the variables change together. A third factor may influence both, so the link need not be causal.
- 1 mark for stating that correlation shows only that the variables change together
- 1 mark for stating that a third factor may influence both variables
- 5.Deduce[3 marks]
Deduce what these results tell a café owner. The regression line of hot drinks sold against daily temperature in is , and .
Show mark schemeHide mark scheme
Model answer
The value shows strong negative correlation. Hot drink sales tend to fall as the temperature rises. The gradient -6 means about 6 fewer drinks for each rise.
- 1 mark for stating that shows strong negative correlation
- 1 mark for stating that sales tend to fall as temperature rises
- 1 mark for interpreting the gradient as about 6 fewer drinks per rise
- 6.Sketch[2 marks]
Sketch a scatter diagram that shows strong positive correlation and contains one outlier.
Show mark schemeHide mark scheme
Model answer
Draw labelled axes. Plot points that rise from left to right and lie close to a straight line, with one point clearly far from the pattern.
- 1 mark for drawing labelled axes with points sloping upwards close to a line
- 1 mark for including one point clearly away from the pattern

Practice this topic
10 quick-fire multiple-choice questions, easy to hard. Tap an answer to see if you're right and the worked explanation. (The written questions with mark schemes are under “Check your recall” above.)
0 right0 of 10 answered
Includes the 2 quick checks in the note above.
EasyIdentify[1 mark]
Show answer and explanation
Answer: C · Worked explanation
MediumCalculate[2 marks]
Show answer and explanation
Answer: B · Worked explanation
MediumIdentify[1 mark]
Show answer and explanation
Answer: B · Worked explanation
MediumDetermine[1 mark]
Show answer and explanation
Answer: D · Worked explanation
MediumIdentify[1 mark]
Show answer and explanation
Answer: A · Worked explanation
Hard[2 marks]
Show answer and explanation
Answer: A · Worked explanation
Hard[2 marks]
Show answer and explanation
Answer: C · Worked explanation
HardDetermine[2 marks]
Show answer and explanation
Answer: C · Worked explanation