6.

2 Cumulative Frequency
In this section we describe how to find the median and quartiles by
using cumulative frequency which is a new way to represent the data.
Cumulative frequency is a count of the total number of data items up to
a certain value.

Example 6.
The following table shows the number of sweets in a pod, for a sample
of 100 pods, taken at random. Construct a cumulative frequency table.
Number of sweets per pod 2 3 4 5 6 7 8
Number of pods 5 10 10 14 19 27 15

Solution.
The cumulative frequency table is obtained as follows:
Number of sweets per pod () 2 3 4 5 6 7 8
Cumulative frequency 5 15 25 39 58 85 100
Example 7.
The frequency table shows the marks of 84 students in a Mathematics
test.
Mark () (0, 20) (20, 40) (40, 60) (60, 80) (80, 100)
Frequency 12 20 34 14 4

(a) Construct a cumulative frequency table and draw a cumulative

frequency curve for the Mathematics marks.
(b) Find the median and interquartile range.
Solution.
The cumulative frequency table is obtained as follows:

Mark ()
Cumulative frequency 12 32 66 80 84
Here the first row of the cumulative frequency table consists of the
upper boundaries of the data groups.

We notice that nobody got mark 0, so (0, 0) is the leftmost point in the
diagram. Therefore we plot the five points (20, 12), (40, 32), …,
(100, 84) from the cumulative frequency table plus the additional point
and join them up with a smooth curve.
(b) The total frequency is 84, so draw lines across from the - axis at
for the lower quartile,
for the median and
for the upper quartile.
Therefore, we estimate the values and on the - axis from the graph.
So,
lower quartile , ( means approximately equal)
the median
upper quartile
the interquartile range .

Example 8.
The table shows the mass, in kilograms, of 65 workers of a factory.
Mass ()
Number of workers 10 20 20 10 5
(a) Construct a cumulative frequency table and draw a cumulative
frequency curve from the given information.
(b) Find the median and the interquartile range.
(c) How many workers have mass 58 kg or less?
(d) How many workers have mass of more than 58 kg?
Solution.
(a) Mass ()
Cumulative frequency 10 30 50 60 65
(b) The total frequency is 65, so draw lines across from the -axis at
for the lower quartile,
for the median,
for the upper quartile.
Therefore, we estimate the values on the - axis on the graph. Thus,
lower quartile
the median
upper quartile
the interquartile range .

(c) Number of workers whose mass less than 58 kg = 22.

(d) Number of workers whose mass greater than 58 kg .

6.3 Correlation
In this section we only explain the linear relation between independent
and dependent variables. Therefore we collect two variables to see
whether a relationship exists between them. This can be illustrate by a
scatter diagram.
A scatter diagram is a graph of the ordered pairs of numbers consisting
of the independent variable and the dependent variable .

A linear correlation is the statistical method used to determine a

relationship between dependent and independent variables.
 A positive correlation is a relationship between two variables
when both variables increase or decrease at the same time. A negative
correlation is a relationship between two variables when one variable
increases, the other variable decreases, and vice versa.

Example 9.
Ten students sat a Mathematics test and a Chemistry test. Both tests
were marked put of 20. Their marks are shown below. Draw a scatter
diagram to represent these data.

Students A B C D E F G H I J
Mathematics mark 5 9 9 13 20 8 12 16 11 10
Chemistry mark 5 8 11 14 20 11 10 16 13 8

Solution.
From the diagram we see that the pints lie approximately on a straight
line, the higher the Mathematics mark the higher the Chemistry mark. It
shows positive linear relationship.
Correlation coefficient
A correlation coefficient is used to determine the strength of the
linear relationship between two variables. The value of varies between
and .
Formula for the Correlation Coefficient

Value of Interpretation
Strong positive linear correlation
No linear correlation
Strong negative linear correlation

Note. The value does not mean that there is no relationship between
the two variables – only that there is no linear relationship.
Example 10.
The scatter diagrams show various degrees of correlation.

Match the diagrams with the correlation coefficients below.

(i) (ii) (iii) (iv)
Solution.
Figure (a) is closed to strong positive linear correlation. So figure (a) is
matched with (iv).
Figure (b) is closed to strong negative linear correlation. So figure (b) is
matched with (ii).
Figure (c) is matched with (iii)
Figure (d) is matched with (i)
Example 11.
Calculate the correlation coefficient of the following summary data:
,,,,,.
Solution.
.
.
.
Therefore, . We have a perfect positive correlation.

Example 12.
Calculate the correlation coefficient of the following summary data:
,,,,,.
Solution.
.
.
.
Therefore, . We have a negative correlation.
Linear Regression
If the value of correlation coefficient is significant, we can find the
equation of the regression line, also known as line of best fit. It is a
statistical method used to determine the strength and character of the
relationship between two variables. The purpose of regression line is tio
make predictions on the basis of the data.
 For each point on a scatter diagram, we can express in terms of
as , where is the vertical distance from the regression line. If the points
are close to the line, (i.e., close to ), we get the better prediction on the
data.
The equation of the regression line is

where

and
Example 13.
The results from an experiment in which different masses were placed on a
spring and the resulting length of the spring measured, are shown below.
Find the equation of the regression line of the data.
Mass ( kg) 20 40 60 80 100
Length ( cm) 40 50 60 60 70
Solution.
, , , .
We compute that
.
.
Therefore, we obtain

and
The regression line is .

Students can get the regression line and the relative correlation coefficient
by using a calculator with inputting the given data into the calculator.