3.

1 The Arithmetic Mean

3.1.1 The Arithmetic Mean of Ungrouped Data

We use average many times to mean ~he arithmetic mean. We compute arithmetic
mean for both ungrouped and grouped data. We also compute arithmetic mean, which
we henceforth in this unit call mean, for both the sample and the population from
where the sample is drawn. Mean computed for the sample is called a statistic and it is
donated by x. The mean computed for the population is called a parameter and is
denoted by u You should note in this course that any measure computed for the
sample is called statistic and any measure computed for the population is called
parameter.

The mean of ungrouped data is the summation of the values in the set of data divided
by the number of values in the set of data.

For a sample the number of values is denoted by n, that is the sample size, and for the
population the population size is given as N.
-
Mean of ungrouped data (if a sample) is x
n
∑×i
Where x = i=1
nn
n
∑ = Summation

×i = Different values of a
variable n = Sample size

Example 5.1
Compute the mean for the following:

Scores:7, 5, 8, 10, 11, 6, 3 ,

4, 10, 9,13, 2.

Mean= x = 7+5+8+10+11+6+3+4+10+9+13+2
12

= 88 = 7.33
 12

You will realize that there is no frequency distribution for this example.
Suppose there is a frequency distribution for the values of a variable, then
how do we calculate the mean?

This is simple. If we are computing the mean for a sample, that is x, the mean
_
x = ∑fixi or ∑fixi
∑fi n

fi = different frequencies of values in the set of

data xi = different values in the set of data

∑ fi or n is equal to the number of values in the set of data.

Example 5.2

The table below gives the frequency distribution of the mark scored by 20
students in a test conducted in Statistics for Management

11
Scores 2
Frequency
12 3
13 5
14 7
15 2
16 1

You are required to calculate the mean of the scores

To solve the problem, you should

(i) Multiply each score by its respective frequency to obtain fixi
(ii) Add up the products of each score and its respective frequency to
obtain ∑fixi
(iii) Add up all the frequencies to obtain ∑fi or n.
 (iv) Divide ∑fixi by ∑fi to obtain the mean.

Scores (xi) Frequency (Fi) Fixi

11 2 22
12 3 36
12 5 65
14 7 98
15 2 30
16 1 16

Total 20 276

Mean = x =∑ n ∑fixi = 276 = 13.35

i = 1 ∑fi 20

There can be another method of computing the mean apart from

using

3.1.2 The Arithmetic Mean of Grouped Data

You have learned how to compute the arithmetic mean of ungrouped data.
When the values are many in a set of data, there is the need to group them
into class intervals. You learned about this in unit 3. We need to take some
time to compute the mean of grouped data.

Example 5.4
The earning per share (in kobo) of some firms is presented below with the
frequency distribution.

Earning per share (in kobo) Frequency

65-69 3
70-74 4
75-79 11
80-84 15
 85-89 9
90-94 5
95-99 3

You are required to calculate the mean of the distribution.

To solve this question, we need to compute the class mark for the class
intervals. The class mark becomes the Xi we will use in the computation.
Immediately this is done, the whole distribution is reduced to the form of an
ungrouped data with frequency distribution. You should recall that class mark
is the mean of the upper and lower class boundaries (or Limits) of a class
interval.

Class Mark I Frequency ∑XIfI

Xi fi Fixi
67 3 201
72 4 288
77 11 847
82 15 1230
87 9 783
92 5 460
97 _3 291
TOTAL 50 4100

Mean x = 4100 = 82
50

The average earning per share of the firms is 82 kobo.

From the computation, you will still realize that the mean of a grouped data
is given as

∑ fi×i
∑fi
the xi here are the various values of the class marks of the class intervals.

Example 5.1

For the distribution below, compute the mean, using the two methods we
have used in this unit.
Class Intervals Frequency
0-9 4
 10-19 6
20-29 18
30-39 11
40-49 6
50-59 5

3.1 The Weighted Mean

Some times we are interested in showing the relative importance of some
items. When we do, we attach weights, apart from the real values of the
items. For instance in the Open University, some courses in a programme are
core courses while the others are auxiliary courses. Core courses tend to have
more credit unit than the others.
When we attach weighs to items we can only compute the weighted mean.
_ ∑wi×i
The weighted mean is given as: ×w
= ∑wi

Where xi = values of the items

Wi = relative weights of the
item Xw = Weighted mean
Example

The percentage scores of a student in some courses and the credit units of the courses are given
below
 Courses % Score Credit Unit
HEM 101 52 3.0
HEM 102 64 3.0
HEM 103 71 3.0
HEM 104 45 3.0
GNS 101 82 2.0
GNS 126 55 1.0

Calculate the weighted percentage mean for the scores. xi = Percentage

Scores

Xi = Percentage Scores
wi = Credit Units

Xi wi Wix
i
52 3.0 156
64 3.0 192
71 2.0 142
45 3.0 135
82 2.0 164
55 1.0 55
14 884

The weighted percentage mean = 884 = 60.29

14

Exercise 5.2

The scores of a student in a university in the courses he took in the first

semester of the year and the units of the courses are presented as follows:

Courses % Scores Unit

CSC 101 75 3.0

CSC 102 70 3.0

CSC103 42 2.0

CSC104 45 3.0
GNS 101 52 2.0
GNS 128 63 1.0
 The University is interested in computing weighted grade point average. The
academic standard of the University shows the following scores, grades and
grade points

Score Grades Grade Point

Above 75 A 4.0
70-74 AB 3.5
65-69 B 3.25
60-64 BC 3.0
55-59 C 2.75
50-54 CD 2.5
45-49 DE 2.25
4044 F 2.0
below 40 0.0

With this information, compute the weighted grade point average for the
student to two decimal places.

3.2 Advantages And Disadvantages Of Arithmetic Mean

3.2.1 Advantages of Arithmetic Mean.

The arithmetic mean has the following advantages

(i) Mean is the best known of all the averages

(ii) Mean can be used for further mathematical process. Mean is used to
perform statistical procedures such as estimation and hypothesis
testing.
(iii) Mean is unique, unlike mode (this will be discussed later], because a
set of data has one and only one mean.

3.2.2 The disadvantages of mean

Arithmetic mean has the following disadvantages.

(i) Since all the values in a set of data are used to compute the mean, the mean
can be influenced by extreme values

For instance the mean of 3, 4, 5, 6 and 7 is 3+4+ 5+6+7 = 5

5
There is no extreme value here. Suppose we have 3, 4, 5, 6, 7, 19, we then
have an extreme value, 19. The mean becomes
3 + 4 +5 + 6 +7 +19 = 44 = 7.33
5 5

The extreme value has greatly influenced the mean.

(ii) A mean may result into an impossible value where the data is discrete.
For instance, If the number of female students in five programmes at
the National Open University are 35, 38, 42, 53, and 66, the mean
value of the female students will be

35+38+42+53+66 = 234 = 46.8 Students

5 5

This is an impossible value.

(iii) We are unable to compute mean for data in which there are open-
ended classes either at the beginning of the distribution or at the end
of the distribution. It will be difficult to know the class mark of the
open-ended class.
3.1 MEDIAN

Computation Of The median is a measure that shows the most central item
in a set of data. It is a single value computed from the set of data that
measures the central item in the data. If median is the most central item in the
set of data, half of the values in the set must lie below the median, and the
other half above the median

To calculate the median, there is a need to have a mathematical definition of

the median

( )
th
n+1
Median = the 2 item in the set of data. Where n is the number of
values in the set of data.

For an ungrouped data with odd number of values, the item in the middle
constitutes the median.

Example 7.1

What is the median of 17, 12, 13, 15, 18? These values are not arranged in
any order. To find the median, we need to order the arrangement of the
values. as 12, 13, 15, 17, 18. The value in the middle is 15 and that is the
median.

Using our formula for median

Median = the ( )
n+1
2 item

th

th
5+1 2
( )
 item

= 3rd item
The third item is 15 )th item

Suppose we have an even number of values

such as 8, 4, 3, 5,9,2,11,7., what will be the
median?. We need to order the arrangement of
the values as 2,3,4,5,7,8,9,11

(
n+1
There are 8 values. The median is therefore the 2

( ) 8+1
2
th
item

= 4.5 item

The 4.5th item an only be between 5 and 7. Therefore the median will be
equal to

5+7
2 ) th

=6

Suppose we have ungrouped data with frequency distribution, how do we

compute the median?
We still compute the median the same way we did the two previous examples

Example 7.2
For the distribution below, compute the median

Scores Frequency
2 3
3 4
4 6
5 7
 6 5
7 3
8 2

The sum of the frequencies here is 30. There are people examined. The

median should now be

the ( )
th
320 + 1 item

= 15.5th
5+5 = 5
The median (the 15.5th value) i s
2

To obtain the median for grouped data, we still do the same thing we did
previously.
However, we can decide to adopt the use of formula for the median.
Some authors use the formula shown below

Median = L1 + J x w
fi
Where L1 = Lower class boundary of the median
class fi = frequency of class that contain the median
w= class width of class containing the median
J = n minus sum of all the frequencies up to, but not including the
medium 2
class.
n = sum of all the frequencies in the set of values ∑fi

Some other authors uses the formula below for median

Sample median = 30 + 1 - (f + 1)
2 w + Lm
fm

Where m = total number of items in the distribution

f = sum of all the class frequencies up to, but not including, the median class
fm = frequency of median class
w = class width
 Lm = lower limit of median class interval for the purpose of this course, we
will continue to use

Median = L1 + J x w
fl

Example 7.3

For the distribution below, compute the median

Class interval Frequency

0-499 229
500-999 306
1000-1499 398
1500-1999 298
2000-2499 127
2500-2999 66
3000-more 26

1450 + 1
The sum of the frequency here is 1450 the median is the th
2 item

The median class is then 1000 -1499

Since the 725.5th item lies in the class. From the first two classes we obtain
(229 + 306) items = 535 items.
In the third class containing 398 items, we can get 190.5 items to make up
for the 725.5 items we are looking for.
Since the median class is 1000 -1499, the lower class boundary = 1000,the fi
= frequency of the median class = 398.
The class boundaries for the median class interval are 1000 - 1500. The class
width is equal to 500

J = 1450 – (229 + 306) = 190

2
Median = 1000 + 190 x 500 = 1238.69
398

3.1.1 Estimation Of Median Using Ogive

 We can also estimate the median by plotting ogive for the set of data.

Example 7.4

Plot an ogive for the table below and estimate the median from the ogive
Class I fi
1-1.99 2
2-2.99 5
3-3.99 6
4.499 7
5.599 5
6-6.99 3
7.799 2

We will need to obtain the cumulative frequency distribution to plot the ogive.

Upper Class Cumulative frequency

1.00 0
2.010.00 2
3.00 7
4.00 13
5.00 20
6.00 25
7.00 28
8.00 30

To obtain the ogive, we plot the cumulative frequency against the upper class
boundary as we discussed previously
 ()
n th
The median is taken 2 the median is taken as 15th value
th
as n
Locate the 15.0 value along here. axis and draw it to meet the curve. At
2 the vertical
.
the point it meets the curve, draw a vertical line to meet the horizontal at the
point the line meets the horizontal axis is the median. The estimate of the
median here is 4.5 (from ogive)

3.13 Advantages and Disadvantages of median

The median cannot be influenced by extreme values as in the case of the

means since not all the values are involved in the calculation of the median.
(ii) The median can be calculated from incomplete records and for open
ended classes if the median does not fall into the open ended class.
(iii) It is not difficult to recognize the

median Median also has a number of

disadvantages

(1) Median cannot be subjected to further statistical processing as it is

with mean.
(2) To compute median we need to order the values. This can consume
time.

Exercise 7.1
 For the table below,

i. Compute the median

ii. Draw the ogive and estimate the median

Class Interval Frequency

21-30 2
31-40 6
41-50 9
51-60 9
61-70 11
71-80 6
81-90 4
91-100 3

3.2 Mode

3.2.1 Computation of Mode from Ungrouped and Grouped Data

Mode is another measure of central tendency. It is simply defined as the

value with the highest frequency in a set of values. Unlike mean and median,
the mode can have more than one value in a set of data.

Example 7.5
From the distribution below, find the mode

X1 (ages in years I Frequency

5 5
6 7
7 8
8 11
10 3
11 2
12 1
The mode is 8 because it has the highest frequency of 11. Suppose we replace
the frequency of those in age of 10 years by 11, the distribution will have two
modes, that is 8 and 10 respectively.

We can also calculate the mode from grouped data. To do this, we will use
the formula.

Mode = Ll + dixw where Li = Lower class boundary of the modal class

di + d 2
d1=frequency of the modal class minus frequency of the class just before the
modal class

d2 = frequency of the modal class minus frequency of the class just above the
modal class

W = Class width

Example 7.6 For the distribution below computer the mode.

Class I Frequency

22-26.9 5
27-31.9 8
32-36.9 14
37-41.9 23
42-46.9 19
47-51.9 7
52-56.9 4

The modal class in this question is 37-41.9.

L1= lower class boundary = 37