Measure of

Central
tendency

Dr. Hina Dutt

[email protected]
SEECS-NUST
Measures of Central Tendency
The tendency of observations to
cluster in the central part of the data
Mean is called Central Tendency. The value
of data that summarizes the central
Median tendency or locate in the middle of
Mode the data (in some sense) is called
Measure of Central Tendency. It is also
known as a Measure of Location or
Position.
Measures of Central Tendency

Mean • The average observation

• The value that lies in the

Median middle of the data

• The most frequently

Mode occurring observation
Mean
Mean

The arithmetic mean or simply the mean is defined as the value obtained by
adding up all the observations and dividing by the total number of observations.
𝑛
𝑥1 + 𝑥2 + ⋯ + 𝑥𝑛 𝑥𝑖
𝑥ҧ = =෍
𝑛 𝑛
𝑖=1
Mean; Example 1 (Ungrouped Data)
Mean as a Balancing Point

Mean
Weighted Mean
The multipliers or a set of numbers which express more or less adequately
the relative importance of various observations in a set of data are
technically called the weights.
We assign weights 𝑤1 , 𝑤2 , … 𝑤𝑛 to the observations in a set of data
according to their relative importance, when the observations are not of
equal importance.
The weighted mean, denoted by 𝑥ҧ𝑛 , of a set of 𝑛 values 𝑥1 , 𝑥2 , … 𝑥𝑛 with
corresponding weights 𝑤1 , 𝑤2 , … 𝑤𝑛 is then defined as

𝑥1 𝑤1 + 𝑥2 𝑤2 + ⋯ +𝑥𝑛 𝑤𝑛 σ 𝑥𝑖 𝑤𝑖
𝑥ҧ𝑛 = =
𝑤1 + 𝑤2 + ⋯ + 𝑤𝑛 σ 𝑤𝑖
Weighted Mean; Example 2

Calculate the weighted mean of expenditure from the following data.

Item Expenditure (Rs) Weights

Food 290 7.5
Rent 54 2.0
Clothing 98 1.5
Fuel and Light 75 1.0
Other items 75 0.5
Weighted Mean; Example 2
Item Expenditure 𝒙𝒊 Weights 𝒘𝒊 𝒙𝒊 𝒘𝒊
Food 290 7.5 2175
Rent 54 2.0 108
Clothing 98 1.5 147
Fuel and Light 75 1.0 75
Other items 75 0.5 37.5
Total −− 12.5 2542.5

σ 𝑥𝑖 𝑤𝑖 2542.5
𝑥ҧ = = = 𝑅𝑠. 203.4
σ 𝑤𝑖 12.5
Combined Mean
If 𝑘 subgroups of data consisting of 𝑛1 , 𝑛2 , 𝑛3 ….. 𝑛𝑘 , 𝑛 = σ 𝑛𝑖
observations has respective mean 𝑥ҧ1 ,ഥ𝑥2 ,ഥ𝑥3 ,…. 𝑥ҧ𝑘 , then 𝑥ҧ𝑐 , the mean
for all the data, is given by

𝑛1 𝑥ҧ1 + 𝑛2 𝑥ҧ2 + ⋯ + 𝑛𝑘 𝑥ҧ𝑘

𝑥ҧ𝑐 =
𝑛1 + 𝑛2 + ⋯ + 𝑛𝑘
σ 𝑛𝑖 𝑥ҧ 𝑖
𝑥ҧ𝑐 = (𝑖 = 1, 2, 3, … … … . , 𝑘)
𝑛
Combined Mean; Example 3
The means scores and the number of students in three sections of a statistics
class are given in the table.
Section No. of students 𝒏𝒌 Mean score 𝒙𝒌
A 40 64.1
B 40 64.8
C 43 63

𝑛1 𝑥ҧ1 +𝑛2 𝑥ҧ2 +𝑛3 𝑥ҧ3 7865

𝑥ҧ𝑐 = = = 63.9 marks
𝑛1 +𝑛2 +𝑛3 123
Mean (Grouped Data)
For grouped data, firstly we convert information of class limits in the form of
mid points. We assumed that the observation in each class is identical with
class midpoints. The formula for mean is given below
σ 𝑥𝑖 𝑓𝑖 σ 𝑥𝑖 𝑓𝑖
𝑥ҧ = =
σ 𝑓𝑖 𝑛
where
𝑓 = frequency in each class
𝑥𝑖 = class mark of each class
𝑛 = total number of observation (σ 𝑓)
Mean; Example 4
Twenty management students, in preparation for graduation, took a course to
prepare them for a management aptitude test. A simulated test provided the
following scores: 77 89 84 83 80 80 83 82 85 92 87 88 87 86 99 93 79 83 81 78.
Calculate the mean score directly from the data and from the data grouped into a
frequency distribution.

Solution:
Calculations based on ungrouped data:

σ𝑛𝑖=1 𝑥𝑖 1696
𝑥ҧ = = = 84.8
𝑛 20
Mean; Example 4
Calculations based on grouped data:

Class Boundaries Frequency Class Marks 𝒙𝒊 𝒇𝒊 𝒙𝒊

75 − 80 3 77.5 232.5
80 − 85 8 82.5 660
85 − 90 6 87.5 525
90 − 95 2 92.5 185
95 − 100 1 97.5 97.5
Total 20 −− − 1700

σ 𝑥𝑖 𝑓𝑖 σ 𝑥𝑖 𝑓𝑖 1700
𝑥ҧ = = = = 85
σ 𝑓𝑖 𝑛 20
Mean (Change of origin and scale)
To reduce the computational labor and to save time, a change of
origin and scale can be made. If 𝑥𝑖 denotes the mid points, ‘𝑎’ an
arbitrary origin (usually chosen the class mid point) and ℎ denotes the
class interval.

Let 𝑥𝑖 = 𝑎 + ℎ𝑢𝑖 ,

Then 𝑥ҧ = 𝑎 + ℎ𝑢ത ,

σ 𝑓𝑖 𝑢𝑖
where 𝑢ത =
𝑛
Mean (Change of origin and scale); Example 5
Given the following frequency distribution of weights of 60 apples. Calculate the
mean weight.

Weight 65-84 85-104 105-124 125-144 145-164 165-184 185-204

(g)
f 9 10 17 10 5 4 5
Mean (Change of origin and scale); Example 5
Weight 𝒇𝒊 Midpoints 𝒙𝒊 𝒖𝒊 𝒇𝒊 𝒖𝒊
65-84 9 74.5 -3 -27
85-104 10 94.5 -2 -20
105-124 17 114.5 -1 -17
125-144 10 134.5→ 𝑎 0 0
145-164 5 154.5 1 5
165-184 4 174.5 2 8
185-204 5 194.5 3 15
Total 60 --- --- -36
σ 𝑓𝑖 𝑢𝑖 36
𝑢ത = σ 𝑓𝑖
=− = −0.6
60
𝑥ҧ = 𝑎 + ℎ𝑢ത = 134.5 + 20 −0.6 = 122.5 grams
Median
Median
Median is defined as a value which divides an ordered data into
two equal parts, one part comprising of observations greater than
and the other part smaller than it.
How to find Median (Individual Observation)
Rank the items.

If the number of items is odd, the median is the middle item in

𝑛+1
the list i.e. 2 𝑡ℎ value

If the number of items is even, the median is the mean of the

𝑛 𝑛
two middle numbers i.e. average of 2 𝑡ℎ and 2 + 1 𝑡ℎ values
Median; Example 6 (Individual Observation)
The height (in inches) of 10 players of a team are:
65, 67, 62, 63 67, 64, 63 68, 66, 61.
Calculate the median height.

The values are arranged in the ascending order as:

61 62 63 63 64 65 66 67 67 68

Since the number of observations = 10 (even number)

So median is the mean of two middle values i.e., 64 and 65.
Therefore median = 64.5
Median; Example 7 (Ungrouped frequency
Distribution)
The following distribution relates to the number of assistants in 50 retail
establishments.
No. of 0 1 2 3 4 5 6 7 8 9
assistants
f 3 4 6 7 10 6 5 5 3 1

Calculate the median number of assistants.

Median; Example 7 (Ungrouped frequency
Distribution)
Median class
No. of
0 1 2 3 4 5 6 7 8 9
assistants
f 3 4 6 7 10 6 5 5 3 1
Cumulative
3 7 13 20 30 36 41 46 49 50
frequency

Since 50 is an even number, therefore median is the average number of assistant in the two
middle classes, i.e. 25𝑡ℎ and 26𝑡ℎ retail establishments. Looking at the cumulative frequency
row, we find that these two values corresponding to the same value of 𝑥, i.e. 4.
Hence, median number of assistants = 4.
How to Find Median from an Ogive
Median (Grouped Data)
ℎ 𝑛
𝑚𝑒𝑑𝑖𝑎𝑛 = 𝑙 + −𝑐
𝑓 2
𝑤ℎ𝑒𝑟𝑒 𝑛 = ෍ 𝑓

𝑙 = lower class boundary of the median class

ℎ =class interval
𝑓 =frequency of the median class
𝑐 =cumulative frequency corresponding to the group preceding
median class
Median; Example 8 (Grouped Data)

Calculate mean for the distribution of examination marks given below:

Marks 30-39 40-49 50-59 60-69 70-79 80-89 90-99

No. of 8 87 190 304 211 85 20
students
Median; Example 8 (Grouped Data)
Class Limits Class Boundaries Frequency 𝒇 Cumulative
frequency 𝑭
30-39 29.5-39.5 8 8
40-49 39.5-49.5 87 95
50-59 49.5-59.5 190 285
60-69 59.5-69.5 304 589 Median Class
70-79 69.5-79.5 211 800
80-89 79.5-89.5 85 885
90-99 89.5-99.5 20 905

𝑛 905
Median = Marks obtained by 2 𝑡ℎ student i.e. 2
= 452.5𝑡ℎ student
ℎ 𝑛 10
Median = 𝑙 + 𝑓 − 𝐶 = 59.5 + 304 452.5 − 285 = 65 marks
2
Mode
Mode

Mode is defined as the value

which occurs most frequently in
a set of data, that is it indicates
the most common result.
Mode can not be Unique

No Mode • {1,0,5,9,12,8}

Unique • {4,5,5,5,9,20,30}, Mode=5

Mode
More than • {2,2,5,9,9,15}, Mode=2
one Mode and 9
Mode (Ungrouped Frequency Distribution)

In an ungrouped frequency distribution

with classes consisting of single values, the
mode can be immediately located by
examining the distribution. In example 7,
mode is 4, as the frequency for 𝑥 = 4 is
greater than for any other value of 𝑥.
How to Find Mode from Histogram
Mode (Grouped Data)
In a grouped frequency distribution, the mode would lie in the class that carries
the highest frequency. This class is called modal class.

𝑓𝑚 − 𝑓1
𝑚𝑜𝑑𝑒 = 𝑙 + ℎ
𝑓𝑚 − 𝑓1 + 𝑓𝑚 − 𝑓2
where
𝑙 = lower class boundary of the modal class
ℎ =class interval
𝑓𝑚 =frequency of the modal class
𝑓1 = frequency corresponding to the group preceding modal class
𝑓2 = frequency corresponding to the group following modal class
Mode; Example 9 (Grouped Data)
Calculate the mode for the distribution of examination marks in example 8.
Class Limits Class Boundaries Frequency 𝒇
30-39 29.5-39.5 8
40-49 39.5-49.5 87
50-59 49.5-59.5 190
60-69 59.5-69.5 304 Modal class
70-79 69.5-79.5 211
80-89 79.5-89.5 85
90-99 89.5-99.5 20

𝑓𝑚 − 𝑓1
𝑚𝑜𝑑𝑒 = 𝑙 + ℎ
𝑓𝑚 − 𝑓1 + 𝑓𝑚 − 𝑓2
(304 − 190)
= 59.5 + × 10 = 65 marks
304 − 190 + (304 − 211)
 MEAN MEDIAN MODE
• most stable measure • affected by the position • Not based on all
• based on all of the items but not the observations
observations of the values • unaffected by the
data • unaffected by the extreme values
• affected by the extreme extreme values • exist as a data point in
values • may not exist as a data the set
• may not exist as data point in the set • can be calculated for
point in the set • can be calculated for the qualitative data
• calculated only for the the ranked qualitative • may have more than
quantitative data data one value
• not an appropriate • appropriate measure of • May not exist
measure of average for average for a highly
a highly skewed data skewed data
Reference Book: Probability and Statistics for
Engineers, 9th edition by Richard A. Johnson

Practice Questions:
Q # 2.32, 2.34(a), 2.39(a,b), 2.44(a), 2.45(a), 2.53, 2.54,
2.56, 2.57