Measurement

of Central
Tendency
By
Muhammad Zikriya
Mphil Statistics
Introduction
 Usually, when two or more different data sets are to be compared it is necessary
to condense the data, but for comparison, the condensation of the data set into a
frequency distribution and visual presentation is not enough.

 It is then necessary to summarize the data set in a single value. Such a value is
usually somewhere in the center and represents the entire data set hence it is
called the measure of central tendency or averages.

 Since a measure of central tendency (i.e. an average) indicates the location or

the general position of the distribution on the X-axis therefore it is also known
as a measure of location or position.

2
Types of Measures of
Central Tendency Arithmetic Mean or Average 01
02 Geometric Mean

Harmonic Mean 03
04 Median

Quartiles 05
06 Deciles

Percentiles 07
08 Mode
Arithmetic Mean or Average
The mean, also called the arithmetic mean, is the most frequently
used measure of central tendency.
For ungrouped data, the mean is obtained by dividing the sum of all
values by the number of values in the
data set:

𝑆𝑢𝑚 𝑜𝑓 𝑉𝑎𝑙𝑢𝑒𝑠
Mean =
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑉𝑎𝑙𝑢𝑒𝑠

The mean calculated for sample data is denoted by x (read as “x

bar”), and the mean calculated for population data is denoted by μ
(Greek letter mu). We know that the number of values in a data set
is denoted by n for a sample and by N for a population. we learned
that a variable is denoted by x, and the sum of all values of x is
denoted by Σx. Using these notations, we can write the following
formulas for the mean.

4
Arithmetic Mean or Average
Calculating Mean for Ungrouped Data The mean for ungrouped
data is obtained by dividing the sum of all values by the number of
values in the data set. Thus,

Σx
Mean for population data: μ=
𝑁

Σx
Mean for sample data: x=
𝑛

where Σx is the sum of all values, N is the population size, n is the

sample size, μ is the population means, and x is the sample mean.

5
Arithmetic Mean or Average

Example: The variable in this example is the 2008 total sales for a
company. Let us denote
this variable by x.

Σx 149+406+183+107+426+97 1368
x= = = = 228
𝑛 6 6
Thus, the mean 2008 sales of these six companies was 228, or $228
billion.
6
Arithmetic Mean or Average
Calculating Mean for grouped Data We learned that the mean is
obtained by dividing the sum of all values by the
number of values in a data set. However, if the data are given in the
form of a frequency table, we no longer know the values of
individual observations. Consequently, in such cases, we cannot
obtain the sum of individual values. We find an approximation for
the sum of these values using the below procedure. The formulas
used to calculate the mean for grouped data follow.

Σx Σfx
Mean for population data: μ= =
𝑁 Σ𝑓

Σx Σfx
Mean for sample data: x= =
𝑁 Σ𝑓

where Σfx is the sum of all values, Σf (N) is the population size,
Σf(n) is the sample size, μ is the population means, and x is the
sample mean.
7
Arithmetic Mean or Average
Example: T
Calculate Arithmetic Mean for the given data.
x 12 14 16 18 20 22
F 1 4 6 10 7 2

Solution:
x f xf

Σfx
12 1 12
528
14 4 56 x= = = 17.6
Σ𝑓 30
16 6 96
. So the A.M of the given data is 17.6
18 10 180

20 7 140

22 2 44

Total 30 528
8
Arithmetic Mean or Average
Example: T
Calculate Arithmetic Mean for the given data.
Height 60 – 62 63 – 65 66 – 68 69 – 71 72 - 74

F 5 18 42 27 8
Solution:
x f x xf

60 – 62 5 61 305 Σfx
x=
63 – 65 18 64 1152 Σ𝑓
6745
66 – 68 42 67 2814
x =
69 – 71 27 70 1890 100
72 - 74 8 73 584 x = 67.45
. So the A.M of the given data is 67.45
Total 100 6745

9
Arithmetic Mean or Average
Advantages of A.M
i. It is very simple to understand and to calculate.
ii. It is familiar to all and most frequently used.
iii. It considers all observation.
iv. It is capable of further mathematical treatment.
v. It is determined for almost every kind of quantitative data.

Disadvantages of A.M
i. It is greatly affected by extreme values in the data.
ii. It cannot be calculate for qualitative data.

10
Median
Another important measure of central tendency is the median. It is defined as
follows.

Definition:
The median is defined as a value that divides data into two equal
parts.

The calculation of the median consists of the following two steps:

1. Rank the data set in increasing order.
2. Find the middle term. The value of this term is the median.

Computation of Median for individual series are ungrouped data

Case (i): When n is odd

𝑛+1
Median = th
2

11
Median
Computation of Median for individual series are ungrouped data

Example: The marks obtained by 7 students on an achievement test are; 17,

47,15,35,25,39,44. Calculate median marks.

Solution:
Step (1) Arrange the data in ascending order.
x = 15, 17, 25, 35, 39, 44, 47
Here n = 7
𝑛+1
Median = 2 𝑡ℎ

7+1 8
Median = 𝑡ℎ = 𝑡ℎ = 4th value
2 2
Median = 35

12
Median
Computation of Median for individual series are ungrouped data

Case (i): When n is Even

Arrange the data in ascending or descending order and select the middle
two values to add them and divided them by 2.

Example: Find the median for the distribution of scores; 8, 12, 10, 14, 11, and 15.

Solution:
Step (1) Arrange the data in ascending order.
x = 8, 10, 11, 12, 14, 15
Here n = 6
x = 8, 10, 11, 12, 14, 15

11+12 23
Median = =
2 2

Median = 11.5

13
Median
Computation of Median in Discrete Series are Grouped Data

Step (i): Arrange the data in ascending order along with frequencies.

Step (ii): Compute the cumulative frequencies.

Σ𝑓
Step (ii): Median = th value if n = Σf is even.
2

Σ𝑓+1
Median = th value if n = Σf is odd.
2

14
Median
Computation of Median in Discrete Series or Grouped Data
Example:
During the moth of March, a teacher recorded the number of absentees for each
students in a class of n = 20 and obtained the following distribution.
x 2 0 3 5 1 4
f 5 2 7 1 3 2
Solution:
Step (i): Arrange the data in ascending order along with frequencies.

x f cf Step (ii): Compute the cumulative frequencies.

0 2 2
Σ𝑓
1 3 5 Step (ii): Median = th
2
2 5 10 20
Median = th = 10th value.
2
3 7 17
so over median is 2
4 2 19
5 1 20
Σ𝑓 20

15
Median
Computation of Median in Continuous Series or Grouped Data

Step (i): Arrange the data in ascending order along with frequencies.

Step (ii): Compute the cumulative frequencies and Class boundaries.

Step (iii): Use formula of discrete case to locate the median class.

ℎ Σ𝑓
Step (iv): Calculate the median by formula, Median = l + 𝑓 − 𝑐
2

Where
l = Lower class boundaries in which median lies.
h = class interval of the median class.
f = Median class frequency.
c = cumulative frequency of the preceding class.

16
Median
Computation of Median in Continuous Series or Grouped Data
Example: Find out median of the following data.
Wages 20 - 29 30 – 39 40 – 49 50 – 59 60 – 69
f 3 15 20 10 5
Solution:
Since Σ𝑓 = 53 is odd, Therefore,
Wages f c.f c.b Σ𝑓+1 53+1
Median = 2 = 2 𝑡ℎ = 27𝑡ℎ
20 – 29 3 3 19.5 – 29.5
this implies that the median lies in
30 – 39 15 18 c 29.5 – 39.5
the group (39.5 – 49.5)

ℎ Σ𝑓
40 – 49 20 f 38 l 39.5 – 49.5 Median = l + 𝑓 − 𝑐
2
10 53
Median = 39.5 + 20 2 − 18
50 – 59 10 48 49.5 – 59.5

60 – 69 5 53 59.5 – 69.5
Median = 39.5 + 4.25
Median = 43.75
Total 53

17
Median
Advantages of Median
i. It is not affected by extreme values because it is a positional average.
ii. It can be located even if the data are incomplete.
iii. It can be located even for qualitative data.
iv. It is an appropriate average for a highly skewed distribution.

Disadvantages of Median
i. It is not based on all the observations.
ii. Computation of the median needs the arrangement of data into an array, which
is a difficult job, especially for a large body of data.

18
Quartiles, Decile & Percentile
Quartiles
The values which divide a series of ordered data into four equal parts are called quartiles
They are denoted by Q1, Q2, and Q3.

Decile
The values which divide a series of ordered data into ten equal parts are called deciles
They are denoted by D1, D2, D3, D4, D5, D6, D7, D8, D9.

Percentile
The values which divide a series of ordered data into hundred equal parts are called
percentiles. They are denoted by P1, P2,… P99.

These all are positional averages and are used to know the position of an observation
relative to the other values in a data set.
Note that
i. Median = Q2 = D5 = P50
ii. Quartiles, Deciles & Percentiles are collectively called Quantiles.

19
Quartiles, Decile & Percentile
Computation of Quartiles, Decile & Percentile
Case (i) Individual series or ungrouped data

If n is Even If n is Odd
𝑛 𝑛+1
𝑄𝑖 = 𝑖 th item i =1,2,3 𝑄𝑖 = 𝑖 th item i =1,2,3
4 4

𝑛 𝑛+1
𝐷𝑖 = 𝑖 th item i =1,2,3,…,9 𝐷𝑖 = 𝑖 th item i =1,2,3,…,9
10 10

𝑛 𝑛+1
𝑃𝑖 = 𝑖 th item i =1,2,3,…,99 𝑃𝑖 = 𝑖 th item i =1,2,3,…,99
100 100

20
Quartiles, Decile & Percentile
Example: Find Q3, D7 & P20 for the data 68, 50,32, 21, 54, 38, 59, 66 & 14.

Solution:
Arrange the data.
x = 21, 32, 38, 44, 50, 54, 59, 66, 68 Here n = 9 is odd, therefore
2nd 7th

9+1 10
𝑄3 = 3 𝑡ℎ = 𝑡ℎ = 7.5th
4 4
66+59
So then add the 7th & 8th values in the data and divided them into = 62.5
2

9+1 10
𝐷7 = 7 10
𝑡ℎ =7 10
= 7th observation =

𝐷7 = 59

9+1 10
𝑃20 = 20 𝑡ℎ = 20 = 20 × 0.1 = 2nd
100 100

𝑃20 = 32

21
Quartiles, Decile & Percentile
Computation of Quartiles, Decile & Percentile
Case (ii) Discrete series or grouped data

If n is Even If n is Odd
Σ𝑓 Σ𝑓+1
𝑄𝑖 = 𝑖 th item i =1,2,3 𝑄𝑖 = 𝑖 th item i =1,2,3
4 4

Σ𝑓 Σ𝑓+1
𝐷𝑖 = 𝑖 th item i =1,2,3,…,9 𝐷𝑖 = 𝑖 th item i =1,2,3,…,9
10 10

Σ𝑓 Σ𝑓+1
𝑃𝑖 = 𝑖 th item i =1,2,3,…,99 𝑃𝑖 = 𝑖 th item i =1,2,3,…,99
100 100

22
Quartiles, Decile & Percentile
Example: Calculate Q2, D6 & P25 for distribution of the value of 140 fruits given below
x 30 25 10 36 50 44 40 55 60

f 14 8 3 18 22 23 27 17 8

Solution:
Σ𝑓 140
x f cf 𝑄2 = 2 𝑡ℎ = 2 𝑡ℎ = 70th
4 4
10 3 3 𝑄2 = 40
25 8 11
Σ𝑓 140
30 14 25 𝐷6 = 6
10
𝑡ℎ= 6
10
= 84th
36 18 43
𝐷6 = 44
40 27 70
44 23 93 Σ𝑓 140
𝑃25 = 25 𝑡ℎ= 25 𝑡ℎ = 35th
100 100
50 22 115 𝑃25 = 36
55 17 132
60 8 140
total 140
23
Quartiles, Decile & Percentile
Computation of Median in Continuous Series or Grouped Data

Step (i): Arrange the data in ascending order along with frequencies.

Step (ii): Compute the cumulative frequencies and Class boundaries.

Step (iii): Use the formula of a discrete case to locate the Quartiles, Decile &
Percentile classes.

Step (iv): Calculate the Quartiles, Decile & Percentile by formula,

ℎ Σ𝑓+1
Quartiles =l+𝑓 𝑖 − 𝑐
4
ℎ Σ𝑓+1
Decile =l+𝑓 𝑖 − 𝑐
10
ℎ Σ𝑓+1
Percentile =l+ 𝑖 − 𝑐
𝑓 100
Where
l = Lower class boundaries in which the Quartiles, Decile & Percentile lie.
h = class interval of the Quartiles, Decile & Percentile class.
f = Quartiles, Decile & Percentile class frequency.
c = cumulative frequency of the preceding class.
24
Quartiles, Decile & Percentile
Example: Calculate Q1, D8 & P60 for distribution of the value of 140 fruits given below
Classes 0–4 5–9 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34

f 4 6 10 16 12 8 4
Solution:
Σ𝑓 60
Class f c.F C.B 𝑄1 = 2 𝑡ℎ = 1 𝑡ℎ = 15th
4 4
0–4 0 – 4.5 ℎ 60 5
4 4 𝑄1 = l + − 𝑐 = 9.5 + 15 − 10
𝑓 4 10
5–9 6 10 4.5 – 9.5
𝑄1 = 12
10 – 14 10 20 9.5 – 14.5
Σ𝑓 60
15 – 19 16 36 14.5 – 19.5 𝐷8 = 6
10
𝑡ℎ= 8
10
= 48th
ℎ 5
20 – 24 12 48 19.5 – 24.5 𝐷8 = l + 48 − 𝑐 = 19.5 + 48 − 36
𝑓 12
25 – 29 8 56 24.5 – 29.5 𝐷8 = 24.5
30 – 34 4 60 29.5 – 34.5
Σ𝑓 140
60 𝑃60 = 25 𝑡ℎ= 25 𝑡ℎ = 35th
100 100
ℎ 5
𝑃60 = l + 𝑓 35 − 𝑐 = 14.5 + 16 35 − 20
𝑃60 = 19.5
25
Mode
Mode is the Frech word that means “Fashion” and has been adopted to
convey the idea of “most frequent”.

The mode is defined as a value that occurs most frequently in a set of

data, data is it indicates the most common results.

A set of data may have more than one mode or no mode at all when each
observation occurs the same number of times.

Calculating mode for individual series or ungrouped data

The only inspection method is used in this case.
Example:
The following data gives the speeds (in miles per hour) of eight cars that
were stopped by police for speeding violations 77, 82, 74, 81, 79, 84, 74, and 78. Find
the mode.
Solution:
As 74 occurs twice and remaining values occur only once, so Mode = 74
miles per hour.

26
Mode
Calculating mode for discrete series or grouped data
For the discrete series, the inspection method is used.
Example:
Calculate the mode for the following discrete frequency distribution.
x 12 14 16 18 20 22

f 1 4 6 10 7 2

Solution:
x f
By looking at the data, the value of the variable 18 has maximum
12 1 frequency, so Mode = 18.
14 4
16 6
18 10
20 7
22 2

27
Mode
Calculating mode for continuous series or grouped data
For the continuous series, the given formula is used.

𝑓 −𝑓
Mode = 𝑙1 + 2𝑓 𝑚− 𝑓 −𝑓
0
×ℎ
𝑚 0 1

𝑙1 = Lower limit of the model class chosen by inspection method.

𝑓𝑚 = frequency of the model class.
𝑓0 = frequency of the preceding class.
𝑓1 = frequency of the following class.
ℎ = class interval of the modal class.

28
Mode
Example:
Calculated mode for the following data:
Marks 10 – 24 25 – 39 40 – 54 55 – 69 70 – 84 85 – 99
f 6 20 44 26 3 1
Solution:
𝑓𝑚 − 𝑓0
Marks f C.B Mode = 𝑙1 + ×ℎ
2𝑓𝑚 − 𝑓0 −𝑓1
10 – 24 6 9.5 – 24.5
ℎ = 54.5 – 39.5 = 15
25 – 39 𝑓0 20 24.5 – 39.5
44 −20
Mode = 39.5 + × 15
40 – 54 𝑓𝑚44 39.5 – 54.5 2(44)−20−26
𝑙1

55 – 69 54.5 – 69.5 24×15

𝑓126 Mode = 39.5 +
88−46
70 – 84 3 69.5 – 84.5
Mode = 39.5 + 8.5
85 – 99 1 84.5 – 99.5
Mode = 48
29
Mode
Advantages of Mode

i. It can be determined both for quantitative and qualitative data.

ii. It is a value that actually exists within the given set.
iii. It is not affected by extreme values.

Disadvantages of Mode

i. It does not consider all values of the data.

ii. Sometimes a distribution may have no mode or may have more than one mode.

30
Empirical relationship among
Mean, Median, & Mode
Knowing the values of Mean, Median & Mode can give us some idea about the shape
of a frequency distribution.

For Symmetrical Distribution

. The values of Mean, Median, & Mode are identical and they lie at the
center of the distribution.
Mean = Median = Mode

For Asymmetrical or Moderately Skewed Distribution

Mode = 3Median – 2Mean

31
Geometric Mean
The nth root of the product of n positive values is called Geometric Mean.
That is,
𝑛
G. M = 𝑥1 . 𝑥2 . , , , . 𝑥𝑛
1
G. M = 𝑥1 . 𝑥2 . , , , . 𝑥𝑛 𝑛
Example:
G.M of the value 2,4, & 8 is;
1 1 1
G.M = 2 × 4 × 8 3 = (64)3 = (43 )3 = 4
G.M by this formula is difficult to calculate when the figure is larger in size and greater
in number. So we use the given formula,
Σ𝑙𝑜𝑔𝑥
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔 (for ungroup data)
𝑛
Σ𝑓𝑙𝑜𝑔𝑥
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔 (for group data)
Σ𝑓

32
Geometric Mean
Computation of G.M for individual Series or Ungrouped data.

Example:
Calculate the geometric mean for the data: 50, 51, 52, 53, & 54.
Solution:

Σ𝑙𝑜𝑔𝑥
x log x G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔
𝑛
50 1.6990
8.5793
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔
51 1.7076 5

52 1.7160
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔 1.71586
53 1.7243
G. M = 51.991
54 1.7324

Total 8.5793

33
Geometric Mean
Computation of G.M for Discrete Series or Grouped data.

Example: Calculate the Geometric mean for the following data.

x 3 4 5 6 7 8 9 10 11
f 2 5 9 14 15 8 6 3 1

Solution:
x f log x f log x
Σ𝑓𝑙𝑜𝑔𝑥
3 2 0.4771 0.9542 G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔
Σ𝑓
4 5 0.6021 3.0105
5 9 0.6990 6.2910 50.8184
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔
6 14 0.7782 10.8948 63
7 15 0.8451 12.6765
8 8 0.9031 7.2248
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔 0.8066
9 6 0.9542 5.7252
G. M = 6.406
10 3 1.0000 3.0000
11 1 1.0414 1.0414

Total 63 50.8184
34
Geometric Mean
Computation of G.M for Discrete Series or Grouped data.

Example: Calculate the Geometric mean for the following data.

Classes 10 – 19 20 – 29 30 – 39 40 – 49 50 – 59

f 15 35 45 25 12
Solution:
Classes F X f log x
Σ𝑓𝑙𝑜𝑔𝑥
10 – 19 15 14.5 17.4205 G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔
Σ𝑓
20 – 29 35 24.5 48.6208
197.2890
30 – 39 45 34.5 69.2019
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔
132
40 – 49 25 44.5 41.2090
G. M = 𝑎𝑛𝑡𝑖𝑙𝑜𝑔 1.4946
50 – 59 12 54.5 20.8368

Total 132 197.2890 G. M = 31.23

35
Geometric Mean
Advantages of Geometric Mean.

i. It takes into account all values of a series.

ii. It gives more weight to smaller values and less weight to largest
values. Hence it is better measure than A.M.
iii. It is not affected by fluctuations of sampling.

Disadvantages of Geometric Mean.

i. It is easy to calculate.
ii. If one value in the data set is zero, the G.M will zero.
iii. If any one or more values are negative, either G.M cannot be
calculated or and absurd value will obtained.
Harmonic Mean
The reciprocal of the arithmetic mean of the reciprocals of the values is
called harmonic mean.

𝑛
H.M = 1 (Formula for ungrouped data)
Σ
𝑥

Σ𝑓
H.M = 𝑓 (Formula for ungrouped data)
Σ𝑥
It is also mathematical average but not very popular. It is used in those
places where largest weight is given to small items. It is suitable for averaging time and
speed.

37
Harmonic Mean
H.M in case of Individual series or ungrouped data
Example:
Five tractor are running at the rate of 10, 20, 30, 40, 50 km/hour respectively. What would
be the average speed? Use Harmonic Mean.

Solution:
x 𝟏
𝑛
𝒙 H.M = 1
Σ
10 0.1 𝑥

20 0.05 5
H.M =
30 0.03 0.23

40 0.03 H.M = 21.9 km/hour.

50 0.02

Total 0.23

38
Harmonic Mean
H.M in case of Discrete Series or Grouped Data
Example: Calculate Harmonic mean of the following distribution.
x 12 14 16 18 20 22
f 1 4 6 10 7 2

Solution:
x f 𝒇 Σ𝑓
𝒙 H.M = 𝑓
Σ𝑥
12 1 0.0833

14 4 0.2857 30
H.M =
1.7406
16 6 0.3750

18 10 0.5556 H.M = 17.24

20 7 0.3500

22 2 0.0909

Total 30 1.7406

39
Harmonic Mean
H.M in case of Discrete Series or Grouped Data
Example: Calculate Harmonic mean of the following distribution.
Classes 0–9 10 – 19 20 – 29 30 – 39 40 – 49
f 5 10 15 7 3
Solution:

𝒇 Σ𝑓
Classes f x H.M = 𝑓
𝒙 Σ𝑥
0–9 5 4.5 1.1111
40
10 – 19 10 14.5 0.6897 H.M =
12.6833
20 – 29 15 24.5 0.6122
H.M = 14.91
30 – 39 7 34.5 0.2029
40 – 49 3 44.5 0.0674
Total 40 2.6833

40
Harmonic Mean
Merits of Harmonic Mean:
i. It considers all the given data.
ii. It is not affected by fluctuation of sampling.
iii. It is a good averages for a highly variable series.

Demerits of Harmonic Mean:

i. It cannot be calculated if any one of the observation is zero.
ii. It gives less important to large values and more important to small values.
iii. Its value is generally not a member of the series.

Relationship among A.M, G.M, & H.M

i. A.M ≥ G.M ≥ H.M
Equality will hold only when all the observations are same.

ii. G.M = 𝐴. 𝑀 (𝐻. 𝑀)

If two measures are known, the third one can be approximately calculated by this
relation.

41
Thank You

42