You are on page 1of 70

Measure of central tendency

Averages:
A single representative value of the data is called an
average. Where the most of the values of the data have
a tendency to lie around it.
A measure of central tendency is a single value that
attempts to describe a set of data by identifying the
central position within that set of data.
measures of central tendency are sometimes called
measures of central location.
Types of Averages
The most common types of averages are
• Mean(Arithmetic mean)
• Median
• Mode
These averages are also called the first order
averages
Arithmetic mean
The mean (or average) is the most popular and well known
measure of central tendency. It can be used with both discrete
and continuous data
The mean is equal to the sum of all the values in the data set
divided by the number of values in the data set that is:
𝑆𝑢𝑚 𝑜𝑓 𝑎𝑙𝑙 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠
Arithmetic mean= 𝑥 =
𝑁𝑜.𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛

𝑥 (read as x bar).The arithmetic mean is sometimes referred to


as arithmetic average or as the mean or the Average
Arithmetic mean
Un-Grouped data Grouped data
⅀𝑥 ⅀𝑓𝑥
𝑥= 𝑥=
𝑛 ⅀𝑓
Where ⅀𝑥 =sum of all Where x: stands for mid-point
observation f stands for frequency
n= No. of observations
Ungrouped data

X
Example 1: In a batch of
7
10 students, the marks
8
obtained by them are as
5
follows 6
7,8,5,6,7,5,5,4,6,7 7
calculate the average 5
marks 5
⅀𝑥 60
𝑥= = =6 4
𝑛 10 6
𝑥 =6 marks 7
Total ⅀𝑥= 60
Example 2: For the data given, find the A.M
2,7,5,-4,0,3,11,9,-2,1
n=11
⅀𝑥 2:7:5;4:0:3:11:9;2:1
𝑥= =
𝑛 11

𝑥= 3.2
Example 3: if the mean of the following numbers are 17, find
the value of k
12,18,21,k,13
⅀𝑥
since 𝑥 =
𝑛
12:18:21:𝑘:13
17=
5
85= 12+18+21+k+13
85=64+k
K=21
Grouped data
Example 1: Compute A.M from the following distribution
C.I 2-6 7-11 12-16 17-21 22-26 Total
f 2 4 11 9 3 20
Solution:
C.I f X(mid-point) fx
2-6 3 4 12
7-11 4 9 36
12-16 11 14 154
17-21 9 19 171
22-26 3 24 72
Total 30 445
⅀𝑓𝑥
𝑥=
⅀𝑓

445
𝑥= = 14.83
30
Median

The most middle value of an arranged


data is called median. Median divides
the data into two equal parts
Median
Grouped Data Un-grouped Data
ℎ ⅀𝑓 𝑛:1
Median=𝑥 = 𝑙 + − 𝐶. 𝐹 Median= 𝑥 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
𝑓 2 2
Where l=lower boundary of Where n= no. of observation
median class
h= width of median group
⅀𝑓= total frequency
C.F= cumulative frequency of
median class.
Median for un-Grouped data
Case 1: if number of obs. are Case 2: if number of obs. are
odd even
Example: find median of the Example: find median of the
following data: following data:
7,7,2,3,4,2,7,9,31,11,18 36,41,27,32,29,38,39,43,55,42
Solution: Solution:
Arrange the data in ascending Arrange the data in of
order( array) Magnitude
2,2,3,4,7,7,7,9,11,18,31 27,29,32,36,38,39,41,42,43,55
N=11 N=10
11:1 10:1
Median = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 Median = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
2 2
= 6th value = 5.5th value
=7 5𝑡ℎ 𝑣𝑎𝑙𝑢𝑒:6𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
=
2
Median = 7
38:39
= = 38.5
2
Median=38.5
Median from frequency table
Example: find median from the following data

X 2 3 4 5 6 7
F 7 11 16 20 22 24
Solution:
x f C.f
2 7 7
3 11 18
4 16 34 50.5
5 20 54
6 22 76
7 24 100
Total N=100
𝑛:1
Median= 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
2
100:1
= 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
2
= 50.5
Since 50.5 lies in the cumulative frequency
corresponding to 5, therefore the Median=5
Example: the following table gives the marks of 20
students in a class out of 100 marks. Find out the
median

marks 50 55 58 60 63 66 75
No. of students 1 2 2 5 4 5 1
Solution:
x f C.f
50 1 1
55 2 3
58 2 5
60 5 10 Median
63 4 14 classes
66 5 19
75 1 20
Total N=20
𝑁:1 20:1
Median= = =10.5th value
2 2
Therefore, 10.5th is the mean value of two middle
values i.e.,(10th and 11th ) in the given data
10𝑡ℎ 𝑣𝑎𝑙𝑢𝑒:11𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 60:63
Hence, median= =
2 2

= 61.5 marks
Grouped Data
Example: Find Median for the following frequency
distribution.

C.I 2-4 5-7 8-10 11-13 14-16


F 2 8 12 4 2
Solution:
C.I C.B f C.F
2-4 1.5-4.5 2 2
Median class
5-7 4.5-7.5 8 10
8-10 7.5-10.5 12 22
11-13 10.5-13.5 4 26
14-16 13.5-16.5 2 28
Total N=28
⅀𝑓
now determine group by the formula th value
2
therefore,
⅀𝑓 28
We get = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 14𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
2 2
Which is in the interval (7.5-10.5).
This interval Is called Median group
ℎ ⅀𝑓
Hence; Median = 𝑙 + − 𝐶. 𝐹
𝑓 2
3
= 7.5+ 14 − 10
12
3
=7.5+ + 4
12
=7.5+1.0
=8.5
Median =𝑥= 8.5
Median by graph:
Median is that value on the x-axis which
⅀𝑓
corresponds to a C.F when an Ogive is
2
drawn.
Median by graph
90
80
70
60
⅀𝑓
= 42
2 50
40
30
20
10
0
14 18 22 26 30

Median=21.9
Mode
Mode is the most repeated value in a set of data
OR
Mode is that value which has a maximum
frequency than others in its neighborhood.
A distribution may have one or two modes. If a
distribution has only one mode than it is called
Uni-Modal. If the dist. has two modes it is called
Bi-Modal.
Mode

Un-Grouped Grouped
The most repeated value in a 𝑓𝑚 ;𝑓1
Mode=𝑥=𝑙 + ×ℎ
given set of data. 2𝑓𝑚 ;𝑓1−𝑓2

Where l=width of the group


𝑓𝑚 = maximum frequency
𝑓1 = frequency preceding to
modal class
𝑓2 = frequency following to
modal group
Ungrouped data
Example: find mode for the following data
9,10,12,13,14,11,15,11,9,10,11,16
Solution:
First we array the data
9,9,10,10,11,11,11,12,13,14,15,16
Since 11 is the most repeated value, therefore,
Mode =11
Example: determine mode of the following data
2,5,5,5,6,7,7,8,8,8,9
Solution:
Since 5 and 8 are the most repeated values
therefore there are two mode 5 and 8
The distribution is said to be Bi-Modal
Grouped Data
Example: Calculate Mode for the following data
C.I 2-4 5-7 8-10 11-13 14-16
F 2 6 11 4 1

In order to calculate mode for


the above data we must first
find class boundaries. Since
the third class has highest
frequency, therefore it is our
modal class
Solution:
f Class interval Frequency Class boundaries
2-4 2 1.5-4.5
5-7 6 4.5-7.5 𝑓1
8-10 11 7.5-10.5 𝑓𝑚
11-13 4 10.5-13.5 𝑓2
14-16 1 13.5-16.5

L=7.5, h=3 , 𝑓𝑚 = 11, 𝑓1 = 6, 𝑓2 = 4


By putting all the values in the formula
𝑓𝑚 ;𝑓1
𝑥=𝑙 + ×ℎ
2𝑓𝑚 ;𝑓1−𝑓2
11;6
Mode= 7.5+ × 3
2 11 ;6;14
5
Mode=7.5+ × 3= 7.5+1.25=8.75
12
Mode= 8.75
Mode by graph
A histogram is used to determine the mode graphically.
The steps are as follows
1. Draw a histogram of the given distribution
2. The highest rectangle will be modal class
3. Join the top right and left corners of this rectangle with
the top right corner of the rectangle of preceding class
and the top left with the following class respectively
4. From the point of intersection of both lines draw a
perpendicular on the x-axis, the point where the
perpendicular meets the x-axis is the value of the Mode
Example: Draw a histogram for the following data and measure
the modal value. Verify it by actual calculations.

Class interval Frequency


0-10 5
10-20 11
20-30 19
30-40 21
40-50 16
50-60 10
60-70 8
70-80 6
80-90 3
90-100 1
Solution:
Mode by graphical method
25

20

15

10

0
10 20 30 40 50 60 70 80 90 100
Mode=32.86
Quartile, Percentile and Deciles
Although quartile, percentiles and deciles are
not averages, they are included here because
they are of same nature as Median. In other
words we can say that median is a family
member of Quartile, percentile and deciles
QUARTILE
Quartiles are the values that divide a list of
numbers into quarters or four equal parts. These
values are denoted by 𝑄1 , 𝑄2 𝑎𝑛𝑑 𝑄3 .
Where,
𝑄1 = first quartile
𝑄2 =Second quartile
𝑄3 = Third Quartile
Quartile
Un-Grouped Data Grouped Data
For n values arranged in array For grouped data quartiles are
the quartiles are computed as computed as
𝑛:1 ℎ ⅀𝑓
𝑄1 =
4
𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝑄1 = 𝑙 + − 𝑐. 𝑓
𝑓 4
𝑛:1 ℎ 2⅀𝑓
𝑄2 = 2 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝑄2 = 𝑙 + − 𝑐. 𝑓
4 𝑓 4
𝑛:1 ℎ 3⅀𝑓
𝑄3 =3 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝑄3 = 𝑙 + − 𝑐. 𝑓
4 𝑓 4
Quartile for un-grouped frequency
distribution
In case of a discrete (or un-grouped) frequency
distribution the quartiles are computed as
⅀𝑓:1
𝑄1 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
⅀𝑓:1
𝑄2 =2 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
⅀𝑓:1
𝑄3 =3 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
Ungrouped data
Example: find 𝑄1 , 𝑄2 , 𝑄3 for the following data
2,3,3,9,6,6,12,11,8,2,3,5,7,5,4,4,5,12,9 (n=19)
Solution: arrange the data in ascending order
2,2,3,3,3,4,4,5,5,5,6,6,7,8,9,9,11,12,12

10th value 15th value


5thvalue
𝑛:1 19:1 20
Then 𝑄1 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = = = 5𝑡ℎ 𝑣𝑎𝑙.
4 4 4
𝑄1 = 3
𝑛:1 19:1
𝑄2 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒=2 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒= 10th value
4 4

𝑄2 = 5

𝑛:1 19:1
𝑄3 = 3 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 3 = 15th value
4 4

𝑄3 = 9
Example 2: find quartiles for the following data
X 0 1 2 3 4 5 6 7 8 9 10

f 2 2 3 4 6 11 15 10 6 3 1

Solution: we know that


⅀𝑓:1
𝑄1 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
⅀𝑓 + 1
𝑄2 =2 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
⅀𝑓 + 1
𝑄3 =3 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
Marks frequency c.f
0 2 2
1 2 4
2 3 7
3 4 11
4 6 17 16th value

5 11 28
6 15 43
32th value
7 10 53
8 6 59 48th value
9 3 62
10 1 63
TOTAL 63
⅀𝑓:1 63:1
𝑄1 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒= = 16th value
4 4
2⅀𝑓:1 63:1
𝑄2 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 2 =32th value
4 4
3⅀𝑓:1 63:1
𝑄3 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒=3 = 48th value
4 4

𝑄1 = 4
𝑄2 = 6
𝑄3 = 7
Grouped data
Example: from the following grouped frequency
distribution . Calculate 𝑄1 , 𝑄3
Wages 150-170 170-190 190-210 210-230 230-250
No. of 30 50 80 30 10
workers

Solution:
⅀𝑓
Since 𝑄1 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 50𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
⅀𝑓
𝑄3 = 3 = 150𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
4
C.B f C.F
150-170 30 30
170-190 50 80 50th value
190-210 80 160
150th value
210-230 30 190
230-250 10 200
TOTAL 200
ℎ ⅀𝑓
𝑄1 = 𝑙 + − 𝑐. 𝑓
𝑓 4
20
𝑄1 =170+ 50 − 30 = 𝟏𝟕𝟖
50
ℎ 3⅀𝑓 20
𝑄3 = 𝑙 + − 𝑐. 𝑓 =190+ (150-80)=207.5
𝑓 4 80
Decile
The value which divides the arranged data into
ten equal parts. The values are denoted by 𝐷1 ,
𝐷2 , 𝐷3 , … , 𝐷9 where 𝐷1 is the first decile 𝐷2 is
the second decile, 𝐷3 is the third decile and
lastly 𝐷9 is the last decile
All values are arranged in ascending order
DECILE
Un-Grouped Grouped
𝑛:1 ℎ ⅀𝑓
𝐷1 = 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝐷1 = 𝑙 + − 𝑐. 𝑓
10 𝑓 10
𝑛:1 ℎ 2⅀𝑓
𝐷2 =2 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝐷2 = 𝑙 + − 𝑐. 𝑓
10 𝑓 10
𝑛:1
𝐷3 = 3 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝐷3 = 𝑙 +
ℎ 3⅀𝑓
− 𝑐. 𝑓
10
𝑓 10
. .
. .
. .
𝑛:1 ℎ 9⅀𝑓
𝐷9 =9 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 𝐷9 =𝑙 + − 𝑐. 𝑓
10 𝑓 4
un-grouped data
Example: Find 𝐷4 and 𝐷6 from the following weights in
kg.
19,27,24,39,57,44,56,59,67,62,42,47,60,26,34,57,51,59
,45,50
Solution: first we array the data
19,24,26,27,34,39,44,44,45,47,50,51,56,57,57,59,59,60
,62,67 n=20
𝑛:1 20:1
𝐷4 =4 =4 th value = 8.4 th value
10 10
= 44+0.4(9th vlaue-8th value)
= 44+ 0.4(45-44)
=44.4kg
20:1
𝐷6 =6 th value= 12.6 th value
10
= 51+0.6(56-21)= 54Kg
𝐷4 = 44.4 kg
𝐷6 = 54kg
Grouped data
Example: Calculate 𝐷2 and 𝐷3 from the following data

X 0-5 5-10 10-15 15-20 20-25


f 7 18 25 30 20

Solution:
⅀𝑓
Since 𝐷2 = 2 = 20𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
10
⅀𝑓
𝐷3 = 3 = 30𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
10
C.I F C.F
0-5 7 7
5-10 18 25 20th value
10-15 25 50
15-20 30 80 30th value
20-25 20 100
Total 100
ℎ 2⅀𝑓 5
𝐷2 = 𝑙 + − 𝑐. 𝑓 =5+ (20 − 7)=8.6
𝑓 10 15
ℎ 3⅀𝑓 5
𝐷3 = 𝑙 + − 𝑐. 𝑓 = 10+ (30-25)= 11.0
𝑓 10 25
Decile in frequency distribution
In case of a discrete (un-grouped) frequency
distribution Decile are calculated as:
⅀𝑓:1
𝐷1 = th value
10
⅀𝑓:1
𝐷2 =2 th value
10
⅀𝑓:1
𝐷3 = 3 th value
10

⅀𝑓:1
𝐷9 =9 th value
10
Example: find D2 and D6 the following frequency
distribution
X 4 5 6 7 8 9 10 11 12 13
f 2 4 13 35 51 54 69 41 27 54

Solution
⅀𝑓:1
𝐷2 =2 th value
10
⅀𝑓:1
𝐷6 =6 th value
10
X F C.F
4 2 2
5 4 6
6 13 19
7 35 54
8 51 105 70.2 th value
9 54 159
10 69 228 210.6th value
11 41 269
12 27 296
13 54 350
TOTAL 350
⅀𝑓:1 350:1
𝐷2 =2 th value = 2 = 70.2th value
10 10

𝑫𝟐 =8

⅀𝑓:1 350:1
𝐷6 =6 th value =6 = 210.6th value
10 10

𝑫𝟔 = 10
Percentile
The values which divides the arranged data in 100
equal parts are called percentile. These values are
denoted by 𝑃1 , 𝑃2 , 𝑃3 ,…, 𝑃99 . Where 𝑃1 is the first
percentile 𝑃2 is the second percentile 𝑃3 is the third
percentile and 𝑃9 is the last percentile.
Percentile
Un-Grouped Grouped
𝑛:1 ℎ ⅀𝑓
𝑃1 = th value 𝑃1 =𝑙 + − 𝑐. 𝑓
100 𝑓 100
𝑛:1 ℎ 2⅀𝑓
𝑃2 = 2 th value 𝑃2 =𝑙 + − 𝑐. 𝑓
100 𝑓 100
𝑛:1 ℎ 3⅀𝑓
𝑃3 = 3 th value 𝑃3 =𝑙 + − 𝑐. 𝑓
100 𝑓 100

ℎ 99⅀𝑓
𝑛:1 𝑃99 =𝑙 + − 𝑐. 𝑓
𝑃99 =99 th value 𝑓 100
100
Un-Grouped
Example: For the following data calculate 𝑃25 , 𝑃59 ,𝑃80
10,11,15,16,10,12,13,14,15,14,16,17,20,14,13 and 10
Solution:
First arrange the data
10,10,10,11,12,13,13,14,14,14,15,15,16,16,17,20
Here n=16
𝑛:1 16:1
𝑃25 = 25 th value = 25 =4.25th value
100 100
4th value+0.25(5th value- 4th value)

=11+0.25(12-11)=11.25

𝑷𝟐𝟓 =11.25

𝑛:1 16:1
𝑃59 =59 th value=59 = 10.03th value
100 100

=10th value+0.03(11th value-10th value)


=14+0.03(15-14)
𝑷𝟓𝟗 =14.03
𝑛:1 16:1
𝑃80 =80 th value=80 =13.6th value
100 100
=13th value+0.6(14th value-15th value)
=16+0.6(16-16)
𝑷𝟖𝟎 =16
Grouped data
Example: find 𝑃20 , 𝑃60 , from the following frequency
distribution.
C.I F
2-12 2
12-22 5
22-32 8
32-42 12
42-52 15
52-62 20
62-72 16
72-82 14
82-92 10
92-102 8
Solution
C.I F C.F
2-12 2 2
12-22 5 7
22-32 8 15
32-42 12 27 22nd value
42-52 15 42
52-62 20 62
62-72 16 78 66th value
72-82 14 92
82-92 10 102
92-102 8 110
TOTAL 110
⅀𝑓 110
since, 𝑃20 = 20 th value=20 = 22nd value
100 100
ℎ 20⅀𝑓 10
𝑃20 =𝑙 + − 𝑐. 𝑓 =32+ 22 − 15 =37.83
𝑓 100 12
⅀𝑓 110
Since, 𝑃60 = 60 th value=60 =66th value
100 100
ℎ 60⅀𝑓 10
𝑃60 =𝑙 + − 𝑐. 𝑓 =62+ 66 − 62 =64.50
𝑓 100 12
𝑷𝟐𝟎 =37.83
𝑷𝟔𝟎 = 64.50
Percentile of ungrouped frequency
distribution
Example: Find out 𝑃10 , 𝑃40 and 𝑃90 from the
following frequency distribution
Wages Employees
100 3
125 6
150 10
175 15
200 24
225 42
250 75
275 90
300 79
325 55
350 36
375 26
400 19
425 13
450 8
⅀𝑓:1
𝑃10 =10 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 50.2𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
100
𝑃10 =200
⅀𝑓:1
𝑃40 = 40 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 200.8𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
100
𝑃40 =275
⅀𝑓:1
𝑃90 = 90 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒 = 451.8 𝑡ℎ 𝑣𝑎𝑙𝑢𝑒
100
𝑃90 =375
Situations when Mean, Median and
Mode are used
• Mean is used for general purpose i.e., to find average
height, avg. weight, avg. marks etc. and it should be
avoided in case of skewed distribution
• Median is used when the distribution have open end
classes or are highly skewed, e.g. Distribution of wages,
Dist. Of incomes, dist. Of wealth etc. it is also used
when the values in the series are really ranks (like
intelligence or honesty)
• Mode is used when most common value is required
such as size of ready made clothes, size of shoes, N0. of
children per household

You might also like