X 1 + X 2 + ... + X n
(i) X =
n
or X =
∑ X
n
Case (ii) For Discrete Series
(ii) X= A+
∑ fd
XC
N
Xi − A
=
where di = ∀ i 1, 2,3....n
C
A = Assumed Mean
Solution:
Example 3.
• No. of students. 3 4 5 7 9 8
Solution
Weights in Kgs. No. of Mid –Values (X) fX
Students
(f)
44-48 3 46 138
49-53 4 51 204
54-58 5 56 280
59-63 7 61 427
64-68 9 66 594
69-73 8 71 568
=X
∑=
fX 2211
= 61.42 kgs
N 36
Example 4.
• Frequency : 23 38 58 82 65 31 11
Solution
Class Frequenc Mid- fd
X i − A X i − 419.50
intervals y (f) Values =di =
C 20
(X)
X= A+
∑ fd
XC
N
( −43)
= 419.50 + X 20
308
= 416.71
Properties of AM
i.e. ∑( X − X ) =
i 0 for ungrouped data. and
∑ f (X − X ) =
i i 0 for grouped frequancy distribution.
i.e. if
=d
X−A
= and d
∑ fd
C N
then X= A + C.d
Properties of AM
If there are two groups having n1 and n2
observations and
X 1 and X2
as their respective AMs , then
the AM of combined group is given as
n1 X 1 + n2 X 2
X =
n1 + n2
Example 5.
N=30 670
Solution: AM
=
X
∑=
fX 670
= 22.33
N 30
Example.7
• X: 3 5 Y 6
• f: 1 3 2 4
Solution. AM is given by
=X
∑
=
fX 1 × 3 + 3 × 5 + 2Y + 6 × 4 42 + 2Y
=
N 1+ 3 + 2 + 4 10
42 + 2Y
i.e. 5 =
10
or 50 − 42 = 2Y
therefore Y = 4
Question Time
MCQ’s on Arithmetic Mean
Question 1.
AM of 1,2,3,4,5,6 is
a) 2.5
b) 3.5
c) 4.5
d) 5.5
(a) 3
(b) 5
(c) 7
(d) 9
Answer: (b) 5
Question.3.
The algebraic sum of deviations taken about the
AM of 15, 20 and 25 is
(a) 0
(b) 5
(c) -5
Answer: (a) 0
Question 4.
If a variable assumes the values 1,2,3,4,5 with frequencies
as 1,2,3,4,5 respectively, then AM is
(a) 11/3
(b) 5
(c) 4
(d) 4.5
(a) 40%
(b)50%
(c) 60%
(a) 30
(b) 29
(c) 33.68
(d) 34.21
(a) 51.2
(b) 52.2
(c) 53.2
(d) 54.2
• X: 6 4 Y 2
• f: 4 3 2 1
Question 8
(a) 5
(b) 6
(c) 7
(d) 8
Answer: (b) 6
Question.9
(a) 20.5
(b) 21.5
(c) 22.5
(d) 23.5
(a) 115.5
(b) 116.5
(c) 117.5
(d) 118.5
(a) 4, 4
(b) 6, 2
(c) 5, 3
(d) 7, 1
Answer: (b) 6, 2
Question.12
(a) 7
(b) 9
(c) 11
(d) 12
Answer: (c) 11
Median
Median (Md) and other
Partition Values
Median is a positional average, which divides the data
into two equal parts.
Properties of Median
If Y= a +b X
then Md(Y) = a + b Md(X)
Properties of Median
∑ X−A is minimum if A =
Md
Determination of Median
Case (i):
When number of observations is odd, i.e.
n is odd.
n +1
Then Md = th term
2
Example 1 :
Find median of the following data
40, 29, 15, 58, 70
n +1 5 +1
= Md th=
term = rd
3 term
2 2
i.e. Md = 40
Case(ii)
When number of observations is even, i.e. n is even.
First arrange the observations in ascending or
descending order.
1 n n
=
Then Md ( th term ) + ( + 1)th term
2 2 2
Example 2.
Find the Median of the following observations. 5, 25,
11, 14, 26, 30.
1 n n
=Md ( th term ) + ( + 1)th term
2 2 2
1 6 6 1
Then Md= ( th term) + ( + 1)th term = (14 + 25)= 19.5
2 2 2 2
Case (iii)
• X: 60 65 70 75 80 85 90
• f: 7 12 20 22 18 14 9
Solution:
X f Less than type c.f.
60 7 7
65 12 19
70 20 39
75 22 61
80 18 79
85 14 83
90 9 92
Total N=92
Solution
N
− c
l+
2
Then Md = ×h
f
where l = lower limit of the Md class
f = frequency of the Md class
c = cumulative frequency preceeding the Md class
h = class width (besure that the classintervals are exclussivetype, i.e. continuous classes)
Example. 4.
10-20 8 13=c
20-30 10=f 23
30-40 7 30
40-50 6 36
Totals N=36
Solution
The c.f. just greater than N/2 i.e. 18, is 36, which is the
Md class. Hence, Note that if class intervals are not
continuous then they are first made continuous by
adding d/2 to the all upper limits and subtracting d/2
from all the lower limits,
where d is the difference between lower limit of any
class and upper limit of the previous class.
N
− c
l+ ×h
2
Then Md =
f
36
− 13
Md = 20 + ×10 = 25
2
i.e.
10
Quartiles
There are three quartiles viz.
Q1, Q2 and Q3,which divide
the data into four equal parts.
Case (i)
For x1-f1, x2-f2,…xn-fn type data
iN
−c
l+ ×h
4
Then Q i =
f
where l = lower limit of the Q i class
f = frequency of the Q i class
c = cumulative frequency preceeding the Q i class
h = class width (besure that the classintervals are exclussivetype, i.e. continuous classes)
Example 1
Find the third quartile for the following
frequency distribution.
X: 10 20 30 40 50 60 70 80 90 100
f: 3 5 10 20 30 40 45 35 7 5
Solution
X f c.f.
10 3 3
20 5 8
30 10 18
40 20 38
50 30 68
60 40 108
70 45 153
80 35 188
90 7 195
100 5 200
Total N= 200
Solution
Here to calculate third quartile we see the value of X
corresponding to the
is given by
Q3 = 70
Case (ii)
iN
−c
l+ ×h
4
Then Q i =
f
where l = lower limit of the Q i class
f = frequency of the Q i class
c = cumulative frequency preceeding the Q i class
h = class width (besure that the classintervals are exclussivetype, i.e. continuous classes )
Example 2
Find the First and Third Quartiles for the following
frequency distribution.
10-20 8 13
20-30 10 23
30-40 7 30
40-50 6 36
50-60 4 40
Totals N=40
Solution
N
− c
Then Q1 =
l+ 4 ×h
f
40
− 5
i.e. Q1 = 10 + 4 ×10 = 16.25
8
And 3N
− c
l+ ×h
4
Then Q3 =
f
3 × 40
− 30
i.e. Q3 = 40 + 4 ×10 = 40
6
Deciles
X: 10 20 30 40 50 60 70 80 90 100
f: 3 5 10 20 30 40 45 35 7 5
Solution
X f c.f.
10 3 3
20 5 8
30 10 18
40 20 38
50 30 68
60 40 108
70 45 153
80 35 188
90 7 195
100 5 200
Total N= 200
Solution
Here we have formed the column of
cumulative frequencies, now see the
c.f. just greater than
5N/10=5x200/10=100, which is 108,
hence the corresponding value of X is
fifth decile, i.e.
D5 = 60
Case (ii)
For grouped data, i.e. frequencies against
class intervals are given.
iN
− c
Di =l+ 10 ×h
f
where l = lower limit of the Di class
f = frequency of the Di class
c = cumulative frequency preceeding the Di class
h = class width (be sure that the class intervals are exclussive type, i.e. continuous classes )
Example 2
No. of students: 5 8 10 7 6 4
Solution
Marks No. of students c.f.
Class (f)
Intervals
0-10 5 5
10-20 8 13
20-30 10 23
30-40 7 30
40-50 6 36
50-60 4 40
Totals N=40
First we find the seventh decile class
corresponding to the c.f. just greater than
7N/10=28, i.e. 30. Hence
7N
− c
Then D7 =
l+ 10 ×h
f
7 × 40
− 23
i.e. D7 =30 + 10 ×10
7
i.e. D7 =30 + 7.14 = 37.14
Percentiles:
Percentiles divide the
whole data into 100 equal
parts. They are 99 in
number and denoted by
P1,P2,…P99
Case (i)
For x1-f1, x2-f2,…xn-fn type data.
X: 10 20 30 40 50 60 70 80 90 100
f: 3 5 10 20 30 40 45 35 7 5
Solution
X f c.f.
10 3 3
20 5 8
30 10 18
40 20 38
50 30 68
60 40 108
70 45 153
80 35 188
90 7 195
100 5 200
Total N= 200
The 65th Percentile
The 65th percentile is the value of X for which
corresponding cumulative frequency is just greater
than 65N/100=65x200/100=130,
which is 153 .
Hence P65 = 70
Case (ii)
For grouped data, i.e. frequencies are
given corresponding to class intervals.
th
First obtain the percentile
i
class which is corresponding to
the cumulative frequency just
greater than iN/100.
Percentile
Then ith (i=1,2,…99) percentile (Pi) is given
by
iN
−c
l+ ×h
100
Pi =
f
where l = lower limit of the Pi class
f = frequency of the Pi class
c = cumulative frequency preceeding the Pi class
h = class width (be sure that the class intervals are exclussive type, i.e. continuous classes )
Example 2.
Find the 52nd percentile for
the following frequency distribution.
• Marks: 0- 10 10-20 20-30 30-40 40-50 50-60
• No. of students: 5 8 10 7 6 4
Solution
Marks No. Of students c.f.
Class Intervals (f)
0-10 5 5
10-20 8 13
20-30 10 23
30-40 7 30
40-50 6 36
50-60 4 40
Totals N=40
The 52nd percentile
52 × N
− c
l+ ×h
100
P52 =
f
52 × 40
− 13
or P52 =20 + 100 ×10
10
P52 =20 + 7.8 =27.8
Partition Values
for Ungrouped Data:
50, 56, 65, 75, 79, 80, 82, 90, 120, 130
3
Q3 = ( n + 1) × th value
4
3
i .e. = (10 + 1) × th value = 8.25 th value
4
i.e. = 8th value + 0.25(9th value − 8th value)
i .e . = 90 + 0.25(120 − 90)
i .e . = 90.5
Partition Values
for Ungrouped Data:
50, 56, 65, 75, 79, 80, 82, 90, 120, 130
7
D7 = ( n + 1) × th value
10
7
i .e. = (10 + 1) × th value = 7.7 th value
10
i.e. = 7 th value + 0.7 (8th value − 7 th value)
i .e. = 82 + 0.7(90 − 82)
i . e . D 7 = 87.6
Partition Values
for Ungrouped Data:
50, 56, 65, 75, 79, 80, 82, 90, 120, 130
73
P73 = ( n + 1) × th value
100
73
i .e. = (10 + 1) × th value = 8.03 rd value
100
i.e. = 8th value + 0.03 (9th value − 8th value)
i .e. = 90 + 0.03(120 − 90)
i . e . P73 = 90.9
Question Time
M.C.Q's. on Median and other
Partition Values
Question 1
What is the Median of 5, 8, 6, 9, 11 and 4.
a) 6
b) 7
c) 8
d) None of these
Answer: (b) 7
Question 2
What is the value of the first Quartile for the
observations 15, 18,10, 20, 23, 28, 12, 16?
(a) 17
(b) 15.75
(c) 12
(d) 12.75
(a) 13
(b) 10.70
(c) 11
(d)11.50
(a) 45
(b) 36
(c) 58
(d) 42
Answer: (b) 36
Question 5
Find the 73rd percentile of 4, 12, 36, 45, 3, 58, 65?
(a) 54.92
(b) 55.92
(c) 56.92
(d) 57.92
(a) 9
(b) 11
(c) 10
(d) 7
Answer: (c) 10
Question 7
The absolute sum of deviations is minimum when it
is measured about
(a) Mean
(b) Median
(c) Mode
(d) None of these
(a) 3
(b) 6
(c) 18
(d) 36
Answer: (d) 36
Question 9
What is the value of the first quartile for observations
15, 18, 10, 20, 23, 28, 12, 16?
(a) 17
(b) 16
(c) 15.75
(d) 12
(a) 13
(b) 10.70
(c) 11
(d) 11.50
(b) 17.79
(c) 11
(d) 11.50
• e.g. Mo of 5, 3, 8, 9, 5, 6 is 5 which
is the most frequent value.
Case (ii)
For grouped data i.e. if frequencies corresponding to
class intervals are given, then Mo is given as
f 0 − f −1
Mo = l + × h
2 f 0 − f −1 − f1
where
l = lower limit of modal class
f 0 = frequency of the modal class
f −1 frequency of pre − modal class
f1 frequency of post − modal class
h = class width (note that class intervals must be continuous)
Example:
Find the Mo for the following frequency
distribution.
350-369 349.5-369.5 23
370-389 369.5-389.5 38
390-409 389.5-409.5 58
410-429 409.5-429.5 82
430-449 429.5-449.5 65
450-469 449.5-469.5 31
470-489 469.5-489.5 11
Solution
f 0 − f −1
Mo = l + ×h
2 f 0 − f −1 − f1
82 − 58
Mo = 409.5 + × 20
2 × 82 − 58 − 65
i.e. Mo = 409.5 + 11.71 = 421.21
Relation between Mean, Median
And Mode:
Mean-Mode=3(Mean-Median)
Question Time
MCQ’ on Mode
Question 1
What is the Modal value for the numbers
5, 8, 6, 4, 10,15, 18, 10, 5, 10?
(a) 5
(b) 10
(c) 14
Answer: (b) 10
Question 2
The Mode for the following frequency distribution:
Class Intervals: 350-369, 370-389, 390-409, 410-429, 430-449, 450-469
Frequency : 15 27 31 19 13 6
(a) 390
(b) 394.86
(c) 395.86
(d) 396
(a) 45
(b) 46
(c) 47
(d) 48
Answer : (d) 48
Question 7
Measures of Central Tendency for given set of
observations measures
(b) The data are put in the form of grouped frequency distribution
GM is defined if no observation
is zero or negative.
Geometric Mean (G)
GM of n positive observations
is defined as the nth root of
their product.
Geometric Mean (G)
If Z=X.Y then
GM(Z)=GM(X).GM(Y)
Geometric Mean (G)
If Z=X/Y then
GM(Z)=GM(X)/GM(Y)
GM for Individual series
GM of X1, X2,…Xn is given by
1
G = (X1 . X 2 ... X n ) n
1
or log G =
n
∑ log X
GM for Discrete Series
1
or l og G =
N
∑ f log X
Example 1
GM of 2, 3, and 4 is given as
1 1
G= ( 2 × 3 × 4) 3 = ( 64 ) 3 = 4
Example 2.
Find the GM of the following frequency
distribution.
• X : 2 4 8 16
• f: 2 3 3 2
Solution
1
G = (X 1
f1
.X 2
f2
... X n
fn
) N
1
or (
G = 2 .4 .8 .16
2 3 3 2
) ( 2 + 3+ 3+ 2)
1
or (
G = 2 .2 .2 .2
2 6 9 8
) 10
1
or =
G ( 2 )=25 10 2=
2.5
4 2
G = 5.66
Harmonic Mean
Harmonic Mean (H)
HM is defined if
no observation is zero.
Harmonic Mean (H)
( n1 + n2 )
H =
n1 n2
+
H 1 H 2
Case (i)
HM of X1, X2, … Xn is
n
H =
1
∑ X
Example 1
HM of 2, 4, and 6 is
n 3 3 36
=
H = = = = 3.27
1 1 1 1 11 11
∑ X ∑ 2 + 4 + 6 12
Example 2
X : 2 4 8 16
f : 2 3 3 2
Solution :
N ( 2 + 3 + 3 + 2)
=H = = 4.44
f 2 2
∑X
3 3
∑ 2 + 4 + 8 + 16
Case(ii)
If the data is of X1-f1, X2-f2,…Xn-fn type
data, then HM is given by
N
H =
f
where N ∑ f
∑X
Relation Between AM, GM and HM:
(G=
) 2 ( A) × ( H )
Moreover
AM ≥ GM ≥ HM
Note that equality holds if all the observations are equal.
Weighted Average
Weighted Average:
AM =
∑ WX
∑W
Weighted GM is
∑ W log X
GM = Ante log
∑W
Weighted HM is
HM =
∑W
W
∑ X
Example 1
Find the weighted AM and weighted
HM for the following data
X : 1 2 3 4 5 6 7 8 9 10
W : 12 22 32 42 52 62 72 82 92 102
Solution
=
Weighted AM
∑
=
WX 1 × 12 + 2 × 2 2 + ... + n × n 2
∑W 12 + 2 2 + ... + n 2
n(n + 1)
2
13 + 23 + ... + n3
2
= =
12 + 2 2 + ... + n 2 n(n + 1)(2 n + 1)
6
3 n(n + 1)
or AM =
2(2 n + 1)
Solution
Weighted =
HM
∑
=
W 12 + 2 2 + ... + n 2
W 12 22 n2
∑X 1
+
2
+ ... +
n
n(n + 1)(2 n + 1)
1 + 2 + ... + n
2 2 2
6
= =
1 + 2 + ... + n n(n + 1)
2
(2 n + 1)
or HM =
3
Question Time
MCQ’ on GM & HM
Question 1
Which measure of central tendency is used to
find average rates.
(a) AM
(b) GM
(c) HM
(d) Both (b) & (c)
Answer: (b) GM
Question 2
Which of the following results hold for a set
of distinct positive observations?
(a) AM ≥ GM ≥ HM
(b) HM ≥ GM ≥ AM
Answer: (a) AM ≥ GM ≥ HM
Question 3
The Harmonic Mean for the numbers 2,3,5 is
(a) 2.00
(b) 3.33
(c) 2.90
(d) -3.33
(a) 24
(b) 12
(c) 36
(d) 18
Answer: (b) 12
Question 5
If the AM and GM for the two numbers are 6.50
and 6 respectively then the two numbers are
(a) 6 and 7
(b) 9 and 4
(c) 10 and 3
(d) 8 and 5
(a) 16
(b) 4.10
(c) 4.05
(d) 4.00
Answer: (d) 4
Question 7
If GM of X is 10 and GM of Y is 15, then
the value of GM of XY is
(a) 150
(b) log 10 x log15
(c) log 150
(d) none of these
Answer: (a) 150
Question 8
If AM of a variable X is 10 and GM of X is
20, then the HM of X is
(a) 40
(b) 50
(c) 20
(d) 30
Answer: (a) 40
Question 9
The HM of 1, 1/2, 1/3 … 1/n is
(a) n
(b) 2n
(c) 2/(n+1)
(d) n(n+1)/2