D

STATISTICS : ITS INTRODuCTION

SOME MEASURES
Related Terms
6.1. Statistics and its
Statistics. An aggregate of facts which are affecetd by a number
causes and which are expressed numerically to Some resonable extent
of accuracy and which are collected in a systematic manner fo
specific purpose is called statistics.
The subject in which we study the characteristics of the facte i
also known as Statistics .
Illustration. Let we collect the facts and figures of the car accident
taken place in Kolkataduring last 15years. Then this collection is a
statistics of 'car accident in Kolkataduring last 15 years'.
Variable. Variable is a symbol e.g. x,y..that can assume any prescribed
value.
If a variable assume only one value then it is called constant.
lIlustration. (i) Let N= number of members in a family of India.
Then Ncan assume any of the values 1, 2, 3, ... ... . So, here Nis a
variable.
(i1) Let X= number of prime ministers in India Then we see X
assumes only one value, [Link] X is a constant.
(11) Let H=weight of a person of city. Then we see H can assume
the values 45 kg, 46.12 kg, 80.0015 kg [Link] here H isa variable.
Discrete and Continuous Variable. A variable that can theoreticaly
assume any value between two given values is called a
variable. continuou
A variable which is not continuous is
called a discrete
In the above illustration the variable cited in (i) and (ii)variable.
are discrete
whereas the variable in (ii) is
Data / Observations. The valuescontinuous.
assumed by a variable are known
as data or
observation. Sometimes,
statistical data or statistical in statistics, these are
regardedas

Illustration. let Y= observation.

marks obtained by the students in Mathematics
in a class. Then Y 5,90,0,81
etc .The data can beis a variable which can assume the datas
:0 presented
81
as
90
Remark: In fact variable,data [Link] be 70defined 65in different [Link]
this text we keep this definition thinking of the relevant concerned
readers.
STATISTICS:ITS INTRODUCTION &SOME MEASURES
229

62Frequency Distribution.
The number of occurence of an observation or data of a
Frequency.
variable is called the frequency of that
data.
Illustration. Letet x be the marks obtained by 30 [Link] x assumes
thevalues
35 31 32 34 31
30
34 42 30 57 68
30
71 20 15 10 51
42
57 51 51 52 51 80

51 57 20 71 35 32

Here we see the data 30 occurs three [Link] frequency of 30 is

3. Similarly the frequency of the data 80,71,68 are 1,2,1respectively.
Simple Frequency Distribution. The simple frequency distribution
of a variable is the statistical table where the observations (assumed by
the variable ) are arranged in order of magnitude and the frequency of
each observation is shown side by side.
lustration. Letx be avariable which takes the value :
3 4 5 3 4
4 3 2 5 6 1
3 4 3 1
5
Then the frequency distribution of x is
1 4 5 6 Total
2 3
2 4 3 2 18
2 5
Ihe table can also be shown in column-wise.
Grouped Frequency Distribution.
When alarge number of datas are available we cannot grasp their
atable. In these
characteristic only by placing them individually insuitable
cases we group the observations into a number of [Link]
atable (or statistical table) these intervals are shown and the

IreqsideuencyThis
by
of the
observations included in each interval are
table is called Grouped frequency distributior
shown side

secured by 70 candidates
Il uscerttrataiionn. examination:
The datas give the
below marks
230
ENGINEERING MATHEMATICS-lA
21 31 35 52 64 74 89 53
22 35 43 67 76 35
42 7
46 26
72 43
32 40
38 41 63 71 28 32 45
15 18 52 73 86 50 S4
39
44 47 12
58 67 85 39 40 50 65 72
57 63 5 56 79 69
37 24 54 82 49
51 54 68 29 34 44 58 62 59
65
Here we see there is a large number of the
almost [Link] group the datas into the observations which are
32,..We see the data 7,5 are included in the intervals 0-10,1 1-21.2.
interval [Link] the
frequency (called class-frequency)of the class interval 0-10 is 2. In this
way we have the following grouped frequency
: distribution
Marks secured :0-10 11-21 22-32 33-43 44-54
55-65 66-76 77-87 88-98
Frequency : 2 4 8 14 15 12 10 4 1
Note. Frequency distribution is nothingbut
Terms associated with Grouped Frequency
quantitative classification.
Distribution.
(1) Class interval : The group of datas into a
number of suitable
intervals are called class interval. In the previous
11-21 etc. are class intervals. example 0-10
(2) Class limits: The two extreme values
specifying a class interval
are called class limit. In the previous example the
lower class limit (LCL)
and the upper class limit (UCL) of the class
22 and 32. 22-32 are respectivey
(3) Class Boundaries : The class boundaries of a
class are detined a
Lower Class Boundary (LCB) = LCL of the class
where d= LCL of the class -UCL of the
previous class.
Upper class Boundary (UCB) = UCL of the class + d
2
where d= LCL of the next class -UCL of the class. In the previous
example LCB of the class 22-32 is
22 -21
22 - = 21·5: UCB of the
class 22-32 is
32-
33-32
= 32-5,
2
STAT7STICS:- 17S INTRODUCTION, &SOME MEASURES, 231

(4) Class Mark or Mid Value: Class Mark of a class

-(LCL+VCL) of the class.

In the previous example, class Mark of the class 66-76 is
76) =71,
(5)Width of a class : Width of a class = (UCB-LCB) of the class.
In the previous example, width ofthe class 22-- 32 is 32-5-21-5 =11.
Cumnmulative Frequency.
For a siriple frequency distribution the total frequency of the
observations lesser or equal to an observation is called "less (s) than
type"cummulative frequency of the observation.
For a grouped frequency distribution the total frequency of
observations lesser or equal to the observation in a class is called the
"less (S) than type" cummulative frequency of the observation.
Illustration.
(i) In the simple frequency distribution
2 4 9 11
f: 3 6 4 1

the "less than (s) type" cummulative frequency of 9 is

3+6+4= 13; the "greater than () type" cummulative frequency of
9 is 4+1l=5.
(1) Inthe grouped frequency distribution
10-14 15-19
Class 0-4 5-9
6 4
Frequency 3
the class 10- 14
e less than (S) type" cummulative frequency of
is 3+6+4 =13: the
Cgainst the upper boundary 14-5 of this class
the class 10- 14 or
3eater than type (2)" cummulative frequency of
8ainst the LCB 9.5 of this class is 4+1=5.
6.3. Mean.
or may not be among the datas assumed
typical value which may of all the datas, For
by a variable is considered as a representative
7 can bc treated as
example among the datas 2,5,[Link],7 the value
 232
ENGINEERING MATHEMATICS-A

that [Link] this representative-value

centrally within the set of datas .This
value is measured tends
by
way. Following is the one of best such measurements
Arithmetic Mean. The Arithmetic mean (A.A.M) or briefly
dif erem
of the values (datas) x,X),...x, is defined as
the mean

If the datas have the frequency shown in the following table

Variable
Frequency (): f
then their A.M is +=fi t x,fat....x,f) = f
N
where = f +f+..f,= Lfi
N
i=1
n
For agrouped-frequency distribution the A.M, * = xf where
N il
x; is the mid-value of each
class-interval.
Illustration. If x 2 4 3 Total
3 2 4 10
be the frequency
distribution of avariable xthen its A.M,

I-2x3+4x2+1x4+3xi)
10

or ,~=6+8+4+3) 1
10 =x21
10 =2.1
Theorem 1. If the
two x and y are related by the equation
then ù = X-cvariables
d d where c and d are any number.
Proof. We consider the
frequency distribution of x:
Total

N
Since the values of x frequency
distrbution of y would be are changed to those of y sothe

..

Total
N
STATISTICS.
:/TS INTRODUCTION &SOME MEASURES
233

whereJV; =
d
Now the A.M 1of y F=fy-2
ofv, ù=
N i=l N jzI d

Nd i=l
Nd j=1

1
Nd \i=l
Nd \i=l i=1

d d N d
Nd i=l i=l d d
Note. The above theorem is very much helpful to determine the A.M
of the variable assuming large data. This is shown in the following
Illustration.
Illustration. We are given the following Grouped frequency distribution :
Weight in gms (x): 110-119 120-129 130-139 140-149 150-159
5 7 12 20 16
Frequency:
170-179 180-189
Weight in gms(x): 160-169
10 3
Frequency
To find the A.M we construct the following table :
Calculation of Mean
I;-1545
Class-interval Midpoint Frequency -154.5 10
= yi
() ()
-4 -20
110-119 114.5 5 -40
-3 -21
120-129 124.5 7 -30
-2 -24
130-139 134.5 12 -20
-1 -20
140-149 144.5 20 -10

150-159 154.5 16 0
10
160-169 164.5 10 10
2 14
170-179 174.5 20
3 )
180-189 184.5 3 30
-52
Total 80
234 ENGINEERING MATHEMATICS.

Here N =80 andy,=-[Link] the

i=l

A.M ofy, y= X-52= -0.65

80

A;-154.5 So V = I-1545
Since y; = 10
10

i-154.5
Or, -0.65= or, = 148.
10

Thus the required A.M, = 148.

Theorem. 2. (On Mean of Composite Group)
Let j,p..*, be the A.M. of rgroups containig n.n,," ,
observations respectively. Then their combined mean or composite
mean is given by
n tn, t..+n,~,
n +n t t n r
Proof. Omitted.
Illustration. Suppose the mean wage of 60 labourers in morning shit
is Rs 80 and the mean wage of 40 labourers working in evening shit
is Rs 70. Then the mean wage of all labourers (of both shift)
60× 80+ 40x70
=76
n + 60 + 40
and
Note. There are other type of means viz., Geometric mean
Harmonic mean. But study of these is beyond the scope ofthe book.
6.4. Median.
the
Median of a set of observations is the middle most value when of
order
observations are arranged in increasing or decreasing
magnitude.
Thus to find the median of a set of observation it is necessrary
arrange the observations in order of magnitude.
Calculation of Median. of
number
Calculation of median may be confusing for even / odd
classity
the
observation: for grouped frequency distribution. So we
:
procedure of Calculation of Median in the following three cases
STATISTICS:1/TS
INTRODUCTION& SOME MEASURES 235

CaseI. (For simple distribution i.e. without

having any freguency).
.ange the given n nunmber of
descending order of magnitude. observations in ascending |
(i) If n isodd then
n+1
Median = th observation of the arranged set.
2
(ii) If n is even then
n
Median =2 th observation th observation).
Jllustration. (i) f we are required to find the median of the set {5,
8.7, 20, 13, 3, 11) then we first arrange the datas in increasing order
ie. 3,5,7,8, 11, 13, 20. Here the number of observations, n=7(odd).
7+1
So its median = -th = 4th observation = 8.
2
(i) If we are required to find the median of the set (5, 4,7, 3, 21,
12), we first arrange the datas in ascending order of magnitude (3, 4,
5,7, 12, 21). Here the number of observation, n=6 (even).
6 6
So its median = th observation + +1| th observation)
2

{3rd observation +4 th observation) =(5+7)

2 =6.
Case 2. (For Simple Frequency Distribution).
Arrange the observations in ascending order of magnitude.
N+1
Construct "less (S) than type" cumulative frequency. Calculate 2
eTe N 1s total frequency. Then Median = the observation
N+1
or next heigher
corresponding
N+1
to the cummulative frequency 22
(if
is not acummulative frequency ).
Illustration.
Consider the following frequency distribution 60 70 80
Marks 30 40 50
16 8
No of students 8 15 23
The "lesS (s) than type" Cummulative Frequency.
236
ENGINEERING MATHEMATICS-|A
Marks

(r)
Frequency
()
Cummulative Frequency
(1) (2)
(less(S)
(3)
than type)
30
40 15 23
50 23 46
60 16 62
70 70
80 75
Total N=75
N+1 75+ 1
Here =38.
2 2
We see there is nocummnulative frequency '38 in column 3. Next
higher figure than 38 in column 3 is 46.
Required Median is the observation corresponding to the
cummulative frequency 46 = 50.
Note : N+1
may be fraction. The procedure in same is that case also.
2
Case 3. (For Grouped Frequency
Distribution)
Here also construct the "less (S) than type"
N
cummulative frequly
agairst class boundaries. Calculate where N is total frequency. nu
2
the Median-Class, i.e. the class corresponding to the cummulative
N
N
frequency 2 or next higher (if
2 1S not a cummulative frequency
N
- F
Then Median = ,, + 2 xi

where L,, = lower boundary of

Median-Class
N=Total frequency
Mediut
F=
Class. Cummulative frequency
lhe
of the class preceedingto
STATISTICS: ITS INTRODUCTION &SOME
MEASURES 237

frequency of median class.

width of the median class.
Illustration.

Consider the following grouped frequency distribution :

Class-interval
130-134
Frequency
135-139 15
140-144 28
145-149 24
150-154 17
155-159 10
160-164

The "less than type" Cummulative frequency against Class

boundaries.
Class-interval Frequency Upper Class-BoundaryCummulative
Frequency (< type)
(1) (2) (3) (4)
130-134 134.5 5
135-139 15 139.5 20*
140-144 28 144.5 48
145-149 24 149.5 72
150-154 17 154.5 89
155-159 10 159.5 99
160-164 164.5 100
Total |N=100
* It
means there are 20 observations which are less or equal to
139.5. N
Now, 2 =30. There is no cummulative frequency 50 in the 4th
column of the above table. In the 4th column the next higher figure is
12. This
Corresponds to the median class 145-149.
238 ENGINEERIN MATHEMATICS-IJA

Therefore, ,, = lower, boundary of the median class =

F= the cummulative frequency of the class
144-5.
preceeding the median class = 48.
fm =frequency of the median class = 24.
i= width of the median class = 5.

- F 50- 48
So the Median =l, + 2 x i= 144-5+ -x5 =144-92.
24
fm
Note. There is another formula to find the median called "interpolation
formula".
6.5. Mode.
The observation having maximum frequency is called mode.
Calculation of Mode.

Case 1. For a simple frequency distribution the mode is calculated by

simply method of inspection.
Illustration. For the frequency distribution
10 20 30 40 50 70

2 2 3 2 2 1
Frequency
Here we see the observation 30 has heighest frequency 3. So the
mode is 30.
Case 2. (For Grouped Frequency Distribution).
For a grouped frcquency distribution mode can be calculated to
frequency distribution having unique class with heighest frequency anu
with equal class width. First find the modal-class i.e. the class havn;
heighest frequency. Then

Mode = lm t fi-fo Xi
2fi - fo -f2
where lm = lower class boundary of the modal class
fi= frequency of the modal class
fo = frequency of the class preceeding the modal class
fa= frequency of the class succeeding the modal class
class).
i= width of each class (note that this is same for each
 lustration.
CMNdT the gnNt tiyucy tisuibution,
Marks
No, of CandateN

20
S0-59

70-79 )
80-89
9-99

Here we see every class has same width which is 10 and only class
4049 has heighest trequency. So the modal class is 40-49, As we

Mode = ,, + xi
2fi - fo -f:
where = lower class boundary of the class 40-49 - 39.5
f= frequency of the class "40- 49"= 20.
fo = frequency of theclass preceeding the class 40-49 =18
Sa =frequency of the class succeeding 40-49 = 12
i= width of each class = 10.

:. Mode =395+ 20-18

-x10= 41.5.
40-18-12
Relation among Mean, Median and Mode : For adistribution having
Single Mode the relation is Mean- Mode 3(Mean - Median)
Note. For asymetrical distribution mean, median and mode coincide.
6.6.
Variance
and
As we have statedStandard
Set of
Deviation.
in the previous article A.M representsthe entire
[Link]
Ihe A.M is the degree to which the datas tend to spread
to be measured. It is usually measured by variance or
about

slandard deviation which are discussed below:

240 ENGINEERING MATHEMATICS-lJA

Variance.
The mean of the squares of the differences of the
datas) assumed by a variable from their arithmetic observations
mean (A.M) is
variance of the variable.
Standard Deviation.
called
The positive square root of variance is called Standard
Thus (i) if x|, X,.X, be the datas then their Deviation (sd)
variance,
and
n /=l

the standard deviation, o,

(ii) if the datas have the frequency shown in the following table
Variabie (x): ...Xy

Frequency
then the Variance,
Var(s) =f(a-z +f( -}' +..+fa(a,-}
N i=l
flx; - where N = f, +fat..+f, and the standard

deviation, O, =+
If thedatas have grouped frequency distribution then x; will be ie
mid -value of each class-interval and f;would be the correspondi
frequency. the datas:
Illustration : Let x be a variable which assumes

20,85,120,60,[Link] =-(20+ 85+120+ 60+ 40) = 65. To find o

5
variance and standard deviation we go through the following table.

A;-65 (4-65)
20 -45 2025
85 20 400
120 55 3025
60 -5 25
40 -25 625
Total 6100

Here (
i=]
; - 65) =6100
 INTRODUCTION &SOME MEASURES
STATISTICS.:ITS 241

Then
Var(x)=E(, -65)² 5
x6100 =1220.
i=1

andthe S.d,
o= J1220 = 34.93.
heorem 2. If , 2.... N, be the datas then
2

Var(a)= } - }
Proof. Left as an exercise.
Nate. The above theorem can be extended
2
for frequency distribution
[Link] 'ar(1)=f
Nj=l
where N=f +ft..tf,.
Theorem 3. If the two variables x and y are related by the equation
V= then ,= xwhere d is a positive number.
d d
Proof. Beyond the scope of this text.
Note. The above theorem is very much helpful to determine the s.d
of the variable assumning large [Link] is shown in the following
illustration. In practice, we use the result in Th-2 to find the variance /
S.d.
llustration. (i) Let we be given the following observation having
the frequency distribution :
: 240.12 240.13 240.15 240. 16 240.17 240.21 240.22
Frequency: 2 2 1 2

lo find the S.d we go through the following table :

x;-240.16| a; - 240.16
= yi
.01
240.12 -.04 -4 -8 32

240.13 2 -.03 -3 -6 18

240.15 -01 -1
240.16 1 0
240.17 2 .01 2 2

240.21 .05 25

240.22 .06 36
Totl -2 114

MIILA-16
242 ENGINEERING MATHEMATICS A

7 7

Here, N= 10. 2fy, =-2 and y} = I14

i-1 i=l

I14
Now, Var(o)= N N 10 =1136
i. o, = VI 1.36 = 3.37
X, -- 240.16
Since V; =
.01 .01
or, 3.37= .. O, = 0.0337.
.01
(ii) Consider the following grouped frequency
Value
distribution :
: 90-99 80-89 70-79 60-69 50-59 40 49
30-39
Frequency : 2 12 22 20 14 4
To find the variance and standard
deviation of this grouped
frequency distribution we construct the following table :
Class interval|Mid poin frequency X-64-5| y=I-64-5|
10
() ()
90-99 94-5 2 30 18
80-89 84-5 12 20 24 48
70-79 74-5 22 10 1 22 22
60-69 64-5 20
50-59 54-5 14 -10 -l -14 14|
40-49 44-5 4 -20 -2 -8 16|
30-39 34-5 -30 -3 -3
Total 75 127
27

Here N=75, fy, =27, iy=127.

2

Now, Var(y)E-x =1-56,

.:. the standard deviation of y, o, = /56 =|·23.

Now since y = I-64.5 ;. therefore
10
Or, o, =1:23x 10= 12.3.
STATISTICS:1ITS INTRODUCTION &SOME MEASURES 243

[llustrative Examples.
data ;
Ex. l. Find the mean fromn the following
25-29 30-34 35-39 40-44
Daily wages (Rs) .:
No. of workers 16 28 14 12
45 and
tis given that the total wvage for 10workers earning Rs
more is Rs 600.
without the last
First we are to work out the mean for the rest part
table :
class. For that we construct the following
*;-37
class mid point frequency 5
interval (;)
27 16 -2 --32
25-29
32 28 -1 -28
30-34
35-39 37 14
12 1 12
4 044 42
70 -48
Total

.:. Here N=70,fy, =-48

-48 -24
70 35
37
But ù=I .
5
24
i.e., X =37 +5ù =37 ×5
35
=33-57.
DO the total wage of 70 workers = Rs. 33-57x70
= Rs. 2349 -9.
Thus the of total 80 workers is
wage Rs (2349.9+ 600)
= Rs 2949-9.
2949.9
. Mean wage is Rs
80
= Rs 36-87.
244 ENGINEERING MATHEMATICS-IA

Ex. 2. The A.M calculated fromthe following, frequency

known to be 72.5. Find the value of x :
classes :30-39 4049 50-59 60-69
2 3 20
70-79 80- dis89tribu90-tion99 is

Frequency : 25 7
We first construct the following table to find A.M using thhe
datas :
x;-74.5
given
Classes Mid pt Frequency y; =
10
() (5)
30-39 34.5 2 4 -8
40-49 44.5 3 -3 -9
50-59 54.5 11 -2
-22
60-69 64.5 20 -20
70-79 74.5 X 0
80-89 84.5 25 25
90-99 94.5 7 2 14
Total 67+x -20

-20
67+ x

=745+ 10ù =74.5 200

67+x
By the give condition we have.
200
74.5 = 72.5
67+ x
or, 67+x=100
X=33
Thus the missing frequency x is 33.
Ex. 3. Average narks obtained by a class of 70 students [Link]
to be 65. Later it was found that the marks of one student was wrongly
recorded as 85 in place of 58. Find the corrected mean.
Wrongly calculated mean = 1 70 ; =6s
70
70
or, >n; = 4550
i=l
 INTRODUCTION &SOME MEASURES 245
. (TS
STATISTICS.

sum of Wrongobservations =4550-

ie.
of corrected observations =4550 85 +58= 4523
4523
mean - 64-61.
So the corfected 70

Ex.4. Following is a frequency distribution lacking two class

mean is 7.74.
frquency. Findthem if the S-7 7-9 9-11 ||-13 Total
value 3-5
32 57 25 200
frequency
Let the two missing frequencies be fi and f respectively. We
Construct the following table :
Class-interval| Frequency Class-Mark
32 128
3-5
5-7 6 6f
7-9 57 8 456
9-11 10 10f2
11-13 25 12 300

Total 200=N 884+ 6f +10f2

Now, 32+ f +57+fa +25 =200
(1)
or, f+f = 86

The mean,

or, 7-74=x(884+6fi +10f2)

200
(2)
or, 3f +5f, =332
the two missing
Solving (1) and (2) we get f =49, f, =37 so
[Link]
5.
es are 49 and 37.
3xt4y= 21. A. M. of x
is 3. Two variables x and y are related by
Find A. M. of y. 21-3r
From the given relation we get y= 4
21-35 21--3x3 12 =3
4 4 4
246 ENGINEERING MATHEMATICS.

Ex. 6. Two variables x and y are related by x=2y +S

x is 25. Find the median of y. The
From the given relation we have y= 2
median of
Here if x increases y also does So.
Median of x-5 25- 5 = 10.
So the median of y= 2 2
Ex. 7. The number of observations of two groups are in the
and their A. M. are 8 and 128 respectively. Find the A. ratio 2:1
combined group. M. of the
Let the number of observations of the two groups be
) = [Link] combined A. M.. =nx t ng 2k x8 +kx128 2k,k. =8,
2k +k
144k
Or, r =48.
3k
Ex. 8. Find the median of the following frequency distribution
5 10 15 20 25 30 35 40
f 7 10 15 18 23 21 17 8
The cumulative frequency distribution table is given below:
f C. f
5 7 7
10 10 17
15 15 32
20 18 50
25 23 73
30 21 94
35 17 111
40 8 119
Here N=119
N+1
= 60.
2
N+l andthe
So the cumulative frequency just greater than 0s73
2
value ofx correspondingto c.f 73 is 25. Hence the median is 25.
STATISTICS:ITS INTRODUCTION & SOME MEASURES
247

Calculate mean, median and hence find the

Ex. 9.
approximate
ofthe modefromthe following frequency distributions : value
Heighi (inches) :60-63 64-67 68--71 72-75 76-79 80-83
No of students 8 3 18 6 16 8
To find mean we first construct the following table :

*, -73.5
classes Mid pt frequency c.f Y; =
4
() (G)
60-63 61.5 8 -3. -24
64-67 65.5 3 11 -2 -6
68-71 69.5 18 29 -1 -18
72-75 73.5 6 35 0
76-79 77.5 16 51 16
80-83 81.5 8 59 2 16

Here N = f =59 -16

.:. y= Lyf. -16 =-0.2712

59
*;-73.5
Now, y;= or, x, = 73.5+4y;
4

*=735+ 4ù = 73.5+4(-0.2712)
=72.4152
. Mean of the given distribution = 72.4152·
N
Now 2 =295
So the
median class is 72-75
i. m=715, F =29, fm
=6
and /=755715=4
N
:. -F
Median =Im + 2 -xi

7164 29.5-294= 7183.

=715+ 6
(Mean-Median) wehave
Using the relation Mean-Mode =3
 248 ENGINEERING MATHEMATICS4|lA

72.4152-Mode ={72.4152 -71.83) or.

72.4152-mode =1.75 6
Ex. 10, Calculate the mode of the following data : Mode =10.659%
1. 12, 5, 8, 12, 13, 8, 1, 4, 8, 7, 8, 5.
Let us arrange the given variaties with corresponding
given below frequencies as
2

2
7
4
12 2
13 1
As the variate 8ocurs 4 times which is
maximum, so,
mode =8.
Ex. 11. Calculate the mode from the following
distribution :
class 10-15 15-20 2U-25 25-30 30-35
frequency 6 7 7
Here the greatest frequency 11 lies in the class 20-25. Hence modal
class is 20-25.
.:. = lower class boundary of the modal class
=20.

f= frequency of the class.=|1

Jo= frequency of the procceding the modal class
=9.
f= frequency of the class succeding the modal class =7.
i= width of each
class =6.
.". Mode = +

=l2f-fo-S2
Xi

= 20+ 11-9
22 -9 -7 X6 = 22.
STATZSTICS:17S INTRODUC TION &SOME MEASURES 249

medianand
and mode of the following frequency distribution
Ex.12. The
knoWn to be 27 and 26 respectively. Find the values of a and ß:
are
classinterval 0-10 10-20 20--30 30-40 40-50
3 20 12 B
frequency
Since mode = 26, so it lies within the class 20-30

Mode = m t
fi-fo Xi
2f; - fo - fa
20
= 20+ -X10
40- a-12

20
= 20+ x 10,
28- a

20+
10(20- c«) =26
28-a

Or,
10(20 a) 6
28- ,
a=8.

So the total frequency is N =43+B.

But media - 27, so it lies within the class 20-30.
Then we construct the following frequency table :
class interval frequency cf. frequency
0-10 3 3

10-20 11
20-30 20 31
30-40 12 43

40-50 B 43+ß

N
- F
Now median =,l
+2 -Xi
 ENGINEERINGGMATHEMATICS-UA
250

-2
43+B11-X10
.. 27= 20+ 20

oT, 21+1
4

or, 21+ß= 28
. B=7.
variance and standard deviation of the followine
Ex. 13. Findthe
frequency distribution :
3640 41-45 46-50 51-55 S6-60 61-65 66.7%
Weight (in kg): 33 50 37 25
14 26 40
No of persons
table:
We construct the following
;-53
class Mid pt frequency
|interval () -3 -42 126
3640 38 14
-2 -52 104
41-45 43 26
40 40
48 40 -1
46-50
51-55 53 33 0
50 50
58 50 1
56-60
74 148
61-65 63 37
225
25 3 75
66-70 68 693
225 65
Total
Here N = 225,fiy; =65, f,y = 693

Varly)x-Ex
65
x 693
225 225

= 2.996.

.. S.D of y, o, = V2.996 = 1.731

SZATISTICS.:ITS INTRODUCTION &SOME MEASURES 251

5
L731 x 5= 8.655

. 14 The mean and standard deviation of marks of 70 students

be 65 and 5.2respectively. Later it was detected that
werfound to 58.
he marks of one student was wrongly recorded as 85 instead of
Obtain the correct s.d.
marks.
Let x.,3: be the
The "incorrect;"=65x70= 4550.

. the "correct"=4550-85+58 = 4523

4523
the 'correct mean =64-61.
70

We know (sd)' =--3 or, Sx'-nla +*).

n

Again the "incorrect x"-70(5-2) +(65) =297642-8.

=293781-8
I henthe correct )=297642-8- 85 +58
293781-8
Hence the correct (sa' -(64-61) =22-43
70

So the correct s.d= J22.43 =4.74

similar products are such
Ex. [Link] factories A, B, C producing with a
Ais Rs. 100and
that the mean daily wage of workers of factory s.d
mean wage is Rs. 150
s.d of Rs. 10, whereas in factory B, the is Rs. 150and s.d
is Rs.
is Rs. 12 and in factory C, the mean wage
in respect of
the daily wage of
10. Which is most consistent
their workersfactory
? 10
unit-mean =I00 =0:|.
The s.d of workers of factory Aper 12
=08
unit-mean =I50
The s.d of workers of factory B per
10
- 0-07.
unit-mean
I50
The s.d of workers of factory C per
workers of factory Cis most
This shows that the daily fuctory A.
wage of the

Consistan. Variability is highestfor