CHAPTER

ij 1

INFERENTIAL STATISTICS

21.1 POPULATION AND SAMPLE

If a grain merchant wishes to buy wheat from a farmer, he assesses the quality of wheat by
taking a handful of it from a bag and then decides to purchase or not. Wheat in the bag is
considered as a population and handful of wheat taken for inspection is called a sample.

w
Similarly, in a factory producing electric lamps, some lamps are picked up randomly to check
the quality by the quality control department. Lamps chosen for inspection form a sample and

F lo
the totality of the lamps manufactured is called the population.
Tlie following is the formal statistical definition of population.

ee
POPULATION The population is an aggregate of objects, animate or inanimate, under stud}/.

Fr
Objects (animate or inanimate) in the population are also called statistical individuals.
The population may be finite or infinite according as the number of objects in it is finite or
for
infinite.
ur
In the above discussion wheat in the wheat bag form an infinite population whereas the totality
of electric lamps manufactured by the factory over a period of time form a finite population.
s
ook
Yo

If the population is infinite or the number of statistical individuals (objects) in the population is
eB

extremely large, then for any statistical investigation complete enumeration of the population is
not possible. Even if the population is finite, complete enumeration is impracticable because of
administrative and financial implications, time factor etc. So, for any statistical investigation, we
our
ad

take the help of sampling. Sampling is quite often used in our day-to-day practical life.
SAMPLE A finite subset of statistical individuals (objects) in a population is called a sample.
SAMPLE SIZE The number of statistical individuals (objects) in a sample is called the sample size.
Y
Re

The process of selecting samples from a population is called sampling. The purpose of sampling
nd

is to draw inference about the population by examining the sample. For the purpose of
Fi

determining population characteristics, instead of enumerating the entire population, the

individuals in the sample are observed to determine sample characteristics. These sample
characteristics are used to estimate the population characteristics. For example, by examining
some samples of certain items, we may arrive at a decision of accepting or rejecting those items.
However, the acceptance or rejection based on the characteristic s of the sample or samples give
rise to an error called the sampling [Link] sampling error is inherent and unavoidable in any
and every sampling scheme.
21.2 PARAMETERS AND STATISTICS

PARAMETERS The Statistical constants or measures of the population, like mean(\.i), variance (g^) etc.
are called the parameters of the pjopulation.
STATISTICS The statistical measures or constants computed from the sample observatio7is alone, like
mean (X), variance (s ) etc. are called statistics.
21.2 APPLIED MATHEMATICS-XII

<y
In practice, values of population parameters viz. p, cr etc. are not known and the corresponding
values of statistics obtained from samples are used for the analysis of the population. However,
the statistics based on different samples can vary from one sample to another sample. One of the
fundamental problems of sampling theory is to find out whether these variations in the statistic
obtained from different samples are significant or insignificant.
21.2.1 SAMPLING DISTRIBUTION

If we select a number of independent random samples of a definite size from a given population
and calculate some statistic (like mean, mode, median, standard deviation etc) from each
sample, we shall get a series of values of the statistic. These values obtained from the different
samples can be put in the form of a frequency distribution as given below.

Sample Number 1 2 3 4

w
n

Statistic (say mean) X1 ^2 ^3 ^3 X,

Flo
The distribution so formed of all possible values of a statistic is called the sampling distribution
or the probability distribution of that statistic.

e
Thus, if we draw 150 random samples from a given population and calculate their means, we

re
shall get a series of 150 means which would form a frequency distribution. This distribution is

rF
called the sampling distribution of the means.
In general, if , $2, S3 3re values of a statistic S (like mean, variance etc.) obtained from
ur
n independent random samples of a definite size chosen from a_ given population, then
fo
S^, $2, S3,..., form a sampling distribution of statistic [Link] mean(S) and variance of statistic
ks
Sare given by
Yo
1 1
S = -n : ^ Sj and Var (S) Z (Si-s)^
oo

n
j = l j = l
B

STANDARD ERROR (S.E.) The standard deviation of the sampling distribution of a statistic is known as
re

its standard error.

u
ad

21.3 STATISTICALINFERENCES
Yo

The principal objective of statistical inference is to draw inferences about a population

parameters from the analysis of a sample drawn from that population. Statistical inference
consists of two major areas (i) estimation and (ii) testing of hypothesis. We shall study about
d
Re

testing of hypothesis as the theory of estimation is beyond the scope of this book. Hypothesis
in

testing begins with an assumption called null hypothesis that we make about a population
F

parameter. The null hypothesis asserts that there is no significant difference in the sample
statistic and the corresponding population parameter or between two sample statistics. It is a
hypothesis of no difference. Null hypothesis is usually denoted by Hq. In case of a single
statistic, Hq will be that the sample statistic does not differ significantly from the hypothetical
parameter value and in the case of two sample statistics, Hq will be that the sample statistics do
not differ significantly.
Having set up the null hypothesis, we compute the probability P that the deviation between the
observed sample statistic and the hypothetical parameter value might have occurred due to
fluctuations or sampling. If the deviation comes out to be significant (as measured by a test of
significance) null hypothesis is rejected at the particular level of significance adopted and if the
deviation is not significant, null hypothesis may be retained at that level.
Any hypothesis which is complementary to the null hypothesis is called an alternate
hypothesis, usually denoted by Hj. For example, if we want to test the null hypothesis that the
population has a specified mean p q (say) i.e. Hq : p = p q , then the alternative hypothesis could
be:
INFERENTIAL STATISTICS 21.3

(i) Hi :p (ii) Hj :m >Po (iii) Hi :|i <|iQ

The alternative hypothesis in (i) is known as a two-tailed alternative and alternative in (ii) and
(iii) are known as right-tailed and left-tailed alternatives respectively. The setting of alternative
hypothesis is very important since it enables us to decide whether we have to use a single-tailed
(right or left) or two-tailed test.
21.3.1 SIGNIFICANCE LEVEL, CONFIDENCE LEVEL AND CONFIDENCE INTERVAL

The significance level also called the alpha level is a term used to test a hypothesis as defined
below.
SIGNIFICANCE LEVEL
In a hypothesis test, the significance level, a, is the probability of making wrong
decision xohen the null hypothesis is true.
For example, significance level 0.05 indicates a 5% risk of concluding that a difference exists
(between the population parameter and sample statistic or between statistics of two samples)
when there is no actual difference. In other words, if we use the same sampling method to such

w
different samples, then 5% of the samples drawn do not include the population parameter.

F lo
Thus, significance level a means that 100a% of the samples drawn do not include the
population parameter and so (100 - 100a) % = 100 (1 - a) % of the samples drawn include the
population parameter.

e
Fre
To graph a significance level of 0.05 in a two-tailed distribution, we need to shade the 5% of the
distribution, that is furthest away from the null hypothesis. In the following graph two shaded
areas are equidistant from the null hypothesis value and each has a probability of 0.025, for a
for
total of 0.05. These shaded areas are called the critical region for a two-tailed test. The critical
region defines how far away our sample statistic must be from the null hypothesis value before
r
You

we can say it is unusual enough to reject the null hypothesis. If sample statistic falls within the
s
ook

critical region, it indicates that it is statistically significant at the 0.05 level. In Fig. 21.1, the
sample statistic does not fall within the critical region representing 0.01 level of significance. So,
eB

we cannot reject the null hypothesis at 0.01 level of significance. This comparison tells us why
we need to choose the significance level before accepting or rejecting the null hypothesis.
our
ad dY
Re

|.i (Null hypothesis

Fin

Sample statistic
value)
Fig. 21.1

CONFIDENCE LEVEL A confidence level refers to the percentage of all possible samples that can be
expected to include the true population parameter.
For example, a 95% confidence level means that in 95% of the samples drawn include the
population parameter and remaining 5% samples drawn do not include the population
parameter.
Clearly, Confidence level = 1 - Significance level
CONFIDENCE INTERVAL A confidence interval is a range that could be expected to contain the
population parameter of interest.
Confidence intervals are intrinsically connected to confidence levels. When we say that
confidence interval for population parameter is [a, b] with a 95% confidence level, it means that
in 95% of the samples drawn the population parameter falls within the confidence interval.
21.4 APPLIED MATHEMATICS-Xil

21.4 STUDENT'S f-DISTRIBUTlON

In 1905 Sir William Gossett gave a test popularly known as Mest. Gossett was employed by the
Guiness Brewery in Dublin, Ireland, which did not permit employees to publish research
findings under their own name. So, Gossett adopted the pen-name 'Student' and published his
findings under this name. Thereafter, the f-distribution is commonly called Student's
f-distribution or simply Student's distribution.
The ^distribution is used when sample size is 30 or less and the population standard deviation
is unknown.

Let .Vj, .\'2,..., x„ be a random sample of size n from a normal population with mean and variance a"^.
Then Student's t is defined b^ the statistic
f =
X-p
s/Vm

w
where X = - ^ Xj is the sample mean and — S
n-^
unbiased estimate of

F lo
n
/ = ! /=!

population variance.
It follows Student's /-distribution with v =(??-!) degrees of freedom with probability density

ree
function

F
v+r

f{t)=C 1+- 2 - 00 <f < OD. for

V
ur
where C is a constant required to make the area under the curve equal to unity.
s

REMARK 1 The number of degrees offreedom of a statistic is the number of independent varieties used to
ook
Yo

compute that statistic. For example, if a small sample has n observations with m constraints on these
values i.e. m values are already available, then the number of degrees offreedom is v = n-m.
eB

REMARK 2 //.Vj, are n obseivations_ in a sample, then for computing sample mean X, we use
all the values Xi, X2,...,x„.Therefore, the mean X has n degrees offreedom. Since the standard deviation of
r

the sample depends on the mean, therefore the standard deviation has (n -1) degrees offreedom.
ad
ou

GRAPH OF t-DISTRIBUTlON The probability density function of a random variable X folloiving

t-distribution is given by
Y

v+ T
Re

2^
nd

t . 2 ,
m=c i-i-- ,-CO </ <00
Fi

We observe that/(-/) =/(/), so the probability curve is symmetric about the line / = 0.
As f —»■ CO, /(f) 0 rapidly. So, f-axis is an asymptote of the curve.
The curve resembless the standard normal probability curve and is bell shaped as shown in
Fig. 21.2.
Standard normal m
distribution \
j /-distribution for y = 15
2
/-distribution for i’ = 8
3
3 3
●2-
2 1
1 O

Fig. 21.2 t-distribution

INFERENTIAL STATISTICS 21.5

As the number of degrees of freedom increases, the f-distribution curve moves closer to the
standard normal probability curve.
PROPERTIES OF f-DlSTRiBUTlON (i) The variable t of t-distribiition ranges from - co fo oo.
(ii) The probability curve is symmetric about the line f = 0 and it resembles with the standard
normal probability curve and is bell shaped as shoW' i in Fig. 21.2. As tlie number of degrees of
freedom increases, the f-distribution curve moves closer to the standard normal probability

ow
curve,

(iii) The variance of f-distribution is greater than one, but approaches one as the number of
degrees of freedom and therefore the sample size becomes large.
The values of t- distribution have been tabulated extensively. The values of fy(a) (two tailed)

e
have been tabulated for a =0.10, 0.05, 0.025, 0.01, 0.005 and v=l, 2, 3,..., 29, 30 where a) is

re
such that the area to its right under the curve off-distribution with v degrees of freedom is equal

Flr
to a (see Fig 21.3). That is i^ia) is such that if the random variable f has f-distribution with v

F
degrees of freedom, then
P(|fl>fv(a))=a
ou
sr
The table does not contain values of fy(ct) for a>0.50, because 4(1-a) =-fy(ct) as the

fo
probability density function/(f) is symmetrical about f = 0. When vis more than 30 probabilities
k
related to f-distribution are usually approximated witli the use of normal distributions.
oo
Y

Rejection Rejection
reB

Accep ance
region a/2 region (1-a) region a/2
uY

0
-fy(a) fi,(a)
Fig. 21.3 Critical values of f-distribution
ad
do

CONFIDENCE OF FIDUCIAL INTERVALS FOR POPULATION MEANp 7/4(0.05) is the tabulated value of t
in

for v = ?j -1 degrees offreedom at 5% level of significance, then

Re

f \ f \
P |f|> 4(0.05) =0.05 =>P I fl <4(0.05) =0.95
F

Thus, the confidence interval for population mean p with a 95% confidence level is given by
I f| <4(0-05)
X-p
<4 (0.05)
S/4n

p-X
<4(0.05)
Sf-sln

|p-Xl<^4 (0-05)
X -A
sfn
fy (0.05) < p < X + 4= fy (0.05)
n
 21.6
APPLIED MATHEMATICS-Xn

Hence, the confidence interval with a 95% confidence level or at 5% level of significance is
X-A^^(0.05), X+^
v»
fv(0.05) , where is the tabulated value of t for v=(»-l)
degrees of freedom.
Similarly, the confidence interval with a 99% confidence level or at 1% level of significance is

^ ~^^v(0-01)/ ^ +^fv(0.01) , where tv([Link]) is the tabulated value of t for v=(n-l)

degrees of freedom.

If instead of S the sample variance s = — ^ (x; -X)^ is given then confidence interval at 5%

ow
level of significance X -
=^fv(0.05),
y/fJ-l X +-^== (0.05) . Similarly, the confidence
intervals at 1% level of significance is X - = fv(0-0i), x + ^i=t^(0.0i).

e
Vn-1 Vn-1

re
Frl
21.5 APPLICATIONS OF f-DISTRiBUTION

F
The f-distribution has a wide range of applications in Statistics.
In this chapter, we will discuss the following:
ou
or
(i) To test if the sample mean (X) differs significantly from the hypothetical value p of the
population mean,
kfs
(ii) To test the significance of the difference between two sample means.
oo
21.5.1 TEST OF SIGNIFICANCE OF THE DIFFERENCE BETWEEN MEAN OF A RANDOM SAMPLE
AND POPULATION MEAN
Y
B

In order to determine whether the meanXof a small random sample X2, X3,...,x„ drawn
from a normal population deviates significantly from the hypothetical value p of the population
re

mean, when standard deviation ct of the population is known, we proceed as follows.

De/me Null hypothesis Hq : There is no signiifcant difference between the sample mean X and the
oYu

STEP!
ad

population mean p i.e. The sample has been drawn from the population with mean p
STEP II Define the statistic
d

t =
X-p
in

S/ffn
Re
F

where,
1 ^
^ sample mean, p = Hypothetical mean of the population, n = Sample size

2
1 If
n-1 n
i=l

^2
2 1
■, di = Xj - A, A is the assumed mean.
n \

Sample statistic follows t-distribution with v = (n-l) degrees offreedom.

sicpjn Find (f\ and tabulated value oft with (n-l) degrees offreedom at a certain level of significa nee.
INFERENTIAL STATISTICS 21.7

If calculated | f | < tabulated (0.05), the null hypotl^sis Hq may be accepted at 0.05 level of
significance and we say that the difference between X and p is not significant and hence the
sample might have been drawn from a population with mean p.
If calculated 111 < tabulated t,,_i (0.01), the null hypoUiesis Hq may be accepted at 0.01 level of
significance and we say that the difference between X and p is not significant.
If calculated \t\> tabulatedf„_i (0.05), the null hypothesis Hq may be rejected at 0.05 level of
significance and we say that the difference between X and p is significant at 5% level.
If calculated | f | > tabulated (0.01), the null hypothesis Hq may be rejected at 0.01 level of
significance and we say that the difference between X and p is significant at 1% level.

ow
REMARK Let s^be the sample variance, then

n
i=l

e
n
=>

re
n-l

rFl
,=i

F
n
s'^ =S^, where is the unbiased estimate of population variance.
H-1

S2

r
s S
ou
n-\ n yjn-\ 4n
fo
ks
X-p X-p
t = => f =
S/i^ s/^Jn-i
oo

Thus, we have
Y
B

X-p X-p
t = or, t =
S/i^ s/^ln-1
re

1
where, S^=^— Y (x'-X)^ and
ou
Y
ad

n ■
/=1 i=l

ILLUSTRATIVE EXAMPLES
d

[iXAMPLE 1 A company has been producing steel tubes of mean inner diameter of 2 cm. A sample of 10
in
Re

tubes gives an inner diameter of 2.01 cm and a variance of 0.004 ctn^. Is the difference in the values of
F

itieans significant? (Given fg (0.05) = 2.262)

SOLUTION It is given that:
p = Population mean = 2 cm, X = Sample mean = 2.01 cm
«= Sample size = 10 and. = Sample variance = 0.004 cm^
We define.
Null Hypothesis Hq: The difference between the sample mean X and the population mean p is
not significant.
Alternate hypothesis H.y The difference between the sample mean X and the population mean p
is significant.
Let t be the test statistic given by
X-p (X - p) ^^n^
t = or, t =
s/fn-1 s
 21.8
APPLIED MATHEMATICS-Xl!

t =
(2.01 - 2.00) 0.01 X 3 10
— = 0.476
V0.004 0.063 21

The test staristic't' follows Student's f-distribution with (10 -1) = 9 degrees of freedom. We shall
now
compare this calculated value with of [ / [ the tabulated value of t for 9 degrees of freedom
and at a certain level of significance. It is given that f9(0.05) =2.262.
Clearly, 111 </9 (0.05) i.e. calculated 111 < tabulated U) (0.05).
So, the null hypothesis Hq is accepted at 5% level of significance. Hence, the difference in the
values of sample mean and population mean is not significant.
EXAMPLE; 2 A machinist is making engine parts with axle cliaf7ieter of 0.7 inch. A random sample of 10
parts shows mean diameter 0.742 inch with a standard deviation of 0.04 inch. On the basis of this sample,
would you say that the zuork is inferior? (Given (0.05) = 2.262)

ow
SOLUTION It is given that:
|j = Population mean = 0.7, X = Sample mean = 0.742
n = Sample size = 10 and. s = Sample standard deviation = 0.04
We define.

e
Fl
re
Null Hypothesis Hq: There is no significant difference between sample mean X and the
population mean p or, the product is not inferior.

F
Alternate hypothesis Hp The difference beh\'een the sample meanXand the population meanp is
ur
r
significant i.e. p X or the product is inferior.
Let t be the test statistic given by fo
ks
t = or, t =
Yo
sZ/tt-l s
oo

(0.742-0.7) 0.042 126

VT^ = X 3 = = 3.15
B

0.04 0.04 40
e

The test statistic't' follows Student's f-distribution with (10 -1) = 9 degrees of freedom. We shall
ur

now
compare this calculated value with the tabulated value of t for 9 degrees of freedom and at a
certain level of significance. It is given that tg (0.05) = 2.262.
ad
Yo

We observe that

|f| = 3.15>2.262 =t9 (0.05)

d

Calculated | f | > Tabulated tg (0.05)

Re

I.e.
in

So, the null hypothesis is rejected at 5% level of significance or the alternative hypothesis is
F

accepted at 5% level of significance. Hence, sample mean X differs significantly from population
mean p i.e. the work is inferior.

nXAMl'I.t 3 A soap manufacturing company was distributing a particular brand of a soap throng}, 2 a
large number of retail shops. Before a heavy advertisement campaign, the mean sales per week per shop
zvas 140 dozens. After the campaign a sample of 26 sfiops was taken and mean sales was found to be
147 dozens with standard deviation 16. Can you consider the advertisement effective'^ (Given
t25 (0.05) =2.06)
SOLUTION It is given that
p = Population mean = 140, X - Sample mean = 147
n = Sample size = 26 s =16.

We define.

Null Hypothesis Hq: There is no significant difference in the mean sales before and after
advertisement.
K'JFERENTIAL STATISTICS 21.9

Alteruate hypothesis There is significant difference in the mean sales before and after
advertisement.

Let t be the test statistic given by

X-^i X-M
t = or, t =
S/i^ s/^ln-1

t = (l^^V2iPr=^.5=^
16 16 16
= 2.187
The sample statistic't' follows student's f-distribution with v = (26-1) = 25 degrees of freedom.
We shall now compare this calculated value with the tabulated value of t for 25 degrees of

ow
freedom and at a certain level of significance. It is given that ^25 (0.05) = 2.06.
We observe that

|f|=2.187>2.06=f25 (0.05)
i.e.
Calculated | f | > Tabulated ^25 (0.05)

e
re
rFl
So, we reject the null hypothesis and accept the alternate hypothesis. Hence, we conclude that
advertisement is effective for sales.

F
i-;[Link] !● ; A random sample of size 16 has 53 as mean. The sum of the squares of the deviations taken
from mean is 150. Can this sample be regarded as taken from the population having 56 as mean? (Given

r
(0.01) = 2.95)
fo
ou
SOLUTION We have,
ks
n = Sample size = 16, X = Sample mean = 53, |i = Population mean - 56
oo

16

and, ^ (x, -X)^ =150, where Xj, X2, ^re sample observations
Y
eB

/ = 1

1 16 _ o 1
S2 = Y (.^i-X)^=—Xl50=10
ur

(16-1) ^ 15 i=l
ad
Yo

We define,

Null Hypothesis Hq: The sample is drawn from tlie population having 56 as mean.
d
Re
in

Alternate hypothesis Hp The sample is not drawn from the population having 56 as mean.
Let the sample statistic f be given by
F

X -p 53-56 -3x4 12V1O -12x3.162

t - f = = -3.794
S/4n ViO/16 VlO 10 10

The sample statistic follows Student's f-distribution with v = (16 -1) = 15 degrees of freedom.
We shall now compare this calculated value with the tabulated value of f for 15 degrees of
freedom at a certain level of significance. It is given that t^^ (0.01) = 2.95.
Calculated\t\ = 3.794 >2.95 = (0.01)
i.e. Calculated | f | > Tabulated (0.01)
So, we reject the null hypothesis. Consequently, the alternate hypothesis is accepted at 0.01 level
of significance. Hence, the sample is not taken from the population having 56 as mean.
[Link]; s A random sample of 17 values from a normal population has a mean of 105 cm and the sum of
the squares of deviations fi-om this mean is 1225 cm^. Is the assumption of a mean of 110 cm for the normal
populatio)! reasonable? Test under 5% and 1% levels of significance. Also, obtain the 95% and 99%
confidence limits. (Given (0.05) = 2.12 and (0.01) = 2.921).
21.10 APPLIED MATHEMATICS-XII

SOLUTION We have,
|i = Population mean = 110, X = Sample mean = 105
17 _ „
n = Sample size = 17 and. £ (Xi -X)2 =1225.
/=1

5^=1” 1 £
=1

= 72.0588 => s = V72.0588 = 8.4887

17

We define.

w
Null Hypothesis Hq: There is no significant difference between sample mean and population
mean i.e. assumption that mean of the population is 110 cm is valid.
Alternate hypothesis Hj: Assumption that mean of the population is 110 cm is not valid.

Flo
Let t be the test statistic given by

e
X-p 105-110 -5x4
=> t = xVi^ = = -2.3561

re
s/Vn-1 8.4887 8.4887

F
I f| =2.3561
The sample statistic follows Student's f-distribution with v = (17 -1) =16 degrees of freedom.
ur
r
fo
We shall now compare this calculated value with the tabulated value of t for 16 degrees of
freedom at 5% and 1% levels of significance.
ks
At 5% level of significance: It is given that (0.05) = 2.12. We find that
Yo

Calculated | f | = 2.3561 > 2.12 = (0.05)

oo

i.e.
Calculated | f | > Tabulated fjg (0.05)
B

So, we reject the null hypothesis at 5% level of significance. Hence, the assumption that the
re

population has mean 110 cm is not correct.

The confidence limits at 5% level of significance are
u
ad
Yo

X-
VM—1
(0-05) and X +
^*>6(0.05)
8.4887 8.4887
d

or. 105- x2.12 and 105 + x2.12

Re

4 4
in

or. 105 - 4.499 = 100501 and 105 + 4.499 = 109.499

F

The confidence interval is [100501,109.499]

At 1% level of significance: It is given that (0.01) = 2.921.
Clearly, calculated 111 < tabulated (0.01).
So, we accept the null hypothesis at 1% level of significance. Hence, the assumption that the
mean of the population is 110 cm is valid.
The confidence limits at 1% level of significance are
X -
(is (0.01) and X + (16(0.01)
8.4887 8.4887
or. 105- X 2.921 and 105 + X 2.921
4 4

or. 105-6.199=98.801 and 105 + 6.199=111.199

The confidence interval at 1% level of significance or at 99% confidence level is [98.801,111.199].

INFERENTiAL STATISTICS 21.11

Ten students are selected at random from a college and their heights arefound to be 100,104,
●Mi’i.l-i'

108,110,118,120,122,124,126 and 128 cms. In the light of these data, discuss the suggestion that the
mean height of the students of the college is 110 cms (Given (0.05) = 2.262).
SOLUTION We define

Null Hypothesis Hq: There is no significant difference between the sample mean and hypothe
tical population mean 110 cm.
Alternate hypothesis Hj: The sample mean is not same as the population mean.
Let the sample statistic t be given by
t =
X-p

ow
S/-Jn
Let us now compute the sample mean (X) and S.
Computation ofX and S

.r,--X

e
X;
!

re
100

Fl -16 256

F
104 -12 144
ur
108 -8 64

r
110 -6 fo 36
ks
118 2 4
Yo
120 4 16
oo

122 6 36
eB

124 64

126 10 100
ur

128 12 144
ad

10
Yo

ZXf =1160 £ (x--Xf=864

/=!
d
Re
in

X-M
t =
S/4n
F

116-110
=> f = X Vl0=-^x 3.162=1.94 [v 1.1 =110]
9.798 9.798
10

We find that ^ = 1160, and n -10.

j = 1

1 10 1160
X=- = 116
10
” /=1
10

From the table, we find that ^ (Xj -X)^ = 864

j = l

1 10 1 1 29.393
=
£ (Xi-X)2=>s2=-x864^S = 3 3
= 9.798
^1 = 1 ^
21.12 APPLIED MATHEMATICS-XI!

Tlie sample statistic follows student’s f-distribution withv =(10 -1) = 9 degrees of freedom. We
shall now compare tliis calculated value with the tabulated value of t for 9 degrees of freedom at
a certain level of significance. It is given that (0.05) = 2.262.
Calculated 11 \ =1.94 < 2.262 = tg (0.05)
i.e.
Calculated | f | < tabulated tg (0.05)
So, we accept the null hypothesis. Hence, the sample mean is same as the population mean.
Consequently, the mean height of the students of the college is 110 cm.
EXAMPLE: 7 A random sample of 10 boys had the following LQ's: 70,120,110,101,88,83,95,98,107,
100. Do these data support the assumption of a population mean I.Q. of 100? Find a reasonable range in

ow
which most of the mean I.Q. values of samples of 10 boys tie. (Given tg (0.05) = 2.262)
SOLUTION We have,

p = Population mean = 100, n = Sample size = 10

We define

e
Null Hypothesis Hq: Tlie data are consistent with the assumption of a mean I.Q. of 100 in the

re
population.

rFl
F
Alternate hypothesis H.p Tlie mean I.Q. of population ^ 100.
Let the sample statistic t be given by

r
ou
«

t =
X-p
S/fiT
, where -
1
Z (.V, -X)^ fo
ks
n-l ■
/ = 1

Let us now compute X and S^.

oo

Computation ofX and S

Y
eB

X;
I dj = Xj - 90
r

70 -20 400
ou
ad
Y

120 30 900

no 20 400
d

101 11 121
Re
in

88 -2 4
F

83 -7 49

95 5 25

98 8 64

107 17 289

100 10 100

[Link] =72 Z di^ = 2352

Here, if =Xj - 90
1 1
X=90 +
10
'£f=90 + —
10
= 972 Using:X=A+-J]4 n

(72)2
=1^ 2352-
1833.6
S2 = = 203.73
10 9
INFERENTIAL STATISTICS 21.13

X 97.2-100 -2.8 -2.8

t = ^=>t = = -0.62
S/4ii 203.73 "V2037 4514
10

=> \t\= 0.62

The sample statistic t follows student's f-distribiition with v = (10 -1) = 9 degrees of freedom. It is
given thatf9 (0.05) = 2.262.
Calculated\t\< tabulated tg (0.05)
So, the null hypothesis may be accepted at 5% level of significance.
Hence, the assumption of a population mean I.Q. of 100 is valid.
The 95% confidence limits within which the mean I.Q. values of samples of 10 boys will lie are

X
Jn
fq (0.05) and X + -^
V/i
to (0.05)

w
203.73 20373
X 2.262 and 97.2 + X 2.262
V 10
or, 97.2-
V 10
or.

or,
97.2 - 4514 X 2.262 and 97.2 + 4514x 2.262
97.2-10.21 and 97.2 + 10.21

F lo
e
Fre
or. 86.99 and 107.41

Hence, the required 95% confidence interval is [86.99,107.41]. for EXERCISE 21.1
r
1. Ten cartons are taken at random from an automatic filling machine. The mean net weight of
You

the cartons is 11.8 kg and the standard deviation 0.15 kg. Does the sample mean differ
oks

significantly from the intended weight of 12 kg? (Given fq (0.05) = 2.262).

eBo

2. A machine is designed to produce insulating washers for electrical devices of average

thickness of 0.025 cm. A random sample of 10 washers was found to have an average
thickness of 0.024 cm with a standard deviation of 0.002 cm. Test the significance of the
our
ad

deviation. (Given fq (0.05) = 2.262).

3. A random sample of size 25 from a normal population has the mean 47.5 and standard
deviation 8.4. Does this information refute the claim that the mean of the population is 42%.
dY
Re

(Given f24 (0.05) = 2.06).

4. A process of marking certain bearings is under control if the diameter of the bearings have
Fin

the mean 0.5 cm. What can we say about tliis process if a sample of 10 of these bearings has a
mean diameter of 0.506 cm and standard deviation of 0.004 cm? (Given fq (0.05) = 2.262).
5. A machine is supposed to produce washers of mean thickness 0.12 cm. A sample of 10
washers was found to have a mean thickness of 0.128 and standard deviation 0.008. Test
whether the machine is working in proper order at 5% level of significance. (Given
fq (0.05) =2.262).
6. A random sample of 16 values from a normal population showed a mean of 41.5 and sum of
squares of deviations from mean equal to 135. Can it be assumed that the mean of the
population is 43.5? (Given fj5 (0.01) = 2.95).
7. A sample of size 9 from a normal population X = 15.8 and = 10.3. Find 99% confidence
interval for population mean. (Given fg (0.01) = 3.335).
 21.14
APPLIED MATHEMATICS-XII

8. A random sample of size 16 has 53 as mean. The sum of the squares of deviations taken
from mean is 150. Find 95% and 99% confidence intervals for population mean. (Given
tl5 (0.01) = 2.95 and (0.05) = 2.13).
9. A random sample of 16 values from a normal population showed a mean of 41.5 inches and
the sum of squares of deviations from this mean equal to 135 square inches. Show that the
assumption of a mean of 43.5 inches for the population is not reasonable. Obtain 95% and
99% confidence intervals for the same. (Given (0.05) = 2.131 and (0.01) = 2.947).
1 (). The annual rainfall at a certain place is normally distributed with mean 45 cm. The rainfall
during the last five years are 48 cm, 42, cm, 40 cm, 44 cm and 43 cm. Can we conclude that
the average rainfall during the last five years is less than the normal rainfall? (Given
f4 (0.05) = 2.132).
r ■. The heights of 8 males participating in an athletic championship are found to be 175 cm, 168
cm, 165 cm, 170 cm, 167 cm, 160 cm, 173 cm and 168 cm. Can we conclude that the average

w
height is greater than 165 cm? (Given tj (0.05) = 1.895).
\ 2. The mean weekly sales of chocolate bar in general stores was 146.3 bars per store. After an

F lo
advertising the mean weekly sales in 22 stores for typical week increased to 153.7 bars and
showed a standard deviation of 17.2. Was the advertising campaign successful? (Given
/2i (0.05) =2.08).

ee
13. The foreman of ABC mining company has estimated the average quantity of iron ore

Fr
extracted to be 36.8 tonnes per shift and the sample standard deviation to be 2.8 tonnes per
for
shift, based upon a random selection of 4 shifts. Consider a 90% confidence interval around
this estimate. (Given tg (0.1) = 2.353).
ur
14. A random sample of size 20 from normal population gives a sample mean of 42 and
s

standard deviation of 6. Test the hypothesis that the population mean is 44. (Given
ook
Yo

fl9 (0.05) = 2.09).

eB

1 - . A random sample of 10 boys had the following I.Q.s.: 70,120,110,101,88,83,95,93,107,100

Do these data support the assumption of a population mean I.Q.s of 100? (Given
fg (0.05) = 2.262). Also, find the 95% confidence interval for the population mean.
our
ad

1 f,. The manufacturer of a certain make of electric bulbs claims that his bulbs have a mean life
of 25 months with a standard deviation of 5 months. A random sample of 12 such bulbs
gave the following values:
Y
Re

Life in months: 24 26 32 28 20 18 23 27 29 34 20 28
nd

Can you regard the producer's claim to be valid at 1% level of significance?

Fi

_ANSWERS
1. Yes 2. Not significant Yes 4, Process is not under control
5. No 6. Yes 7 [11.99,19.61]
8. [51.32,54.68], [50.67,55.33] 9. [39.902, 43.098], [39.29, 43.71]
10. No ■1. Yes 12. Yes )2. [34.5,39.10]
14. True 15. Yes; [87.494,107.906] 16. Yes

21.5.2 TEST OF SIGNIFICANCE OF THE DIFFERENCE BETWEEN THE MEANS OF TWO

RANDOM SAMPLES (INDEPENDENT SAMPLES)
Suppose two independent random samples .Vj, X2,...x,, and t/j, \/2'--^¥n2 sizes and ;i2
respectively have been drawn from the same normal population. Let and X2 be the means
and let and S2 be their standard deviations. In order to test whether the two sample means X 1
and X2 differ significantly or not, we calculate the statistic t given by
INFERENTIAL STATISTICS 21.15

t =
X1-X2 or, t = Xi-X2^. »1»2
1 1 S fl-j + II2
S I- +
V»1 «2

, where S =
+ »2S2^
^ n-i + »2 - 2
The statistic t follows i-distribution with v = +112 - 2 degrees of freedom.

KLMAHK Wc/jrtW —"li ^= l (Xj -X-if rtJirf S2^ ^£(./,--X2)^

”2/=]

ow
II'
"1

Z (-Vi-Xi)" + IVi-Xlf
1 =1 i =l

+«2“2

e
are taken from assumed means. In such cases, S is

re
When the actual means are infraction the deviations

Frl
given by

F
\2 \2
”1 1 fll "2
1 2 1
S2 = 1
+ I I

”1 +”2 “ 2 ;_i ”1 1=1 n

/ =! ,i = l
ou
;/ i = 1,2,... ,n-[ and dj = i/j - A2 i = 1, 2,..., 112
or
xvhere diI = X;I - A 1
kfs
ILLUSTRATIVE EXAMPLES
oo

EXAMl’Ui ! for the follozoing data examine if the means of two samples differ significantly:
Y

Mean Standard deviation

B

Size
6 40
Sample I:
re

5 50 10
Sample II:
oYu

(Given t^{0.05) =2.262)

ad

SOLUTION We have,
H-j = 6, X| = 40, = 8, ?i2 =5, X2 =50 and $2=10
d

"1^1^ +”2^2^
in
Re

^ H-] + }t2 ~ 2
F

6x 64 + 5x100 884 29.732

S = = 9.910
6 + 5-2 9 3

We define

Null hypothesis Hq : The difference in the means of two samples is not significant.
Alternate hypothesis Hi: Means of two samples differ significantly.
Let f be the sample statistic given by
X1-X2-X «1”2
t = —i
s ill + '^2

t =
40-50 ,, 675 -10 30 -10x 1.651
= -1.666
9.910 ^\6+5~ 9.910 V 11 9.910
The sample statistic t follows f-distribution with v=(6 + 5-2)=9 degrees of freedom. We shall
now compare this calculated value of f with the tabulated value for 9 degrees of freedom at a
given level of significance. It is given that tg (0.05) = 2.262.
 21.16
APPLIED MATKEMATICS-Xn

We find that; calculated 11 \ =1.666 < tabulated fg (0.05). So, we accept the null hypothesis at 5%
level of significance. Hence the difference in sample means is not significant.
‘ AMI-U-,: Tzvo batches of the same product are tested for their mean life. Assuming that the lives of the
product follow a normal distribution with an unknozun variance, test the In/pothesis that the mean life is
the same for both the branches, given the follozuing information:
Batch Sample size Mean life in hrs Standard deviation
Batch I 10 750 12
Batch II 8 820 14
(Use t^(,{0.05)^ 2.2120)
SOLUTION We define

Null In/pothesis Hq : Mean life is same for both the batches.

Alternate hypothesis : Two batches have different mean

w
lives.

The sample statistic t is given by

F lo
t = X
, where S =
s + "2 +772-2

ee
We have.

Fr
77] -10,7?2 - 8, Xj =750, X2 =820,S|=12 and $2=14
2 2
"1^1 +”2^2
^ 77| + 772 ~ ^
for
ur
10x144 + 8x196 3000
S = = VIM =13.711
s
=>
ook

10 + 8-2 16
Yo

750-820 10x8 -70 80 -70 -140

-Vio =
eB

t = X
X 3.162= -10.762
13.711 VlO+8 13.711 V18 13.711 3 41.133

The sample statistic follows Student's f-distribution with v =(77i+?72-2) =16 degrees of
our

freedom. Let us now compare the calculated value of t with the tabulated value of f at a given
ad

level of significance. It is given that fjg(0.05) =21.20.

We observe that Calculated | f| =10.762 > 2.120 = Calculated (0.05). So, we reject the null
Y

hypothesis at 5% level of significance and hence accept the alternate hypothesis. Hence, the
Re
nd

mean life for both the batches is not the same.

Fi

liXAMPM :
Samples oftivo types of electric light bulbs were tested for length of life andfolloiving data
were obtained:

Type I Type II
Sample size 77j = 8 7?2 =7
Sample means Xi = 1234 hrs X2 =1036 hrs
Sample S.D's = 36 hrs S2 = 40 hrs
Is the difference in the means sufficient to zvarrant that type I is superior to type II regarding length of
life? (Given ti3 (0.05) = 2.216)
SOLUTION We define

Null hypothesis Hq : Two types I and II of electric bulbs are identical.

Alternate hypothesis : Type I bulbs are superior to type II bulbs.
Let the sample statistic t be given by
 21.17
INFERENTIAL STATISTICS

2 2

t =
X1-X2 X
»1»2
, where S =
”l^l +''2''^2
s \ }J| + H2 \ + ;i2 - 2
Now,
8x 36^+7x40^ I10368 + 11200 = 1659.076=40.731
^ /i-j + JJ2 ”2 ^ 8+7-2 M 13

1234-1036 8x7 198

t = X 1.932 = 9.391
40.731 8+7 40.731

The sample statistic 't' follows Student's f-distribution with v=(8 + 7~2) = 13 degrees of
freedom. Let us now compare this calculated value with the tabulated value of f at a given level
of significance. It is given that f|3 (0.05) = 2.16.
We find that Calculated | f | = 9.391 > tabulate (0.05). So, the null hypothesis is rejected at 5%

w
level of significance. Hence, the two types of electric bulbs differ significantly. Further, since
is much greater thanX2, we conclude that type 1 bulbs are definitely superior to type 11 bulbs.

Flo
Samples of sales in similar shops in towns A ami B regarding a new product yielded the
following information:

ee
For town A : X-i = 3.45 IX;I = 38 = 228 11-1=11

Fr
For town B: Xi1 =4.44 Zy,-=40 = 222 112 - 9

is there any evidence of difference in sales in the two towns? (Given f-jg (0.05) - 2.10)
for
ur
SOLUTION We define

Niillln/pothesis HQ-.H-xereisno significant difference in sales in similar shops in towns/I and S.

s
Alternate hypothesis : Sales in similar shops in towns A and B are not same.
k
Yo

The sample statistic t is given by

oo

2 2
Xi -X2= niii2
eB

t = —I X , where S =
s + )l2 /?! +^2 -2
We have, = 11, X^ = 3.45, 1.^^- = 38, = 228
r
ou
ad

= 9, X2 = 4.44, Ii/,- = 40, = 222

Y

S^ = + ?l2 ^2^
JI-] + ^2 “ 2
Re
nd

Z(j;-X,)^+I(y,--X2)^ lAr;^-HlXi^ + I.y-K2X2^

Fi

?j-j +112 -2 n^ +112-2

^2 228-ri X (3.45)^ + 222 - 9x (4.44)^ 450 -130.93 -177.42 141.65 = 7.869
® " 11+9-2 18 18

S = V7.869 =2.805
X1 X2 »i»2
f = —

S n^ +112

3.45-4.44 11x9
f = ^ 1 =-0.353 X 2.22 =-0.784
2.805 Vll+9
-2=11+9-2 = 18
The sample statistic 't' follows Student's /-distribution with v=«i +JI2
degrees of freedom. Let us now compare the calculated 11 \ with the tabulated value of / at a
given level of significance. It is given that fjg (0.05) = 2.10. We find that the calculated ] f | is less
than the tabulated fig (0.05).So, the null hypothesis is accepted a 5% level of significance. Hence,
there is no evidence of difference in sales in the two towns.
21.18 APPLIED MATHEMATICS-XII

LXAMiM I 5 Two different types of drugs A and B were tried on certain patients for increasing weight, 5
persons were given drug A and 7 persons were given drug B. The increase in weights in pounds is given
below:

Drug A: 8 12 13 9 3
Drug B: 10 8 12 15 6 8 11

Do the two drugs differ significantly with regard to their effect in increasing weight (Given
ilO (0.05) =2.23) ^ ^
SOLUTION We define

Null hypothesis Hq : Two drugs do not differ significantly with regard to their effect of
increasing weight.
Alternate hypothesis Hj: Two drugs differ significantly with regard to their effect of increasing

ow
weight.
The sample statistic t is given by
X1-X2
|z(^-X,)2 + S(K-X2)2j
«1«2 1
t = —

, where =
S +«2 +ri2-2

e
Fl
re
Let us now compute Xj, X2 / ^ -^1)^ and S (y,- -X2)^.

F
Drug A
ur
DrugB
Xi-X^=Xi-9 (^- -Xi)^ y/ or
yi~^2 -10 (» -X2)2
sf
8 -1 1
k
10 0 0
Yo
oo

12 3 9 8 -2 4
B

13 4 16 12 2 4
re

9 0 0 15 5 25
3 -6 36 6 -4 16
u
ad
Yo

8 -2 4

11 1 1
d

IXi =45 Z(:^-Xi)=0 Z(:ti-Xi)2=62

Re

^Vi=70 2(y,-X2)=0 =54

in
F

1 45
1 ^ 70
— =10
n2 7

S^=
«1 +«2 -2 |z(^;-X,)2 +Z(k.-X2)2|
S2 =
5^1-^(62+5.4)=11.6
S = VlL6 =3.406

f = ~^2
S y«j+n2
9-10 5x7 1 35 17078
f = ^ ^ ^ __ = -0501
3.406 V5 +7 3.406 V12 3.406
 21.19
INFERENTIAL STATISTICS

The sample statistic 7' follows Student’s f-distribution with v =(5 + 7-2) =10 degrees of
freedom. Let us now compare the calculated 11 \ with the tabulated value of t at a given level of
significance. It is given that (0.05) =2.23. We find that the calculated | f | is less than the
tabulated t^Q (0.05) = 2.23. So, the null hypothesis is accepted at 5% level of significance. Hence,
the null hypothesis Hq holds true i.e. the drugs A and B do not differ significantly with regard to
their effect in increasing weight.
RIINIAKI- In the above example and X2 come out to be integral values
_ and hence the direct method of
computing I (.r,- -X^)^ and Z (pj -Xj)^ is used. In case X^ and (or) X2 out to be fractional, then
2
the step deviation method is convenient for computmg S .
EXAMlM I !■ The heights in inches of 6 randomly chosen sailors and 10 randomly chosen soldiers are given

as under:
Sailors: 63 65 68 69 71 72

Soldiers: 61 62 65 66 69 69 70 71 72 73

Do these figures show that the soldiers are on an average shorter than sailors? (Given t-^^ (0.05) = 2.15)

w
We define

F lo
SOLUTION

Null hypothesis Hq : Soldiers are not on average shorter than sailors.

Alternate hypothesis : Soldiers are on an average shorter than sailors.
The sample statistic t is given by

ree
t = X »1»2 ^
for F
s 111 + »2'
n2 n2

f u-)
1 ”1 1 1
r
where = di
You

"2^/ = l
oks

■'■”2 1 = ] "1
eBo

, where dj = Xj - A-i; i =1, 2 iij and dj =yj - Aj }i = 1, 2 U2

— — 2
Let us now compute X^, X2 and S .
ad
our

Sailors Soldiers

X-t i1j=Xi-A-^=Xi-68 Vi 4'=i/,--A2=y,--66

Re
dY

63 -5 25 61 -5 25
Fin

65 -3 9 62 -4 16

68 0 0 65 -1 1

69 1 1 66 0 0

71 3 9 69 3 9

72 4 16 69 3 9

70 4 16

71 5 25

72 6 36

73 7 49

Z^- =0 Zd,-^ =60 ZtV =18 Zd:^ =186

I
21.20
APPLIED MATHEMATICS-XII

Wehave,nj=6 and U2=10.

From the above table, we find that
/4 1
= 6S, A2=66,Tcij =0,Tcij^ =[Link]' =18 and I^^:^=186
1 18
X-[ = /\i H— Z dj = 68 H— = 68, Xt = + I 4' = 66 + = 67.8
6 ^ »2 10

1
S2 =
//-] + ~2 ”1 ”2

=> S2 =
1 0
● 60 — +186
(18)^ 1 213.6
6 + 10-2 6
(60 + 186-32.4) = = 15.2571
10 14 14

ow
S = Vl 5.2571 = 3.9060
f = ^1 ^^2 X
”1»2
s ”l+”2

e
68-67.8 6x10 _ 0.2
t = X 1.9365 = 0.0991

re
3.9060 6 + 10 ~ 3.9060

rFl
F
The sample statistic 't‘ follows Student's f-distribution with v =«-j+«2-2 =(6 +10-2) =14
degrees of freedom. Let us now compare this calculated value of | f | with the tabulated value for
14 degrees of freedom at a given level of significance. It is given that

r
(0.05) = 2.15. Clearly, the
ou
fo
calculated value of|f| is much less than the tabulated value. So, the null hypothesis Hq is
accepted at 5% level of significance. Hence, on an average soldiers are not shorter than sailors.
ks
EXERCISE 21.2
oo

1. The mean life of a sample of 10 electric bulbs was found to be 1456 hours with standard
Y
eB

deviation of 423 hours. A second sample of 17 bulbs chosen from a different batch showed a
mean life of 1280 hours with standard deviation of 398 hours. Is there a significant
difference between the means of the two batches?
ur

2. Strength tests carried out on samples of two yarns spun to the same count gave the
ad

following results:
Yo

Size of sample Smnple mean Sample size Sample Variance

Yarn A
d

4 50 42
Re

Yarn B
in

9 42 56

The strengths are expressed in kg. Is the difference in mean strengths significant of real
F

difference in the mean strengths of the sources from which the samples are drawn?
3. Samples of two types of electric bulbs were tested for length of life and the following data
were obtained :

T\/pe I Type II
Number of bulbs in the sample 8 7
Mean (in hours) 1134 1024
Standard deviation (in hours) 35 40

Is the difference in the sample means significant? (Given (0.05) = 2.16).

4. I.Q. test on tw'o groups of boys and girls gave the following results:
Girls: Xi=78 =50 s, =10
Boys: X2=73 =100 S2=15
Is there a significant difference in the mean scores of boys and girls? (Given ^23 (0.05) = 2.07)
 21.21
INFERENTIALSTATISTICS

5. Two methods of performing a certain operation are compared. The following data are
obtained:
Xj =505 2 =95
n-j =15
»,=12 X2=57.2 S2^=5.7
Is there a significant difference in the means of the two methods at 5% level of significance?
(Given f25 (0.05) = 2.06).
6. The meansof the random samples of sizes 9 and 7 are 196.42 and 198.42 respectively. The
are 26.94 and 18.74
sum of the squares of deviations from the respective means
respectively. Can the samples be considered to have been drawn from the same normal
population?

ow
7. The I.Q.'s (intelligence quotients) of 16 students from one area of a city showed a mean of
107 with a standard deviation of 10 while the I.Q.’s of 14 students from another area of the
city showed a mean of 112 with a standard deviation of 8. Is there a significant difference
between the I.Q.’s of the two groups at (i) 1% and (ii) 5% level of significance?
8. Below are given the gain in weiights (in kg^) of pigs fed on two diets A and B.

e
Gain in weight

re
Diet A: 25 32

Fl
30 34 24 14 32 24 30 31 35 25

F
DietB: 44 34 22 10 47 31 40 30 32 15 18 21 35 39 22
ur
Test, if the two diets differ significantly as regards their effect on increasing in weight

r
((25 (0.05) = 2.06).
9. In a certain experiment to compare two types of animal foods A and
fo B, the following results
ks
of increase in weights were observed in animals;
Yo
oo

2 3 4 5 6 7 8
Animal number 1
47 50 52 53
eB

Food A 49 53 51 52
Increase

weight in lb Food B .52. 55 52 53 50 54.. _ 54 _ 53

Assuming that the two samples of animals are independent, can we conclude that food B is
ur
ad

better than food A?

Yo

lO. The marks obtained by two groups of students in Mathematics test are given below:
Group A Group B
d

15 11
Number of students:
Re
in

42 38
Mean Marks:
15
F

Standard deviation of Marks : 10

On the basis of this data, can it be concluded that there is a significant difference in the
mean marks obtained by two groups? (Given (14 (0.05) = 2.064).
11 . Two kinds of fertilizers were applied to 15 plots of one acre, other conditions remaining the
same. The yields in quintals are given below:
14 20 34 48 32 40 30 44
Fertilizer I:
18 22 28 40 26 45
Fertilizer II: 31

Examine the significance of difference between the mean yields due to the use of difierent
kinds of fertilizers. (Given (^3 (0.05) = 2.16)
12. Two different types of drugs A and B were tried on certain patients for increasing weight. 6
persons were given drug A and 8 persons were given drug B. The increase in weight in
pounds is given below:
7 10 13 12 4
Drug A:
6 18 16 9 3
12 8
Drug A:
21.22
APPLIED MATHEMATICS-XII

Do the drugs differ significantly with regard to their effect i

in
increasing weight? (Given
fl2 (0.05) =2.16).
13. A group of 5 paHents treated with medicine A weigh 42,39,48,60 and 41 kg; second group
of/patients from the same hospital treated with medicine B weigh 38,42,56,61,68,69 and
62 kg. Do you agree with the claim that medicine B increases the weight significantly^
(Givenfjo (0.05) =2.281). & y-
ANSWERS
1. No
-■ Not significant 3. Sigiiificant

ow
4. Significant
>- Significant 6. No
9. Yes
7. Not significant 8. Do not differ significantly
10. Not significant n ● No significant difference
12. No 13. Yes

e
re
F
Frl
ou
sr
kfo
oo
Y
reB
uY
ad
do
in
Re
F