Testing Of iypothesis

1.41
1.2 TESTING OF HYPOTHESIS FOR MEAN AND
DIFFERENCE OF MEANS USING
t-DISTRIBUTION.
1.2 (a) t-test
a. Student's t
Let x1, z, ..n be a random sample of size n from a
normal
population with mean and variance o.
The student's t test is áefined in the
statistics as
t =
S where x = - ) X , is the
Vn ni=1
n
1
sample mean and S = Wi (; - x) is an unbiased estimate of
n=1

the population variance ¡ and t follows student's t-distribution with

v=n- 1 degrees of freedom with probability density function
vt1
2

f)= Yo,

where y =n -1and yo is a constant determined by f f() dt = 1.

This is known as student's t distribution' or t-distribution

Here, v =n -1is called the number of degrees of freedom.
Note : The range of t-distribution is -oo to o
II. Uses and its applications [A.UN/D 2011]
1. It is used to test the significance of the difference of the mean
of a random sample and the mean of the population
2. It is used to test the significance of the difference between two
sample means.
 3.
Testing of Hypothes
To test the
significance of an observed sample corrclation
and sample regression cocfficient. coefficic
4 To test the
significance of observed partial and multiple correlati
coefficients.

Il. Properties of t-distribution

1 The variable ranges from -oo to o

v+1
2
2. In the 1+2|
distribution y () =f() = Yo
where t =
S Sf) dt = 1
Vn Vn - i
Total area under the curve is 1.

3. t-distribution is symmetrical about t = 0 and has a mean zero.

4. The variance of (-distribution is greater than 1, but tends to 1 as
n> o, As v’ o, t-distribution becomes normal.

5. Variance if y> 2 and u>1 always.

V-2

Test for a specified mean

Given a random sample of size n (n < 30) with sample mean
and thc population standard deviation is not known and we want to
test whether the population mean has a specified value, then we apP'Y
the sane procedure as in the case of a large sample.
Testing of Hypothesis 1.43
Working Procedure :
For the small samples (n < 30), a known,
decision is based on
the t-distribution with y=n-1degrees of freedom.
1. Null hypothesis HÍ : 4= Mo
2. Aliternative hypothesis H, :u uo (or) >uo (or) < Mo
3. Level of significance : a
4. Critical region
(a) If u uo, then the test is twO-tailed test for the given a.
The critical values are -t/) and t/) from the t-distribution table
with d.f. = n -1

-ta/2 ta/2

(b) If > u, then the test is one-tail test (right) for the given a
The critical values is t with d.f. = n- 1.

ta/2

(c) If u < un then the test is one-tailed test (left) for the given a.
The critical value is -t, with d.f = n - 1.

-te/2
 1.44

The test statistic

Testing of Hypothesi

Vn - 1
where x ’ mean of the sample
sample size
sample s.d.
6. Conclusion :
(a) If -lal2 <l < lal2 , then we accept Ho;
otherwise,
we reject HÍ
(b) If t <l¡o then accept HÙ,
otherwise, we reject H,
(c) If -t, < I, then accept Ho, otherwise, we
reject Ho
1.2(a) Problems under t-test single mean
Example 1.2a(1).
A sample of ten house owners is drawn and the
following values of
their incomes are obtained, mean Rs. 6,000 standard deviation Rs. 650.
test the hypothesis that the average income of house owners of the
town is Rs. 5,500.
Solution : Given : n = 10 < 30 (So, we can apply (-test)
S= 650, x = 6000, u = 5500
1. HÍ : u= 5500
HË : u 5500 [Two-tailed test]
3 a= 5% with d.f. = n -1 10 - 1 =9
4
Critical region :

2262 =2.262
Testing of Hypothesis 1.45

5. The test statistic

6000 - 5500 S00
= 2.31
S 650 216.67
V10-I

6. Conclusion

If -ta/2<t<la/2, then we accept Ho; otherwise, we reject Ho.

Here, -2.262 < 2.31 2.262
So, we reject Ho.

Example 1.2a(2).
A machinist is making engine parts with axle diameters of 0.7 inch.
A random sample of 10 parts shows a mean diameter of 0.742 inch
with astandard deviation of 0.04 inch. Compute the statistic you would
use to test, whether the work is meeting the specification.
[A.U M/J 2006]
Solution : Given :n= 10, x = 0.742, s = a =0.04, u = 0.7
1. HÍ: u = 0.7

2. H : u 0.7 Two-tailed test]

3. a= 5o, d.f. = n-1 = 10 -1 = 9
4. Critical region :

-ta/2 Lal2
5. =-2.262 =2.262
The test statistic :

0.742 -0.7
= 3.15
0.04
Vn -+ V9
 1.46
Testing of Hypothesi
Conclusion :

If -te/2 <I<la2 then we accept HÍ ; otherwise, we

reject H
Here, -2.262 < 3.15 <2.262. So, we reject HÍ
i.e., The product is not conforming the specification.
Example 1.2a(3)
A certain injection administered to each of 12 patients resulted in th.
following increases of blood pressure :
5, 2, 8, -1, 3, 0, 6, -2, 1, 5, 0, 4
can it be concluded that the injection will be, in general,
accompanied
by an increase in B.P ? [A.U. N/D 20071
SoBution : Given : n = 12 Total
2 -1 3 0 6 -2 1 5 0 4 31

2 8 - 1 3 06-2 1Ps 0 4' 185

31
= 2.58 i.e., x = 2.58
12

185
-

12 - (2.58) = 15.417 - 6.656

ie., s = 8.761 ’ S=2.96
1 HÍ : u =0 |No increase of blood pressure]
2. H : u >0 [One-tailed test (right)]
3 a = 5o, d.f. n --1 = 12 -1 = 11
4.
Critical region

ta
1.796
Testing of Hypothesis 1.47

5 The test statistic

2.58 - 0
= 2.89
S 2.96
Vn -1 V11
6. Conclusion :
If t<to then accept H; otherwise, we reject Ho
Here, 2.89 <1.796 ie., 2.89 > 1.796
So, we reject Ho
ie., we may conclude that the injection is accompanied by an
increase in B.P.

Example 1.2.a(4)
Arandom 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 percent
and 99 percent fiducial limits for the same.
Solution : Given : n= 16, x= 41.5, u= 43.5
E(r-x) = 135,
135
= = 8.44
n 16
S = 2,9
1. HÍ : u = 43.5
2. HÍ : u 43.5 |Two-tailed test]
3 a= 5% and a = 1% d.f. = n-1 = 16 -- 1 = 15
4 Critical region

-ta/2 la/2
-2.131 (5%) = 2.131 (5%)
-2.947 (1%) = -2.947 (1%)
 SIS
41.5 - 43.5 -2
2.9 0.749 -2.67
V15
Conclusion
If -a/
<I<la/2s We accept HÍ ; otherwise, we
Here, -2.131 -2.67 < 2.131 reject H
So, we reject HÍ at 5%
level of significance.
Here, -2.947 < -2.67 < 2,947
So, we accept Ho at 1%
level of significance.
We conclude that the
7.
assumplion
95% fiducial limits for u:
is reasonable.
(d.f = 15)
S
Vn
41.5 ± 2.131 = 41.5 1598
39.902 < u 43.098
99% fiducial limits for u :
(d.f = 15)
S
= 41.5 +
t 2.947
Vn = 43.71 and 39.29
39.29 < u< 43.71

Example 1.2.a(5)
A random sample of 10 bovs had the
i01, 88, 83, 95, 98, 107, 100. Do these
folowing .Q's : 70, 120, 10
of a
data support the assumpu
populatiog mean LQ of 100 ? Find a
reasonable range in which
most oB the mean 1.Q. values of
samples of l0 beys He.
Solution: |A.U AM 2015 R-8. N/D 2012 R-8, MJ 2016 R-13)
[A.UN/D 2017 R-13}
Testing of Hypothesis
Tota!

98 107 100 972

88 83 95
120 110 101
70 100 96312
952 9g2 1072
1202 1102 1012 882 832
702
M
Ex 972 = 97.2 x = 97.2
i.e.,
10

96312
(97.2)? = 9631.2 - 9447.84
- (r)² 10
n

= 183.36

S = 13.5

1. HÍ : u = 100
100 [Two-tailed test)
2. HË : u
a= 59%. d.f = n - 1 = 10 - 1 = 9
3

4 Critical region

tal2
-ta/2
=-2.262 =2.262

5. The test statistic :

97.2 - 100 -2.8

-0.62
13.5 4.5
Vn V9
6. Conclusion
lI -e2<I<a/2, then we accept Ho : otherwise, wve reject H

Here, -2.262 <K -0.62 <2.262

So, we accept HÍ
 1.50 Testing of Hypothe
We may conclude that the data are consistent with the
of mean I.Q of 100 in the population. assump
95% confidence limits arc given by x ± 2.26 Vn
= 97.2 + (2.26) (4.269)
= 97.2 +9.648 = 106.85 and 87.55
95% confidence limits within which the mean
I.Q values
samples of 10 boys will lie is (87.55, 106.85]
Example 1.2a(6)
Sandal Powder is packed into packets by a
of 12 packets is drawn and machine. A random samp
their weights are found to be (in
0.48, 0.47, 0.48, 0.49, 0.50, 0.51, 0.49, kg) 0.49
if the average weight of 0.48, 0.50, 0.51, and 0.48. Tes.
the packing can be taken as
0.5 kg.
Solution : Given : n = 12, u = 0.5 [A.U. N/D 2003
0.49 0.48 0.47 0.48 0.49
0.49 0.48 0.47 0.482
0.50 0.51 0.49
0.492
0.50 0.51 0.49

0.48 0.50 Total

0.51
0.482 0.48
0.50 0.512
0.48
2.5o
£x 5.88
= 0,49
12

- () = 2.883
12(0.49)2 = 0.00015
= 0.012
1 HÍ : ñ = 0.5

2. H :u 0.5 (Two-tailed test]

 Testing of Hypothesis
1.51
a=5%, d.f = n -1 = 12 - 1 =
11
3.
Critical region :
4.

V
-ta/2 ta/'
=-2.201 =2.201

The test statistic:

5.
0.49 - 0.5
= -2.76
0.012
Vn - VI2-1
6. Conclusion :

If -ta/2 <t<la/2 then we accept Ho; otherwise, we reject HÙ.

Here, -2.201 -2.76 < 2.201

So, we reject H,
:. The average packing cannot be taken to be 0.5 kg.

Example 1.2a(7)
The mean lifetime of a sample of 25 bulbs is found as 1550 hours
With a S.D. of 120 hours. The company manufacturing the bulbs claims
that the average life of their bulbs is 1600 hours. Is the claim acceptable
at 5% level of
significance?
Solution: Given:n =25, =1550, s= 120, u=1600
I. H, : u= 1600
2. H u <1600
|One-tailed test (left)]
3
a= 5%, d.f. = n--1 = 25 -1= 24
1.52 Testing Hypoh
4 Critical region :

-la/
=-1.711

5. The test statistic :

1550 1600 -50

S 120 120 =-2.04
Vn-1 V25-1 V24
6. Conclusion:

If -t,<t, then accept HÍ; otherwise, we

reject Ho.
Here, -1.711 <-2.04
So, we reject H
i.e., The claim of the
significance. Company cannot be accepted at 5% level

Example 1.2a(8)
A
of machinist
is
1.75 cm. A expected to make engine
random parts
1.85 cm, with a S.D. 10 parts with axle
sample of
you say that the
of 0.1
Work of
diameter
cm. On the showS a mean diameter
the basis of this
Solution: Given :n= 10, machinåist inferior? sample,
is would
1. HÍ : u = 1.75 x=185, =175, s =0.1
2.
H: 1.75 [Two-tailed testl
3. a= 5 . df.
=n-i = 10 - 1 = 9
Testing of Hypothesis 1.53
4. Critical region

--tal2 ta/2
=-2.262 = 2.262

5. The test statistic :

1.85 - 1.75
= 3
0.1
Vn -1 V101
6. Conclusion :

If -ta/2 < t<ta/s then we accept H, ; otherwise, we reject Hy

Here, -2.262 < 3 2.262
So, we reject Ho
i.c., the work of the machinist can be assumed to be inferior.

EXERCISE (Test for a specified mean)

1. Certain refined edible oil is packed in tins holding 16 kg each.
The filling machine can maintain this, but with a Standard deviation
a
of 0.5 kg. Samples of 25 are taken from the production line. If
that
sample mean is (i) 16.35 kg (ii) 15.8 kg, can we be 95% sure
the sample has come from a populaion of l6 kg tins ?
Ans. The population mean can be equal to l6 kg.
measurement.
2. A random sample of 9 items is taken from a certain
1584. Find the
From the data, it is found that XX=108. X=
confidence limits for the population mean at 5% level of significance
Table value
and test the hypothesis that the population mean is 8. 2.306 and
of t at 5% level for 8 d.f and 9 d.f are respectively
2.226. mean is 8is valid.
Ans. The population
 Testing of Hypothesis
1.54

is designed to produce insulating washers for electrical

3 A machine 1o
random sample of
devices of average thickness of 0.025 cm. A
thickness of 0.024 cm witl.
washers was found to have an average
significance of tha
a standard deviation of 0.002 cm. Test the
level i
deviation. Value of t for 9 degrees of freedom at 5%
2.262. Ans. The deviation is not significant
inner diameter
4. A company has been producing steel tubes of mean
diameter of
of 2.00 cm. A sample of 10 tubes gives an inner
2.01 cm and a variance of 0.004 cm. Is the difference in the val1le
of mean significant ? Value of t for 9 d.f at 5% level = 2.262
Ans. The means of the population and the sample are not significant.

5. ¢filing machine is expected to fill 5 kg of powder into bags.

A sample of 10 bags gave the weights 4.7, 4.9, 5.0, 5.1, 5.4, 5.2,
4.6, 5.1, 4.6 and 4.7. Test whether the machine is working properly.
Ans. The machine is working properly.
6. Eight individuals arc chosen at random from a population and
then, heights are found to be in cms. 163, 163, 164, 165, 166, 169,
170, 171. In the ight of these data, discuss the suggestion that
the mean height in the universe in 165 cms.
Ans. The mean height can be regarded as 165 cms.
7. Ten students are selected at random ron a college and then
heights are found to be 100, 104, 106, 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 cm.
The table value of t at 5% level for 8
d.f is 2.306 and for 9 d!
is 2.262 and 10 d.f is
2.228 for a two-tail test.
Ans. The mean bheight of the
population can bÝ
regarded as 110 cms.
Afertilizer mixing
machine is set to give 12 kg of nitrate forevery
quintal bag of fertilizer. Ten 100 kg
bags are examined and
Testing of Hypothesis 1.55
percentage of nitrate is as follows :
11, 14, 13, 12, 13, 12, 13, 14, 11, 12
Is there reason to believe that the machine is defective?
Ans. The machine is not defective

9. The heights of 10 males of a given locality are found to be 17s,

168, 155, 170, 152, 170, 175, 160, 160 and 165 cms. Based on this
sample of 10 items test the hypothesis that the mean height of
males is 170 cms.
Ans. The mean height of males can be regarded as 170 cms.
10. A machinist is making engine parts with axle diameters, if 0.700
inch. A sample of 10 parts shows, a mean diameter of 0!742 inch
with a S.D of 0.040 inch. Compute the statistic you would use to
test, whether the work is meeting the specification.
Ans. The product is not meeting the specification

1.2.b t-distribution for difference of means

Suppose, we want to test, if two independeni samples x1, X2, ... n
and y1, y2 ..'a, of sizes n, and n, have been drawn from two normal
population with means u and , respectively.
Under the null hypothesis, HÍ that the samples have been drawn
from the normal population with means u1 and u are under the
assumption that the population variance are equal. (i.e., o =, =o)
The statistic t, where t =
(r-y) - (u1- e)
1 1
+

where, nË
y=

and
th,
1
- 22 - +Eo-p
n
IS an
unbiased estimate of the population variance o.
 1.56
Testing of H
t follows -distribution with degrees of freedom + Hypothes
Under the null hypothesis HÍ, that () samples have n-
2.
from the population with the same mcans 1.e., uË=u2, or (Y been dra,
sample means x and y do not differ significantly, take the statict:
with degrees of
freedom, nËt ny-2,
n1 n2
If | < table value of ta accept HÍ, at a level of significance.
Assumption : The following assumptions are made in using this tes
(i) Parent populations, from which the samples have been drawm
ar
normally distributed.

() Population variances are equal and unknown

(ii) The two samples are random and independent.
Since. $²- E-²+2( -y
=

nË t n - 2

we have, S ns t ngs where,

n t n, -2

S1, S) are standard deviations of the two samples. Therefore, statisti:

to be tested is

n1 sít n2 S~ 1 1
nj t n - 2 n1

statistic
Note 1. If n=n, =n, t can be used as a test

n -1
correlated)
Note 2. If the pairs of values are in some way associated (or
findthe
we cannot adopt the case under Note 1. Then, we have to meal

differences of the associated pairs of values and apply for single

Testing of Hypothesis 1.57

ic., ( = , to test, if the means of the differences is significantly

s/ Vn - T'
different from zero. Then, the degrce of freedom is n - 1.
d
Here, the test statistic is t = Or
s/vn - T S/Vn

where d is the mean of the paired differences.

i.c., d, = x, -y, and d =x -y,
where (, v;) are the paired data, i = 1, 2, .. n
Working Procedure
Concerning difference between two means, with unknown o and o
but cqual. (o =¡ = o). For the small samples (n < 30, ny < 30)
drawn from two normal population.
1. Null hypothesis HÍ :1 =M
2. Alternative hypothesis H : u * uy (or) 41> 2
(or) 1 <W2
3. Level of significance : c, d.f. = n t n - 2
4. Critical region :
(a) If u # us, then the test is two-tailed test for the given a.
The critical values are -ta/) and t/) from the t-distribution table
with d.f. = n + n - 2

-ta/2 ta/2

(D) II u,>u, then the test is one-tailed test (right) for the given c,

The critical value is t, with d.f. = n tny -2

 i58 Testing of Hypothesis

ta/2

(c) If u < uz, then the test is one-tailed test (left) for the given
The critical value is -tg with d.f. = n t n, -2

-la/2

5 The test statistic

where S =
n sí + n, s5
1 1
nË t n, -2

6. Conclusion :

(2) II -ta/2 <t<la/'2, then we accept HÍ for two-tailed test; otherwise,

we reject Ho

(b) Ift< la, then we accept Ho for one-tailed test (right); otherwise,
we reject HÍ

(C) I1 -la <t, then we accept Ho for one-tailed test (left); otherwSC,
we reject Ho
Testing of Hypothesis

Example 1.2b(1)|
tine (in seconds)
Two horses A and B were tested accofding to the
:
to run a particular race with the foilowing results
28 30 32 33 33 29 34
Horse A :
Horse B 29 30 30 24 27 29

level.
Test whether the horse Ais running faster than B at 5%
[Anna, May 2001, N/D 20091
Solution : Given : n 7, n, = 6
= 219
£x = 28 + 30 + 32 + 33 + 33 + 29 + 34

E= 28' + 30 + 32 + 33 + 33 + 29+ 34 = 6883

29 = 169
Ex = 29 + 30 + 30 + 24 + 27 +

Er; = 29 + 30 + 30 + 24' + 27² + 292

4787

219 = 31.29
7

169
28.17
6

6883 979.06 = 4.23

- (31.29) =983.27
7

4787
(28.17) = 797.83 793.55 = 4.28
6 -9
29.61 + 25.68
7(4.23) + 6 (4.28)
n t n -2 7+6 - 2 11

55.29
= 5,03
11

2. HË: u1 > u, [Horse A is faster than BÊ

 160
Testing of Hypothes
a=$%, d.f R tn, - 2 = 7+6-- 2 = 11
4 Critical rcgion :

ta=1.796

5. The test statistic

X1 - X 31.29 -- 28.17
= 2.498
1
V(so3)
6. Conclusion
If 1< Las then we accept Ho ; otherwise, we reject H)
Here, 2.498 1.796
So, we reject HÍ at 5% level of
significance.
Example 1.2b(2)
A group of 10 rats fed on
ldiet A and another group of 8 rats fed
on diet B, recorded the
following
Diet A : 5, 6, 8, , 12, 4, 3, 9, 6, increase in weight (gms).
10
Diet B: 2, 3, 6, 8, 10, 1, 2, 8
Does it show superiority of diet A over diet B.
[A.U. N/D 2011]
Solution : Given : nË = 10, n, = 8
2 = 5+ 6 t 8 + 1 +
12 t 4 t 3 + 9 + 6+
10 = 64
Ex = 5 + 6 +8'+1 + 12+ 4
+3 + 9 + 6 + 10 = 512
2I, = 2+3 + 6 + 8 + 10 + 1
+ 2+8 = 40
E= 2+3 +6 +8² +10 + 1² +2 +
8² = 262
 Testing of Hypothesis S 1.61

Ex 64
= 6.4
10 10

40
=5

2
512
10 (6.4) 10.24

282
25 =10.25

10 (10.24) + 8 (10.25) = 11.525

nË tny -2 10+ 8 - 2
1. HÍ : 1 = W2
2. H :1 > uy [One-tailed test (right)]
3. a = 5% d.f. = n, + n, 2 = 10 + 8- 2 = 16
4. Critical region

la=1.75

5. The test statistic :

XË - X 6.4 - 5

1 1 1 1
11.525
10 8

1.4
= 0.869
1.6103
6. Conclusion :
It<tas then we accept Ho; otherwise, we reject Ho
 1.66
Testing of Hypothe

Example 1.2b(5)
measurements of the
The following random samples are heat produg
capacity (in millions of caBories per ton) of specimen's of coals fr
two mines.

Mine 1: 8,260 8,130 8,350 8,070 8,340

Mine 2: 7,950 7,890 7,900 8,140 7,920

041
Use the 0.01 level of significance to test whether the difference bet
the means of these two samples is significant.
Solution : |A.U. N/D 201
£ = 8260 t 8130 + 8350 + 8070 + 8340 = 41150

Ex = 8260 + 8130² + 8350 + 8070 + 83402 = 338,727,500

Ex = 7950 + 7890 + 7900 + 8140 + 7920+ 7840 = 47650
= 7950 +7890 + 7900 + 8140 + 7920 + 7840 = 378,3162
41150
= 8230
5 5

47640
7940
6

338,727,500
5 (8230)
67745500 - 67732900

12,600

378316200
n2 (7940)
= 63052700 - 63043600

9,100
 Testing of Hypothesis 1.67

| " tn, -2
(5) (12600) +(6) (9100)
5 +6- 2

117600
9

= 13066.67

1. HÍ : u1 = 2
2. H :u u [Two-tailed test)
3. a= 1%, d.f. = n, + n, -2 = 5 +6-2 = 9

4. Critical region :

-tal' ta/2
=-3.25 =3.25

5. The test statistic :

8230 - 7940
= 4.19

2 1 1
V 13066.67

6. Conclusion :

If -ta/2 <I<la/) then we accept Ho ; otherwise, we reject HÍ.

Here, -3.25 < 4.19 3.25
So, we rcject H.
 Testing of Hypothesi
1.68

Example 1.2b(6)|
independent samples are chosen from two schools A and n
Two
subject. The scores of the students a
a commnon test is given in a
follows :
70 43 94 68 33
School A : 76 68

40 48 92 85 70 76 68 22
School B

Can we conclude that students of school A performed better than:

students of school B? [A.U. A/M 2008)
Soiution :

HÍ : There is no significant difference of performance

of students between school A and school B.

H: The performance of students of school A is bette

than that of school B.

}x = 76 + 68 + 70 + 43 + 94 + 68 + 33 = 452

E = 76+68 + 70²+ 43 + 942 + 68² + 33 = 31698

£x, = 40 + 48 + 92 + 85 + 70 + 76 + 68 + 22 = 501

Ex =40 + 48 + 92 + 85 + 70 + 76 + 68 + 22 = 35377
452 501
= 64.6, 62.6
8

31698
7 (64.6) = 4528.3 -4173.16 =: 355.14

35377
n2 (62.6)² = 4422.13 - 3918.76 = 503.3/

n sí+ns5 7(355.14) + 8(503.37)

7+8-- 2 500,995
Testing of Hypothesis
1.69
1+1 = 500.995 = 134,195
n2
1
+ = 11.58
n2
1 HÍ : 1 =2
2. HË : 1> W|One-tailed test (right))
3. 5%, d.f = n1t n -2 = 7+8 - 2 = 13
4 Critical region

ta=1.771

5. The test statistic :

64.6 - 62.6
= 0,173
11.58
1 1

n2
6. Conclusion :
If t<ta then we accept Ho ; otherwise, we reject Ho
Here, 0.173 < 1.771
So, we accept HÍ

Example 1.2b(7)|
Ihe following table gives the values of protein from Kangeyam cow's
milk and bufalo's milk. Examine if these differences are significant.
Cow's milk : 1.90 1.95 2.00 2.02 1.85 1.80
Buffalo's milk 2.12 2.00 2.20 2.45 2.20 2.10