ence totin

hteal of SAMPLING THEORY

10 0.1
and Z=
V4/9 x 5/9(1/1000 +1/800 )
= 4.243

Zos= 1.96
Z = 4.243 >
-01 =2.58
Thus the hypothesis H, isrejected both at 5% and 1% levels of significance.

Student's t distribution /test

31. Find the student's t for the following variable values in a sample of eizht :
-4, -2, -2,0, 2, 2,3,3, taking the mean of the universe to be zero.
>> t=
S

By data u = 0 and we have n = 8

1
+= ( - 4 -2-2+0 +2 +2+3+3) = = 0.25
4
1
s =
n-1 i=1

1
=-(-4.25 )' +(-2.25)+(-2.25 )
+(-0.25 ) +(1.75 )' +(1.75) +(2.75 )' +(2.75 '
=,(49.5 ) = 7.07 S = 2.66

0.25 -0
Thus t= V8 = 0.266
2.66

Note :The expression for s can also be put in the followingform.

1
1
2
1 1
n-1 1 1 1

2
1-1 1

1
H-1

Bccording to thisformula we have in thegiven example

SAMPLING THEORY & SAMPLING DISTRIBUTIONS 419

1|
7 (49.5 ) = 7.07
Remark : We canemploy this formula when is not an integer.

32. Amachine is expected to produce nails of length 3 inches. A random sanple of 25 nails
8«ve an average length of 3.1 inch with standard deviation O.3. Can it be said that the
machine is producing nails as per specification? (toos for 24 d.f is 2.064 )
>> By data we have,
u =3,X = 3.1, 1 = 25, s = 0.3
0.1
V25 = 1.67 < 2.064
0.3
Thus the hypothesis that the machine is producing nails as per specification is accepBed
at 5% level of significance.

33. Ten individuals are chosen at random from a population and their heights in inches are
found to be 63, 63, 66, 67, 68,69, 70, 70, 71,71. Test the hypothesis that the mean height
of the universe is 66 inches. ( tgs = 2.262 for. 9d,f)
>> We have u = 66, n = 10
678
M66
X= = 67.8
10

n-1
X(r-)?
1
=l(63-67.3) +. +(71-678|=9.067
9
S = 3.011

We have =
(67.8- 66) V10 = 1.89 < 2.262
3.011

Thus the hypothesis is accepted at 5% level of significance.

34. A sanple of 10 measurenments of the diameter of asphere gave a mean of 12cm anda
standard deviation 0.15cm. Find the 95% confidence linits for the actual dianneter.
>> By data n = 10, X = 12, s = 0.15
Also os for 9d-f = 2.262
Confidence lumits for the actual diameter is given by
0.15
0s = 12 +
V10 (2.262 ) = 12 ± 0.1073
Thus 11.393cm to 12.107em is the
conii: sital dianeter.
420 SAMPLING THEOHY

35. Acertain stimulus administered to each of the 12 patients resulted in the follouwino
change in blood pressure. 5, 2, 8, -1,3, 0, 6, -2, 1,5, 0,4. Can it be
concluded that the stimulus will increase the blood presSure?
(Ios for 11 d-f = 2.201)
2x 31
= 2.5833
12

1 1
n-1 n-1

1
11 185-(31 = 9.538 .:. s = 3.088

We have, t = Vn

Let us suppose that the stimulus administration is not accompanied with increase in
blood pressure, we can take = 0
2.5833 0
t= V12 = 2.8979 = 2.9 > 2.201
3.088

Hence the hypothesis is rejected at 5% level of significance.We conclude with 95%

confidence that the stimulus in general is accompanied with increase in blood
pressure.

36. A group of boys and girls were given an

intelligence test. The mean score, S.D score and
numbers in each group are as follows.
Boys Girls
Mean 74 70
SD 10

12 10

Is the difference between the means of the two

groups significant at 5% level of significancc
(t o5 = 2.086 for 20 df)

>> We have by data x = 74, s, = 8, n, = 12 | Boys ]

y= 70, S, =10, l, =10 |Girls |

Also we have f =
S
 Ve
have,
SAMPLING THEORY & SAMPLING DISTRIBUTIONS

M
1
where s =
"; tn, -2 I=1

Or
n, + n, -2
12 ( 64 ) + 10 (100 ) 1768
Now = =88.4 S = 9.402 9.4
20 20

74 -70
Hence t= = 0.994
9.4 Vi/12 + 1/10
t =0.994 < toE = 2.086
Thus the hypothesis that there is a difference between the means of the two groups
isaccepted at 5% level of significance.
37. Asample of 11rats from acentral population had an average blood viscosity of 3.92 with
astandard deviation of 0.61. On the basis of this sanple, establish 95% fiducial limits for
the mean blood viscosity of the central population (os = 2.228 for 10 df)
>> By data x = 3.92, s = 0.61, n = 11
95% fiducial limits for u are x t
Vn
0.61
i.e., = 3.92+ (2.228 )
V11
= 3.92 t 0.41 = 3.51 and 4.33
Thus 95% confidence limits for u are 3.51 and 4.33.

38. Two types of batteries are tested for their length of life and the following results were
obtained.

Battery A: n, = 10, x, = 500 hrs , o = 100

Battery B: 1, = l0, x, = 500 hrs , o, = 121
Compute Student's t and test whether there is a significant difference in tlhe tvo means.
n,o t n, o
I1, +1, -2

(C0 x i 0G ) (10x 121

18
 SAMPLING THEORY
(7,-)
ve have,
Vi/n, +1/n
60
t= = 12.1081 = 12.11
11.0805 V0.1+0.1
This value of t is greater than the table value of t for 18 d-f at all levels of
significance.
The null hypothesis that there is no significant difference in the two means is
rejected at allsignificance levels.
39. Agroup of 10 boys fed on a diet Aand another group of 8 boys fedon adifferent diet Bfor
aperiod of 6 months recorded the following increase in weiglhts (lbs.)
Diet A: 5 6 8 1 12 4 3 9 6 10
Diet B: 2 3 6 8 10 1 28
in weight.
Test whether diets Aand Bdiffer significantly regarding their effect on increase
y to the diet B.
>> Let the variable x correspond to the diet A and
o40
64
X= =6.4 ; y= =5
10

-M -M
"2
) (r-7 = 102.4 ; 5E (y-7 - 82
1

"

1
E (r-+ (y-)
-2 1
|1

1 184.4 S = 3.395
(102.4 + 82 ) = 16 = 11.525
16

Consider t=
Ni/n, +1/n,
1.4 = 0.86935 0.87
3.395 V1/10 + 1/8
2.12 from the tables. t = 0.87 is less than the table value for
But tog for 16dj=
16 d-f at 5% level ofsignificance.
not differ significantly regarding their
Thus we conclude that the two diets do
effect on increase in weight.
 SAMPLING THEORY & SAMPLING DISTRIBUTIONS 423

40. Tuo horses Aand B were tested acordingto the time (in seconds) to runa particular
race with the following results.
Horse A: 28 30 32 33 33 29 34

Horse B: 29 30 30 24 27 29

Test wiether you can discrininate between the two horses.

>> Let the variables x and y respectively correspond to horse A and horse B.

219 169
" 7 = 31.3, y= = 28.2

-M
"2
(x-7 =31.43, 1
(y- = 26.84

1
X (x-7)# 1
(y-ù

(31.43 + 26.84 ) = 5.2973 S= 2.3016

Consider t =
s V1/n, +1/n,

(31.3- 28.2 ) = 242

2.3016 Vi/7+ 1/6

But tos = 2.2 and to = 2.72 for 11df

t= 2.42
> 05 = 2.2

<o = 2.72

The discrimination between the horses is significant at 5% level but not at 2%

level of significance.
424 SAMPLING THEORY

Chi-Square distribution
41. Adie is thrown 264 times and the number appearing on the face (x) follows the
frequency distribution. following
1 2 3 5 6
40 32 28 58 54 60

Calculate the value of x.

>> The frequencies in the given data are the observed frequencies. Assuming that the
dice is unbiassed the expected number of frequencies for the numbers 1,2, 3, 4,5,6 to
264
appear on the face is = 44 each. Now the data is as follows :
6

No. on the dice 1 2 3 4 5 6

Observed frequency (0,) 40 32 28 58 54 60

Expected frequency (E;) 44 44 44 44 44 44

(0,-E,}
E,
(4044 ) (32 44 ) (60 - 44 )
44 44 44
1 968
44
16+ 144+ 256 + 196 + 100 + 256 = 22
44

Thus x=22

42. Five dice were thrown 96 times and the numbers 1, 2, or 3 appearing on the face of the dice
follows the frequency distribution as below.

No. of dice showing 1,2 or 3 5 4 3 2 1

Frequency 7 19 35 24 8 3

Test the hypothesis that the data follows abinomial distribution. (Xáos = 11.07 for 5 d.f)

>> The data gives the observed frequencies and we need to calculate the expected
frequencies.
SAMPLING THEORY & SAMPLING
DISTRIBUTIONS
Probability of asingle dice throwing 1, 2 or 3 is p= 3/6 =1/2 .. q = 1-p
The binomial distribution of fit is, N(q+p) =96 ( 1/2 + 1/2 )
The theoretical
frequencies
the successive terms of
of the getting 5,4,3,2, 1,0 successes with 5dice are
binomial expansion. respectively
1
They are respectively 96 x 96 x 1 1
5 5C, X 25.96 or 3, 15, 30, 30,15, 3.
2
We have the table of
observed and expected
0 7
-..........

frequencies.
19 35
24 3
E 3 15 30
30 15 3

(0,-E,'
16 16 25 +
36
+ + 49
3 15 +
30 30 15 T= 11.7

x =11.7 >Xú0s = 11.07

Thus the hypothesis that the data
follows a binomial distribution is
reiected.
43. A sample analysis of
students had failed, 170 hadexamination results of 500 students was made. It was
first class. Do these figures secured third class 90 had secured second class and found that 220
support the general examination result ohich is in 20had secured
4:3:2:1 for the respective categories (zó 05 = 7.81 for 3d.) the ratio

>> Let us take the

ratio 4:3:2:1. hypothesis that these figures support to the general
result in the
The expected frequencies in the respective category are
3 2
10
x 500, 10
x 500, x 500, x 500
'10 10 Or
200, 150, 100, 50.
 table. SAMPLING THEORY
Ve nave the following
220
170
90
200 20
E 150
100
50
(0,-E,
E;
400 400 100 +
900
200 150 100 50

x = 23.67 > xos = 7.81

Thus the hypothesis is rejected.

44. 4 coins are tossed 100 times and the following results were
obtained. Fit abinomial
distribution for the data and test the goodness of fit (Xó0s= 9.49 for 4 d,fp
Number of heads 1 2 4

Frequency 5 29 36 25 5

>> Referring to Problem-28 in Unit -VII, we have obtained the theoretical frequencies
equal to 7, 26, 37, 24, 6 respectively.
We have the following table.
5 29 36 25 5
0,
7 26 37 24 6

(0,-E,

4 9 1 1 1
+ + = 1.15
7 26 37 24 6

x = 1.15 < Xoos = 9.49

Thusthe hypothesis that the fitness is good can be accepted.
 SAMPLING THEORY & SAMPLING
DISTRIBUTIONS 427
45. Fit a Poisson distribution for the
2 following data and test the goodness of fit given that
Xoos =7.815for 3d.f
2 3 4
122 60 15 2 1

>> Referring to Problem-32 in Unit - VIl we have

obtained the theoretical frequencies
equal to 121, 61, 15, 3, 0. Since the last of the expected
with the previous one. frequency is 0 we shall club it

We have the following table.

122 60 15 2+1= 3
E 121 61 15 3+0 = 3

(0,-E?

1 1
121 61 + 0+0 0.025

x= 0.025 < Kons = 7.815. The fitness is considered good.

Thus the hypothesis that the fitness is good can be
accepted.

46. The number of accidentsper day ( x )as recorded in a

textile industry over a period of 400
days is given below. Test the goodness of fit in respect of Poisson distribution offit to the
given data (Ko0s = 9.49 for 4d,f)
0 1 3 4 5
f 173 168 37 18 3
 SAMPLING THEOAY
428

>> Referring to the Problem-33 in Unit- VIl, the corresponding theoretical frequencies
are 183, 143,56, 15, 3, 0. We shallclub the last two frequencies to have the following
table.

0 173 168 37 18 3+1 = 4

183 143 56 15 3+0 = 3

(0,-E,
E
100 625 361 9 1
+ + = 12.297 12.3
183 143 56 15 3

y =12.3 > ns =9.49. The fitness is not good.

Thus the hypothesis that the fitness is good is rejected.

EXERCISES

1. Arandom sample of size 2is drawn from the population 3,4, 5. Find the sampling
distribution of the sample mean. (a) with replacement (b) without
replacement. Find the sample mean and sample variance in these two cases.
2. 500 ball bearings have amean weight of 142.30 gms. and S.D of 8.5 gms. Find the
probability that arandom sample of 100 ball bearings chosern from this group will
have a combined weight (a) between 14,061 and 14,175 gms. (b) more than
14,460 gms.
3. The weights of packages received by adepartment store have a mean of 136 kgs.
and aS.D of 22.5 kgs. What is the probability that 25 packages received at random
and loaded on an elevator will exceed the safety limit of the elevator quoted as
3720 kgs.
4. A die' was thrown 9000 times and a throw of 5 or 6 was obtained 3240 times. On
the assumption of random throwing, do the data indicate an unbiassed die?