Sampling Theory and Testing Hypothesis

1. Collection of individuals – Population (or) Universe

2. Population may be finite (or) infinite.

3. Finite subset of individuals in a population – Sample

4. Number of individuals in a sample – Sample size

5. Statistical measure [ mean and variance ] calculated on the basis of Population

values of x - Parameters

6. Statistical measure computed on the basis of sample observations ( x1 , x 2 ,..., x n ) -

Statistics.

7. A procedure for deciding whether to accept (or) to reject a null hypothesis {and
hence to reject (or) to accept the alternative hypothesis} – Testing of Hypothesis.

8. Assuming that there is no significant difference between sample statistic and

corresponding population parameter (or) between two sample statistics – Null
Hypothesis H o .

9. Hypothesis different from null hypothesis – Alternative Hypothesis, H1

10. A procedure for testing the hypothesis whether the difference between population
parameter θ o and corresponding sample statistic θ is significant (or) not – Test
of Significance.

11. Errors in Sampling:

i) Type I Error (Producer’s Risk): Reject H o when it is true.

ii) Type II Error (Consumer’s Risk): Accept H o when it is wrong

Actual Decision
Accept H o Reject H o
H o is true Correct decision Wrong
(no error) Type I Error
Probability = (1 − α ) Probability α
H o is false Wrong Correct decision
Type II Error (no error)
Probability β Probability = (1 − β )
 12. One Tailed and Two Tailed Tests:

• A test of any statistical hypothesis where the alternative hypothesis is one – tailed
(right tailed or left tailed) is called a One – Tailed test.

For Example: For testing the mean of population in a single tailed, we assume that
the
null hypothesis :H0 : µ = µ 0 against the
alternative hypothesis : H1 : µ > µ 0 (Right tailed)
H1 : µ < µ 0 (Left tailed) is called One – tailed test.

• In a test of statistical hypothesis where the alternative hypothesis is 2- tailed, we

assume that the
null hypothesis : H0 : µ = µ 0 against the
alternative hypothesis :H1 : µ ≠ µ 0 is called 2- tailed test.

• Suppose if we want to test if the bulbs produced by new process ( µ 2 ) have higher
average life than those produced by standard process ( µ1 ) then we’ve H0 :
µ1 = µ 2 and H1 : µ1 < µ 2 ( left tail test)

• If we want to test whether the product of new process ( µ 2 ) is inferior to that of

standard process ( µ1 ) then H0 : µ1 = µ 2 and H1 : µ1 > µ 2 ( right tail test)

13. Critical values of Zα :

Critical values Level of Significance ( α )

( zα ) 1% 2% 5% 10%
Two tailed test zα = 2.58 zα = 2.33 zα = 1.96 zα = 1.645
Right tailed test zα = 2.33 zα = 2.054 zα = 1.645 zα = 1.28
Left tailed test zα = -2.33 zα = -2.054 zα = -1.645 zα = -1.28

14. Procedure for Testing of Hypothesis:

a. Set up Null hypothesis H o : µ = µ o

b. Set up Alternative hypothesis H 1 : µ ≠ µ o
t − E (t )
c. Test Statistic (Calculated Value) Z =
S .E (t )
d. Level of Significance (Tabulated value) Z α : 1% (or) 5%
e. Conclusion: a) If Z < Z α then accept H o
b) If Z > Zα then reject H o
15. A region corresponding to a statistic t in the sample space S which lead to the
rejection of H o - Critical Region (or) Rejection Region

16. The region which lead to the acceptance of H o - Acceptance Region

17. Test of Significance of Small Samples:

When sample size is less than 30 (n < 30) then the sample is called as Small
sample.

1. Student’s – t test for single mean:

Null hypothesis : H o : µ = µo
Alternative hypothesis: H1 : µ ≠ µ o
x−µ
Test statistic: t = (S.D is not given directly)
S2
n
∑ xi
Where x = = sample mean; µ = population mean
n
2
2 ∑ ( xi − x )
S = ; n = sample size
n −1

x−µ
ii) Test statistic : t = (S.D is given directly)
S .D
n −1
Degree of freedom(d.f) : n − 1

Confidence (or) Fiducial limits for µ

1. The 95% confidence limit of the mean of the population is given by

S
x ± t 0.05 .
n
2. The 99% confidence limit of the mean of the population is given by
S
x ± t 0.01
n

Problems:

1. The mean life time of a sample 25 fluorescent light bulbs produced by a

company is computed to be 1570 hours with a S.D of 120 hours. The
company claims that the average life of the bulbs produced by the
 company is 1600 hours. Use the level of significance of 5%. Is the claim
acceptable?

Solution: Given: Sample size n = 25 , Sample mean x = 1570

S.D (S) = 120 , Population mean µ = 1600
Degree of freedom = n − 1 = 24

Null hypothesis : H o : µ = 1600 hrs

Alternative hypothesis: H 1 : µ ≠ 1600 hrs

x−µ 1570 − 1600 − 30

Test statistic (C.V) :t= = = = -1.22
S .D 120 24.49
n −1 24
t = 1.22

Tabulated Value (T.V) : t 0.05 with 24 d.f is 2.06

Conclusion : C.V < T.V

we accept the null hypothesis H o .
The claim is acceptable

2. A machinist is making engine parts with axle diameters of 0.700 inch. A

random 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.

Solution: Given: Sample size n = 10 , Sample mean x = 0.742

S.D (S) =0.040, Population mean µ = 0.700
Degree of freedom = n − 1 = 9

Null hypothesis : H o : µ = 0.700

Alternative hypothesis: H 1 : µ ≠ 0.700

x−µ 0.742 − 0.700

Test statistic (C.V) :t= = = 3.15
S .D 0.040
n −1 9
t = 3.15

Tabulated Value (T.V) : t 0.05 with 9 d.f is 2.26

Conclusion : C.V > T.V

 we reject the null hypothesis H o .

3. A random sample of size 16 valves from a normal population showed a

mean of 53 and a sum of squares of deviation from the mean equals to
150. Can this sample be regarded as taken from the population having 56
as mean? Obtain 95% confidence limits of the mean of the population.

Solution: Given: Sample size n = 16 , Sample mean x = 53

2
∑ ( x − x ) = 150 , Population mean µ = 56
2
2 ∑ (x − x) 150
Therefore, S = = = 10
n −1 15
⇒ S = 10

Degree of freedom = n − 1 = 15

Null hypothesis : H o : µ = 56

Alternative hypothesis: H1 : µ ≠ 56

x−µ 53 − 56
Test statistic (C.V) :t= = = -3.79
S 10
n 15
t = 3.79

Tabulated Value (T.V) : t 0.05 with 15 d.f is 2.13

Conclusion : C.V > T.V

we reject the null hypothesis H o .
The 95% confidence limit of the mean of the population is given by
S
x ± t 0.05 = 53 ± 2.13 × 0.79
n
= 53 ± 1.6827
Hence 95% confidence limit is [51.31, 54.68]

4. A random sample of 10 boys had the following I.Q’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 which most of
the mean I.Q. values of samples of 10 boys lie.

Solution:
x x−x (x − x)2
 70 -27.2 739.84
120 22.8 519.84
110 12.8 163.84
101 3.8 14.44
88 -9.2 84.64
83 -14.2 201.64
95 -2.2 4.84
98 0.8 0.64
107 9.8 96.04
100 2.8 7.84
972 1833.60

∑x 972
Mean x = = = 97.2
n 10
2
2 ∑ ( xi − x ) 1833.60
We know that, S = = = 203.73
n −1 9
Standard deviation, S = 203.73 = 14.27

Null hypothesis : H o : µ = 100

Alternative hypothesis: H1 : µ ≠ 100

x−µ 97.2 − 100

Test statistic (C.V) :t= = = -0.62
S 14.27
n 10
t = 0.62

Tabulated Value (T.V) : t 0.05 with 9 d.f is 2.26

Conclusion : C.V < T.V

we accept the null hypothesis H o .
The 95% confidence limit of the mean of the population is given by
S
x ± t 0.05 = 97.2 ± 2.26 × 4.514
n
= 53 ± 10.20
Hence 95% confidence limit is [87.00, 107.40]
2. Students t – test for difference of means:

To test the significant difference between two means x1 and x 2 of sample of

sizes n1 and n2 .

Null Hypothesis: H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2
Test statistic: t =
1 1
s +
n1 n2
2 2
∑ ( x1 − x1 ) + ∑ ( x 2 − x 2 )
Where s 2 = [ S.D is not given directly]
n1 + n2 − 2
n s 2 + n2 s 22
s2 = 1 1 [ S.D is given directly]
n1 + n2 − 2
s1 , s 2 - sample standard deviation.
Degrees of freedom d . f . = n1 + n2 − 2

1. The average number of articles produced by two machines per day are 200
and 250 with standard deviations 20 and 25 respectively on the basis of
records of 25 days production. Can you regard both the machines equally
efficient at 1% level of significance?

Solution: Given: n1 = 25 , x1 = 200 , s1 = 20

n2 = 25 , x 2 = 250 , s 2 = 25
n s 2 + n2 s 22 25(20) 2 + 25(25) 2
Now, s 2 = 1 1 = = 533.85
n1 + n2 − 2 48
⇒ s = 23.10

Null Hypothesis: H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2 200 − 250
Test statistic (C.V) : t = = = -7.65
1 1 1 1
s + 23.10 +
n1 n2 25 25
t = 7.65
Tabulated value (T.V): t 0.01 for 48d.f. is 2.58
Conclusion : C.V > T.V
We reject the null hypothesis H o .
2. The means of two random samples of size 9 and 7 are 196.42 and 198.82
respectively. The sum of the squares of the deviation from the mean are
26.94 and 18.73 respectively can the sample be considered to have been
drawn from the same normal population.

Solution: Given: n1 = 9 , x1 = 196.42 , ∑ ( x1 − x1 ) 2 = 26.94

n2 = 7 , x 2 = 198.82 , ∑ ( x 2 − x 2 ) 2 = 18.73
2 2
∑ ( x1 − x1 ) + ∑ ( x 2 − x 2 ) 26.94 + 18.73
Now, s 2 = = = 3.26
n1 + n2 − 2 9+7−2
⇒ s = 1.81

Null Hypothesis: H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2 196.42 − 198.82
Test statistic (C.V) : t = = = -2.63
1 1 1 1
s + 1.81 +
n1 n2 9 7
t = 2.63
Tabulated value (T.V): t 0.05 for 14d.f. is 2.15
Conclusion : C.V > T.V
We reject the null hypothesis H o .

3. The nicotine content in milligrams of two samples of tobacco were found

to be as follows:

Sample A 24 27 26 21 25 -
Sample B 27 30 28 31 22 36

Can it be said that two samples come from normal populations having the
same mean.

Solution:
Sample Sample B x − x (x − x)2 y− y ( y − y)2
A (x) ( y)
24 27 -0.6 0.36 -2 4
27 30 2.4 5.76 1 1
26 28 1.4 1.96 -1 1
21 31 -3.6 12.96 2 4
25 22 0.4 0.16 -7 49
36 7 49
123 174 0 0 108
 123 174
Mean x = = 24.6 , y = = 29
5 6
2 2
∑ ( x1 − x1 ) = 21.2 , ∑ ( x 2 − x 2 ) = 108
2 2
∑ ( x1 − x1 ) + ∑ ( x 2 − x 2 ) 21.2 + 108
Now, s 2 = = = 14.35
n1 + n2 − 2 5+6−2
⇒ s = 3.78

Null Hypothesis : H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2 24.6 − 29
Test statistic (C.V) : t = = = -1.92
1 1 1 1
s + 3.78 +
n1 n2 5 6
t = 1.92
Tabulated value (T.V): t 0.05 for 9d.f. is 2.262
Conclusion : C.V < T.V
We accept the null hypothesis H o .

3. F- test: ( To test Equality of variance)

Null Hypothesis : H0 : σ 12 = σ 22 {(ie) Population variance are same}

Alternative hypothesis : H1 : σ 12 ≠ σ 22

s12
Test Statistic: F= , s12 > s 22
s22
where s12 =
∑ (x − x)
2
, s 22 =
∑ ( y − y) 2

{S.D is not given directly}

n1 − 1 n2 − 1
n1 s12 n s2
S12 = , S 22 = 2 2 { S.D is given directly}
n1 − 1 n2 − 1
Degree of freedom : γ 1 = n1 − 1 and γ 2 = n 2 − 1

Greater var iance

Note:. F =
Smaller var iance

1. In one sample of 10 observations from a normal population, the sum of the

squares of the deviations of the sample values from the sample mean is 102.4
and in another sample of 12 observations from another normal population, the
sum of the squares of the deviations of the sample values from the sample
mean is 120.5. Examine whether the two normal populations have the same
variances.
 Solution:
2
∑ ( x1 − x1 ) 102.4
Given: n1 = 10 , ∑ ( x1 − x1 ) 2 = 102 .4, s12 = = = 11.37
n1 − 1 9
2
2 ∑ ( x2 − x2 ) 120.5
n2 = 12 , ∑ ( x 2 − x 2 ) = 120.5 .4, s 22 = = = 10.95
n2 − 1 11
Null Hypothesis : H0 : σ 12 = σ 22
Alternative hypothesis : H1 : σ 12 ≠ σ 22
s 2 11.37
Test Statistic (C.V) : F= 1 = = 1.038
s 22 10.95
Tabulated Value (T.V) : F0.05 (9,11) d.f is 2.91
Conclusion : C.V < T.V
We accept the null hypothesis H o .

2. Is is known that the mean diameters of rivets produces by two firms A and B
are practically the same but the standard deviations may differ. For 22 rivets
produced by firm A, the standard deviation is 2.9 mm, where 16 rivets
manufactured by firm B, the standard deviation is 3.8 mm. Compute the
statistic you would use to test whether the products of firm A have the same
variability as those of firm B and test its significance.

Solution: Given : n1 = 22 , s1 = 2.9 mm

n2 = 16 , s 2 = 3.8 mm
Here the S.D’s of the samples s1 and s 2 are given.
Therefore, the population variance S12 and S 22 are obtained by using the
relations. n1s12 = (n1 − 1) S12 and n2 s 22 = (n2 − 1) S 22
n s2 n s2
⇒ S12 = 1 1 ⇒ S 22 = 2 2
n1 − 1 n2 − 1
22 × (2.9) 2 16 × (3.8) 2
= =
21 15
= 8.8105 = 15.403
Null Hypothesis : H0 : σ 12 = σ 22
Alternative hypothesis : H1 : σ 12 ≠ σ 22
s12 15.403
Test Statistic (C.V) : F= = = 1.75
s 22 8.8105
Tabulated Value (T.V) : F0.05 (15,21) d.f is 2.20
Conclusion : C.V < T.V
 We accept the null hypothesis H o .

3. The nicotine contents in milligrams in two samples of tobacco were found to

be as follows:

Sample A 24 27 26 21 25 -
Sample B 27 30 28 31 22 36
Can it be said that two samples come from same normal population?
Solution: Hint: To test whether the two samples come from the same normal
population.
That is to test i) the equality of variances
ii) equality of means

Chi – Square test ( The square of a standard normal variate)

A test for testing the significance of difference between experimental

values and the theoretical values obtained under some theory or hypothesis.
This test is known as ψ 2 - test of goodness of fit.

Test Statistic:

(O − E ) 2
ψ =∑2
where O - Observed frequency
E
E – Expected frequency
Note:

1. It is used to test the difference between observed and expected frequencies are
significant.
2. If the data is given in a series of ‘n’ number then degrees of freedom = n-1
3. In case of a .Binomial distribution d.f = n-1
b. Poisson distribution d.f. = n-2
c. Normal distribution d.f. = n-3

1. The following table gives the number of aircraft accidents that occurred during
the various days of the week. Test whether the accidents are uniformly distributed
over the week.

Days Mon Tue Wed Thur Fri Sat

No. of 14 18 12 11 15 14
accidents

Solution:

Null hypothesis Ho : The accidents are uniformly distributed over the week.
 84
Expected frequencies of the accidents on each of the days = = 14
6
Observed Expected (O − E ) 2 (O − E ) 2
Frequency Frequency E
(O) (E)
14 14 0 0
18 14 16 1.143
12 14 4 0.286
11 14 9 0.643
15 14 1 0.071
14 14 0 0
2.143

(O − E ) 2
Test Statistic (C.V) :ψ2 =∑ = 2.143
E
Degrees of freedom (d.f) : n −1 = 6-1 = 5
Tabulated Value (T.V) : ψ 2 at 5% level for 5 d.f. = 11.07
Conclusion : C.V < T.V
We accept the null hypothesis H o .
Therefore, the accidents are uniformly distributed over the week.

2. A sample analysis of examination results of 500 students was made. It was found
that 220 students had failed, 170 had secured a third class, 90 were placed in
second class and 20 got a first class. Do these figures commensurate with the
general examination result which is in the ratio of 4: 3: 2: 1 for the various
categories respectively.

Solution:
Null hypothesis Ho : The observed results commensurate with the general
examination results.
Expected frequencies are in the ration 4: 3: 2: 1
Total frequency = 500
Expected frequency = Dividing the total frequency in the ratio 4: 3: 2: 1
4 3 2
= × 500 = 200 , × 500 = 150 , × 500 = 100
10 10 10
1
× 500 = 50
10

Class Observed Expected (O − E ) 2 (O − E ) 2

Frequency Frequency E
(O) (E)
Failed 220 200 400 2.000
Third 170 150 400 2.667
 Second 90 100 100 1.000
First 20 50 900 18.000
23.667

(O − E ) 2
Test statistic (C.V) :ψ2 =∑ = 23.667
E
Degrees of freedom (d.f) : n − 1 = 4-1 = 3
Tabulated Value (T.V) : ψ 2 at 5% level for 3 d.f. = 7.81
Conclusion : C.V > T.V
We reject the null hypothesis H o .
Therefore, the observed results are not commensurate with the general
examination results.

Chi- Square Test for Independence of Attributes:

Attribute means a quality of characteristics.( Eg: dancing, honesty etc.,)

An attribute may be marked by its presence (position) or absence in a number of a

given population.

Let us consider 2 attribute A & B. A is divided into 2 classes and B is divided into 2
classes. The various cell frequencies can be expressed in the following table known as
2 x 2 contingency table.
A a b
B c d

a b a+b
c d c+d
a+c b+d N

The Expected frequencies are given by

(a + c)(a + b) (b + d )(a + b) a+b

E(a) = E (b) =
N N
(a + c)(c + d ) (b + d )(c + d ) c+d
E (c ) = E (d ) =
N N
a+c b+d N
(Total frequencies)

Note:

1. Degree of freedom : (r − 1)( s − 1) where r – number of rows, s – number of

columns
2. Null hypothesis H0 : Attributes are independent.
3. Testing that if 2 attributes A & B under consideration are independent (or) not.

1. The following table gives a classification of a sample of 100 plants of their flower
colour and flatness of leaf.

Flat leaves Curved leaves Total

White flower 40 20 60
Red flower 10 30 40
Total 50 50 100

Test whether the flower colours is independent of the flatness of leaf.

Solution:
Null hypothesis H0 : They are independent.
Expected frequencies are

50 × 60 50 × 60 60
= 30 = 30
100 100
50 × 40 50 × 40 40
= 20 = 20
100 100
50 50 100

Now,
Observed Expected (O − E ) 2 (O − E ) 2
Frequency Frequency E
(O) (E)
40 30 100 3.333
20 30 100 3.333
10 20 100 5.000
30 20 100 5.000
16.666

(O − E ) 2
Test statistic (C.V) :ψ2 =∑ = 16.666
E
Degrees of freedom (d.f) : = (r − 1)( s − 1) = (2-1)(2-1) = 1
Tabulated Value (T.V) : ψ 2 at 5% level for 1 d.f. = 3.84
Conclusion : C.V > T.V
We reject the null hypothesis H o .
Large Samples

If the sample size n > 30 then the sample is called Large sample.

1. Test of Significance for Single Proportion:

To test the significant difference between the sample proportion p and population
proportion P.

Test Statistic:

p−P
Z= where n – sample size
PQ
n
Note:
1. Limits for population proportion P are given by p ± 3 pq where q = 1 - p
n
2. 98% confidence limits for population proportion are p ± 2.33 pq
n

1. In a city, a sample of 1000 people were taken and out of them 540 are vegetarians
and the rest are non-vegetarians. Can we say that both habits of eating (vegetarian
or non-vegetarian) are equally popular in the city at 1% level of significance.

Solution:
540
Given: n = 1000, p = sample proportion of vegetarians = = 0.54 ,
1000
1
P = population proportion of vegetarians = =0.5
2
Null hypothesis H0 : Both habits are equally popular. ( H o : P = 0.5 )
Alternative hypothesis : H1 : P ≠ 0.5
p−P 0.54 − 0.5
Test Statistic (C.V) : Z = = = 2.532
PQ 0 .5 × 0 .5
n 1000
Tabulated value (T.V) : Z at 1% level of significance is 2.58
Conclusion : C.V < T.V
We accept null hypothesis H o .

2. A die was thrown 9000 times and of these 3220 yielded a 3 or 4. Is this consistent
with the hypothesis that the die was unbiased?
Solution: Given: n = 9000,
 3220
p = Proportion of successes of getting 3 or 4 in 9000 throws = = 0.3578 ,
9000
1 1 2
P = population proportion of successes = P(getting a 3 or 4) = + = =0.33
6 6 6
1
Null hypothesis H0 : The die is unbiased. ( H o : P = )
3
1
Alternative hypothesis : H1 : P ≠
3
p−P 0.3578 − 0.3333
Test Statistic (C.V) : Z = = = 4.94
PQ 0.3333 × 0.6667
n 9000
Tabulated value (T.V) : Z at 1% level of significance is 2.98
Conclusion : C.V > T.V
We reject null hypothesis H o .

2. Test of Significance of Difference of Proportions:

To test the significant difference between the sample proportions p1 and p 2 .

Null hypotheisis : H0 : p1 = p 2
Alternative hypothesis : H1 : p1 ≠ p 2

a) To test the significant difference between the sample proportions p1 & p2

p1 − p 2 n p + n2 p 2
Test Statistic : Z= where p = 1 1
⎛ 1 1 ⎞ n1 + n2
pq⎜⎜ + ⎟⎟
⎝ n1 n2 ⎠
and q = 1 – p

b) To test the significant difference between p1 & p

p1 − p n1 p1 + n2 p 2
Test Statistic : Z = where p =
n2 pq n1 + n2
n1 (n1 + n2 )
P1 − P2
c) If the sample proportions are not known then Z =
P1Q1 P2 Q2
+
n1 n2

1. Random samples of 400 men and 600 women were asked whether they would like
to have a flyover near their residence. 200 men and 325 women were in favour of
 the proposal. Test the hypothesis that proportions of men and women in favour of
the proposal are same, ar 5% level.

Solution:
Given: sample sizes n1 = 400, n2 = 600.
200
Proportion of men = p1 = = 0.5
400
325
Proportion of women = p 2 = = 0.541
600
Null hypothesis : H0 : p1 = p 2
Alternative hypothesis : H1 : p1 ≠ p 2
p1 − p 2 0.5 − 0.541
Test Statistic : Z= =
⎛ 1 1 ⎞ ⎛ 1 1 ⎞
pq⎜⎜ + ⎟⎟ 0.525 × 0.425⎜ + ⎟
⎝ n1 n2 ⎠ ⎝ 400 600 ⎠
− 0.041
= = −1.34
0.032
z = 1.34

200 325
400 × + 600 ×
n p + n2 p 2 400 600 = 525 = 0.525 }
{where p = 1 1 =
n1 + n2 400 + 600 1000

Tabulated value (T.V) : Z at 5% level of significance is 1.96

Conclusion : C.V < T.V
We accept null hypothesis H o .

2. In a random sample of 400 student of the university teaching department, it was

found that 300 students failed in the examination. In another sample of 500
students of the affiliated colleges the number of failures in the same examination
was found to be 300. Find out whether the proportion of failures in the university
teaching departments significantly greater than the proportion of failures in the
university teaching departments and affiliated colleges taken together.

Solution:
300 300
Given: n1 = 400, n 2 = 500, p1 = = 0.75 , p 2 = = 0.6
400 500
⎛ 300 ⎞ ⎛ 300 ⎞
400⎜ ⎟ + 500⎜ ⎟
n1 p1 + n 2 p 2 ⎝ 400 ⎠ ⎝ 500 ⎠
p= = = 0.667 , q = 0.333
n1 + n 2 400 + 500
 Null hypothesis H o : Assume that there is no significant difference between p1 &
p
p1 − p 0.75 − 0.667
Test Statistic : Z= = = 4.74
n2 pq 500 × 0.667 × 0.333
n1 (n1 + n2 ) 400(400 + 500)
Tabulated value : z at 5% level of significance is 1.96
Conclusion : C.V > T.V
We reject the null hypothesis H o .
Therefore, the proportion of failures in the affiliated colleges is greater than the
proportion of failures in university departments and affiliated colleges taken
together.

3. In two large populations, there are 30% and 25% respectively of fair haired
people. Is this diffrence likely to be hidden in samples of 1200 and 900
respectively from the two populations.

Solution:
Given: n1 = 1200, n2 = 900 ,
30
P1 = Proportion of fair hairedd people in the first population = = 0.3
100
25
P2 = Proportion of fair hairedd people in the second population = = 0.25
100
Null hypothesis H o : Assume that the sample proportions are equal.
H0 : p1 = p 2
Alternative hypothesis : H1 : p1 ≠ p 2
P1 − P2 0.3 − 0.25
Test Statistic : Z= = = 2.55
P1Q1 P2 Q2 0.3 × 0.7 0.25 × 0.75
+ +
n1 n2 1200 900
Tabulated value : z at 5% level of significance is 1.96
Conclusion : C.V > T.V
We reject the null hypothesis H o .
Therefore, the proportion are not equal.

3. Test of Significance for Single Mean:

To test whether the given sample of size n has been drawn from a population
with mean µ .

Null Hypothesis: There is no difference between sample mean x and population

mean µ
 x−µ
Test Statistic: Z = where σ - S.D of the population
σ
n
Note:

1. If the population S.D is not known then the test statistic

x−µ
Z= where s – sample S.D
s
n
σ
2. 95% confidence limits for µ are x ± 1.96
n
σ
3. 98% confidence limits for µ are x ± 2.33
n
σ
4. 99% confidence limits for µ are x ± 2.58
n

1. A sample of 900 members has a mean of 3.4 cms and a S.D. 2.61 cms. Is the
sample from a large population of mean 3.25cm and a S.D. 2.61 cms. If the
population is normal and its mean is unknown find the 95% confidence limits of
true mean.

Solution: Given : n = 900 , µ = 3.25 , x = 3.4cm , σ = 2.61 , s = 2.61

Null hypothesis : H o : µ = 3.25
Alternative hypothesis: H 1 : µ ≠ 3.25
x−µ 3.4 − 3.25
Test statistic (C.V) : z = = = 1.724
S 2.61
n 900
Z = 1.724

Tabulated Value (T.V) : Z at 5% level of significance is 1.96

Conclusion : C.V < T.V
we accept the null hypothesis H o .
Therefore, the samples has been drawn from the population mean µ = 3.25
σ
The 95% confidence limit of the mean of the population is given by x ± 1.96
n
2.61
= 3.4 ± 1.96 × = 3.4 ± 0.1705
900
Hence 95% confidence limit is [3.2295, 3.57]

4. Test of Significance for Difference of Mean:

1. To test the significant difference between x1 and x 2

 x1 − x 2
Test Statistic: Z=
σ 12 σ 22
+
n1 n2
2. If the samples have been drawn from the same population then σ 12 = σ 22 = σ 2

x1 − x 2
Test Statistic: Z=
σ2 σ2
+
n1 n2
n1 s12 + n2 s 22
3. If σ is not known, we can use a estimate of σ 2 given by σ 2 =
n1 + n2

1. The means of 2 large samples 1000 and 2000 members are 67.5 inches and 68.0
inches respectively. Can the samples be regarded as drawn from the same
population of S.D. 2.5 inches.

Solution : Given: Sample sizes n1 = 1000, n2 = 2000.

Sample mean x1 = 67.5inches , x 2 = 68inches
Population S.D σ = 2.5 inches
Null Hypothesis : H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2 67.5 − 68
Test statistic (C.V) : z = = = -5.16
σ2 σ2 (2.5) 2 (2.5) 2
+ +
n1 n2 1000 2000
z = 5.16
Tabulated value (T.V): z at 5% level of significance is 1.96
Conclusion : C.V > T.V
We rejet the null hypothesis H o .
Therefore, the samples are not drawn from the same populaiton.

Problems for Practice:

1. A sample of 100 students is taken from a large population. The mean height of the
students in this sample is 160. Can it be reasonably regarded that, in the
population the mean height is 165cm and S.D is 10cm?
Solution: Given :n = 100, µ = 165 , x = 160 , σ = 10 ,
Null hypothesis : H o : µ = 165
Alternative hypothesis: H1 : µ > 165
 x−µ
Test statistic (C.V) : z= = -5 ; rejected
σ
n
2. In a random sample of 60 workers, the average time taken by them to get to work
is 33.8minutes, with a S.D of 6.1 minutes. Can we reject the null hypothesis
µ = 32.6 minutes in favour of alternative hypothesis µ > 32.6 minutes at tabulated
value 2.58?
Solution: Given :n = 60, µ = 32.6 , x = 33.8 , σ = 6.1 ,
Null hypothesis : H o : µ = 32.6
Alternative hypothesis: H1 : µ > 32.6
x−µ
Test statistic (C.V) : z = = 1.5238 ; Accepted
σ
n

3. The mean and S.D of a population are 11795 and 14054 respectively. If n=50,
find 95% confidence limit for mean.
σ
Solution: x ± 1.96 = (7899.4, 15690.6)
n
4. An ambulance service claims that it takes on the average less than 10 minutes to
reach its destination in emergency calls. A sample of 36 calls has a mean of 11
minutes and the variance of 16 minutes. Test the significance at 5% level of
significance.
Solution: Given :n = 36, µ = 10 , x = 11 , σ = 16 = 4 ,
Null hypothesis : H o : µ = 10
Alternative hypothesis: H1 : µ ≠ 10
x−µ
Test statistic (C.V) : z = = 1.5 ; Accepted
S
n

5. The average marks scored by 32 boys are 72 with a standard deviation of 8, while
that for 36 girls is 70 with a standard deviation of 6. Test at 1% level of
significance whether the boys perform better than girls.
Solution: Null Hypothesis : H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2
Test statistic (C.V) : z = = 1.15, H0 accepted
σ2 σ2
+
n1 n2

6. An examination was given to two group of students of two colleges:

College A : mean = 75; S.D = 9; n1 = 50
College B : mean = 79; S.D = 7; n2 = 60
Is there a significant difference in the performance at two colleges at 5%level
Solution: Null Hypothesis : H o : µ1 = µ 2
Alternative hypothesis: H1 : µ1 ≠ µ 2
x1 − x 2
Test statistic (C.V) : z = = -2.56, H0 rejected
σ2 σ2
+
n1 n2